Germline precursors in most metazoans segregate from somatic
lineages specified during early stages of embryonic development
and are irreversibly committed to gamete production (Santos and
Lehmann, 2004). By contrast, studies in animals that can propagate
asexually suggest that germline precursors can also originate in
adults, from either pluripotent stem cells that can give rise to somatic
or germline fate, or de- and/or transdifferentiation of cells and tissues
(Blackstone and Jasker, 2003; Extavour et al., 2005; Nieuwkoop and
Sutasurya, 1979; Nieuwkoop and Sutasurya, 1981).
We are studying germline precursors in the colonial ascidian
Botryllus schlosseri. Ascidians are sessile marine invertebrate
chordates that develop from swimming tadpole larvae with
characteristic chordate features, and they are closely related to
vertebrates (Bourlat et al., 2006; Delsuc et al., 2006).
Embryogenesis in solitary ascidians is the textbook example of
determinative or mosaic development, with most cell specification
being due to the inheritance of cytoplasmic determinants (Conklin,
1905; Imai et al., 2006). Following a free-swimming phase, the
larvae settle and metamorphose into a sessile, filter-feeding form,
called an oozooid, wherein most chordate characteristics are lost. In
most species, this initial metamorphosis is followed by growth and
sexual maturity of the solitary individual. By contrast, colonial
ascidian species grow by asexual propagation, resulting in a colony
of genetically identical individuals (Swalla, 2006). This regenerative
ability appears to be due to the presence of self-renewing progenitors
(Laird et al., 2005; Sabbadin and Zaniolo, 1979), but the nature of
these cells (e.g. pluripotent or lineage restricted), and the
mechanisms by which they contribute to asexual reproduction are
To date, germline development has been studied in solitary
ascidians, such as Ciona intestinalis, and progenitors have been
found to segregate from somatic lineages early in development,
resulting in maternally specified germ cells that migrate into the
gonad rudiments during metamorphosis, and that later mature into
gametes (Fujimura and Takamura, 2000; Shirae-Kurabayashi et al.,
2006; Takamura et al., 2002; Yamamoto and Okada, 1999).
Embryos in colonial species are brooded and lineage-tracing studies
lag behind those in the more accessible free-spawning solitary
species. Nevertheless, determinative development is likely to be a
conserved characteristic in ascidians, and we hypothesize that
colonial species retain long-lived germline progenitors specified
B. schlosseri is a colonial ascidian with three co-existing
generations arranged spatially into a star-shaped group called a
system, containing both adults (zooids) and buds, all connected by
a common vasculature (Fig. 1). The center of each system is
occupied by the zooids, which are actively feeding and capable of
sexually reproducing. They are joined peripherally by ‘primary
buds’, which are completing their development of both somatic and
germline tissues. In turn, these are connected to ‘secondary buds’,
which are in the initial stages of development. The zooid has a
lifespan of only one week, after which it dies in a massive wave of
apoptosis called ‘takeover’ (Lauzon et al., 2002). Development is
coordinated throughout the colony, and during takeover the zooid
Early lineage specification of long-lived germline precursors
in the colonial ascidian Botryllus schlosseri
Federico D. Brown1,2,3,*,†, Stefano Tiozzo3,*, Michelle M. Roux3,*, Katherine Ishizuka3, Billie J. Swalla1,2and
Anthony W. De Tomaso3,‡
In many taxa, germline precursors segregate from somatic lineages during embryonic development and are irreversibly committed
to gametogenesis. However, in animals that can propagate asexually, germline precursors can originate in adults. Botryllus
schlosseri is a colonial ascidian that grows by asexual reproduction, and on a weekly basis regenerates all somatic and germline
tissues. Embryonic development in solitary ascidians is the classic example of determinative specification, and we are interested in
both the origins and the persistence of stem cells responsible for asexual development in colonial ascidians. In this study, we
characterized vasa as a putative marker of germline precursors. We found that maternally deposited vasa mRNA segregates early in
development to a posterior lineage of cells, suggesting that germline formation is determinative in colonial ascidians. In adults,
vasa expression was observed in the gonads, as well as in a population of mobile cells scattered throughout the open circulatory
system, consistent with previous transplantation/reconstitution results. vasa expression was dynamic during asexual development in
both fertile and infertile adults, and was also enriched in a population of stem cells. Germline precursors in juveniles could
contribute to gamete formation immediately upon transplantation into fertile adults, thus vasa expression is correlated with the
potential for gamete formation, which suggests that it is a marker for embryonically specified, long-lived germline progenitors.
Transient vasa knockdown did not have obvious effects on germline or somatic development in adult colonies, although it did result
in a profound heterochrony, suggesting that vasa might play a homeostatic role in asexual development.
KEY WORDS: Asexual reproduction, Botryllus schlosseri, Blastogenesis, Heterochrony, Regeneration, Budding, Colonial ascidian,
Coloniality, Germ cells, Germline, Stem cells, vasa
Development 136, 3485-3494 (2009) doi:10.1242/dev.037754
1Biology Department, Center for Developmental Biology, and Institute for Stem Cell
and Regenerative Medicine, University of Washington, Seattle, WA 98195, USA.
2Friday Harbor Laboratories, University of Washington, Friday Harbor, WA 98250,
USA. 3Molecular, Cellular and Developmental Biology, UC Santa Barbara, Santa
Barbara, CA 93106, USA.
*These authors contributed equally to this work
†Present address: Evolutionary Biology – Division 4, Max Planck Institute for
Developmental Biology, 72076 Tübingen, Germany
‡Author for correspondence (firstname.lastname@example.org)
Accepted 17 August 2009
bodies undergo apoptosis and are removed via phagocytic cells in
the blood, the primary bud migrates into the newly vacated region
of the colony, opening its siphons and becoming a zooid, the
secondary bud becomes the primary bud, and a new secondary bud
begins to develop. Thus, the life history of Botryllus consists of a
constant succession of individual zooids, each with a three-week
lifespan (Lauzon et al., 2002).
Asexual development (blastogenesis) takes 14 days under
laboratory conditions, and can be divided into seven distinct visual
stages (Fig. 1; stages A-1 through D) (Lauzon et al., 2002). A new
generation starts as a secondary bud, first visible as a thickening of
the peribranchial epithelium of a primary bud (Fig. 1; stage A1),
which evaginates and forms a closed vesicle (Fig. 1; stage A-2
through B-2). Next, a series of epithelial invaginations and
protrusions (Fig. 1; stage C-1) differentiate into somatic tissues and
organs (Fig. 1; stage C-2). After seven days (Fig. 1; stage D), the
secondary bud transitions to a primary bud and continues to develop.
At day 14, the siphons open and the primary bud becomes a filter-
feeding adult zooid. Each zooid can generate multiple buds each
week, so the colony will eventually expand asexually. While
interconnected, the zooids and buds develop independently, and
individuals can be separated from the colony without disturbing their
growth (i.e. subcloning), thus multiple experiments can be done on
a single genotype.
