Developmental Cell Perspective Tracing Cells for Tracking Cell Lineage and Clonal Behavior Margaret E. Buckingham 1, * and Sigole ` ne M. Meilhac 1, * 1 Molecular Genetics of Development Unit, CNRS URA 2578, Department of Developmental Biology, Institut Pasteur, 28 rue du Dr Roux, 75015 Paris, France *Correspondence: [email protected](M.E.B.), [email protected](S.M.M.) DOI 10.1016/j.devcel.2011.07.019 Reconstructing the lineage of cells is central to understanding development and is now also an important issue in stem cell research. Technological advances in genetically engineered permanent cell labeling, together with a multiplicity of fluorescent markers and sophisticated imaging, open new possibilities for prospective and retrospective clonal analysis. History and Concepts Cell lineage analysis is intimately connected with the emergence of developmental biology as a field of scientific research (see Galperin, 1998). In the mid-19th century, identification of cells as the basic units of life (Schleiden, 1838; Schwann, 1839) led to the realization that cells come from pre-existing cells (Virchow, 1858). Before the end of the 19th century, the work of Charles Whitman (1887) and later of Edmund Wilson (1892) on leech and annelid embryos led to the formulation of the term ‘‘cell lineage.’’ This early work inspired the Wood’s Hole School at the Marine Biology Laboratory in Massachusetts, where pioneer- ing research in invertebrate embryos led to important concepts for lineage analysis. Thus E. Wilson viewed lineage in terms of the fate of cells and E.G. Conklin (1905), another major figure, made the distinction between determinate and indeterminate types of cleavage in ascidians, leading to the concept of invariant and noninvariant cell lineages. Breakthroughs in vertebrate fate mapping came from the systematic use of vital staining of groups of cells (Vogt, 1929) and from grafting experiments (Spemann and Mangold, 1924) in the amphibian embryo. In addition to embryological approaches, the work of A.H. Sturtevant, based on genetic studies initiated by T.H. Morgan and others on spon- taneously generated mosaicism in insects, led to retrospective analyses in which cell lineage and gene function were associated (Sturtevant, 1929). Many of the conceptual issues of today were evident when cell lineages were first explored. Lineage studies, then as now, aim to establish which cells, and how many cells, in the early embryo will give rise to a structure and, as development proceeds, from which part of a structure a substructure derives. These interrogations now extend to the origin of stem cells that permit the regeneration of an adult structure as well as its initial forma- tion. Clonal analyses, which describe the derivatives of a single cell, provide insight into the mode of growth of a tissue and its regionalization with potential clonal boundaries (Garcia-Bellido et al., 1973) between compartments, or with segregation between distinct cell lineages, which do not necessarily corre- spond to distinct differentiated cell types but rather to topo- graphical subdivisions (Lescroart et al., 2010). Analyses of clones can also provide information about cell death and prolif- eration, cell competition, cell movement and dispersion, and tissue polarity. Experimentation in a growing number of tissues and model organisms reveals the diversity of cell behavior that underlies progression along a lineage tree and has led to the elaboration of conceptual frameworks for cell lineage analysis (e.g., Garcia-Bellido, 1985; Petit et al., 2005; Stent, 1985). In the context of embryonic development, many invertebrates have invariant lineages, meaning that a blastomere not only has a predictable future but also has a reproducible position and a defined group of neighbors from one individual to another. This is illustrated by C. elegans, for which a complete lineage tree has been defined (Sulston et al., 1983). In contrast, in the early mouse embryo, for example, more cell mixing takes place and cells in the inner cell mass of the blastocyst retain pluripo- tency and plasticity (Cockburn and Rossant, 2010). In the case of such noninvariant (regulative) development it is more chal- lenging to analyze cell lineages. Intertwined with the concept of lineage is that of cell commit- ment. Cell lineage follows the normal fate of a cell and its daugh- ters, leading to the formulation of genealogical trees of cells with increasingly restricted cell fate choices as development proceeds. Unlike lineage, commitment can only be established by experimental challenge, such as ectopic grafting or in vitro manipulation, showing that the cell has acquired a restricted cell fate potential (e.g., Tam et al., 1997). As G. Stent (1985) pointed out, cell lineage plays a role in cell commitment by the unequal partitioning of cell determinants in daughter cells in successive cell divisions, as illustrated by the ascidian, Ciona (Nishida, 1987), and by the orderly placement of cells relative to intercellular signals as development proceeds, which is a major feature of vertebrate embryogenesis. It has become increasingly clear that even differentiated cells retain plasticity, as demonstrated by the spectacular phenomenon of induced pluripotency (Yamanaka, 2009). Caution should be exercised in equating cell fate restriction with gene expression. Character- izing lineage progression in these terms provides a genetic complement to cellular studies but can also lead to experimental pitfalls as discussed later. Experimentalists today face many of the same dilemmas that confronted embryologists a hundred years ago—namely, how to label cells and subsequently analyze their contribution to the embryo based on the perdurance of the label, without perturbing the development of the organism. New technological develop- ments now facilitate detailed analysis of complex situations. 394 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
Transcript
Developmental Cell
Perspective
Tracing Cells for TrackingCell Lineage and Clonal Behavior
Margaret E. Buckingham1,* and Sigolene M. Meilhac1,*1Molecular Genetics of Development Unit, CNRS URA 2578, Department of Developmental Biology, Institut Pasteur, 28 rue du Dr Roux,75015 Paris, France*Correspondence: [email protected] (M.E.B.), [email protected] (S.M.M.)DOI 10.1016/j.devcel.2011.07.019
Reconstructing the lineage of cells is central to understanding development and is now also an importantissue in stem cell research. Technological advances in genetically engineered permanent cell labeling,together with a multiplicity of fluorescent markers and sophisticated imaging, open new possibilities forprospective and retrospective clonal analysis.
History and ConceptsCell lineage analysis is intimately connected with the emergence
of developmental biology as a field of scientific research (see
Galperin, 1998). In the mid-19th century, identification of cells
as the basic units of life (Schleiden, 1838; Schwann, 1839) led
to the realization that cells come frompre-existing cells (Virchow,
1858). Before the end of the 19th century, the work of Charles
Whitman (1887) and later of Edmund Wilson (1892) on leech
and annelid embryos led to the formulation of the term ‘‘cell
lineage.’’ This early work inspired the Wood’s Hole School at
theMarine Biology Laboratory inMassachusetts, where pioneer-
ing research in invertebrate embryos led to important concepts
for lineage analysis. Thus E. Wilson viewed lineage in terms of
the fate of cells and E.G. Conklin (1905), another major figure,
made the distinction between determinate and indeterminate
types of cleavage in ascidians, leading to the concept of invariant
and noninvariant cell lineages. Breakthroughs in vertebrate fate
mapping came from the systematic use of vital staining of groups
of cells (Vogt, 1929) and from grafting experiments (Spemann
and Mangold, 1924) in the amphibian embryo. In addition to
embryological approaches, the work of A.H. Sturtevant, based
on genetic studies initiated by T.H. Morgan and others on spon-
taneously generated mosaicism in insects, led to retrospective
analyses in which cell lineage and gene function were associated
(Sturtevant, 1929).
Many of the conceptual issues of today were evident when cell
lineageswere first explored. Lineage studies, then as now, aim to
establish which cells, and how many cells, in the early embryo
will give rise to a structure and, as development proceeds,
from which part of a structure a substructure derives. These
interrogations now extend to the origin of stem cells that permit
the regeneration of an adult structure as well as its initial forma-
tion. Clonal analyses, which describe the derivatives of a single
cell, provide insight into the mode of growth of a tissue and its
regionalization with potential clonal boundaries (Garcia-Bellido
et al., 1973) between compartments, or with segregation
between distinct cell lineages, which do not necessarily corre-
spond to distinct differentiated cell types but rather to topo-
graphical subdivisions (Lescroart et al., 2010). Analyses of
clones can also provide information about cell death and prolif-
eration, cell competition, cell movement and dispersion, and
tissue polarity. Experimentation in a growing number of tissues
394 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
and model organisms reveals the diversity of cell behavior that
underlies progression along a lineage tree and has led to the
elaboration of conceptual frameworks for cell lineage analysis
(e.g., Garcia-Bellido, 1985; Petit et al., 2005; Stent, 1985).
In the context of embryonic development, many invertebrates
have invariant lineages, meaning that a blastomere not only has
a predictable future but also has a reproducible position and
a defined group of neighbors from one individual to another.
This is illustrated by C. elegans, for which a complete lineage
tree has been defined (Sulston et al., 1983). In contrast, in the
early mouse embryo, for example, more cell mixing takes place
and cells in the inner cell mass of the blastocyst retain pluripo-
tency and plasticity (Cockburn and Rossant, 2010). In the case
of such noninvariant (regulative) development it is more chal-
lenging to analyze cell lineages.
Intertwined with the concept of lineage is that of cell commit-
ment. Cell lineage follows the normal fate of a cell and its daugh-
ters, leading to the formulation of genealogical trees of cells
with increasingly restricted cell fate choices as development
proceeds. Unlike lineage, commitment can only be established
by experimental challenge, such as ectopic grafting or in vitro
manipulation, showing that the cell has acquired a restricted
cell fate potential (e.g., Tam et al., 1997). As G. Stent (1985)
pointed out, cell lineage plays a role in cell commitment by the
unequal partitioning of cell determinants in daughter cells in
successive cell divisions, as illustrated by the ascidian, Ciona
(Nishida, 1987), and by the orderly placement of cells relative
to intercellular signals as development proceeds, which is
a major feature of vertebrate embryogenesis. It has become
increasingly clear that even differentiated cells retain plasticity,
as demonstrated by the spectacular phenomenon of induced
pluripotency (Yamanaka, 2009). Caution should be exercised in
equating cell fate restriction with gene expression. Character-
izing lineage progression in these terms provides a genetic
complement to cellular studies but can also lead to experimental
pitfalls as discussed later.
