FASEB J.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-7337com.
 (The FASEB Journal. 2007;21:1335-1344.)
© 2007 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
fj.06-7337comv1
21/7/1335    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Voskoboynik, A.
Right arrow Articles by Weissman, I. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Voskoboynik, A.
Right arrow Articles by Weissman, I. L.

Striving for normality: whole body regeneration through a series of abnormal generations

Ayelet Voskoboynik*,1, Noa Simon-Blecher{dagger},{ddagger}, Yoav Soen§, Baruch Rinkevich{dagger}, Anthony W. De Tomaso*, Katherine J. Ishizuka* and Irving L. Weissman*

* Department of Pathology, Stanford University School of Medicine, Stanford, California, USA; Department of Biology, Hopkins Marine Station, Pacific Grove, California, USA;

{dagger} National Institute of Oceanography, Oceanographic and Limnological Research, Tel-Shikmona, Haifa, Israel;

{ddagger} Faculty of Life Science, Bar Ilan University, Ramat Gan, Israel; and

§ Department of Biochemistry, Stanford University School of Medicine, Stanford, California, USA

1Correspondence: Department of Pathology, Stanford University School of Medicine, Stanford, CA 94305, USA. E-mail: ayeletv{at}stanford.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Embryogenesis and asexual reproduction are commonly considered to be coordinated developmental processes, which depend on accurate progression through a defined sequence of developmental stages. Here we report a peculiar developmental scenario in a simple chordate, Botryllus schlosseri, wherein a normal colony of individuals (zooids and buds) is regenerated from the vasculature (vascular budding) through a sequence of morphologically abnormal developmental stages. Vascular budding was induced by surgically removing buds and zooids from B. schlosseri colonies, leaving only the vasculature and the tunic that connects them. In vivo imaging and histological sections showed that the timing and morphology of developing structures during vascular budding deviated significantly from other asexual reproduction modes (the regular asexual reproduction mode in this organism and vascular budding in other botryllid species). Subsequent asexual reproduction cycles exhibited gradual regaining of normal developmental patterns, eventually leading to regeneration of a normal colony. The conversion into a normal body form suggests the activation of an alternative pathway of asexual reproduction, which involves gradual regaining of normal positional information. It presents a powerful model for studying the specification of the same body plan by different developmental programs.—Voskoboynik, A., Simon-Blecher, N., Soen, Y., Rinkevich, B., De Tomaso, A. W., Ishizuka, K. J., Weissman, I. L. Striving for normality: whole body regeneration through a series of abnormal generations.


Key Words: vascular budding • blastogenesis • development • tunicate


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
EMBRYOGENESIS AND ASEXUAL REPRODUCTION are two disparate developmental processes that share essential developmental stages including the establishment of body axes, morphogenetic patterning, organ formation, and cell differentiation (1 2 3 4 5 6) . The life cycle of Botryllus schlosseri, a colonial tunicate (family Styelidae, subfamily Botryllinae), includes both sexual and asexual reproduction pathways (1 2 3) . Despite differences in initiation, these two reproduction modes give rise to essentially the same adult body plan (1 2 3 4 5 6 7) . Sexual reproduction starts with egg fertilization and progresses through classic embryonic stages into a tadpole larva featuring chordate characteristics such as a tail, notochord, neural tube, and striated musculature (8) . Upon hatching, the motile tadpole swims off from the natal colony, settles on a suitable substratum, and metamorphoses into the sessile body plan (9 , 10) . In this invertebrate form, B. schlosseri constitutes a colony of individuals (zooids and buds) undergoing cycles of asexual reproduction by budding. All zooids and buds are connected through a vasculature network and are embedded within a gelatinous matrix (tunic). The common vasculature terminated in sausage-shaped protrusions (ampullae; 1 , 2 ; Fig. 1 ; Supplemental Video S1). A mature colony includes three overlapping generations: an adult zooid, a primary bud, and a secondary bud. Each zooid is an autonomous, filter-feeding individual comprising a heart, digestive tract, endostyle, branchial sac, neural complex, oral and atrial siphons and gonads (1 , 2 , 11) . Asexual (palleal) budding begins with the formation of a small vesicle that breaks off from the parent epidermis and peribranchial epithelium and segregates into a blastula-like structure (1) . As the cell proliferates, the vesicle undergoes a series of invaginations, differentiates, and develops into an adult zooid (1 , 12 , 13) ; Fig. 1 ; Supplemental Video S1). The replacement of zooids’ generation occurs through a synchronized wave of massive apoptosis in which the adult zooids are destroyed and taken over by their primary buds (14 , 15 ; video S1). Palleal budding is synchronized throughout the colony and is divided into four major developmental stages (A–D; ref. 16 ; Fig. 1 ; Supplemental Video S1), a cycle that continues throughout the life span of the colony. Thus, unlike most species, where the body is long-living and maintained by cellular replacements (e.g., ref. 17 ), B. schlosseri regenerates new colonial units on a weekly basis while its connective vasculature and tunic remains intact (1 , 2 , 14 , 15) .


Figure 1
View larger version (66K):
[in this window]
[in a new window]

  		Figure 1. Normal palleal budding in B. schlosseri. Stereomicroscopic images of palleal budding cycle are shown in panels A–D corresponding to developmental stages A–D, respectively. Coincident development of primary and secondary buds is outlined in panels E–H (microscopic images). In stage A colonies (A, E), the primary buds almost completed organogenesis and the secondary buds initiated as a thickening of the atrial wall of the primary bud (see also Fig. 3A ). In stage B (B, F), the heart beats in the primary bud and the secondary bud forms a closed sphere that will soon begin a series of invaginations reminiscent of gastrulation. Stage C (C, G) is characterized by growth of the primary bud and organogenesis in the secondary bud. In stage D (D, H), zooids die in a wave of apoptosis while the primary buds mature and replace them. amp, ampulla; sb, secondary bud; res, zooid-resorbing zooid; bv, blood vessel; end, endostyle; h, heart; ds, digestive system; bs, branchial sac. Scale bars: A–D) = 1 mm, E–H) 100 µm.