Following metamorphosis, colonies undergo at least 8-12
developmental cycles prior to the first appearance of gametes
(sexual maturity). In addition, populations show seasonal fertility,
and in the lab cycle in and out of reproductive (fertile) and non-
reproductive (infertile) states. However when the colony is fertile
development of the gametes is synchronized with somatic
development (Mukai, 1977; Mukai and Watanabe, 1976; Sabbadin
and Zaniolo, 1979). The first appearance of gonads occurs in the
secondary bud (stage B), when mobile progenitors in the blood
migrate to a region between the inner epithelium and the epidermis
and begin to proliferate. Concurrently, oocytes at various stages of
development also appear (Fig. 1). Over the next 10 days, the medial
region of the blastema will differentiate into the lobular testis, while
the lateral region will become the ovary (Sabbadin and Zaniolo,
1979). For the latter, one or several oocytes will become fixed on the
epithelia of the peribranchial chamber, and an oviduct will form
from the outer follicular layer. Upon transition to the adult zooid,
mature eggs will immediately ovulate into the peribranchial
chamber, be fertilized by exogenous sperm, and develop in situ.
Several hours to days later the testes will complete development, and
sperm will be released into the peribranchial chamber and flushed
into the water column, fertilizing neighboring colonies (Johnson and
Yund, 2004). The time lag between ovulation and sperm release
(protogyny) prevents self-fertilization.
Given this plasticity, it appears that germline precursors are
mobile, and can migrate to a niche in the secondary bud then expand
and differentiate into gametes. Moreover, at least for oocytes,
intermediate stages migrate between each successive asexual
generation until maturity, where they become fixed and contribute
to sexual reproduction in that generation (Izzard, 1968; Mukai,
1977; Sabbadin and Zaniolo, 1979). In summary, each asexual
generation probably initiates gametogenesis of both male and female
gametes, as well as harboring intermediate stages of oocytes.
The mobility of germline precursors has also been demonstrated
independently. B. schlosseri undergoes a natural transplantation
reaction that can result in vascular fusion, uniting the circulation of
two individuals. In these chimeras, both long-lived germline and
somatic chimerism has been observed, even months after surgical
Development 136 (20)
Fig. 1. Outline of asexual development (blastogenesis) in Botryllus schlosseri. Animals were staged according to Lauzon et al. (Lauzon et al.,
2002). Each stage (A-D) represents 1 day under laboratory conditions. Drawings represent dorsal views of zooids (green frame), primary buds
(yellow frame) and secondary buds (red frame). For simplification, each adult zooid carries only one set of buds. A secondary bud appears as a
thickening of the epidermis and the peribranchial chamber leaflet of the primary bud (stage A1), which evaginates into a closed vesicle (stage B2),
followed by organogenesis (stages C1-D). Gonadogenesis occurs in the secondary bud from mobile precursors (blue; stages B1-C2). During
takeover (stages C2-D), the secondary bud becomes the primary bud and a new blastogenic cycle begins for the next secondary bud. After the
second takeover event, the primary bud opens its siphons and becomes a functional adult (zooid). In fertile colonies (as illustrated here), the
hermaphroditic gonad fully matures on both sides of the zooid.
separation of the two individuals (Sabbadin and Zaniolo, 1979).
Moreover, when stem cells of two genotypes mix, the populations will
compete, and stem cells from one genotype will often replace the
germline and somatic tissues of another, in a process called ‘stem cell
parasitism’(Buss, 1982; Stoner and Weissman, 1996). Germline and
somatic precursors can be prospectively enriched in B. schlosseriand
show long-term reconstitution ability upon transplantation. In
addition, limiting dilution experiments show that transplantation of
single cells results in either somatic or germline chimerism, but not
both, which suggests that both germline-committed and somatic-
committed stem cell populations exist in adults (Laird et al., 2005).
Our long-term goal is to isolate and study these stem cells in order to
understand the cellular and molecular basis of asexual development
(regeneration) and stem cell parasitism.
In this study, we characterized the expression and function of
vasa as a putative lineage-specific marker for germline
progenitors, and tested for the presence of functional germline
progenitors in juveniles. vasa encodes an ATP-dependent RNA
helicase, and its expression is restricted to primordial germ cells
(PGCs) in most phyla studied to date (Sengoku et al., 2006).
However, in some metazoans, vasa has also been implicated in
aspects of somatic development (Dill and Seaver, 2008; Extavour
et al., 2005; Mochizuki et al., 2001). In Botryllus, our results
suggest that vasa labels long-lived, functional germline
progenitors specified during embryogenesis that are also found in
enriched stem cell populations, and these results correlate with
functional reconstitution following transplantation (Laird et al.,
2005). By contrast, transient siRNA-mediated knockdown of vasa
had no effect on germline formation during blastogenesis, but did
produce heterochronic shifts of somatic growth in the colony,
suggesting that vasa-positive (vasa+) cells play a role in somatic
regeneration as well.
MATERIALS AND METHODS
Botryllus schlosseri colonies were raised, staged, crossed and screened at
18-20°C according to Boyd et al. (Boyd et al., 1986).Embryos were isolated
and prepared for in situ hybridization as previously described (Brown and
Swalla, 2007). Following metamorphosis, colonies are considered juveniles
until the first appearance of gametes (sexual maturity). Following this time,
colonies with gametes are defined as fertile, those without are referred to as
cDNA was synthesized from embryonic and adult stages as described
(Tiozzo and De Tomaso, 2009). Quantitative PCR (qPCR) was carried out
(Tiozzo et al., 2008) with the following set of Botryllus specific vasa
primers: 5?-GGCGGATTTAGCGATGATGAG-3?and 5?-TTCCCCCATA -
GC GACTGTTAGAC-3?. Analysis of q-PCR was performed using 2-ΔΔCt
according to Livak and Schmittgen (Livak and Schmittgen, 2001). Fold
change of mRNA was normalized to either actin (ActinF, 5?-CTA -
TACGCTTCCGGCAGAAC-3?; ActinR, 5?-CAAGAGCGACATAGC -
ACAGC-3?) or elongation factor 1 alpha (EFF, 5?-CGTGGTC ATT -
GGCCACGTAGATTCCGGAAA-3?; EFR, 5?-ATGAAATCACGA TGA -
CCGGGAGCGTCGATG-3?) as specified. Individual experiments were
then normalized to experiment-specific reference mRNA quantity (see
Results). Each experiment was repeated at least three times from two
different genotypes of every stage analyzed, with samples run in triplicate.