Experimentalists today face many of the same dilemmas that
confronted embryologists a hundred years ago—namely, how to
label cells and subsequently analyze their contribution to the
embryo based on the perdurance of the label, without perturbing
the development of the organism. New technological develop-
ments now facilitate detailed analysis of complex situations.
In the following sections we shall discuss the current state of
this art and future developments, where temporal as well as
spatial regulation of the onset of labeling, simultaneous detec-
tion of several lineages, systematic labeling of all progenitors
of a structure, visualization of the dynamics of lineage progres-
sion, and linking lineage to gene function are the underlying
issues. In this era of genetic tools for cell tracing, we will focus
on Drosophila, zebrafish, and mouse, with reference to avians,
amphibians, and plants, as well as to the other invertebrate
models that have provided important insights into lineage anal-
ysis. Approaches currently employed for following the history
of a cell, which we discuss here, are summarized in Table 1.
Prospective Lineage Analysis: Selectionof Labeled ProgenitorsProspective lineage analysis is a classic approach in which cell
labeling is performed at a known position and stage and the
contribution of the cell’s descendants to a structure is subse-
quently analyzed. In this context, cell fate mapping can be
achieved by grafting experiments where labeled cells are intro-
duced into the embryo and their subsequent contributions moni-
tored. Following on from the pioneering work of Mangold and
Spemann in amphibians (Spemann and Mangold, 1924) and of
Waddington in the chick (Waddington, 1932), the analysis of
chick-quail chimaeras, in which it is possible to distinguish nuclei
between these closely related species, underlies important
aspects of our understanding of cell fates in the amniote embryo,
as exemplified by neural crest cell derivatives (Le Douarin and
Barq, 1969). In an alternative, generally applicable approach,
radioactive labeling of transplanted cells has been used
(Weston, 1963) and was instrumental in mapping the heart-form-
ing fields in the chick embryo (Rosenquist and De Haan, 1966).
This approach was also used in challenging experiments on
the mouse embryo to map the fate of cells in the epiblast
(Beddington, 1981). Lypophilic carbocyanine dyes, such as DiI,
which can be introduced into a small region of the embryo
have been, and continue to be, used to trace groups of cells
and examine cell movements. Less invasive than grafting exper-
iments, these vital dyes intercalate into the cell membrane
and are easily visualized. In early experiments with this method
of cell marking, the migration pathways of neural crest cells
and the temporal order in which they contribute to their deriva-
tives were refined for the chick embryo (Serbedzija et al.,
1989). Numerous fate maps have been established using this
technique, including more recently in the lamprey (McCauley
and Bronner-Fraser, 2003) and in cultured mouse embryos (Galli
et al., 2008).
Single-cell labeling, which permits lineage analysis, is more
challenging but can be achieved by microinjection. Classically,
horseradish peroxidase (HRP) or dextran linked fluorescent
dyes, which are too large to diffuse between cells, have been
used as intracellular markers. Pioneering experiments using
HRP pressure mediated microinjection in the leech embryo,
where there is little cell migration (Weisblat et al., 1978), showed
that teloblasts, which are the founder cells of segments, give rise
to topographically invariant lineages that consist of different cell
types. Interestingly, in this case, unlike that of Drosophila (Gar-
cia-Bellido, 1985), morphological segment boundaries do not
necessarily correspond to borders of clonal restriction. In the
De
frog embryo, the progeny of different blastomeres were shown
to contribute to distinct clonal domains with well-defined bound-
aries in the central nervous system (Hirose and Jacobson, 1979),
although the descendants of a blastomere are not restricted to
a single neural, or other, cell fate, indicative of global cell mixing
(Moody, 1987). As distinct from amphibians, early planes of
cleavage are not related to the plane of bilateral symmetry of
the zebrafish embryo, and descendants of a single blastomere
tend to remain associated initially (Kimmel and Law, 1985a) until
they disperse at the onset of epiboly (Kimmel and Law, 1985b).
During gastrulation, cell mixing decreases and tissue-specific
lineages have been observed from this stage (Kimmel and
Warga, 1986). In the mouse embryo, where there is extensive
cell mixing, single-cell labeling by iontophoresis microinjection
of HRP (pioneered by Ba1akier and Pedersen, 1982) in cells of
the epiblast has led to fate maps in which the probability of
descendants of a cell contributing to a particular tissue was
determined (Lawson et al., 1991). In the absence of stereotyped
lineages and despite geometrical differences, topological fate
relationships at the stage of gastrulation are conserved between
mammals, birds, amphibians, and zebrafish.
The advent of fluorescent proteins as markers (Shimomura
et al., 1962) has had a major impact on fate mapping and cell
tracking, as they are genetically encoded. Furthermore, fusion
proteins, inwhich thefluorescentprotein is targeted to thenucleus
(e.g., H2B-GFP) or to the plasma membrane, provide clearer
cellular resolution and additional information, such as mitotic
status or cell shape dynamics. Thus, for example, microinjection
of DNA encoding a fluorescent protein, GFP, demonstrated that
a single cell in the chick somite is bipotent and revealed how its
descendants progressively acquire a dermal or muscle cell fate
(Ben-Yair and Kalcheim, 2005). In another example, microinjec-
tion of mRNAs encoding membrane-bound fluorescent proteins
into a single cell of the inner cell mass of the mouse blastocyst,
followed by time-lapse imaging in relation to a chromosomal
marker, has shown how segregation between epiblast and
primitive endoderm lineages is accompanied by extensive cell
movement and, coupled with early markers of these cell types,
supports the conclusion that primitive endoderm formation
involves cell sorting and position-dependant induction (Meilhac
et al., 2009). In this case, the characteristics of clones were
used to test computer models of mechanisms for lineage segre-
gation. In invertebratemodels, too, fluorescent proteins are being
used to track cells. In the leech embryo, an analysis based on
injection of a plasmid encoding H2B-GFP now indicates a transi-
tion from tightly regulated tomorestochastic cell division, pointing
to a less black-and-white distinction between invariant andnonin-
variant lineages (Gline et al., 2009), as observed to some extent
even in C. elegans (Schnabel et al., 1997).
Advances in understanding chromophore photochemistry
have made it possible to engineer photomodulatable fluorescent
proteins (see Piatkevich et al., 2010), which have tended to
replace caged molecules that require chemical synthesis, for
marker activation. The value of such a caged dyewas first shown
in an experiment in Drosophila that revealed clonal restriction
anteriorly but not posteriorly when the dye was activated at the
site of establishment of an Engrailed 1-positive parasegment
(Vincent and O’Farrell, 1992). In the last few years a range of
photoconvertible fluorescent proteins, which undergo a spectral
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 395
Table 1. An Assessment of Currently Used Cell-Tracing Techniques
Organism
Current Methods Requirements Merits Limitations M C F Z D I P
Prospective
DiI vital staining, embryo
or explant culture
targeted, easier than
microinjection
nonclonal, dilution,
accessibility to cells
x x x x
HRP, dextran
microinjection
micromanipulation,
embryo or explant culture
clonal, targeted dilution, invasive,
accessibility to cells
x x x x x
DNA, RNA
microinjection
micromanipulation,
embryo or explant culture
clonal, targeted,
amplification of
the marker
dilution, invasive,
accessibility to cells
x x x x x
Uncaging or
photomodulation of
a fluorescent protein
laser fluorescent
microscopy, embryo or
explant culture, injection
of the marker or
genetically modified line
targeted, no
micromanipulation
dilution, phototoxicity,
accessibility to cells
x x x x
Genetic tracing /
tissue-specific
recombinase
genetically modified lines noninvasive,
permanent
nonclonal, dependent
on gene/promoter
expression and potential
integration site effects
x x x
Transplantations micromanipulation,
labeled donor
permanent if genetic
marker, easier than
microinjection
nonclonal, invasive,
accessibility to cells
x x x x
Mosaic
Early chimaeras
with cell mixing
micromanipulation, lines
with distinct phenotypes
permanent, sparse
labeling
nonclonal, invasive x
DNA electroporation electroporation, cell
tracking
multicolor, sparse
labeling
nonclonal, dilution,
accessibility
x x
X-inactivation genetically
modified lines
spontaneous,
permanent, sparse
labeling
nonclonal x
Multicolor genetic
mosaics: MADM,
twin-spot,
brainbow/confetti
genetically modified
lines, resolving
color hues
multicolor,
permanent,
sparse labeling
nonclonal x x
Retrospective
Spatially Random Labeling
Retrovirus library of tagged
retroviruses, isolation
of cells for PCR/
sequencing analysis
time-control,
permanent, clonal
differential infectivity
of cells, potential
integration site effects
x x
Inducible
recombinase:
temperature,
hormone or
antibiotics
genetically modified
lines, temperature shift
or inducer molecule
concentration, control
background levels
time-control, dose
control of clone
frequency
toxicity, partial activity
of the Cre,
reproducibility
x x x x
Inducible transposon
mobility: temperature,
inducer molecule
genetically modified
lines, temperature shift
or inducer molecule
concentration, control
background levels
time-control, dose
control of clone
frequency
instability, potential
integration site effects
x
Spatially and Temporally Random Labeling
Microsatellites isolation of cells,
sequencing analysis
systematic,
spontaneous
large number of
observations for
statistical analysis
x
Mitotic recombination
with a laacZ-like
reporter
genetically
modified line
systematic,
spontaneous
large number of
observations for
statistical analysis
x x
396 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
Developmental Cell
Perspective
Table 1. Continued
Organism
Current Methods Requirements Merits Limitations M C F Z D I P
4D Imaging
Time-lapse imaging embryo or explant
culture, high-resolution
microscopy,
computing capacity
dynamic, direct,
comprehensive
limited developmental
window, penetration,
complexity of image
analysis
x x x x x x
This table is mainly based on papers discussed in the text. M, mouse; C, chick; F, frog; Z, zebrafish; D, Drosophila; I, other invertebrates; P, plants.