Previous studies in B. schlosseri (18 19 20 21) and in some other botryllid tunicates (22 23 24 25 26 27 28) revealed another regeneration pathway known as vascular budding, involving an asexual regeneration of colonies from blood vessels (rather than regeneration from buds and zooids). In Botryllus primigenus, vascular budding begins with the formation of a hollow blastula-like structure originating from vascular epithelium and cells circulating in the blood (23) . This structure develops into a vascular bud capable of regenerating an entire colony (23) . In some species, vascular budding occurs concurrently with palleal budding (Botryllus primigenus and Botryllus delicates; refs. 23 , 28 ), whereas in others it is initiated in the absence of palleal budding when the colony recovers from hibernation or aestivation (Botrylloides leachi and Botrylloides gascoi; refs. 22 , 24 25 26 27 ). Normal patterns and stages of development that resemble asexual reproduction (palleal budding) patterns were observed in these studies (23 , 24 , 26 27 28) .

Unlike other botryllid tunicates (23 24 25 26 27 28) , vascular budding in B. schlosseri is an induced phenomenon that requires the removal of zooids and buds (18 19 20 21) . The stages of this regeneration process have never been characterized. Using in vivo imaging and histological sections, we analyzed the regeneration of vascular budding from initiation and over successive generations, and compared it with palleal budding. Unlike palleal budding and vascular budding in other botryllid species, which maintain normal patterns of development throughout the process (23 , 24 , 26 27 28) , we found that vascular budding in B. schlosseri involves progression through a sequence of abnormal generations that undergo gradual conversion toward a morphologically normal colony. Through various experimental manipulations, we studied the parameters required to induce vascular budding. We discovered that successful regeneration requires colonies to have an intact vasculature, a minimal colony size, and an active blood flow. In addition, vascular budding can be induced only when buds and zooids are removed during a specific stage of the asexual life cycle. Incomplete removal of all buds and zooids or the presence of intact zooids and buds next to the treated colonies prevents the regeneration of zooids from vascular origin. This indicates a suppression mechanism mediated by soluble signals.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Animals
Colonies of Botryllus schlosseri (Pallas) were collected from the Monterey Marina (Monterey, CA, USA). Hatched larvae were settled and maintained as described previously (29) . Colonies were raised on 5 x 7.5 cm2 glass slides and kept vertically in slots of glass staining racks (3 slides per rack), with up to 18 colonies per 17 L standing seawater aquariums. Colonies were kept at a 14:10 h light:dark regimen under constant 20°C temperature and fed four times a week with Liquifry Marine (Liquifry Co., Dorking, UK). Vascular budding experiments were performed on 12 healthy colonies that reached the minimum size of 10 zooidal systems. These colonies were further subcloned and divided into 10 experimental groups. Each of the four largest colonies (1225c, 1196j, 1301d, and 1210d) had one or more subclones in any of the experimental groups. Each of eight colonies (1105e, 1185d, 1255a, 1248a, 1114a, 1129a, 1264a, 1264d) had subclones in four or more different experimental groups.

Experimental design and surgical manipulations
We have studied the process of vascular budding in B. schlosseri by surgically removing buds and zooids with a wheeler dissection knife (Miltex, York, PA, USA). All surgical manipulations were performed under a stereomicroscope (Wild Heerbrugg M7A, Gais, Switzerland). The removal of buds was confirmed using an inverted microscope (Diaphot 200, Nikon, NY, USA). The influence of different conditions on vascular budding was examined using 10 experiments, each having only one factor changed (compare with the relevant control). All other conditions (tank size: 4 L, temperature: 20°C, light/dark regimen: 14/10 h, frequency of cleaning and water replacement: once in every 1–2 days, feeding regimen: no feeding) remained invariant. In experiment 1, both zooids and buds were removed from 10 colonies at stage D, with the number of zooids ≥8. Experiment 2 tested different developmental stages (n=7, stages A–C, number of zooids≥8), the effect of the size of a colony was tested in experiment 3 (n=6, stage D, number of zooids≤8). Experiments 4 and 5, respectively, tested the removal of the peripheral ampullae and the central vasculature in addition to the removal of zooids and buds (stage D: n=4 and number of zooids≥8, each). Experiment 6 was designed to examine the effect of the presence of intact colonies in the same tanks with the manipulated colonies. In this experiment zooids and buds were removed from 12 subclones (8 different genotypes, stage D; colony size≥8 zooids). Manipulated and intact colonies were placed in six different 4 L plastic tanks, with two manipulated colonies and one intact colony in each tank. In two of the six tanks, the intact colony were placed on the same slide as the manipulated colonies at a distanced of 2 cm. In the other four tanks, the intact and manipulated colonies were placed on separate slides. Experiments 7 and 8 examined the effects of partial removal of buds or zooids on vascular budding by removing only the buds in experiment 7 (stages A–C: n=10, number of zooids≥8) and removing all zooids and primary buds, leaving up to four secondary buds intact in experiment 8 (stages B–D: n=10, number of zooids≥8). In experiment 9, cells were drawn from the vasculature of colonies in stage D. Between 1000 to 5000 of these cells were injected into the vasculature of isogeneic subclones at stages A–C (n=8, number of zooids≥8). In experiment 10, buds were removed from colonies at stage D, leaving only the resorbing zooids (stage D: n=8, number of zooids≥8; experiments 1–9, Table 1 ; experiment 10 described in the text).


View this table:
[in this window]
[in a new window]

  		Table 1. Experimental manipulation and regenerative potential (vascular budding) of Botryllus schlosseri coloniesa

Manipulated colonies were placed in 4 L plastic tanks (up to three colonies per tank). When the vascular buds became zooids (filter feeding individuals), they were moved into 17 L tanks and fed. Stereomicroscopic imaging was performed using a stereomicroscope (Wild Heerbrugg M7A, Gais, Switzerland) coupled to a CCD camera (coolpix995 Nikon, NY, USA). Pictures and video clips were taken once a day during the first 2 wk and once every 3–7 days thereafter. The results were statistically analyzed using t test and Friedman ANOVA for nonparametric data.