Error bars were calculated using s.d. and statistical significance using paired
Cell staining and sorting
Botryllus cells were labeled for sorting as outlined in Laird et al. (Laird et
al., 2005). Following incubation, cells were washed, resuspended in
Botryllusbuffer containing 25 mM Verapamil and 1 μg/ml propidium iodine
(Molecular Probes) to assay viability. FACS was carried out using a
FACSAria cell sorter (BD Biosciences, San Jose, CA, USA). Samples were
compensated prior to sorting and dead cells were excluded by gating-out
high propidium iodine signals. Analysis was performed using FlowJo
software (Tree Star, San Carlos, CA, USA). Sorted cells were stained with
0.05% Trypan Blue to verify quantity and viability. SSClowand SSChighgates
were pooled for qPCR analysis.
In situ hybridization and histological analysis
In situ hybridization (ISH)was performed as described (Brown and Swalla,
2007). The sequence used for probes is shown in Fig. S1 in the
supplementary material. Prehybridization proteinase K concentrations used
were: 500 μg/ml for oozoid and adult whole mounts, 100 μg/ml for tadpole
embryos and larvae, and 1 μg/ml for sections. Results are from at least three
independent whole-mount ISH replicates, and at least five different colonies
were sectioned for independent ISHreplicates. Positive cells identified with
either sense or antisense probes were compared in five independent ISH
replicates (see Fig. S2 in the supplementary material) (Brown and Swalla
(2007). Sectioning and image capture were performed as described
previously (Tiozzo and De Tomaso, 2009).
siRNAs to vasawere generated as described (Tiozzo and De Tomaso, 2009;
Tiozzo et al., 2008). Sequences targeted for siRNA are shown in Fig. S1 in
the supplementary material. Samples were observed and collected for
histology and RNA isolation, and results were correlated with vasa
expression levels. Seven independent experimental replicates were done
using different genotypes. For each experiment, a colony was separated into
two subclones and vasa siRNA or control (GFP) siRNA treatment was
initiated at the same stage in the blastogenic cycle. In all experiments, the
level of vasa transcripts declined in the first 3 days and remained low during
the treatment, as demonstrated by qPCR and RT-PCR (see Fig. S3 in the
Juvenile/adult transplantation and analysis of germline
Genetically distinguishable genotypes were crossed, and progeny were fused
into naïve subclones of the parents. We used six different genotypes in three
independent crosses. At each time point, multiple testes were dissected from
the colony and DNA isolated (Laird et al., 2005). Testes were tested for the
presence of three PCR-based genetic markers (e18sp6,L19sscpandH9hdx),
as previously described (De Tomaso and Weissman, 2003; De Tomaso and
Weissman, 2004). Results were equivalent for each marker.
Spatiotemporal expression of vasa during
embryogenesis, metamorphosis and in the adult
A B. schlosseri vasa cDNA clone was isolated, sequenced and
annotated (see Fig. S1 in the supplementary material; GenBank
Accession number FJ890989). Probes were designed to unique
regions of the gene and expression analyzed in eggs and during
embryogenesis (Fig. 2A-O). vasa mRNA concentrates in granular
structures of the fertilized B. schlosseri egg cortex (Fig. 2A).
During the first cleavage, vasa mRNA localized to the vegetal
pole of the embryo (Fig. 2B); after the third cleavage, vasa
segregated in the cleavage furrow of the posterior-most B4.1 pair
of cells of the eight-cell embryo (Fig. 2C), and in the descendant
B5.1 pair of cells of the 16-cell embryo (Fig. 2D). During
gastrulation (Fig. 2E), vasa mRNA was expressed in a pair of cells
found posterior to the blastopore that are likely to correspond to
the precursor germline B7.6 pair (Fig. 2F) (Shirae-Kurabayashi
et al., 2006).
vasalocalization was then characterized during metamorphosis,
in oozooids, and in zooids derived after several rounds of asexual
development, but prior to sexual maturity. During initiation of
vasa in Botryllus schlosseri
metamorphosis, zygotic expression of single vasa+ cells was
scattered throughout the larval head (Fig. 2G). Immediately
following metamorphosis, vasa+ cells were present in the mantle
of the oozooid and within the peripheral vasculature of the tunic
(Fig. 2H). In fertile colonies, vasa expression was observed in
oocytes labeled with an antisense probe (Fig. 2I), but not in those
labeled with a sense control (Fig. 2J). Histological analysis (Fig.
2L-O) showed vasa+ cells in the periphery of the lobules in
developing testes (Fig. 2M), in single hemocytes in ampullae,
throughout the peripheral vasculature (Fig. 2N), and in a region of
mesenchyme surrounding the sinuses and lacunae of the zooid (Fig.
ISH in adult colonies is difficult due to the presence of the tunic,
which shows medium to high non-specific background, but was
necessary to comprehensively characterize vasa localization.
Background was normalized as described (Brown and Swalla,
2007), and results were verified by qPCR. Expression was
quantified in tadpoles, in post-metamorphosis oozooids and in first-
generation asexually derived zooids. Tadpoles contained the lowest
levels ofvasa mRNA expression,oozooids showed a slightly higher
expression, and zooids showed ≥threefold higher levels, regardless
of developmental stage (Fig. 2K).
Additionally, we confirmed that vasa is expressed in the
extracorporeal circulation as well as within the zooid bodies (Fig.
3D). Colonies were dissected and vasa expression was quantified
and compared between the extracorporeal vasculature and the
zooid and bud tissues in fertile colonies (Fig. 3D). vasa+ cells
were observed in the circulation both inside and outside of the
zooid body; however, there were differences in the number of
vasa+ cells in ampullae and in the zooid tissues between fertile
and infertile colonies, most likely due to vasa expression in the
Development 136 (20)
Fig. 2. vasa expression during embryogenesis, metamorphosis and in adult colonies as shown by ISH. (A)Fertilized egg with polar bodies
in the animal pole and granule-like aggregation of maternal vasa (arrowheads) in the cortex. (B)A two-cell-stage embryo with vasa mRNA
(arrowheads) aggregated on one side of the cleavage furrow. (C)An eight-cell-stage embryo with two proximate aggregates of vasa (arrowheads)
at the posterior cortical region of the B4.1 blastomeres. (D)Sixteen-cell-stage embryo with vasa mRNAs (arrowheads) at the posteriormost cortical
region of the B5.1 blastomeres. (E)Gastrula- and neural-plate-stage embryos (past the 110-cell stage) show invagination and blastopore formation
on the vegetal side of the embryo. (F)Gastrula- and neural-plate-stage embryos with a single aggregate of vasa (arrowheads) in the posterior region
of the embryos. (G)Detail of a larval head during metamorphosis. Anterior adhesive papillae (p) can be observed on the right; vasa+ cells
(arrowheads) are scattered throughout the head (h). (H)vasa expression in a newly settled colony is seen in individual cells (arrowheads) in the
oozooid (oz), and within the extracorporeal vasculature. (I)In a sexually mature adult, vasa expression is seen in small oocytes (o) at the site of the
gonads of the zooid (z) and in individual cells (arrowheads) within the ampullae (a). (J)A sexually fertile adult shows general background levels of
autofluorescence in blood and tunic cells, as shown by the sense probe negative control (cf. H and I; antisense probe). (K)Quantitative PCR analysis
of vasa mRNA levels during metamorphosis. Larvae show the lowest mRNA levels, oozooids show a slight increase, and first-generation asexually
derived zooids show a further increase after ten days of settlement. (L)Illustration of a colony (dorsal view) showing location of the testis, ampullae
and zooid body (framed), corresponding to histological sections in M-O. (M)Fluorescence in situ hybridization shows two vasa expressing cells (red;
arrowheads) at the periphery of a mature testis containing highly packed nuclei of sperm precursors (blue). (N)vasa expressing cells (red,
arrowheads) in the ampullae. vasa+ cells are scattered throughout the vasculature and are surrounded by other hemocytes (blue, nuclei
counterstained with DAPI). (O)Cells expressing vasa (red, arrowheads) are also seen in the sinuses and lacunae surrounding the epithelial folds of
the stomach (s) in an adult zooid of the colony. a, ampullae; h, head; o, oocyte; oz, oozoid; p, papillae; s, stomach; t, testis; tu, tunic; z, zooid. Scale
Quantitative analysis of vasa expression during
Asexual somatic development occurs independently of
gametogenesis, and adults will often fluctuate between fertile and
infertile states in both laboratory-reared and natural populations.