Developmental Cell
Perspective
change after exposure to activating light, has become available.
These proteins can be photoconverted by confocal laser micros-
copy or even using regular fluorescence microscopes (Baker
et al., 2010; Stark and Kulesa, 2007). In some instances, photo-
conversion by two-photon microscopy may be applicable (Hatta
et al., 2006) for single-cell labeling, with the advantage of deeper
penetration into the tissue and less phototoxicity. Issues in using
fluorescent proteins for cell tracking include the rapidity and
stability of photoconversion, the brightness of the fluorescence,
and toxicity; it is important to test these parameters for the
organism and developmental stage under study (e.g., Nowot-
schin and Hadjantonakis, 2009). The zebrafish embryo because
of its accessibility, transparency, and rapid development partic-
ularly lends itself to fluorescent cell tracking. Temporal conver-
sion of fluorescent proteins can be used for in vivo birthdating
of tissue types. This is illustrated by the BAPTI system where
a photoconvertible Kaede reporter is under the control of a neural
promoter; after exposure to activating light early-born neurons
B B’
A
early late
Green State Red Stat
Intact polypeptide Backbone cl
Photoconversion
Figure 1. Live Imaging of Kaede Photoconversion in Zebrafish: An Exa(A) Photoconversion of the Kaede fluorescent protein by exposure to UV lighzeiss-campus.magnet.fsu.edu/articles/probes/highlighterfps.html).(B) Early trigeminal sensory neurons born before the photoconversion appear ysynthesized protein (green), whereas the neurons born later remain green.(C) Schematic representation of the mode of action of this BAPTI system on cell
De
are labeled red, whereas later-born neurons, not exposed to
photoconversion, remain green (Figure 1). The BAPTISM system
extends this to include an additional population-specific
reporter. This analysis led to the conclusion that the specification
and function of different classes of trigeminal sensory ganglia
depend on the timing of neurogenesis (Caron et al., 2008).
With the approaches of prospective clonal analysis, fate maps
can be drawn and partial lineages reconstructed. However, the
use of microinjected markers or sequences encoding photomo-
dulatable fluorescent proteins is limited by the problem ofmarker
dilution at each cell division and thus is only applicable for
short-term labeling experiments. Furthermore, their introduction
is invasive, is often challenging technically, and may cause
damage. A major interest of fluorescent proteins is that they
can be employed after stable integration of their coding
sequence into the genome by transgenesis or gene targeting
and can therefore provide permanent cell labeling. This is
classical in mouse and fly and is now becoming practicable in
B’’
405nm
early-born late-born
C e
eaved
mple of Prospective Lineage Analysist induces rupture of a covalent bond (adapted with permission from http://
ellow, due to the presence of the photoconverted red protein and the newly
lineage (Caron et al., 2008, adapted with permission).
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 397
Figure 2. Genetic Manipulations forSpatiotemporal Control of ReporterExpression(A–C) Schematic representation of genetic tracingprocedures based on tissue-specific expression.(A) Schematic representation of themode of actionof such procedures.(B) Genetic inducible fate mapping (GIFM) used inmouse or G-TRACE in Drosophila depends uponactivation of a ubiquitous reporter by a tissue-specific recombinase (Zinyk et al., 1998; Evanset al., 2009). <: target sites for recombination.(C) In the Kaloop system, as used in zebrafish,autoactivation of the fluorescent marker providespermanent labeling after tissue-specific initiationof expression, without recombination (Distel et al.,2009).(D) In theMAZe procedure, developed in zebrafish,transient activation of a heat-shock promoter (hsp)leads to expression of a reporter via the Gal4/UASsystem (Collins et al., 2010). <: loxP sites for Crerecombination.
Developmental Cell
Perspective
zebrafish, based on transposon-mediated integration. Avian
transgenic lines, based on lentiviral-mediated transgene integra-
tion, should also provide new tools for lineage analysis (Sato
et al., 2010). Further improvements for directing transgene
insertion in a range of species can be envisaged with zinc finger
nucleases, meganucleases, or TALE nucleases (Christian et al.,
2010).
Genetic TracingIn order to follow the descendants of a cell, the recombinase
approach to permanent genetic labeling by specific activation
of a conditional reporter is widely used in mouse and fly and is
now available in fish and has also been used for genetic cell
tracing in Xenopus (Satoh et al., 2005). Recombinase activity
should be rapid, efficient, and specific, although there can be
problems with certain loci and with Cre toxicity, even in mice
(Naiche and Papaioannou, 2007). This is a problem in
Drosophila, where Flp recombinases are the preferred tools.
Improved variants of Flp and Cre, together with the identification
of specific target site variants (FRT, lox), have increased the
efficiency and scope of these tools (Turan et al., 2011). In addi-
tion to reporter lines where a single marker is activated on
Figure 3. Examples of Reporter Constructs for Mosaic Analyses(A) In the MADM system in mouse or in the twin-spot system in Drosophila, partial sequences for fluorescent proteins are separated by recombination sites (<),which, when exposed to recombinase, reconstitute a fluorescent reporter. If mitotic recombination takes place, green and red fluorescent reporters markseparate sister cells as a result of allelic segregation. After nonmitotic recombination, cells become yellow (Zong et al., 2005; Griffin et al., 2009).(B) Schematic illustration of cell labeling in the lineage.(C) In Drosophila, the Flybow construct (version FB1.1) of membrane tagged fluorescent reporters can be rearranged by the action of Flp on FRT sites.(D) Generation of mosaics by heat-shock induction of the Flp recombinase, which induces stochastic labeling of cells. Elav-Gal4155 drives expression of the UASconstruct in neurons. Prior to expression the potential fluorescent marker is indicated as an outline; full colored circles indicate expressing cells.(E) An example of a mosaic in an L3 larval optic lobe, showing labeled lineages of medulla (m) and lamina (l) neurons, in blue, red, or yellow, after transient heatshock at early larval stages. EGFP (green) is the default fluorescent protein, when Flp is not active. The merged image and the red channel (detecting red andyellow cells) are shown on the left and right, respectively (Hadjieconomou et al., 2011, adapted with permission).
Developmental Cell
Perspective
cells of the crypt of the mouse intestine. Tamoxifen induction of
Cre at different time points, followed bymathematical analysis of
cell patterns marked with the four randomly generated reporters,
led to conclusions about stem cell turnover without asymmetric
cell divisions. Stochastic adoption of stem or transit amplifying
cell fates depends on neutral competition between cells. Such
sophisticated mathematical analysis of clone distributions has
also been applied to other tissues, and a general theoretical
framework, which discriminates between patterns of long-term
clonal evolution for distinguishing three classes of stem cell
behavior, has been proposed (Klein and Simons, 2011). Adapta-
tions of this system have now been described for Drosophila—
d-Brainbow (Hampel et al., 2011) or Flybow (Hadjieconomou
et al., 2011)—in which cell labeling as a result of different
stochastic recombination events is linked to the UAS/GAL4
system to drive expression of the transgene reporter construct.
In the d-Brainbow application, epitope tagged, aswell as fluores-
cent proteins, were used, thus permitting both imaging and histo-
logical examination of fixed tissue. The Flybowsystemavoids the
problem of Cre toxicity by employing Flp-mediated inversion of
reporter sequences and has been used to address questions
about the formation of neural network architecture (Figures 3C–
3E). As in all mosaic analyses, the challenge is to sort out cells
400 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
following the color code, which can be limited by the spectral
separation of different combinations of fluorescent proteins and
by the light microscopy resolution of subcellular localization.
Mosaics have opened the way to spectacular multicolor
labeling of cells and have given insight into the polyclonal origin
of tissues, their architecture, and the cell dynamics underlying
tissue growth. The approaches discussed above, which inte-
grate tissue-specific or temporal control, can also potentially
be extended to introduce recombinases under cell-cycle control.
Since they are generated genetically, an important application of
mosaic approaches is that they can be combined with functional
analyses, based on the use of mutated alleles.