Time-lapse imaging
Time-lapse imaging was performed by an automated microscopy (ImageXpress, Molecular Devices Corp., Palo Alto, CA, USA). After manipulation, colonies (stages B–D: n=10, number of zooids≥8) were kept on standard glass slides placed in holders that contained 10 ml of seawater. The holders were mounted onto the ImageXpress stage, and kept at room temperature. Regeneration areas of interest at each colony were identified and imaged by an automated custom (Visual Basic) script. Phase contrast images at varying magnifications were performed every 20–30 min during the first 2–3 days and every 60 min on days 4–7. Images were stored in the ImageXpress database. Time-lapse sequences were generated from consecutive images taken from the same location.

Microinjections
Glass needles were prepared using a micropipette puller (Sutter Instruments, Movato, CA, USA). The tapped area was cut diagonally, forming a 50–60 µm diameter sharp tip. Hemolymph samples of 1–3 µl, each containing 1000–5000 cells, were drawn from peripheral ampullae (stage D colonies) using air compressed microinjector (PLI-188, Nikon, NY, USA). The micropipette content was counted under the microscope (Diaphot 200, Nikon, NY, USA) and immediately injected into peripheral ampullae of isogeneic subclones at blastogenic stages A, B, or C (n=8, number of zooids≥8). Microinjections were done under a microscope (Diaphot 200, Nikon, NY, USA). Incorporation of cells into the colony transparent vasculature was confirmed visually using 200x magnification.

Circulation parameters measurements
Images of the colony marginal blood vessels and video clips of circulating hemocytes in the marginal blood vessel were taken using a CCD camera (coolpix995 Nikon, NY, USA) coupled to a microscope (Diaphot 200, Nikon, NY, USA; 100x magnification). Number of circulating hemocytes and rates of blood flow were measured and counted using ImajeJ software, version 1.32j (NIH, USA). The results were statistically analyzed using t test and correlation tests.

Histology
Vascular budding samples (from days 1, 2, 3, 6, 8, 12, 16 of development; n=40) were fixed in 2% paraformaldehyde for 20–30 min at room temperature, dehydrated in a graded ethanol series (70–100%), and embedded in paraplast (Sigma-Aldrich, MO, USA). Sections (5 µm) were cut by hand-operated microtome (Leica, IL, USA) and stained with Azan Heidenhains (30) for general morphology.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Abnormal development through vascular budding
We studied the process of vascular budding in B. schlosseri by surgically removing all developing buds and zooids and analyzing regenerating structures over successive generations of zooids. Time-lapse imaging revealed that, within a few hours after surgery, regeneration began in a few ampullae that harbored aggregations of undifferentiated lymphocyte-like blood cells (as opposed to all other ampullae harboring aggregations of pigment cells; Fig. 2 A; Supplemental Video S2). Histological sections of these vascular buds from the first regenerating stages revealed unique vesicle structures, each comprising an outer endothelial cell layer and an inner mass consisting mainly of lymphocyte-like cells (Fig. 3 D). Buds developed within the ampullae lumen and remained connected to the colony marginal blood vessels (Fig. 2 ; Supplemental Video S2). While growing, each regenerating ampullae was changed morphologically from a sausage-like structure to an oval shape (Fig. 2B ). A beating heart-like organ was observed in the developing bud within 1–2 days after surgery (Fig. 2C, D ). An endostyle-like, pigmented structure extending on the ventral face of the developing bud, stigmata-like structures, and aggregations of lymphocyte and macrophages were observed in the bud 3 days after removal of zooid and bud (Fig. 2F ).


Figure 2
View larger version (53K):
[in this window]
[in a new window]

  		Figure 2. Vascular bud formation. Images were taken at various times during the first 73 h of the vascular bud formation (ventral view). A) Regeneration area 1 h after zooid and bud removal from a colony in stage D. Budding was initiated and developed in ampullae 1 and 2. The tips of these ampullae contained aggregations of lymphocyte-like cells (unlike other ampullae in the colony). B) 7 h after manipulation, the tip of ampulla 1 grew and changed its shape (vascular bud 1). A functional heart was observed in ampulla 1 after 14 h (C). The bud grew in size (D). An endostyle-like, pigmented structure was observed on the ventral face of the developing bud after 54 h (E). F) 73 h vascular buds (1 and 2) contained a heart, endostyle-like, stigmata-like structures, and aggregations of lymphocyte and macrophages. h, hour; vb, vascular bud; endo, endostyle-like; amp, ampulla; bv, blood vessel; mac, macrophages. Scale bar = 100 µm.


Figure 3
View larger version (126K):
[in this window]
[in a new window]

  		Figure 3. Histological analysis of normal blastogenesis (A–C) vs. vascular budding (D–F) sections were stained with azan heidenhain, which stains nuclei and cytoplasm in red, collagenic and reticular fibers as well as mucus in blue or dark blue. A) Day 2 palleal bud. The bud arises from the lateral wall of its parent, creating an epidermal disc that will later pinch off to form a sphere. B) Day 6 palleal bud (encircled). Subdivisions of 3 small spheres have recently been completed; 2 of the subdivisions can be seen in this section. C) A fully developed bud (day 14) before takeover. Its organs—endostyle, branchial sac, and digestive system—can be seen here. D) Day 1 vascular bud within an ampulla connected to the colony marginal blood vessel. The bud (encircled) comprises an outer single layer of epidermis and an inner mass of lymphocyte-like cells. E) Day 6 vascular bud exhibiting early formation of organs (e.g., digestive system, branchial sac, stigmata, and endostyle); all have an abnormal structure (compare to panel C). F) First-generation vascular zooid: 2 days after it opened its siphons are already in a deteriorating state. The zooid is rotated 45° relative to normal position. A second-generation bud (encircled) formed near the endostyle of the vascular zooid. This bud looked abnormal, too. bv, blood vessel; mbv, marginal blood vessel; hemoblast lymphocyte cell aggregation; ec, endothelial cell; full mac, full macrophagel; endo, endostyle; stig, stigmata; ds, digestive system; bs, branchial sac; cl, cloacal. Scale bars: A, D) 10 µm; B, C, E, F) 100 µm.