vasa expression was quantified and compared between fertile and
infertile adult colonies across one complete cycle of asexual
development (Fig. 3A-C). In fertile colonies, vasa expression was
highest in stage A1 and fluctuated throughout the asexual cycle (Fig.
3A). Statistically significant changes (*) occurred at the transition
from A1 to A2 (P=0.002), B1 to B2 (P=0.007) and D to A1
(P=0.04). Variability was highest in stage C1, the point in the cycle
when germline formation is being initiated in secondary buds (Fig.
1, Fig. 3A). In infertile colonies (Fig. 3B), vasa expression was
highest at stage A1 and fluctuated across successive stages of
asexual development,with statistically significant changes detected
between A1 and A2 (P=0.011), and D and A1 (P=0.0005). Fertile
colonies were then compared with infertile colonies (Fig. 3C). In all
stages, fertile colonies had higher levels of vasa expression (Fig.
3C). Changes in vasaexpression most likely reflect cell proliferation
owing to an increase in colony size following takeover in both fertile
and infertile adults, as well as during gamete development in fertile
vasa enrichment in BAAA+ stem cell populations
Previous studies have enriched self-renewing germline and somatic
progenitors 10-fold versus total cells based on ALDH activity using
the reagent BAAA. We tested whether vasaexpression was enriched
in BAAA+ versus total cells from both fertile and infertile colonies
by qPCR (Fig. 4). Botryllus cells were analyzed by FACS,
comparing the unstained population (not treated with BAAA; Fig.
4A) with BAAA+ populations consisting of both side scatter (SSC)
high and low populations (Fig. 4B). Both the BAAA+/SSClowand
the BAAA+/SSChighpopulation were sorted, pooled and used for
qPCR analysis of vasa gene expression. vasa expression was
normalized first to the Elongation factor 1 alpha (EF1α)expression
level, then to pooled cDNA consisting of total mRNA from each
stage (Stage Mix), to determine average vasa expression during
blastogenesis for both states of fertility (dashed threshold line). In
BAAA+ cells from infertile colonies, vasa was highly enriched in
vasa in Botryllus schlosseri
Fig. 3. qPCR analysis of vasa expression levels. vasa
expression levels were analyzed during asexual development
(A-C) and in isolated tissues (D). (A)Fertile colonies (testes,
oocytes and embryos present). Values were normalized as
described in the Materials and methods. Resulting
differences in gene expression are given in fold change with
respect to the threshold (stage A1 normalized to self,
dashed line). (B)Infertile colonies. Values were normalized
and presented as in A. Asterisks indicate statistical
significance (P-value displayed) with respect to the
preceding developmental stage (see Results).
(C)Comparative analysis of vasa expression in fertile versus
infertile colonies. (D)vasa expression compared between
the peripheral vasculature and zooid bodies in both fertile
and infertile colonies.
Fig. 4. vasa gene expression in BAAA+ cells from
fertile and infertile colonies. FACS plots showing
Botryllus cell populations based on Side Scatter (SSC-A,
y-axis) and FITC-A emission (BAAA+, x-axis). (A)No
treatment control showing autoflurescence only.
(B)Cells treated with BAAA+ and PI for viability. Gates
indicate populations sorted for qPCR analysis:
BAAA+/SSC high and low (blue boxes) with total
percentage indicated (?100). (C,D)qPCR analysis of
vasa gene expression in BAAA+ populations from (C)
fertile and (D) infertile colonies. Asterisk indicates
statistically significant differences.
comparison with the average expression (P=0.006; Fig. 4D). By
contrast, BAAA+ cells from fertile colonies showed lower than
averagevasaexpression (Fig. 4C). Results from experiments shown
in Figs 2-4 suggest that the increased vasa expression in fertile
colonies is likely to be due to differentiating gametes,which are not
found in the BAAA+ pool, indicating that at least a proportion of the
vasa+ cells in the BAAA+ population are long-lived progenitors.
Rapid and long-term germline chimerism induced
by juvenile progenitors
The identification of vasa+ cells during embryogenesis and in
juvenile colonies suggested that germline progenitors are specified
early in development. We independently tested for the presence of
functional progenitors by assessing germline chimerism following
the natural transplantation of cells between juveniles and fertile
adults. We crossed two genetically distinct colonies and fused F1
oozooids back into naïve subclones of the parents (Table 1, Fig. 5).
Three to four weeks following transplantation, we tested for the
presence of donor markers in the testes of the recipient. As germline
formation is initiated in the secondary bud at stage B1 (Fig. 1), this
time point is the earliest in which chimerism in the testes could be
detected. Eggs were not tested because they are surrounded by a
somatic follicular layer, and because the time to maturation is not
well defined (Sabbadin and Zaniolo, 1979).
When fertile, each zooid had one or two testes dissected and
motile sperm isolated, free of somatic contamination (Laird et al.,
2005; Stoner et al., 1999). Testes from individual zooids in the
recipient were isolated (6-9 testes/experiment) and chimerism was
assessed using several previously characterized genetic markers (De
Tomaso and Weissman, 2003). In each cross, we observed that one
or two testes from a group of 6-9 contained donor markers, whereas
the rest were derived from the recipient (Table 1, Fig. 5). Control
progeny from the same crosses did not become sexually mature until
10-12 weeks following metamorphosis, which is the normal time to
sexual maturity under laboratory conditions.
On average, a B. schlosserigenotype has a lifespan of 3-9 months,
whereas some laboratory-reared strains can live to be over two years
(De Tomaso et al., 1998). To test whether these germline progenitors
were long lived, we re-analyzed subclones of the same chimeric
individuals at later time points. In all cases, chimerism was detected
in the germline (Table 1, Fig. 5). In summary, sexually immature
juveniles contain long-lived functional germline progenitors that can
contribute to gamete formation immediately upon transplantation
into a fertile colony, and that show long-term self-renewal.