Clonal Analysis and Gene FunctionIn order to relate clonal growth to gene function, mutant clones
can be generated in a wild-type background or clonal analysis
can be performed in a mutant background. Following the work
of C. Stern (1936) onmitotic recombination, Flp-dependent inter-
chromosomal recombination had been used in Drosophila to
generate mutant clones that no longer express a marker (e.g.,
Xu and Rubin, 1993). An adaptation to produce positively
marked clones, mosaic analysis with repressible cell marker
(MARCM) (Figure 4A), leads to segregation of the mutant allele
after mitotic recombination
A1 A2MARCM
Gal80
tub
XX
Gal80
after mitotic recombination
B1 B2Twin-spot MARCM
RFP-miR
GFP-miR
GFP-miR
RFP-miR
X
X
C1 C2
12
lacZ
lacZ
[UAS-GFP] [tub-Gal4]
XX
[UAS-GFP] [tub-Gal4] hsp-Flp
X
X
X
X
hsp-Flp
X
X
RFP-miR
RFP-miR
X
X
[GFP] [RFP]
GFP-miR
GFP-miR
[GFP] [RFP]
[GFP] [RFP]
Gal80
tub
XX
Gal80
[UAS-GFP] [tub-Gal4]
X
hsp-Flp
X
hsp-Flp
1
2 X
X
X
X
X X
ubiquitous
Figure 4. Schematic Representations of Different Approaches to Tracing Cells in Mutant Clones in Drosophila(A1) The MARCM system depends on the elimination, by mitotic recombination, of the Gal80 transgene, which lies on the same chromosome arm as the mutantallele (X). This results in their segregation and thus the activation of the GFP reporter in the homozygote mutant cells (5).(A2) Transient heat-shock activation of the Flp recombinase results in reporter expression in mutant cells (Lee and Luo, 1999).(B1 and B2) In twin-spotMARCM, the two systems are combined (see also Figure 3A), in this case usingmicroRNAs (miR) as repressors ofGFP orRFP expression(Yu et al., 2009).(C1 and C2) To trace lineage progression in mutant clones, the MARCM system of GFP activation is combined with the system of heritable expression of lacZ(Harrison and Perrimon, 1993). A second reporter (lacZ) is activated after a second heat-shock-induced recombination that brings it under the control ofa ubiquitous promoter (Perdigoto et al., 2011). <: FRT sites for Flp recombination.
Developmental Cell
Perspective
from the repressor GAL80, so that a UAS-driven fluorescent
reporter is now activated in a mutated daughter cell and its
descendants (Lee and Luo, 1999). Twin-spot MARCM combines
the two approaches to follow sister cells (Yu et al., 2009). In this
case, repressors are microRNAs that target UAS-dependent
markers and are lost after mitotic recombination (Figure 4B).
This reduces the delay of MARCM derepression, which other-
De
wise depends on GAL80 decay. A potential difficulty is to deter-
mine when the mutation becomes effective, depending on the
perdurance of the endogenous protein. Ideally, one would like
to study lineage progression within clones of mutant cells. This
can be achieved inDrosophila by combining theMARCMsystem
with an independent Flp-induced recombination event that acti-
vates expression of a lacZ reporter (Figure 4C), so that two
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 401
Developmental Cell
Perspective
sequential recombination events can generate clones of
b-galactosidase-positive cells within GFP marked clones of
mutant cells (Perdigoto et al., 2011). In this way, Notch signaling
in the lineage progression of intestinal stem cells has been
shown to restrict their self-renewal as well as affecting the later
stage of terminal differentiation.
Mosaics of cells with distinct genotypes also provide insight
into cell-cell interactions and the importance of cell competition
in selecting the progenitor of tissues and organs (Morata and
Ripoll, 1975). For tracing the descendants of mutant cells in
heterozygote mice, the MADM system can be used with distinct
fluorescent reporters for wild-type and mutant alleles. For this
purpose, a new reporter line, MADM-11, has been developed
to generate clones mutant for a gene located on chromosome
11. With a Cre recombinase under the control of Emx1, which
is expressed in cortical progenitors of the forebrain, it was shown
that components of the Lis1/Ndel1/14.3.3ε complex, which is
defective in lissencephaly syndromes, have distinct cell-autono-
mous functions during different stages of neuronal migration
(Hippenmeyer et al., 2010). The approach in which differently
marked ES cells are introduced into amouse blastocyst to create
mosaics can be combined with functional studies on mutant
ES cells to investigate the cell-autonomous roles of a gene. Alter-
natively, with a specific Cre, genetic tracing of clones can be
achieved. Using a Flk1-Cre integrated into the genome of the
reporter ES cells it was shown that all the endothelial cell
derivatives in the blood islands were derived from progenitors
expressing robust levels of Flk1, whereas most hematopoietic
cells were not (Ueno and Weissman, 2006). Mosaics are also
employed to perform clonal analysis on a mouse mutant back-
ground. Experiments in which insertion of a GFP reporter into
the X chromosome resulted in random X-inactivation made it
possible to follow cell dynamics in the developing limb bud by
live imaging in a Wnt5a mutant embryo (Gros et al., 2010).
Classic single-cell microinjection into the mouse epiblast has
also been carried out with mutant embryos, for example to
show that Otx2 is not required for proliferation of the visceral
endoderm lineage but is essential for anteriorly directed cell
movement (Perea-Gomez et al., 2001).
In addition to examining the effects of mutations on cell fate
choice and associated cell behavior, a related challenge is to
integrate the cellular data with dynamic gene expression, to
understand how genes are expressed or repressed during
progression along a lineage tree. The use of reporters of gene
expression with limited stability, such as the fusion protein
H2B-GFP (Plusa et al., 2008) or destabilized GFP (Harper et al.,
2010), permits reliable imaging of gene expression dynamics.
Integrating quantitative data on transcription factor kinetics
with subsequent lineage patterning is now realizable. Thus
monitoring the nucleocytoplasmic movement of Oct4 fused to
a photoactivatable GFP has demonstrated that differences in
Oct4 kinetics predict the future identity of mouse blastocyst
lineages (Plachta et al., 2011).
Modeling the emergence of different cell types in a lineage is
an emerging theme, as, for example, in C. elegans, for which
a complete lineage tree has been reconstructed and regulatory
genes have been identified. Such a predictive model gives an
indication of the number of regulatory factors required for reca-
pitulating the lineage, the synergistic variation of factors, and
402 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
where, in the cell lineage tree, asymmetry might be controlled
by external influences (Larsson et al., 2011). The superposition
of experimentally determined gene regulatory networks on cell
lineage is beautifully illustrated by pioneering work on the sea
urchin embryo, where lineage is mainly invariant and early
lineage segregation has been examined on a cell-by-cell basis
in terms of transcriptional regulation and cell signaling. In this
way, for example, the endoderm gene regulatory network has
been defined up to the midblastula stage, giving new insight
also into the progressive segregation of endodermal from
mesodermal lineages (Peter and Davidson, 2010).
In most of the mosaic analyses previously discussed, the
lineage is preidentified by the use of tissue-specific regulatory
sequences and in this respect is therefore prospective. The inter-
pretation of mosaics in terms of clones is limited to local events
when growth is coherent and clusters of cells can be considered
as clonal units. In retrospective clonal analysis, it is possible to
lower the frequency of cell labeling to reach clonal conclusions,
even when growth is dispersive and to do this on the scale of the
whole organism.
Retrospective Clonal Analysis: Systematic Analysisof All Progenitors of a StructureProspective lineage analysis depends on a preconceived idea
about the progenitor cell population. Preidentification of the
potential stage and location of the progenitors to label is
required. However, this is not always known and potentially
restricts conclusions on lineage by not considering other coun-
terintuitive options. In contrast, retrospective approaches to
cell lineage depend on analyzing labeled cells at the end point
of the experiment and deducing their interrelationships and
previous history. Retrospective clonal analysis based on the
random genetic labeling of progenitors at a low frequency
constitutes a progenitor screen and permits the systematic anal-
ysis of the potential of any progenitor to colonize a particular
structure. We distinguish two kinds of retrospective clonal
analyses, depending on whether labeling is random in space,
but with temporal control, or random in both space and time.
Spatially Random Labeling
Clones that are induced at a particular time but result from
random spatial labeling have been produced in Drosophila by
X-irradiation induced recombination, pioneered by H.J. Becker
(1957), and subsequently employed to generate, in a heterozy-
gous Minute mutant background, Minute-positive cells with
a cuticular marker, which tend to outgrow their mutant neighbors
(Garcia-Bellido et al., 1973). The distribution of clusters of such
cells demonstrated the existence of internal demarcation lines
in the wing disc, which the clones did not cross. This classic
work led to important concepts and definitions, including
those of clonal compartments and clonal boundaries (Garcia-
Bellido, 1985). In the plant kingdom, elegant experiments
on genetic variegation have manipulated transposon-mediated
gene silencing. For example, a change in temperature led to
low-frequency mobility of a transposon in the promoter region
of the Pal gene, which encodes a red pigment, resulting in
restoration of gene function to give red sectors (clones) on the
ivory background of an Antirrhinum petal. In such experiments,
the sequence of lineage restrictions in the developing floral
meristem has been revealed (Vincent et al., 1995). Sophisticated
Developmental Cell
Perspective
quantitative analysis of sector parameters coupled to computer
modeling led to conclusions on the growth parameters that
are essential to give shape to the flower petal (Rolland-Lagan
et al., 2005).