The timing and morphology of developing structures during vascular budding and thereafter deviated significantly from the normal development of palleal budding. The heart-like organ was the first to develop and was observed in the vascular bud 1–2 days after removal of zooids and buds. This organ controlled blood flow to and from the regeneration area and was necessary for vascular bud development; vascular zooids were developed only from vascular buds that developed hearts (Supplemental Video S3). Vascular buds at days 3–4 were extensively colonized by macrophages (Fig. 2F , histological sections not shown); by day 6 of vascular budding, organs (digestive system, endostyle, branchial sac; Fig. 3E ) were fully developed. In contrast, during regular palleal budding in this species, heart beating was observed only at day 10 of bud development (Fig. 1B ). Likewise, colonization by macrophages and organ development in palleal budding occurred later than in vascular budding (days 15 and 8, respectively). Histological sections of regenerating buds revealed unusual epithelial cell morphology and abnormal cell-cell junctions (Fig. 3C vs. Fig. 3E, F and Fig. 4A, C vs. Fig. 4B, D ). In addition, most of the animal’s cavities lined by the epithelia (e.g., the atrium) were loaded with circulating cells; in palleal budding, cavities were empty (Figs. 3 , Fig. 4 ). Vascular buds also had abnormal size, morphology, and orientation compared with normal buds (Fig. 5 ). By day 6, regenerating buds were significantly larger (0.28±0.26 mm2, n=5, after bud and zooid removal vs. 0.055±0.006 mm2, n=7, in normal, palleal budding t test, P<0.005). The fully developed vascular buds (at the completion of organogenesis) were smaller than regular buds (1.46±0.33 mm2, n=5, vs. 2.58±0.31 mm2, n=6, respectively t test, P<0.005), with an orientation rotated by 45° to 90° vs. orientation of a normal bud (Fig. 5E vs. F). Completion of vascular bud development occurred ~ 12 days after surgery with the formation and opening of spontaneously contracting siphons. Thus, by the end of first regeneration cycle, the structure, size, and histology of the regenerated colony were significantly different from those of the unperturbed colony.


Figure 4
View larger version (161K):
[in this window]
[in a new window]

  		Figure 4. Histological sections of the digestive system and endostyle in normal zooid (A, C) vs. vascular bud (B, D). A) Cross section of normal endostyle near the endostyle groove that contains a series of distinct groups of columnar epithelial cells and a cell free cavity. B) Cross section of the vascular bud endostyle. The epithelial cells of the endostyle groove exhibit unusual morphology and organization as well as irregular cell-cell connections. The putative cavity region contains many cells. C) Cross section of normal digestive system epithelial cells. D) Cross section through the vascular bud digestive system exhibiting fewer organized epithelial structures. Stain, azan heidenhain; scale bar = 50 µm.


Figure 5
View larger version (133K):
[in this window]
[in a new window]

  		Figure 5. Light micrographs of palleal budding (A, C, E; dorsal view) vs. vascular bud development (B, D, F; dorsal view). A) Day 4 palleal bud (outlined in circle) consists of a closed double-layer vesicle comprising an outer single layered epidermal sphere and an inner single-layered vesicle of atrial epithelium. The space between the layers is filled with blood cells. B) Day 3 vascular bud (outlined in circle) is significantly larger and different in structure. C) Day 6 palleal bud looks empty compared with day 6 vascular bud. D) Day 6 organ-containing vascular bud. E) Day 13 normal zooid that has recently opened its siphons. The endostyle, which is extending medially on the zooid ventral face, is marked by a dotted line. Its primary bud is located on the zooid lateral wall. F) Day 12 vascular zooid has recently opened its siphons and differs in orientation, size, and pigmentation pattern. The midsagittal plane of the zooid is perpendicular to normal orientation; as a result, the endostyle (marked in dotted line) is perpendicular to its normal orientation. Its bud is budding from the endostyle area. mbv, marginal blood vessel; amp, ampulla; b, bud; endo, endostyle; os, oral siphon; as, atrial siphon. Scale bars: A–C) 125 µm, D–F) 1 mm.

From that point on, regeneration proceeded through subsequent cycles of budding from regenerated buds, rather than budding from the vasculature. However, the structures and locations of these buds remained different for several generations of asexual reproduction by palleal-like budding. The bud of the second generation originated from the body wall of the vascular bud; unlike normal palleal budding, the initiation site was located near the endostyle region rather than the lateral wall of the zooid (the normal palleal budding site). Since the vascular bud’s position in the tunic was perpendicular to a normal palleal bud, the second-generation bud, had the same net orientation as unperturbed colony with respect to the colony vasculature. As in vascular bud, the development of the second-generation bud was accelerated and exhibited unexpected timing of organogenesis. The second-generation bud took over the vascular zooid within 2 to 4 days (Fig. 6 B) before completion of organ development. After takeover, this bud completed its development and gave rise to a third-generation bud, which also originated near the endostyle area. The third generation developed over 6 days. In 40% of regeneration cases, the morphology, size, and orientation of succeeding generations of buds and zooids gradually became normal. Within an average of 7 ± 0.8 budding cycles, the developed buds no longer differed morphologically from regular palleal budding (Fig. 6 ; successful vascular budding experiment 1, Table 1 ). In 30% of the cases of regeneration, successive generations remained abnormal and the colonies died within 60–70 days after surgery ("abnormal formation only" in experiment 1, Table 1 ; supplemental Fig. 1). In the remaining 30%, only initial buds were formed, but organs (heart, digestive system, branchial sac, endostyle) failed to develop ("initial bud formation," experiment 1 in Table 1 ; supplemental Fig. 1).