Functional analysis of vasa during blastogenesis
Functional analysis of vasa was performed using siRNA-mediated
genetic knockdown. It takes about 3 days to knockdown >95% of
the mRNA for multiple genes (see Fig. S3 in the supplementary
material), therefore in the first set of experiments siRNA treatment
of fertile colonies began in stage C2,so that new buds would initiate
and develop under near complete knockdown conditions, allowing
us to follow an asexual generation throughout the life cycle.
In control experiments, subclones were unaffected, and takeover
and the transition of asexual generations occurred normally (Fig.
6A-D). Experimental colonies also underwent takeover normally
(Fig. 6F);however, vasaknockdown severely affected the timing of
this process, extending it from 24 hours to around 4 days.
Histological analysis showed that despite this delay, there were no
abnormalities in the process (see Fig. S4 in the supplementary
material). This delay can be easily visualized: adult zooids shrink
and the primary buds grow much slower (compare Fig. 6G with 6C):
However, the buds developed normally (Fig. 6H; see also Fig. S5 in
the supplementary material). These experiments were repeated four
times on different genotypes with no deviation.
vasa knockdown had no effect on the development of the
germline in fertile colonies. Eggs and testes of the primary buds
continued normal development (Fig. S6 in the supplementary
material), and, in secondary buds, oocyte and gonad cell recruitment
and blastema formation occurred similar to in untreated colonies
(see Fig. S6E,F in the supplementary material). The adult gonads
(egg and testis) presented no obvious defects. However, the germline
developed in context of the delayed resorption during takeover and
was delayed by four days (see Fig. S6A-D in the supplementary
To further dissect the effect of vasa on the timing of blastogenesis,
siRNA treatment was initiated at stage B2 or C1 and continued for
two consecutive blastogenic cycles. vasa knockdown caused
colonies to undergo an early takeover event, which began 24 hours
following the initiation of siRNA treatment (about 24-48 hours
before in the control). The phenotype was equivalent to that in the
previous experiment: takeover was delayed, with resorbtion of the
zooids and growth of the primary buds, as well as their transition to
Development 136 (20)
Table 1. Genetic analysis of chimeric testes
DonorRecipient Time (weeks)Number testes sampledNumber testes donorNumber testes recipient
Donor and recipient (D and R in Fig. 5) genotypes were crossed and a single F1 individual was fused to a naïve subclone of the recipient parent. Three to four weeks later,
individual testes were removed from the recipient and tested for the presence of donor markers.
Fig. 5. Genetic analysis of chimeric testes. F1 chimerism results from
genotypes 1025f and 1025c (see also Table 1). 1025f was homozygous
for a dominant marker (e18sp6, see Materials and methods). Donor
and recipient genotypes were crossed and a single F1 individual was
fused to a naïve subclone of the recipient parent. Four weeks later,
eight testes were isolated from individual zooids throughout the
subclone and tested for the presence of the donor marker. R, recipient
(1025c); D, donor (1025f); 1-8 are individual testes from the donor
subclone, 4 weeks following fusion, two show the presence of donor
markers. Independent markers gave equivalent results.
an adult delayed by about 48 hours (Fig. 7). Interestingly, forcing the
colonies into early takeover via vasaknockdown coupled to a slower
primary bud to zooid transition caused the transition to occur
simultaneously in control and experimental subclones, and the two
were re-synchronized at the same stage of development five days
later (stage A; Fig. 7E,J). However, once vasa siRNA-treated
subclones had completed this first takeover process (day 5; Fig. 7J),
they immediately entered another takeover phase, around 6-7 days
before the control (Fig. 7K). The zooids began to shrivel with the
siphons closing, but in this case the takeover was further delayed,
and the process took about 5-6 days (the entire cycle) to be
completed (Fig. 7K-O). Despite this further delay, the process was
again completed and the second generation (i.e. the secondary buds
that were developing when siRNA treatment began)became adults,
about 10 days from the start of the experiment, with no apparent
defects (Fig. 7O). At this point, the colony immediately went into
takeover for a third time, but rapidly became disorganized and died
within two days. This was probably due to the fact that the early
takeover phenotype caused the siphons to remain closed, preventing
feeding for about two weeks. A schematic is shown in Fig. 7P. The
same phenotype was observed regardless of the developmental stage
of the colony at the onset of the siRNA treatment (day 0). Thus, it
appears that knockdown of vasa causes the adult generation to
undergo takeover within 24 hours, desynchronizing development
between the zooids and buds. However, development of the primary
and the secondary buds was not delayed, and the colonies re-
synchronized at stage A, in step with the control subclones (Fig.
We next tested the reversibility of the observed phenotypes, as
well as the potential effects of knockdown on succeeding
blastogenic generations. vasa knockdown was initiated at stage B
and the colony underwent early takeover (Fig. 8A,B). Six days later,
vasa siRNA delivery was stopped and the colonies were allowed to
continue development (Fig. 8C). Seven days later, vasaknockdown
colonies had re-established normal coordination between takeover
of the adult zooid and development of the bud, concurrent with vasa
re-expression (Fig. 8D).
In summary, acute vasa knockdown in adults had no effect on
germline development, but caused the colony to go into early
takeover, and in addition the takeover event itself was delayed. The
period of delay was not constant, but was proportional to when
siRNA treatment began. By contrast, development of both primary
and secondary buds was not affected and occurred with normal
timing. Although we do not have the ability to carry out rescue with
transgenic expression, experiments in which siRNA treatment was
stopped revealed no long-term effects of knockdown on somatic or
germline development. Thus, vasa appears to be functionally
involved in the timing of takeover.
B. schlosseri is a basal chordate with the ability to regenerate all
organs and tissues on a weekly basis, but the origins and persistence
of the progenitors responsible are not understood. Previous studies had
revealed three properties of germline progenitors in adults. First,
Sabbadin and Zaniolo demonstrated that progenitors are mobile, and
can naturally transplant between parabiosed individuals resulting in
long-term germline chimerism and parasitism (Sabbadin and Zaniolo,
1979). Second, adults in both natural and laboratory-reared
populations cycle between fertile and infertile states, demonstrating
that germline progenitors that can be quiescent exist within an
individual (Boyd et al., 1986; Sabbadin, 1953). Finally, prospective
isolation studies suggested that both germline and somatic progenitors
could be enriched about 10 times based on ALDH activity (Laird et
al., 2005). These same studies also suggested that distinct germline
and somatic lineages exist. First, single cell transplants from ALDHhi
(BAAA+) populations resulted in germline or somatic chimerism,
vasa in Botryllus schlosseri
Fig. 6. Effects of vasa knockdown on asexual
development. (A-D)Development of control colony,
treated every day (d0-5) with GFP siRNA (d0; stage C).
(E-H)Development of knockdown colony over a 5-day
period. Development is delayed by 2 days compared with
the control colony (compare G and B; H and C).