In amniotes, infection with replication-defective retroviruses
that integrate into the genome provides a spatially random
labeling of progenitors at a defined time. This approach, first
described for themouse (Saneset al., 1986),was refined for clonal
analysis by the development of libraries of individually marked
retroviruses, where each member encodes a reporter and has
a DNA tag (Golden et al., 1995). The complexity of the library
permits evaluation of clonality between labeled cells, based on
the presence of the same tag, identified by PCR. This approach
has been extensively used in birds and mammals, especially for
characterizing lineage segregation in the central nervous system,
as in the early demonstration that clonal derivatives contribute to
more than onemajor subdivision of the telencephalon (Walsh and
Cepko, 1992). In a recent adaptation, used for clonal analysis of
blood cell types, infection with a barcoded retroviral library
carrying a fluorescent marker was followed by separation of indi-
vidual circulating cells by flow cytometry and sensitive sequence-
based characterization of clonally related cells (Gerrits et al.,
2010). Potential problems for random retroviral labeling are inte-
gration site effects and variable infectivity, illustrated by murine
retroviruses that only infect proliferative cells, a limitation partially
overcome by lentiviruses based vectors. As an alternative to
libraries, low-level infection of GFP encoding retro- and lentivi-
ruses has been used to mark single cells—for example, recently
in the zebrafish brain, to show that whereas neuroblasts undergo
a limited amplification, single radial glial cells self-renew and
mines the tissue analyzed at the end point but does not condition
the genetic labeling of the progenitor cells that give rise to it. The
rarity of the event makes clonal analysis possible (Bonnerot and
Nicolas, 1993). Collections of embryos are generated, in which
the frequency of labeling, in the structure under consideration,
is determined. To establish clonality, statistical analyses are
required, based on the frequency of observations, as, for
example, the fluctuation test of Luria and Delbruck (1943), which
estimates the probability of one or more than one recombination
events. The different types of clones that result from random
labeling can be divided into groups based on their characteris-
tics such as size, spatial distribution, and cell type. When similar
clones are observed it can be assumed that the library of clones
has reached saturation. From the collection of clones, derived
from progenitors that have undergone recombination at different
stages, the temporal history can be reconstructed, based on the
premise that subclones have more restricted cell fate potential
than parental clones. In addition to reconstruction of the lineage,
its number of founder cells, and diversification into sublineages,
important aspects of cell behavior such as the formation of clonal
boundaries, or asymmetric stem cell versus symmetric prolifera-
tive modes of cell division, can also be deduced from the prop-
erties of clones within the library (Nicolas et al., 1996; Petit et al.,
2005). An example is provided by analysis of clones in embryos
of an a-cardiac actinnlaacZ/+ mouse line, which led to the demon-
stration of two myocardial cell lineages, which segregate early,
with distinct and overlapping contributions to different parts of
the heart (Meilhac et al., 2004). Subsequent analysis, using the
same mouse line, established sublineages, within the second
myocardial lineage, that contribute to different parts of the arte-
rial pole of the heart and also to different skeletal muscle groups
in the head (Lescroart et al., 2010) (Figures 5G–5L). The nlaacZ
approach has resulted in new lineage insights for many tissues
in the mouse, including segregation of the germ layers during
gastrulation (Tzouanacou et al., 2009), and should be applicable
to other species. A similar reporter, based on duplication in the
b-glucuronidase gene, has been used for clonal analysis in plants
(Swoboda et al., 1994).
The strength of random retrospective clonal analyses is to
reconstruct lineage trees over an extended time scale and to
understand the mode of regionalization of a structure, by target-
ing systematically all progenitors. However, the inference of the
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 403
A
lacZSTOP
ROSA26
low dose of tamoxifen,Cre-ERT2
B
C
E14.5-2084
OFT
E14.5-2901RV
J
1BA2BA
1
3
67
24
5
OFT
RV
1BA
heart head
nlacZ
β-gal
nlaacZ
2
3
56
4
E8.5-156
1
25
6
E8.5-204
first lineage
second lineage
K L
EG H I
F
α-cardiac actin
STOP
α-cardiac actin
spontaneous recombination
nlacZ
nlaacZ
D
loxP
medulla clonecuticle cloneIRS clone
2BA
Figure 5. Retrospective Clonal Analysis in Mouse(A–D) Inducible clonal analysis based on recombination of a conditional ROSA26 reporter (A) activated by low doses of tamoxifen (B). Clones in different layers(color coded; IRS, inner root sheath) of the hair follicle originate from different domains (red arrowheads) of the matrix which is a source of stem cells (C and D).(Legue and Nicolas, 2005, adapted with permission).(E–L) Random retrospective clonal analysis by the laacZ approach (E) The laacZ reporter is rendered nonfunctional by an internal duplication, which canspontaneously recombine into a functional lacZ gene, at a low frequency.(F) Random generation of an lacZ-positive clone (cells outlined in blue), which is detectable in the expression domain of the ac-actin promoter, i.e., in cardiac andskeletal muscles (full blue circles indicate b-galactosidase [b-gal]-positive cells).(G and H) Examples of b-gal-positive clones with an exclusive contribution to region 1 (outflow tract) or 3 (left ventricle) of the myocardium, indicative of lineagesegregation. E8.5, embryonic day 8.5 (followed by clone number).(I) The contributions of the first and secondmyocardial lineages, based on the analysis of 3,629 embryonic heart tubes, are summarized in red and green (Meilhacet al., 2004).(J and K) Examples of b-gal-positive clones of the second myocardial lineage colonizing both skeletal muscles of the head and myocardium of the heart, takenfrom a collection of 2,223 fetuses.(L) The lineage contributing to headmuscles derived from the first branchial arch (1BA), which also contributes to the right ventricle (RV), is shown in blue, while thelineage that contributes to second branchial arch-derived head muscles (2BA) and to the outflow tract (OFT) is represented in pink (Lescroart et al., 2010).
Developmental Cell
Perspective
events that precede the observations can be controversial. To
gain direct access to the dynamics of lineage progression, live
analyses are required, associated with successive observations
at shorter time intervals.
404 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
Four-Dimensional Imaging of Lineage ProgressionA grail of lineage analysis is the complete four-dimensional (4D)
imaging of cells in vivo. This was pioneered in the chick, in which
time-lapse imaging was first attempted over 80 years ago
Figure 6. Lineage Reconstruction for the Early Zebrafish EmbryoBased on 4D Imaging(A–C) Images of mitotic spindles by second harmonic generation (SHG) signal(A), of chromosomes marked by H2B-mcherry (B) and membranes by thirdharmonic generation (THG) signal (C).(D1 and D2) Digital reconstruction of the embryos with color-coded cell line-ages at the 8-cell (D1) and 512-cell (D2) stages.(E) An example of information on cell behavior resulting from this approachshows the orientations of cell divisions, where each color corresponds tosuccessive cell cycles (Olivier et al., 2010, adapted with permission).
Developmental Cell
Perspective
(Wetzel, 1929). A complete lineage tree has been accomplished
for C. elegans using Nomarski optics and direct observation to
track cells (Sulston et al., 1983). This remarkable achievement
was rendered possible by the transparency and small size (fewer
than 1,000 cells) of the nematode. Furthermore, the fact that the
lineage was found to be invariant facilitated the analysis. Inter-
estingly, in spite of the fixed relationship between cell ancestry
and cell fate, the correlation between them lacks obvious
pattern; for example, neurons do not all derive from ectoderm
and only intestinal and germ cells are of monoclonal origin.
With fluorescent markers facilitating in vivo imaging, as dis-
cussed in previous sections, tracking labeled cells has led to
information on cell fate and associated behavior in many organ-
isms. In classic experiments in zebrafish, partial lineages were
reconstructed by video recording, leading, for example, to the
demonstration that specific cell behavior is coupled to particular
cell cycles and appears to account for clonal restriction in neural
De
cell fate (Kimmel et al., 1994). Time-lapse imaging of cortical sli-
ces in the mouse embryo, after in utero infection with GFP-ex-
pressing retroviruses, showed that radial glia generate neurons
by asymmetric cell division (Noctor et al., 2001). However, with
the exception of experiments on invertebrates, only fragmentary
information on specific lineages has been obtained, mainly
limited by the cell-labeling procedure.
For more systematic lineage reconstructions, pioneering
experiments in plants, where the absence of cell migration and
also of apoptosis facilitate cell tracking, have used fluorescent
markers expressed in all cells, with a subcellular resolution,
targeting the membrane or chromatin or revealing cell-cycle
stages. In this case, confocal microscopy with long-term
(12 days), as well as short-term, imaging was employed to
analyze the development of the flower primordium from the
meristem in Arabidopsis. Image registration algorithms were
developed to assist lineage reconstruction (Reddy et al., 2004).
In the plant field, the challenge of quantitative analysis of growth
parameters over time is being met by new tools for image pro-
cessing and reconstruction to track cell lineages (Fernandez
et al., 2010). During animal development, confocal and multi-
photon microscopy are currently extensively used for imaging
cells. However, problems of light scattering and resolution are
particularly critical for cell tracking. With the current interest in
stem cells in the adult, as well as during development, accessing
cells that are located deep within an organism can be a major
problem. Advances in light sheet microscopy (e.g., SPIM,
DSLM), successfully used on zebrafish embryos at early (Keller
et al., 2008) or late (Swoger et al., 2011) developmental stages,
hold out new promise for minimally invasive, high-resolution
images with good penetration depth and fast acquisition
(Huisken and Stainier, 2009). By following fluorescently labeled
nuclei, with an automated image segmentation procedure, Keller
et al. (2008) provide a resource of ‘‘digital embryos,’’ for lineage
analysis over 24 hr, from early cleavage stages until the onset of
organogenesis. Another new development, which is based on
label-free multiphoton technology with spiral scanning to opti-
mize resolution, penetration, and photoperturbation, has led to
complete lineage reconstruction of zebrafish early development,
up until the 1,000-cell (blastula) stage (Olivier et al., 2010)
(Figure 6). This remarkable technical feat exploited the intrinsic
optical nonlinear properties of the sample (harmonic genera-
tion)—namely, the oriented microtubules of the spindle and
aqueous/lipidic surfaces such as membranes—together with
two-photon excitation of a fluorescent chromosomal marker en-
coded by a transgene. This analysis required sophisticated
processes of image acquisition, as well as algorithms for auto-
mated lineage reconstruction, which nevertheless still relied on
time-consuming visual verification. A challenge for these new
approaches is the requirement for higher-resolution multicolor
cell imaging and for optimization of algorithms for fast and auto-
mated image reconstruction in dense environments, so as to
allow unambiguous cell tracking. A practical problem for extend-
ing long-term imaging to later stages is the need to immobilize
the animal without impeding development. It is not clear how
far 4D analysis will progress to map cell behavior from the outset
of development, but it also provides the potential to extend
fragmentary imaging of later cell lineage choices into more
comprehensive documentary films of cell history.