Figure 6
View larger version (149K):
[in this window]
[in a new window]

  		Figure 6. Gradual acquisition of normal patterns in a regenerating colony. This colony, whose vascular bud formation is presented in Fig. 5B, D, F , became normal after eight abnormal generations. A) A fully developed vascular bud, its orientation, shape, size, pigmentation, and organ morphology are all abnormal. The next generation bud is developing near the endostyle (dotted line). B) First-generation zooid and its bud: the second-generation bud is taking over the vascular zooid only 3 days after its siphon opening (compared with 6–7 days normal duration) and before completion of its organ development. The bud endostyle (dotted line) orientation is rotated in 90° relative to normal endostyle orientation. C) Fifth-generation zooid (still unusual). D) Seventh-generation zooid: the shape, size, and general morphology appear more normal. Orientation of the zooids is slightly rotated relative to normal (~25°); the next generation still buds from an unusual position. E) Ninth generation. The colony is heavily pigmented but otherwise appears to be normal; endostyle orientation (dotted line) is normal. F) Eleventh generation exhibits a normal colony. endo, endostyle; amp, ampulla; as, atrial siphon; os, oral siphon; sec, secondary. Scale bar = 1 mm.

Successful regenerating colonies survived under our culturing conditions for 221–260 days after surgery, thereby demonstrating conversion of abnormal budding process into an unperturbed body plan of B. schlosseri.

Conditions for a successful vascular budding
To characterize the conditions required for successful vascular budding, we removed buds and zooids at different blastogenic stages and conditions. In some, the vasculature was further manipulated by removing the peripheral ampullae or the central vasculature. Experiments were carried out on colonies containing one to four zooidal systems. Removal of buds and zooids yielded four types of responses (Table 1) : 1) no regeneration, 2) initiation of a vascular bud that did not complete its development (initial bud formation, supplemental Fig. 1) and bud initiation, followed by successive generations of abnormal budding 3) with or 4) without conversion into a normal colony (successful vascular budding, and abnormal zooid formation, respectively). Successful vascular budding was observed only when all buds and zooids were removed during blastogenesis stage D (the apoptotic stage, Fig. 1D ). The chance of zooid formation after bud and zooid removal during this stage was 70% (n=10, experiment 1, Table 1 ). Removal of buds and zooids during blastogenic stages A–C, though capable of inducing initial vascular bud development (~57%, n=7, experiment 2, Table 1 ), failed to develop a functional heart, and regeneration then ceased. Successful vascular budding also depends on colony size, it did not occur in colonies with less than 9 zooids, not even in colonies in stage D (n=6, experiment 3, Table 1 ). Removal of either the peripheral ampullae (leaving only the central vasculature; n=4, experiment 4, Table 1 ) or the central vasculature (leaving only the peripheral ampullae’s n=4, experiment 5, Table 1 ) completely inhibited successful regeneration, indicating the importance of vasculature integrity in this type of vascular budding. The potential of regeneration in colonies with ≥8 zooids whose buds and zooids were removed at stage D (Table 1 , experiment 1) was significantly higher (P<0.05, Friedman 2-way analysis of variance test for nonparametric data) than those with buds and zooids removed under any other experimental condition (Table 1 , experiments 2–6, 9).

Inhibition of a successful vascular budding by zooids and buds
Since vascular budding does not occur under normal conditions, we tested the effects of intact or ectopically introduced zooids or buds on the regeneration process. Placing intact zooids or buds next to (on the same slide) or in the same tank (4 L, on a different slide) with stage D zooid- and bud-excised colonies, inhibited vascular budding (n=12, experiment 6, Table 1 ), suggesting that diffusible factors may be involved in its suppression. Similarly, removal of buds alone during blastogenic stage A, B, or C while leaving the zooids and the vasculature intact resulted in exclusively normal regeneration of the remaining zooids via palleal budding (n=10, experiment 7, Table 1 ). Keeping stage B–D secondary buds after the removal of primary buds and zooids (n=10, experiment 8, Table 1 ) led in some cases to the development of normal and left-right inverted (situes inversus viscerum, n=2 of 10) coexisting zooids, but never to complete vascular budding. Finally, removal of all primary and secondary buds during the takeover phase led to a reversed takeover process, where resorbing zooids reattached to the colony marginal vessel and budding of the body wall of the degenerating zooid occurred (n=8). We therefore concluded that the presence of any type of buds, zooids, or deteriorating zooids in the same tank, even without physical contact with the colony, prevents regeneration of vascular zooids through vascular budding.

Correlation between circulation parameters and regeneration
Bud initiation was associated with active angiogenesis and remodeling of the colony vasculature manifested by extensive migration, fusion, and branching of nonregenerating ampullae (Fig. 7 A–C vs. Fig. 7D-F ; Supplemental Video S4).


Figure 7
View larger version (52K):
[in this window]
[in a new window]

  		Figure 7. Vascular remodeling during successful (A–C) and unsuccessful vascular budding (D–F) A) Ampullae and marginal blood vessel in a stage D colony 1 h after complete removal of buds and zooids. Ampullae exhibit a sausage-like shape; each ampulla is connected through a short blood vessel to the colony marginal blood vessel. B) 2 days later, the ampullae migrated within the tunic and created a new vascular connection by fusing with each other. C) 3 days after zooid and bud removal, the ampullae fuse with the marginal blood vessel and form a wider vessel able to support rapid cell movement to the regeneration area. D) Ampullae and blood vessels in a stage C colony 1 h after zooid and bud removal. Ampullae and blood vessels contain fewer cells and differ in shape from stage D vessels (plate A). E) One day later, fewer ampullae fusions were formed compared with successful regeneration. F) On day 4, the vasculature is degenerated and blood flow ceases. Scale bar = 100 µm.