(I,J)Schematic representation of zooid development in
control and treated colonies: vasa knockdown causes a
desynchronization in development between adult and
buds, as illustrated on the black timeline. Asterisks
indicate when the treated samples lost synchrony and the
zooids underwent an early takeover. cz, contracted zooid;
nz, new zooid; pb, primary bud; rz, regressing zooid;
z, zooid. Scale bar: 1 mm.
never both (Laird et al., 2005). In addition, following natural
transplantation, parasitism of the germline is repeatable, hierarchical
and appears to be genetically controlled. By contrast, somatic
chimerism is random and does not correlate with germline results. In
fact, following the fusion of two individuals, the resulting chimera is
often composed of soma from one genotype, and germline from the
other (Sabbadin and Zaniolo, 1979; Stoner et al., 1999), thus germline
and somatic development seem to occur from independent
populations of cells. The results presented here show that vasa
expression is strongly correlated with germline precursors.
Early germline specification of long-lived
The constant development and turnover of zooids in colonial
ascidians, coupled to the determinative development characteristic
of the Tunicata, suggest that long-lived progenitors are specified
during embryogenesis that contribute to development throughout
life, but the nature of these cells (pluripotent or lineage-restricted) is
not well understood. Our spatiotemporal localization of vasa in B.
schlosseri embryos revealed expression patterns nearly equivalent
to those found in solitary ascidians (Brown and Swalla, 2007;
Shirae-Kurabayashi et al., 2006; Tanaka et al., 2000), with
maternally deposited vasa segregating into a pair of posterior
blastomeres, suggesting an early specification of germline-restricted
progenitors. At later time points, vasa+ cells are seen scattered in
the head of the tadpole, a region equivalent to the circulatory
distribution in the adult body plan following metamorphosis.
Independently, we found that cells in newly metamorphosed
individuals are competent to reconstitute the germline following
transplantation into fertile colonies, and that this chimerism is
maintained for the life of the adult recipient (up to 112 weeks in one
pairing). In both fertile and infertile adults, vasais expressed in cells
in the extracorporeal vasculature, as well as near lacunae
surrounding the zooid body. Circulatory vasa+ cells have been
observed in all botryllid species examined to date (Brown and
Swalla, 2007; Rosner et al., 2009; Sunanaga et al., 2006; Sunanaga
et al., 2008). vasa expression is also enriched in BAAA+ cells,
which can reconstitute the germline (Laird et al., 2005).
Finally, in recent studies, we have analyzed the contribution of
individual genotypes to germline and somatic tissues in chimeras
made from juvenile individuals immediately following
metamorphosis. We found that when these chimeras reached sexual
maturity, patterns of reconstitution matched those of chimeras made
from adults, and included complete germline parasitism and a lack
of correlation between somatic and germline outcomes. Thus, even
if chimeras were made months prior to sexual maturity, germline
progenitors acted independently of somatic progenitors (M. C.
Carpenter and A.W.D., unpublished).
Therefore, our data support the hypothesis that self-renewing,
lineage-restricted germline progenitors are specified during
embryogenesis that become functional immediately following
metamorphosis and contribute to gametogenesis throughout the life
Development 136 (20)
Fig. 7. vasa knockdown during two consecutive blastogenic cycles. (A-E)Development of control colony treated every day (d0-11) with GFP
siRNA (d0; stage B). (F-O)Development of knockdown colony over a two week period. Control colony develops and initiates takeover (stage D)
normally (D; day 4), and the process is completed in about 24 hours (E, day 5). By contrast, the vasa knockdown colony goes into takeover within
24 hours (G), and disrupted development continues over the next 3 days (H,I), but terminates in a transient recovery of developmental synchrony
(compare J with E). Although the control colony continues development, the vasa knockdown colony immediately goes back into takeover (K),
which lasted the entire asexual cycle, but terminated development synchronously for a second time (compare O with E) in early stage A.
(P)Schematic of blastogenic development during vasa knockdown showing the shifts in synchrony (shifts in black timeline) that occur between the
zooids and buds. The asterisk indicates when the treated samples lost synchrony and the zooids underwent an early takeover, but synchrony was
briefly reestablished after both takeover events. cz, contracted zooid; nz, new zooid; pb, primary bud; rz, regressing zooid; z, zooid. Scale bar:
of the colony. This suggests that fertility is based on developmental
cues that cause these progenitors to move from the circulation to
regions within the secondary bud and initiate germline formation.
vasa expression and enrichment of germline
We compared the enrichment of vasain BAAA+ cells isolated from
fertile and infertile adults to independently correlate vasaexpression
with long-lived progenitors. We found dramatic changes between the
two. In infertile colonies, vasa expression was highly enriched in
BAAA+ cells. By contrast, fertile colonies had lower vasaexpression
in BAAA+ versus total cells. This suggests that vasa expression is
upregulated in cells of the developing gonads that are not part of the
BAAA+ population. The BAAA+ populations from both fertile and
infertile adults contain an enriched pool of stem cells (Laird et al.,
2005), so our results provide strong evidence that a proportion of
vasa+ cells in the BAAA+ population are germline progenitors.
However, these data do not rule out the existence ofvasa-negative
(vasa–)germline progenitors, as studies in two related species have
revealed that surgical ablation of regions containing vasa+ cells did
not affect germline development, and that new vasa+ cells appeared
(Takamura et al., 2002; Sunanaga et at., 2006). These reports suggest
that vasa– progenitors exist that become vasa+ at later stages of
development. This would be analogous to the situation in mice,
where PGCs are specified at day E6, but vasa expression does not
turn on until day E9, when PGCs are migrating to the genital ridge
(Lacham-Kaplam, 2004; Tanaka et al., 2000). However, in B.
schlosseri, vasa+ cells are present at all times, as are functional
Previous studies suggested that vasa+ somatic cells exist in
ascidians. During Ciona embryogenesis, high vasa expression is
detected in the head; however, the fate of these cells following
metamorphosis is unknown, as in the adult vasa expression is
restricted to the gonads (Takamura et al., 2002). Although vasa+
circulating somatic stem cells could exist in B. schlosseri, the only
tissue outside of these circulatory cells in which vasaexpression was
seen is the gonad. In summary, our study cannot discount the
possibility of vasa– germline progenitors nor of vasa+ somatic
and/or pluripotent progenitors. However, the lack of congruity
between somatic and germline chimerism following transplantation
(either natural or experimental) suggests that a germline-committed
population exists, and the distribution of vasa+ cells correlates
exactly with previous limiting-dilution studies (Laird et al., 2005).