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 405
Developmental Cell
Perspective
In conclusion, the technological developments discussed here
have opened up new horizons. Genetic tools, which are
becoming available for an increasing number of organisms,
have solved the old problem of marker dilution at cell division
and have introduced sophisticated methods for spatio-temporal
targeting of labeling. Developments in the range of fluorescent
markers now permit direct and multicolor observations, which
can be orchestrated at will by genetic engineering. The classical
problem of clonality of the labeling is addressed either by
combining several markers or by lowering the frequency of
labeling. Advances in microscopy are crucial for recording clonal
data with increasing resolution in four dimensions. Sophisticated
computational methods are being developed to analyze the
large data sets generated by clonal analyses and to provide
in-depth understanding of the cellular mechanisms leading to
the observed clonal patterns. A challenge for the next decade
is to grasp the significance of changes in cell behavior followed
at a single-cell level and to integrate the cellular with the
molecular dimension to understand lineage choices and lineage
progression. In the future, in vivo imaging and genetic manip-
ulation of markers will be widely applicable to the diversity
of species already apprehended in Evo/Devo type studies, no
doubt leading to unexpected conceptual lineage developments
and revealing the cellular aspects of evolutionary ‘‘tinkering’’
(Jacob, 1977).
ACKNOWLEDGMENTS
We thank P. Herbomel, E. Hirsinger, J.-F. Nicolas, and F. Schweisguth forinsightful discussions and anonymous reviewers for comments. Work onlineage analysis in the Buckingham lab is supported by the Institut Pasteur,the CNRS (URA 2578) and by the EU through the CardioCell (FP7 - HEALTH-2007-2.4.2-5) project. S.M.M. is an INSERM research scientist.
REFERENCES
Arques, C.G., Doohan, R., Sharpe, J., and Torres, M. (2007). Cell tracingreveals a dorsoventral lineage restriction plane in the mouse limb bud mesen-chyme. Development 134, 3713–3722.
Baker, S.M., Buckheit, R.W., 3rd, and Falk, M.M. (2010). Green-to-redphotoconvertible fluorescent proteins: tracking cell and protein dynamics onstandard wide-field mercury arc-based microscopes. BMC Cell Biol. 11, 15.
Ba1akier, H., and Pedersen, R.A. (1982). Allocation of cells to inner cell massand trophectoderm lineages in preimplantation mouse embryos. Dev. Biol.90, 352–362.
Becker, H.J. (1957). Roentgen mosaic spots & defective mutations at theeye of Drosophila & the evolutive physiology of the eye. Z. Indukt. Abstamm.Vererbungsl. 88, 333–373.
Beddington, S.P. (1981). An autoradiographic analysis of the potency ofembryonic ectoderm in the 8th day postimplantation mouse embryo. J. Em-bryol. Exp. Morphol. 64, 87–104.
Ben-Yair, R., and Kalcheim, C. (2005). Lineage analysis of the avian dermo-myotome sheet reveals the existence of single cells with both dermal andmuscle progenitor fates. Development 132, 689–701.
Boniface, E.J., Lu, J., Victoroff, T., Zhu, M., and Chen, W. (2009). FlEx-basedtransgenic reporter lines for visualization of Cre and Flp activity in live zebra-fish. Genesis 47, 484–491.
Bonnerot, C., and Nicolas, J.F. (1993). Clonal analysis in the intact mouseembryo by intragenic homologous recombination. C. R. Acad. Sci. III 316,1207–1217.
406 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as ameans ofaltering cell fates and generating dominant phenotypes. Development 118,401–415.
Caron, S.J., Prober, D., Choy, M., and Schier, A.F. (2008). In vivo birthdating byBAPTISM reveals that trigeminal sensory neuron diversity depends on earlyneurogenesis. Development 135, 3259–3269.
Christian, M., Cermak, T., Doyle, E.L., Schmidt, C., Zhang, F., Hummel, A.,Bogdanove, A.J., and Voytas, D.F. (2010). Targeting DNA double-strandbreaks with TAL effector nucleases. Genetics 186, 757–761.
Christoffels, V.M., Grieskamp, T., Norden, J., Mommersteeg, M.T., Rudat, C.,and Kispert, A. (2009). Tbx18 and the fate of epicardial progenitors. Nature458, E8–E9, discussion E9–E10.
Cockburn, K., and Rossant, J. (2010). Making the blastocyst: lessons from themouse. J. Clin. Invest. 120, 995–1003.
Collins, R.T., Linker, C., and Lewis, J. (2010). MAZe: a tool for mosaic analysisof gene function in zebrafish. Nat. Methods 7, 219–223.
Conklin, E.G. (1905). The organization and cell lineage of the ascidian egg. J.Acad. Nat. Sci. Phila. 13, 1–119.
Distel, M., Wullimann, M.F., and Koster, R.W. (2009). Optimized Gal4 geneticsfor permanent gene expression mapping in zebrafish. Proc. Natl. Acad. Sci.USA 106, 13365–13370.
Evans, C.J., Olson, J.M., Ngo, K.T., Kim, E., Lee, N.E., Kuoy, E., Patananan,A.N., Sitz, D., Tran, P., Do, M.T., et al. (2009). G-TRACE: rapid Gal4-basedcell lineage analysis in Drosophila. Nat. Methods 6, 603–605.
Feil, R., Wagner, J., Metzger, D., and Chambon, P. (1997). Regulation of Crerecombinase activity by mutated estrogen receptor ligand-binding domains.Biochem. Biophys. Res. Commun. 237, 752–757.
Fernandez, R., Das, P., Mirabet, V., Moscardi, E., Traas, J., Verdeil, J.L., Ma-landain, G., and Godin, C. (2010). Imaging plant growth in 4D: robust tissuereconstruction and lineaging at cell resolution. Nat. Methods 7, 547–553.
Galli, D., Domınguez, J.N., Zaffran, S., Munk, A., Brown, N.A., and Bucking-ham, M.E. (2008). Atrial myocardium derives from the posterior region of thesecond heart field, which acquires left-right identity as Pitx2c is expressed.Development 135, 1157–1167.
Galperin, C. (1998). From cell lineage to developmental genetics. Hist. Philos.Life Sci. 20, 301–344.
Garcia-Bellido, A. (1985). Cell lineages and genes. Philos. Trans. R. Soc. Lond.B Biol. Sci. 312, 101–128.
Garcia-Bellido, A., Ripoll, P., and Morata, G. (1973). Developmental compart-mentalisation of the wing disk of Drosophila. Nat. New Biol. 245, 251–253.
Gardner, R.L., and Rossant, J. (1979). Investigation of the fate of 4-5 day post-coitum mouse inner cell mass cells by blastocyst injection. J. Embryol. Exp.Morphol. 52, 141–152.
Gerrits, A., Dykstra, B., Kalmykowa, O.J., Klauke, K., Verovskaya, E., Broe-khuis, M.J., de Haan, G., and Bystrykh, L.V. (2010). Cellular barcoding toolfor clonal analysis in the hematopoietic system. Blood 115, 2610–2618.
Gline, S.E., Kuo, D.H., Stolfi, A., and Weisblat, D.A. (2009). High resolutioncell lineage tracing reveals developmental variability in leech. Dev. Dyn. 238,3139–3151.
Golden, J.A., Fields-Berry, S.C., and Cepko, C.L. (1995). Construction andcharacterization of a highly complex retroviral library for lineage analysis.Proc. Natl. Acad. Sci. USA 92, 5704–5708.
Gossen, M., and Bujard, H. (1992). Tight control of gene expression inmammalian cells by tetracycline-responsive promoters. Proc. Natl. Acad.Sci. USA 89, 5547–5551.
Griffin, R., Sustar, A., Bonvin, M., Binari, R., del Valle Rodriguez, A., Hohl, A.M.,Bateman, J.R., Villalta, C., Heffern, E., Grunwald, D., et al. (2009). The twin spotgenerator for differential Drosophila lineage analysis. Nat. Methods 6,600–602.
Gros, J., Hu, J.K., Vinegoni, C., Feruglio, P.F., Weissleder, R., and Tabin, C.J.(2010). WNT5A/JNK and FGF/MAPK pathways regulate the cellular eventsshaping the vertebrate limb bud. Curr. Biol. 20, 1993–2002.
Developmental Cell
Perspective
Hadjieconomou, D., Rotkopf, S., Alexandre, C., Bell, D.M., Dickson, B.J., andSalecker, I. (2011). Flybow: genetic multicolor cell labeling for neural circuitanalysis in Drosophila melanogaster. Nat. Methods 8, 260–266.
Hampel, S., Chung, P., McKellar, C.E., Hall, D., Looger, L.L., and Simpson,J.H. (2011). Drosophila Brainbow: a recombinase-based fluorescence labelingtechnique to subdivide neural expression patterns. Nat. Methods 8, 253–259.
Harper, C.V., Featherstone, K., Semprini, S., Friedrichsen, S., McNeilly, J.,Paszek, P., Spiller, D.G., McNeilly, A.S., Mullins, J.J., Davis, J.R., and White,M.R. (2010). Dynamic organisation of prolactin gene expression in livingpituitary tissue. J. Cell Sci. 123, 424–430.