Comparative examination of blood flow rate in colonies that failed to regenerate with successful regenerating colonies revealed that the regeneration is associated with higher flow rates and rapid changes in the colony vasculature architecture (Supplemental Video S4A vs. S4B). Average flow rate measured in regenerating colonies 1 day after bud and zooid removal was ~10-fold higher (t test, P<<0.005) than in colonies that failed to regenerate (192±30 cells/min vs. 17±6 cells/min, respectively Fig. 8 ). Correlation coefficient of r = 0.97 was measured between regeneration potential and rate of blood flow. We examined the number and flow rates of hemocytes in blastogenic stage D vs. stages A–C to test the accountability of stage-dependent circulation features for the dependence of regeneration on developmental stage. The number of circulating hemocytes in stage D was 4-fold higher than in stages A–C (6.1±1.9 cells/1000 µm2, and 1.6±0.35 cells/1000 µm2, respectively, with P<0.005 in t test); flow rate was ~10-fold faster in stage D (509±180 cells/min vs. 50±32 cells/min, respectively; t test, P<0.005, Supplemental Video S5A vs. S5B).


Figure 8
View larger version (10K):
[in this window]
[in a new window]

  		Figure 8. Comparison between average blood flow rates of colonies that regenerated vs. colonies that failed to regenerate. Blood flow rates were measured at the vasculature 1 day after bud and zooid removal.

To test the possible contribution of stage-specific cell composition (e.g., stage D-specific presence of circulating stem cells) to successful vascular budding, we removed buds and zooids at stages A–C and injected 1000–5000 circulating cells of stage D collected from isogeneic colonies (number of injected cells was determined according to Laird et al., ref. 31 ). The engrafted cells did not enable development of vascular zooids through vascular budding in colonies at stages A–C (n=8; experiment 9, Table 1 ), suggesting that stage-specific composition of circulating cells is not sufficient to explain the inability to regenerate via vascular budding in stages A–C.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
This study describes the induction and progression of a whole body regeneration phenomenon in a simple chordate, Botryllus schlosseri, in which the entire colony of interconnected reproduction units (buds and zooids) regenerated from the vasculature. Unlike vascular budding in other botryllid ascidians (23 24 25 26 27) , we demonstrated that vascular budding in B. schlosseri progresses through multiple generations of abnormal intermediate body forms, exhibiting gradual conversion into a normal colony.

Striving for normality: an alternative developmental pathway
Conversion into a final configuration of body form is a universal theme in developmental systems and is commonly regarded as a precisely orchestrated program that is deeply embedded in the genome (32 , 33) . This program relies on progression through defined states and is not expected to withstand extreme arbitrary perturbations. For example, when pluripotent stem cells are detached from their normal, regulative environment, they often give rise to progressively abnormal structures (e.g., teratoma, cancer; refs. 34 35 36 ). Therefore, the gradual conversion into a normal colony in B. schlosseri most likely reflects induction of an alternative, embryonic-like, developmental program that is invoked only in the absence of buds and zooids. Our time-lapse records and histological sections suggested that at least some of the ampullae contained a blastogenic tissue that is triggered to develop upon removal of all buds and zooids. The remarkable differences in vascular budding organogenesis compared with palleal budding (e.g., the expedited heart development, and the different histological structures) are also consistent with induction of an alternative developmental pathway.

The differences between vascular and palleal budding developmental programs might reflect in part the differences between the environment inside ampullae and the environment inside buds (or zooids), respectively. Whereas in palleal budding the axes and orientation of a newly formed bud are influenced by positional information provided by the parent bud (12 , 21 , 37 , 38) , in vascular budding the first bud responds to cues in the ampullae. Indeed, the vascular system is capable of affecting bud axes; the posterior end of zooids developed from an early-stage bud engraftment into a vascularized tunic was localized at the intersection of the bud and the entering vessel (21) . In vascular budding, the second-generation buds initiate from the initial (vascular) bud. Thus, the second generation is exposed to different, presumably closer to normal, "bud-like" cues. Subsequent generations are exposed to environments that gradually, from one generation to the succeeding one, appear to be closer to the normal environment, eventually converging into the generation of a normal colony. Thus, the gradual establishment of normal axes and body form might reflect the convergence of positional information set by each parent.

Repression of an unfavorable developmental pathway
The long delay between the onset of vascular budding and establishment of normal colonies might render it unfavorable compared with other budding modes. Assuming that reproduction by vascular budding in B. schlosseri is a viable but unfavorable pathway reserved for loss of other budding modes, there should be a mechanism to prevent it under normal circumstances. We found that when leaving even a single bud in the colony, or if untreated colonies with zooids are placed in the same 4 L tank, no vascular budding occurs. The prevention of vascular budding induced by distant buds or zooids suggests that suppressive diffusible factors are involved. These suppressive factors may be activated by the buds or zooids themselves or could also be activated by the vasculature (or even the tunic) affected by the presence of normal buds or zooids. Inhibition of one budding mode by groups of cells or by a different mode has been reported in colonial tunicates (39 , 40) . Moreover, diffusible substances secreted by parent zooids have been shown to affect the behavior of growing buds (41 42 43) . The defined substances thyroxine and thiourea have been implicated in either suppression or acceleration of budding, respectively (44) . Tunicates, the closest living relative of vertebrate (45) , evolve various patterns of development. While solitary tunicates propagate exclusively by sexual reproduction, colonial tunicates reproduce both sexually and asexually, and are capable of producing buds (2 , 7) . Colonial tunicates use different modes of budding for both proliferation and survival (5 , 6) . Two budding features differentiate botryllid ascidians from other tunicates: Botryllid ascidians are the only tunicates with the capacity to produce palleal buds (in which the bud is derived from the lateral wall of the parent and involves three layers: atrial epithelium, blood, and epidermis). This subfamily is also the only one in which the developmental stages of buds are synchronized in the colony (2) . Whereas in some botryllid ascidians vascular budding occurs spontaneously (either simultaneously with palleal budding: Botryllus primigenus and Botryllus delicates; see refs. 23 , 28 ) or when the colony recovers from aestivation (Botrylloides sp; refs. 24 25 26 27 ), in Botryllus schlosseri vascular budding is strictly an inducible process. Under normal laboratory and field conditions, palleal budding is the only mechanism of asexual development observed in this species.