Effects of vasa knockdown in asexual
Transient knockdown of vasaresulted in no obvious morphological
defects in development of the germline but showed an effect on the
synchrony of asexual development in the colony. One possible
explanation for the absence of specific germline phenotypes is that
only short-term knockdowns (2-3 weeks) were possible before the
colonies degenerated. Testes require at least two weeks to develop,
and oocytes can mature over several generations, therefore our
treatment might not be long enough to assess the potential role of
vasa in germline development. Conversely, vasa knockdown in
different species shows disparate phenotypes, ranging from no
defects [zebrafish (Braat et al., 2001) and flatworms (Pfister et al.,
2008)] to sex-specific defects on male gamete formation, but not
female [mice (Tanaka et al., 2000)], to defects in migration but not
in the identity of the germ cell precursors [medaka (Li et al., 2009)].
Thus,it would not be surprising if vasawas shown not to be required
for germline formation; however, its expression can still be used to
identify precursors, as has been shown in other animals.
Given that vasa+ cells were scattered throughout the circulation,
the global knockdown phenotype was unexpected. However, a
somatic role for vasa would not be surprising. vasa is expressed in
somatic cells and tissues of embryos and adults in species of different
phyla, such as in the head of Ciona ascidian larvae (Fujimura and
Takamura, 2000; Shirae-Kurabayashi et al., 2006), in the somatic
stem cells of planarians and cnidarians(Shibata et al., 1999; Extavour
et al., 2005; Rebscher et al., 2007), andin the embryonic mesodermal
posterior growth zone and multiple somatic tissues of polychaetes
(Rebscher et al., 2007; Dill and Seaver, 2008). Although ultimately
there was no affect on somatic development, we speculate that the
colony might regulate the number of progenitors as it asexually
expands, and that vasa knockdown affects homeostasis. In previous
experiments, it was found that the zooids appear to be monitoring
growth of the buds, and,if delayed,they will undergo early takeover,
presumably to transfer resources to the next generation (Tiozzo et al.,
2008; Lauzon et al., 2002), so vasa+ cells might be involved in this
process. This might also explain the dynamic expression of vasa in
both fertile and infertile colonies, which in general resembles
previous results that characterize telomerase activity during the
budding cycle (Laird and Weissman, 2004).
In conclusion, given the experimental considerations mentioned
above, we cannot yet rule out any germline defects. However, we
did consistently observe a novel and robust phenotype of vasa
knockdown: the breakdown of synchrony between zooid and buds,
but the role vasa plays in these processes is not clear. Whether vasa
is directly involved in regulating short-
communication between cells and tissues (autocrine or endocrine)
of the colony to trigger takeover, or whether vasa+ cell-specific
defects, e.g. migration or determination defects, indirectly affect
takeover remain open questions for future studies.
We would like to thank Karla Palmeri, Tanya McKitrick and Randy Will for their
help and assistance. We thank Dr A. Sabbadin and Dr P. Burighel for valuable
discussions for this manuscript. F.D.B. was supported by a predoctoral grant
vasa in Botryllus schlosseri
Fig. 8. Release of vasa siRNA treatment has no effect on succeeding asexual generations. (A)vasa expression analyzed by RT-PCR at
different timepoints in control and knockdown colonies; siRNA treatment was discontinued at day 6. (B)Colony at the onset of the vasa siRNA
treatment; stage B of blastogenesis; (C) on day 4, zooids are in takeover (stage D), showing the typical vasa knockdown phenotype; (D) on day 13,
zooids contain palleal buds that are synchronized with the zooids, showing that release of vasa siRNA treatment results in the re-synchronization of
asexual development. nz, new zooid; pb, primary bud; rz, regressing zooid; z, zooid. Scale bars: 200μm.
3494 Download full-text
from the American Heart Association, by the Lerner Gray Fund for Marine
Research from the American Museum of Natural History, and by the University
of Washington Biology Department. S.T. was supported by a Stanford Dean’s
Fellowship. M.M.R. was supported by the NIH, individual NRSA
(1F32GM086018-01). Part of this research was funded by a University of
Washington Royalty Research Grant to B.J.S. This work was supported by the
NIH (RO1A104588/R01DK405762) and the Ellison Medical Foundation to
A.W.D. Deposited in PMC for release after 12 months.
Supplementary material for this article is available at
Blackstone, N. W. and Jasker, B. D. (2003). Phylogenetic considerations of
clonality, coloniality, and mode of germline development in animals. J. Exp.
Zoolog. B Mol. Dev. Evol. 297, 35-47.
Braat, A. K., van de Water, S., Korving, J. and Zivkovic, D. (2001). A zebrafish
vasa morphant abolishes vasa protein but does not affect establishment of the
germline. Genesis 30, 183-185.
Bourlat, S. J., Juliusdottir, T., Lowe, C. J., Freeman, R., Aronowicz, J.,
Kirschner, M., Lander, E. S., Thorndyke, M., Nakano, H., Kohn, A. B. et al.
(2006). Deuterostome phylogeny reveals monophyletic chordates and the new
phylum Xenoturbellida. Nature 444, 85-88.
Boyd, H. C., Brown, S. K., Harp, I. L. and Weissman, I. L. (1986). Growth and
sexual maturation of laboratory-cultured Monterey Botryllus schlosseri. Biol. Bull.
Brown, F. D. and Swalla, B. J. (2007). Vasa expression in a colonial ascidian,
Botrylloides violaceus. Evol. Dev. 9, 165-177.
Buss, L. W. (1982). Somatic cell parasitism and the evolution of somatic tissue
compatibility. Proc. Natl. Acad. Sci. USA 79, 5337-5341.
Conklin, E. (1905). Mosaic development in ascidian eggs. J. Exp. Zool. 2, 145-
De Tomaso, A. W. and Weissman, I. L. (2003). Initial characterization of a
protochordate histocompatibility locus. Immunogenetics 55, 480-490.
De Tomaso, A. W. and Weissman, I. L. (2004). Evolution of a protochordate
allorecognition locus. Science 303, 977.
De Tomaso, A. W., Saito, Y., Ishizuka, K. I., Palmeri, K. K. and Weissman, I. L.
(1998). Mapping the genome of a model urochordate. I. A low resolution
genetic map encompassing the Fu/HC locus of Botryllus schlosseri. Genetics 149,
Delsuc, F., Brinkmann, H., Chourrout, D. and Philippe, H. (2006). Tunicates
and not cephalochordates are the closest living relatives of vertebrates. Nature
Dill, K. K. and Seaver, E. C. (2008). Vasa and nanos are coexpressed in somatic
and germ line tissue from early embryonic cleavage stages through adulthood in
the polychaete Capitella sp. I. Dev. Genes Evol. 218, 453-463.
Extavour, C. G., Pang, K., Matus, D. Q. and Martindale, M. Q. (2005). vasa
and nanos expression patterns in a sea anemone and the evolution of bilaterian
germ cell specification mechanisms. Evol. Dev. 7, 201-215.
Fujimura, M. and Takamura, K. (2000). Characterization of an ascidian DEAD-
box gene, Ci-DEAD1: specific expression in the germ cells and its mRNA
localization in the posterior-most blastomeres in early embryos. Dev. Genes Evol.