Harrison, D.A., and Perrimon, N. (1993). Simple and efficient generation ofmarked clones in Drosophila. Curr. Biol. 3, 424–433.
Hatta, K., Tsujii, H., and Omura, T. (2006). Cell tracking using a photoconvert-ible fluorescent protein. Nat. Protoc. 1, 960–967.
Hayashi, S., and McMahon, A.P. (2002). Efficient recombination in diversetissues by a tamoxifen-inducible form of Cre: a tool for temporally regulatedgene activation/inactivation in the mouse. Dev. Biol. 244, 305–318.
Hippenmeyer, S., Youn, Y.H., Moon, H.M., Miyamichi, K., Zong, H., Wynshaw-Boris, A., and Luo, L. (2010). Genetic mosaic dissection of Lis1 and Ndel1 inneuronal migration. Neuron 68, 695–709.
Hirose, G., and Jacobson, M. (1979). Clonal organization of the central nervoussystem of the frog. I. Clones stemming from individual blastomeres of the16-cell and earlier stages. Dev. Biol. 71, 191–202.
Huisken, J., and Stainier, D.Y. (2009). Selective plane illumination microscopytechniques in developmental biology. Development 136, 1963–1975.
Jacob, F. (1977). Evolution and tinkering. Science 196, 1161–1166.
Keller, P.J., Schmidt, A.D., Wittbrodt, J., and Stelzer, E.H. (2008). Recon-struction of zebrafish early embryonic development by scanned light sheetmicroscopy. Science 322, 1065–1069.
Kimmel, C.B., and Law, R.D. (1985a). Cell lineage of zebrafish blastomeres.I. Cleavage pattern and cytoplasmic bridges between cells. Dev. Biol. 108,78–85.
Kimmel, C.B., and Law, R.D. (1985b). Cell lineage of zebrafish blastomeres. III.Clonal analyses of the blastula and gastrula stages. Dev. Biol. 108, 94–101.
Kimmel, C.B., and Warga, R.M. (1986). Tissue-specific cell lineages originatein the gastrula of the zebrafish. Science 231, 365–368.
Kimmel, C.B., Warga, R.M., and Kane, D.A. (1994). Cell cycles and clonalstrings during formation of the zebrafish central nervous system. Development120, 265–276.
Klein, A.M., and Simons, B.D. (2011). Universal patterns of stem cell fate incycling adult tissues. Development 138, 3103–3111.
Larsson, J.K., Wadstromer, N., Hermanson, O., Lendahl, U., and Forchheimer,R. (2011). Modelling cell lineage using a meta-Boolean tree model witha relation to gene regulatory networks. J. Theor. Biol. 268, 62–76.
Lawson, K.A., Meneses, J.J., and Pedersen, R.A. (1991). Clonal analysis ofepiblast fate during germ layer formation in the mouse embryo. Development113, 891–911.
Le Douarin, N., and Barq, G. (1969). Use of Japanese quail cells as ‘‘biologicalmarkers’’ in experimental embryology. C. R. Acad. Sci. Hebd. Seances Acad.Sci. D 269, 1543–1546.
Lee, T., and Luo, L. (1999). Mosaic analysis with a repressible cell marker forstudies of gene function in neuronal morphogenesis. Neuron 22, 451–461.
Legue, E., and Nicolas, J.F. (2005). Hair follicle renewal: organization ofstem cells in the matrix and the role of stereotyped lineages and behaviors.Development 132, 4143–4154.
Lescroart, F., Kelly, R.G., Le Garrec, J.F., Nicolas, J.F., Meilhac, S.M., andBuckingham, M. (2010). Clonal analysis reveals common lineage relationshipsbetween head muscles and second heart field derivatives in the mouseembryo. Development 137, 3269–3279.
De
Livet, J., Weissman, T.A., Kang, H., Draft, R.W., Lu, J., Bennis, R.A., Sanes,J.R., and Lichtman, J.W. (2007). Transgenic strategies for combinatorialexpression of fluorescent proteins in the nervous system. Nature 450, 56–62.
Luria, S.E., andDelbruck,M. (1943). Mutations of bacteria from virus sensitivityto virus resistance. Genetics 28, 491–511.
Marcon, L., Arques, C.G., Torres, M.S., and Sharpe, J. (2011). A computationalclonal analysis of the developing mouse limb bud. PLoS Comput. Biol. 7,e1001071.
McCauley, D.W., and Bronner-Fraser, M. (2003). Neural crest contributions tothe lamprey head. Development 130, 2317–2327.
McGuire, S.E., Le, P.T., Osborn, A.J., Matsumoto, K., and Davis, R.L. (2003).Spatiotemporal rescue of memory dysfunction in Drosophila. Science 302,1765–1768.
Meilhac, S.M., Esner, M., Kelly, R.G., Nicolas, J.F., and Buckingham, M.E.(2004). The clonal origin of myocardial cells in different regions of the embry-onic mouse heart. Dev. Cell 6, 685–698.
Meilhac, S.M., Adams, R.J., Morris, S.A., Danckaert, A., Le Garrec, J.F.,and Zernicka-Goetz, M. (2009). Active cell movements coupled to positionalinduction are involved in lineage segregation in the mouse blastocyst. Dev.Biol. 331, 210–221.
Mintz, B. (1965). Genetic Mosaicism in Adult Mice of Quadriparental Lineage.Science 148, 1232–1233.
Moody, S.A. (1987). Fates of the blastomeres of the 32-cell-stage Xenopusembryo. Dev. Biol. 122, 300–319.
Morata, G., and Ripoll, P. (1975). Minutes: mutants of drosophila autono-mously affecting cell division rate. Dev. Biol. 42, 211–221.
Mosimann, C., Kaufman, C.K., Li, P., Pugach, E.K., Tamplin, O.J., and Zon, L.I.(2011). Ubiquitous transgene expression and Cre-based recombination drivenby the ubiquitin promoter in zebrafish. Development 138, 169–177.
Muzumdar, M.D., Tasic, B., Miyamichi, K., Li, L., and Luo, L. (2007). A globaldouble-fluorescent Cre reporter mouse. Genesis 45, 593–605.
Naiche, L.A., and Papaioannou, V.E. (2007). Cre activity causes widespreadapoptosis and lethal anemia during embryonic development. Genesis 45,768–775.
Nicolas, J.F., Mathis, L., Bonnerot, C., and Saurin, W. (1996). Evidence inthe mouse for self-renewing stem cells in the formation of a segmentedlongitudinal structure, the myotome. Development 122, 2933–2946.
Nishida, H. (1987). Cell lineage analysis in ascidian embryos by intracellularinjection of a tracer enzyme. III. Up to the tissue restricted stage. Dev. Biol.121, 526–541.
Noctor, S.C., Flint, A.C., Weissman, T.A., Dammerman, R.S., and Kriegstein,A.R. (2001). Neurons derived from radial glial cells establish radial units inneocortex. Nature 409, 714–720.
Nowotschin, S., and Hadjantonakis, A.K. (2009). Use of KikGR a photoconver-tible green-to-red fluorescent protein for cell labeling and lineage analysis inES cells and mouse embryos. BMC Dev. Biol. 9, 49.
Nowotschin, S., Eakin, G.S., and Hadjantonakis, A.K. (2009). Live-imagingfluorescent proteins in mouse embryos: multi-dimensional, multi-spectralperspectives. Trends Biotechnol. 27, 266–276.
Olivier, N., Luengo-Oroz, M.A., Duloquin, L., Faure, E., Savy, T., Veilleux, I.,Solinas, X., Debarre, D., Bourgine, P., Santos, A., et al. (2010). Cell lineagereconstruction of early zebrafish embryos using label-free nonlinear micros-copy. Science 329, 967–971.
Perdigoto, C.N., Schweisguth, F., and Bardin, A.J. (2011). Distinct levels ofNotch activity for commitment and terminal differentiation of stem cells inthe adult fly intestine. Development, in press. 10.242/dev.065292.
Perea-Gomez, A., Lawson, K.A., Rhinn,M., Zakin, L., Brulet, P., Mazan, S., andAng, S.L. (2001). Otx2 is required for visceral endoderm movement and forthe restriction of posterior signals in the epiblast of the mouse embryo.Development 128, 753–765.
Peter, I.S., and Davidson, E.H. (2010). The endoderm gene regulatory networkin sea urchin embryos up to mid-blastula stage. Dev. Biol. 340, 188–199.
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 407
Developmental Cell
Perspective
Petit, A.C., Legue, E., and Nicolas, J.F. (2005). Methods in clonal analysis andapplications. Reprod. Nutr. Dev. 45, 321–339.
Piatkevich, K.D., Hulit, J., Subach, O.M., Wu, B., Abdulla, A., Segall, J.E., andVerkhusha, V.V. (2010). Monomeric red fluorescent proteins with a largeStokes shift. Proc. Natl. Acad. Sci. USA 107, 5369–5374.
Plachta, N., Bollenbach, T., Pease, S., Fraser, S.E., and Pantazis, P. (2011).Oct4 kinetics predict cell lineage patterning in the early mammalian embryo.Nat. Cell Biol. 13, 117–123.
Placinta, M., Shen, M.C., Achermann, M., and Karlstrom, R.O. (2009). A laserpointer driven microheater for precise local heating and conditional generegulation in vivo. Microheater driven gene regulation in zebrafish. BMC Dev.Biol. 9, 73.
Plusa, B., Piliszek, A., Frankenberg, S., Artus, J., and Hadjantonakis, A.K.(2008). Distinct sequential cell behaviours direct primitive endoderm formationin the mouse blastocyst. Development 135, 3081–3091.