Role of blood circulation in supporting the program of vascular budding
The high correlation between blood flow rate and the ability to regenerate via vascular budding suggest that high flow rate is essential for successful regeneration. Dependency on high flow rate might account for the failure of regenerating at stages A–C, which have sluggish blood flow rates compared with stage D (Fig. 1D ; video S5). Enhanced blood flow can encourage regeneration by increasing the supply of oxygen and nutrients to the regenerating area. High demand for oxygen and nutrients by the developing bud can also suggest the necessity of heart development at early stages of regeneration in the first 2 days of vascular budding compared with 10 days during normal palleal budding.

In the absence of buds and zooids, the ampullae (46 , 47 ; Supplemental Video S1) and the vasculature keep the circulation by synchronized contractions (Supplemental Video S1). These contractions may be sufficient to drive the initial blood circulation required for the formation of a heart. Further regeneration is likely to depend on a proper blood circulation maintained by the heart. This need for high flow rate can explain the accelerated heart formation, as observed in this work.

Multiple routes to the same body form: B. schlosseri as a model for alternative embryonic programs
B. schlosseri appears to exhibit at least three independent embryonic programs (i.e., sexual reproduction, asexual palleal budding and vascular budding), all progressing through distinct stages and body forms and leading eventually to indistinguishable colonies. The intriguing convergence of three distinct programs into a single morphological module, makes it an attractive model for studying alternative specifications of positional information. This includes mechanisms for context-dependent regulation of developmental process, and mechanisms for switching between reproduction modes (e.g., from palleal budding to sexual reproduction or to vascular budding).

The relatively high abundance of colonies that fail to complete the program of vascular budding (86% n=51 Table 1 , experiments 1–6, 9) is likely to assist in understanding the mechanisms underlying the various stages of vascular budding. Further characterization of the stepwise variations in the spatial patterns of positional clues and hox gene expression in B. schlosseri would likely shed new light on the mechanisms for maintaining and switching between alternative modes of embryonic development.


   ACKNOWLEDGMENTS
 
We thank Elizabeth Moiseeva for histological work of art, Karla Palmeri, Amir Voskoboynik, and Stefano Tiozzo for critical reviews of the manuscript, and Libue Jerberak, Chris Patton, and Randy Will for technical and mariculture assistance. This study was supported by the National Institutes of Health (R01-DK54762) and by a grant from the U.S.-Israel Binational Science Foundation (2003/010). N.S.-B. was a doctoral fellow of the Ch. Clore Foundation.