Imai, K. S., Levine, M., Satoh, N. and Satou, Y. (2006). Regulatory blueprint for
a chordate embryo. Science 312, 1183-1187.
Izzard, C. S. (1968). Migration of germ cells through successive generations of
pallial buds in Botryllus schlosseri. Biol. Bull. 135, 424.
Johnson, S. L. and Yund, P. O. (2004). Remarkable longevity of dilute sperm in a
free-spawning colonial ascidian. Biol. Bull. 206, 144-151.
Laird, D. J. and Weissman, I. L. (2004). Telomerase maintained in self-renewing
tissues during serial regeneration of the urochordate Botryllus schlosseri. Dev.
Biol. 273, 185-194.
Laird, D. J., De Tomaso, A. W. and Weissman, I. L. (2005). Stem cells are units
of natural selection in a colonial ascidian. Cell 123, 1351.
Lauzon, R. J., Ishizuka, K. J. and Weissman, I. L. (2002). Cyclical generation and
degeneration of organs in a colonial urochordate involves crosstalk between old
and new: a model for development and regeneration. Dev. Biol. 249, 333-348.
Letunic, I., Copley, R. R., Pils, B., Pinkert, S., Schultz, J. and Bork, P. (2006).
SMART 5, domains in the context of genomes and networks. Nucleic Acids Res.
Li, M., Hong, N., Xu, H., Yi, M., Li, C., Gui, J. and Hong, Y. (2009). Medaka
vasa is required for migration but not survival of primordial germ cells. Mech.
Dev. 126, 366-381.
Livak, K. J. and Schmittgen, T. D. (2001). Analysis of relative gene expression
data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
Methods 25, 402-408.
Mochizuki, K., Nishimiya-Fujisawa, C. and Fujisawa, T. (2001). Universal
occurrence of the vasa-related genes among metazoans and their germline
expression in Hydra. Dev. Genes Evol. 211, 299-308.
Mukai, H. (1977). Comparative studies on the structure of reproductive organs of
four botryllid ascidians. J. Morphol. 152, 363-380.
Mukai, H. and Watanabe, H. (1976). Studies on formation of germ-cells in a
compound ascidian Botryllus primigenus Oka. J. Morphol. 148, 337-362.
Nieuwkoop, P. D. and Sutasurya, L. A. (1979). Primordial germ cells in the
chordates. Cambridge: Cambridge University Press.
Nieuwkoop, P. D. and Sutasurya, L. A. (1981). Primordial Germ Cells in the
Invertebrates: from epigenesis to preformation. Cambridge: Cambridge
Pfister, D., De Mulder, K., Hartenstein, V., Kaules, G., Borgonie, G., Marx, F.,
Morris, J. and Ladurner, P. (2008). Flatworm stem cells and the germ line:
developmental and evolutionary implications of macvasa expression in
Macrostomum lignano. Dev. Biol. 319, 146-159.
Rebscher, N., Zelada-Gonzales, F., Banisch, T. U., Raible, F. and Arendt. D.
(2007). Vasa unveils a common origin of germ cells and of somatic stem cells
from the posterior growth zone in the polychaete Platynereis dumerilii. Dev. Biol.
Rosner, A., Moiseeva, E., Rinkevich, Y., Lapidot, Z. and Rinkevich, B. (2009).
Vasa and the germ line lineage in a colonial urochordate. Dev. Biol. 331, 113-128.
Sabbadin, A. (1953). Note preliminare sulle cellule del sangue di Botryllus
schlosseri. Bulletin de l’istitut Oceanographique 50, 1-12.
Sabbadin, A. and Zaniolo, G. (1979). Sexual differentiation and germ-cell
transfer in the colonial ascidian Botryllus schlosseri. J. Exp. Zool. 207, 289-304.
Santos, A. C. and Lehmann, R. (2004). Germ cell specification and migration in
Drosophila and beyond. Curr. Biol. 14, R578-R589.
Sengoku, T., Nureki, O., Nakamura, A., Kobayashi, S. and Yokoyama, S.
(2006). Structural basis for RNA unwinding by the DEAD-box protein Drosophila
Vasa. Cell 125, 287-300.
Shibata, N., Umesono, Y., Orii, H., Sakurai, T., Watanabe, K. and Agata, K.
(1999). Expression of vasa(vas)-related genes in germline cells and totipotent
somatic stem cells of planarians. Dev. Biol. 206, 73-87.
Shirae-Kurabayashi, M., Nishikata, T., Takamura, K., Tanaka, K. J.,
Nakamoto, C. and Nakamura, A. (2006). Dynamic redistribution of vasa
homolog and exclusion of somatic cell determinants during germ cell
specification in Ciona intestinalis. Development 133, 2683-2693.
Stoner, D. S. and Weissman, I. L. (1996). Somatic and germ cell parasitism in a
colonial ascidian: possible role for a highly polymorphic allorecognition system.
Proc. Natl. Acad. Sci. USA 93, 15254-15259.
Stoner, D. S., Rinkevich, B. and Weissman, I. L. (1999). Heritable germ and
somatic cell lineage competitions in chimeric colonial protochordates. Proc. Natl.
Acad. Sci. USA 96, 9148-9153.
Sunanaga, T., Saito, Y. and Kawamura, K. (2006). Postembryonic epigenesis of
Vasa-positive germ cells from aggregated hemoblasts in the colonial ascidian,
Botryllus primigenus. Dev. Growth Differ. 48, 87-100.
Sunanaga, T., Satoh, M. and Kawamura, K. (2008). The role of Nanos
homologue in gametogenesis and blastogenesis with special reference to male
germ cell formation in the colonial ascidian, Botryllus primigenus. Dev. Biol. 324,
Swalla, B. J. (2006). Building divergent body plans with similar genetic pathways.
Heredity 97, 235-243.
Takamura, K., Fujimura, M. and Yamaguchi, Y. (2002). Primordial germ cells
originate from the endodermal strand cells in the ascidian Ciona intestinalis. Dev.
Genes Evol. 212, 11-18.
Tanaka, S. S., Toyooka, Y., Akasu, R., Katoh-Fukui, Y., Nakahara, Y., Suzuki,
R., Yokoyama, M. and Noce, T. (2000). The mouse homolog of Drosophila Vasa
is required for the development of male germ cells. Genes Dev. 14, 841-853.
Tiozzo, S. and De Tomaso, A. W. (2009). Functional analysis of Pitx during
asexual regeneration in a basal chordate. Evol. Dev. 11, 152-162.
Tiozzo, S., Voskoboynik, A., Brown, F. D. and De Tomaso, A. W. (2008). A
conserved role of the VEGF pathway in angiogenesis of an ectodermally-derived
vasculature. Dev. Biol. 315, 243-255.
Yamamoto, M. and Okada, T. (1999). Origin of the gonad in the juvenile of a
solitary ascidian, Ciona intestinalis. Dev. Growth Differ. 41, 73-79.
Development 136 (20)