Reddy, G.V., Heisler, M.G., Ehrhardt, D.W., and Meyerowitz, E.M. (2004).Real-time lineage analysis reveals oriented cell divisions associated withmorphogenesis at the shoot apex of Arabidopsis thaliana. Development 131,4225–4237.
Rolland-Lagan, A.G., Coen, E., Impey, S.J., and Bangham, J.A. (2005). Acomputational method for inferring growth parameters and shape changesduring development based on clonal analysis. J. Theor. Biol. 232, 157–177.
Rosenquist, G.C., andDeHaan, R.L. (1966). Migration of precardiac cells in thechick embryo: a radioautographic study. Contrib. Embryol. 263, 111–121.
Rothenaigner, I., Krecsmarik, M., Hayes, J.A., Bahn, B., Lepier, A., Fortin, G.,Gotz, M., Jagasia, R., and Bally-Cuif, L. (2011). Clonal analysis by distinct viralvectors identifies bona fide neural stem cells in the adult zebrafish telenceph-alon and characterizes their division properties and fate. Development 138,1459–1469.
Sanes, J.R., Rubenstein, J.L., and Nicolas, J.F. (1986). Use of a recombinantretrovirus to study post-implantation cell lineage in mouse embryos. EMBOJ. 5, 3133–3142.
Sato, Y., Poynter, G., Huss, D., Filla, M.B., Czirok, A., Rongish, B.J., Little,C.D., Fraser, S.E., and Lansford, R. (2010). Dynamic analysis of vascularmorphogenesis using transgenic quail embryos. PLoS ONE 5, e12674.
Satoh, A., Sakamaki, K., Ide, H., and Tamura, K. (2005). Characteristics ofinitiation and early events for muscle development in the Xenopus limb bud.Dev. Dyn. 234, 846–857.
Schleiden, M.J. (1838). Beitrage zur Phytogenesis. In Muller’s Archiv furAnatomie, Physiologie und wissenschaftliche Medicin (Berlin: von Veit),pp. 137–176.
Schnabel, R., Hutter, H., Moerman, D., and Schnabel, H. (1997). Assessingnormal embryogenesis in Caenorhabditis elegans using a 4D microscope:variability of development and regional specification. Dev. Biol. 184, 234–265.
Schwann, T. (1839). Mikroskopische Untersuchungen uber die Ubereinstim-mung in der Struktur und dem Wachstume der Tiere und Pflanzen, G.E.Reimer, ed. (Berlin).
Serbedzija, G.N., Bronner-Fraser, M., and Fraser, S.E. (1989). A vital dye anal-ysis of the timing and pathways of avian trunk neural crest cell migration.Development 106, 809–816.
Shimomura, O., Johnson, F.H., and Saiga, Y. (1962). Extraction, purificationand properties of aequorin, a bioluminescent protein from the luminous hydro-medusan, Aequorea. J. Cell. Comp. Physiol. 59, 223–239.
Sinha, D.K., Neveu, P., Gagey, N., Aujard, I., Le Saux, T., Rampon, C., Gauron,C., Kawakami, K., Leucht, C., Bally-Cuif, L., et al. (2010). Photoactivation of theCreER T2 recombinase for conditional site-specific recombination with highspatiotemporal resolution. Zebrafish 7, 199–204.
Snippert, H.J., van der Flier, L.G., Sato, T., van Es, J.H., van den Born, M.,Kroon-Veenboer, C., Barker, N., Klein, A.M., van Rheenen, J., Simons, B.D.,and Clevers, H. (2010). Intestinal crypt homeostasis results from neutralcompetition between symmetrically dividing Lgr5 stem cells. Cell 143,134–144.
408 Developmental Cell 21, September 13, 2011 ª2011 Elsevier Inc.
Spemann, H., and Mangold, H. (1924). Uber Induktion von Embryonalanlagendurch Implantation artfremder Organisatoren. Wilhelm Roux Arch. Entw.Mech. Org. 100, 599–638.
Stark, D.A., and Kulesa, P.M. (2007). An in vivo comparison of photoactivat-able fluorescent proteins in an avian embryo model. Dev. Dyn. 236, 1583–1594.
Stent, G.S. (1985). The role of cell lineage in development. Philos. Trans. R.Soc. Lond. B Biol. Sci. 312, 3–19.
Stern, C. (1936). Somatic Crossing over and Segregation in DrosophilaMelanogaster. Genetics 21, 625–730.
Sturtevant, A.H. (1929). The claret mutant type of Drosophila simulans: Astudy of chromosome elimination and cel lineage. Z Wiss Zool Abt A 135,323–356.
Sulston, J.E., Schierenberg, E., White, J.G., and Thomson, J.N. (1983). Theembryonic cell lineage of the nematode Caenorhabditis elegans. Dev. Biol.100, 64–119.
Swoboda, P., Gal, S., Hohn, B., and Puchta, H. (1994). Intrachromosomalhomologous recombination in whole plants. EMBO J. 13, 484–489.
Swoger, J., Muzzopappa, M., Lopez-Schier, H., and Sharpe, J. (2011). 4Dretrospective lineage tracing using SPIM for zebrafish organogenesis studies.J Biophotonics 4, 122–134.
Tam, P.P., Parameswaran, M., Kinder, S.J., and Weinberger, R.P. (1997).The allocation of epiblast cells to the embryonic heart and other mesodermallineages: the role of ingression and tissue movement during gastrulation.Development 124, 1631–1642.
Turan, S., Galla, M., Ernst, E., Qiao, J., Voelkel, C., Schiedlmeier, B., Zehe,C., and Bode, J. (2011). Recombinase-mediated cassette exchange(RMCE): traditional concepts and current challenges. J. Mol. Biol. 407,193–221.
Tzouanacou, E., Wegener, A., Wymeersch, F.J., Wilson, V., and Nicolas, J.F.(2009). Redefining the progression of lineage segregations during mammalianembryogenesis by clonal analysis. Dev. Cell 17, 365–376.
Ueno, H., and Weissman, I.L. (2006). Clonal analysis of mouse developmentreveals a polyclonal origin for yolk sac blood islands. Dev. Cell 11, 519–533.
Vincent, J.P., and O’Farrell, P.H. (1992). The state of engrailed expressionis not clonally transmitted during early Drosophila development. Cell 68,923–931.
Vincent, C.A., Carpenter, R., and Coen, E.S. (1995). Cell lineage patterns andhomeotic gene activity during Antirrhinum flower development. Curr. Biol. 5,1449–1458.
Virchow, R. (1858). Die Cellularpathologie in ihrer Begrundung auf physiologi-sche und pathologische Gewebelehre, A. Hirschwald, ed. (Berlin).
Vogt, W. (1929). Gestaltungsanalyse am Amphibienkeim mit ortlicher Vitalfar-bung. II. Teil. Gastrulation une Mesodermbildung bei Anuren und Urodelen.Wilhelm Roux Arch. Entw. Mech. Org. 120, 384–706.
Voiculescu, O., Bertocchini, F., Wolpert, L., Keller, R.E., and Stern, C.D. (2007).The amniote primitive streak is defined by epithelial cell intercalation beforegastrulation. Nature 449, 1049–1052.
Waddington, C.H. (1932). Experiments on the development of chick and duckembryos. Philos. Trans. R. Soc. Lond. B Biol. Sci. 221, 179–230.
Walsh, C., and Cepko, C.L. (1992). Widespread dispersion of neuronal clonesacross functional regions of the cerebral cortex. Science 255, 434–440.
Wasserstrom, A., Adar, R., Shefer, G., Frumkin, D., Itzkovitz, S., Stern, T., Shur,I., Zangi, L., Kaplan, S., Harmelin, A., et al. (2008). Reconstruction of celllineage trees in mice. PLoS ONE 3, e1939.
Weisblat, D.A., Sawyer, R.T., and Stent, G.S. (1978). Cell lineage analysis byintracellular injection of a tracer enzyme. Science 202, 1295–1298.
Weston, J.A. (1963). A radioautographic analysis of the migration and localiza-tion of trunk neural crest cells in the chick. Dev. Biol. 6, 279–310.
Developmental Cell
Perspective
Wetzel, R. (1929). Untersuchungen am Huhnchen. Die Entwicklung des Keimswahrend der ersten beiden Bruttage. Wilhelm Roux Arch. Entw. Mech. Org.119, 188–321.
Whitman, C.O. (1887). A contribution to the history of germ layers in Clepsine.J. Morphol. 1, 105–182.
Wilson, E.B. (1892). The Cell-lineage of Nereis. A contribution to the cytogenyof the Annelid body. J. Morphol. VI, 368.
Xu, T., and Rubin, G.M. (1993). Analysis of genetic mosaics in developing andadult Drosophila tissues. Development 117, 1223–1237.
De
Yamanaka, S. (2009). A fresh look at iPS cells. Cell 137, 13–17.
Yu, H.H., Chen, C.H., Shi, L., Huang, Y., and Lee, T. (2009). Twin-spot MARCMto reveal the developmental origin and identity of neurons. Nat. Neurosci. 12,947–953.
Zinyk, D.L., Mercer, E.H., Harris, E., Anderson, D.J., and Joyner, A.L. (1998).Fate mapping of the mouse midbrain-hindbrain constriction using a site-specific recombination system. Curr. Biol. 8, 665–668.
Zong, H., Espinosa, J.S., Su, H.H., Muzumdar, M.D., and Luo, L. (2005).Mosaic analysis with double markers in mice. Cell 121, 479–492.
velopmental Cell 21, September 13, 2011 ª2011 Elsevier Inc. 409