Received for publication September 27, 2006. Accepted for publication December 25, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Berrill, N. J. (1941) The development of the bud in Botryllus. Biol. Bull. 80,169-184[Abstract/Free Full Text]
  2. Berrill, N. J. (1950) Sessile Tunicates. Berrill, N. J. eds. The Tunicata with an Account of the British Species ,1-265 Adlar and Son, Ltd. Ray Society London.
  3. Berrill, N.J. (1975) Chordata: Tunicata. Giese, A. C. Pearse, J. S. eds. Reproduction of marine invertebrates 2,241-282 Academic Press New York.
  4. Brien, P. (1968) Blastogenesis and morphogenesis. Adv. Morphog. 7,151-203[Medline]
  5. Nakauchi, M., Kawamura, K. (1986) Asexual reproduction in ascidians. Porchet, M. Andries, J. C. Dhainaut, A. eds. Advances in Invertebrate Reproduction ,49-54 Elsevier Science Publishers New York.
  6. Nakauchi, M., Kawamura, K. (1990) Problems on the asexual reproduction of ascidians. Hoshi, M. Yamashita, O. eds. Advances in Invertebrate Reproduction ,49-54 Elsevier Science Publishers New York.
  7. Manni, L., Burighel, P. (2006) Common and divergent pathways in alternative developmental processes of ascidians. Bioessays 28,902-912[CrossRef][Medline]
  8. Millar, R. (1971) Russell, R. H. Young, M. eds. Advances in Marine Biology ,1-100 Allen and Unwin London.
  9. Satoh, N. (1994) Developmental biology of colonial ascidians. Satoh, N. eds. Developmental Biology of Ascidians ,169-202 Cambridge Univ. Press Cambridge.
  10. Mukai, H., Watanabe, H. (1976) Relation between sexual and asexual reproduction in the compound ascidian, Botryllus primigenus. Sci. Rep. Fac. Edu. Gunma Univ. 25,61-79
  11. Burighel, P., Cloney, R. A. (1997) Urochordata: Ascidiacea. Harrison, F. W. Ruppert, E. E. eds. Microscopic Anatomy of Invertebrates ,221-347 Wiley-Liss, Inc. NY.
  12. Berrill, N. J. (1941) Size and morphogenesis in the bud of Botryllus. Biol. Bull. 80,185-193[Abstract/Free Full Text]
  13. Izzard, C. S. (1973) Development of polarity and bilateral asymmetry in the palleal bud of Botryllus schlosseri (Pallas). J. Morphol. 139,1-26[CrossRef]
  14. Lauzon, R. J., Ishizuka, K. J., Weissman, I. L. (1992) A cyclical, developmentally regulated death phenomenon in a colonial urochordate. Dev. Dyn. 194,71-83[Medline]
  15. Lauzon, R. J., Patton, C. W., Weissman, I. L. (1993) A morphological and immunohistochemical study of programmed cell death in Botryllus schlosseri (Tunicata, Ascidiacea). Cell Tissue Res. 272,115-127[CrossRef][Medline]
  16. Watanabe, H. (1953) Studies on the regulation in fused colonies in Botryllus primigenus (Ascidiae compositae). Sci. Rep. Tokyo Bunrika Daigaku Sec. B 7,183-198
  17. Moore, K. H., Lemischka, I. R. (2006) Stem cells and their niches. Science 311,1880-1885[Abstract/Free Full Text]
  18. Watkins, M. J. (1958) Regeneration of buds in Botryllus. Biol. Bull. 115,147-152[Abstract/Free Full Text]
  19. Milkman, R., Byrne, S. (1961) Recent observations on Botryllus schlosseri. Biol. Bull. 121,376
  20. Milkman, R. (1967) Genetic and developmental studies on Botryllus. schlosseri. Biol. Bull. 132,229-243[CrossRef]
  21. Sabbadin, A., Zaniolo, G., Majone, F. (1975) Determination of polarity and bilateral asymmetry in palleal and vascular buds of the Ascidian Botryllus schlosseri. Dev. Biol. 46,79-87[CrossRef][Medline]
  22. Bancroft, F. W. (1903) Aestivation of Botrylloides gascoi Della Valle. Mark Anniv. Vol. Article VIII,147-166
  23. Oka, H., Watanabe, H. (1957) Vascular budding: a new type of budding in Botryllus. Biol. Bull. 112,225-240[Abstract/Free Full Text]
  24. Oka, H., Watanabe, H. (1959) Vascular budding in Botrylloides. Biol. Bull. 117,340-346[Abstract/Free Full Text]
  25. Burighel, P., Brunetti, R., Zaniolo, G. (1976) Hibernation of the colonial ascidian Botrylloides leachi (Savigny): histological observations. Boll. Zool. 43,293-301
  26. Rinkevich, B., Shlemberg, Z., Fishelson, L. (1995) Whole-body regeneration from totipotent blood cells. Proc. Natl. Acad. Sci. U. S. A. 92,7695-7699[Abstract/Free Full Text]
  27. Rinkevich, B., Shlemberg, Z., Fishelson, L. (1996) Survival budding processes in the colonial tunicate Botrylloides from the Mediterranean Sea: the role of totipotent blood cells. Maramorosch, K. Loeb, M. J. eds. Invertebrate Cell Culture: Looking Towards the 21st Century ,1-9 Society for In Vitro Biology Largo, MD, USA.
  28. Okuyama, M., Saito, Y. (2001) Studies on Japanese botryllid ascidians. A new species of the genus Botryllus from the Izu Islands. Zool. Sci. 18,261-267[CrossRef]
  29. Rinkevich, B., Weissman, I. L. (1987) The fate of Botryllus (Ascidiacea) larvae co-settled with parental colonies: beneficial or deleterious consequences?. Biol. Bull. 173,474-488[Abstract/Free Full Text]
  30. Gretchen, L. H. (1967) Fixation and specific staining methods. Emerson, R. Kennedy, D. Park, R. B. eds. Animal Tissue Technique ,14-163 W. H. Freeman and Company San Francisco and London.
  31. Laird, D. J., De Tomaso, A. W., Weissman, I. L. (2005) Stem cells are units of natural selection in a colonial ascidian. Cell 123,1351-1360[CrossRef][Medline]
  32. Gilbert, S. F. (2000) Patterns of development. Gilbert, S. F. eds. Developmental Biology ,121-383 Sinauer Associates, Inc. Sunderland, MA, USA.
  33. Wolpert, L. (2001) Wolpert, L. eds. Principles of Development ,484 Oxford University Press New York.
  34. Robertson, E. J. (1987) Robertson, E. J. eds. Teratocarcinomas and Embryonic Stem Cells: A Practical Approach ,1-18 IRL Press Oxford, UK.
  35. Bonnet, D., Dick, J. E. (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat. Med. 3,730-737[CrossRef][Medline]
  36. Rossi, D. J., Weissman, I. L. (2006) Pten, tumorigenesis and stem cell self renewal. Cell 125,229-231[CrossRef][Medline]
  37. Sabbadin, A. (1956) "Situs inversus viscerum" provocato sperimentalmente in Botryllus schlosseri (Pallas). Rend. Acad. Naz. Lincei 20,659-666
  38. Tiozzo, S., Christiaen, L., Deyts, C., Manni, L., Joly, J. S., Burighel, P. (2005) Embryonic versus blastogenetic development in the compound ascidian Botryllus schlosseri: insights from Pitx expression patterns. Dev. Dyn. 232,468-478[CrossRef][Medline]
  39. Freeman, G. (1971) A study of the intrinsic factors which control the initiation of asexual reproduction in the tunicate Amaroucium constellatum. J. Exp. Zool. 178,433-456[CrossRef]
  40. Fujimoto, H., Watanabe, H. (1976) Studies on the asexual reproduction in the polystelid ascidian, Polyzoa vesiculiphora. J. Morphol. 150,607-622[CrossRef]
  41. Nakauchi, M., Kawamura, K. (1974) Behavior of buds during common cloacal system formation in the ascidian Aplidium multiplicatum. Rep. USA Mar. Biol. Stat. 21,19-27
  42. Nakauchi, M., Kawamura, K. (1974) Experimental analysis of the behavior of buds in the ascidian Aplidium multiplicatum. Rep. USA Mar. Biol. Stat. 21,29-38
  43. Nakauchi, M., Kawamura, K. (1978) Additional experiments on the behavior of buds in the ascidian Aplidium multiplicatum. Biol. Bull. 154,453-462[Abstract/Free Full Text]
  44. Fukumoto, M. (1971) Experimental control of budding and stolon elongation in Perophora orientalis, a compound ascidian. Dev. Growth Differ. 13,73-88[CrossRef][Medline]
  45. Delsuc, F., Brinkmann, H., Chourrout, D., Phillippe, H. (2006) Tunicates and not Cephalochordates are the closest living relatives of vertebrate origins. Nature 439,965-968[CrossRef][Medline]
  46. De Santo, R. S., Dudley, P. L. (1969) Ultramicroscopic filaments in the ascidian Botryllus schlosseri (Pallas) and their possible role in ampullar contractions. J. Ultrastruct. Res. 28,259-274[CrossRef][Medline]
  47. Sabbadin, A. (1982) Formal genetics of ascidians. Am. Zool. 22,765-773




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
fj.06-7337comv1
21/7/1335    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Voskoboynik, A.
Right arrow Articles by Weissman, I. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Voskoboynik, A.
Right arrow Articles by Weissman, I. L.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS