Printer Friendly

Hematological Analysis of the Ascidian Botrylloides leachii (Savigny, 1816) During Whole-Body Regeneration.

Abstract. Whole-body regeneration (WBR)--the formation of an entire adult from only a small fragment of its own tissue--is extremely rare among chordates. Exceptionally, in the colonial ascidian Botrylloides leachii (Savigny, 1816) a fully functional adult is formed from their common vascular system after ablation of all adults from the colony in just 10 d, thanks to their high blastogenetic potential. While previous studies have identified key genetic markers and morphological changes, no study has yet focused on the hematological aspects of regeneration despite the major involvement of the remaining vascular system and the contained hemocytes in this process. To dissect this process, we analyzed colony blood flow patterns using time-lapse microscopy to obtain a quantitative description of the velocity, reversal pattern, and average distance traveled by hemocytes. We also observed that flows present during regeneration are powered by temporally and spatially synchronized contractions of the terminal ampullae. In addition, we revised previous studies of B. leachii hematology as well as asexual development using histological sectioning and compared the role played by hemocytes during WBR. We found that regeneration starts with a rapid healing response characterized by hemocyte aggregation and infiltration of immunocytes, followed by increased activity of hemoblasts, recruitment of macrophage-like cells for clearing the tissues of debris, and their subsequent disappearance from the circulation concomitant with the maturation of a single regenerated adult. Overall, we provide a detailed account of the hematological properties of regenerating B. leachii colonies, providing novel lines of inquiry toward the decipherment of regeneration in chordates.


Whole-body regeneration (WBR) is the process whereby an entire functional adult is formed from only a minute fragment of the original organism. Multicellular animals capable of varying degrees of regeneration are distributed widely throughout the metazoa (Sanchez Alvarado and Tsonis, 2006), thus suggesting that this biological phenomenon has a primordial origin. However, this regeneration ability correlates inversely with body and tissue complexity (Alvarado, 2000; Rinkevich et al., 2007).

Among vertebrates, while most adults heal only through scarring, teleost fish and urodele amphibians can regenerate tissues, body parts, or organs after injury (Tanaka and Reddien, 2011; Seifert et al., 2012; Jazwihska and Sallin, 2016). Although the study of these organisms has brought an extensive knowledge on patterning and cellular programs required for regeneration in mature animals after severe injury, no vertebrate has yet been shown to undergo WBR. Therefore, deciphering the mechanisms underlying regeneration in evolutionarily related organisms might provide new implications about the basic principles of cellular plasticity in adult organisms of our phylum.

Tunicates represent the closest phylogenetic relative of vertebrates (Delsuc et al., 2006). In particular, colonial ascidians can undergo WBR as a result of their high blastogenetic potential (Oka and Watanabe, 1957, 1959; Milkman, 1967; Zaniolo et al., 1976; Rinkevich et al., 1993). Belonging to phylum Chordata, class Ascidiacea, these invertebrates have tissue and organ complexity incipient to that of vertebrates (see Fig. 1), including a well-developed notochord, dorsal nerve cord, and postanal tail during their free-living tadpole larval stage. In particular, BotryUoides leachii (order Stolidobranchia, family Styelidae; Savigny, 1816) can undergo WBR in as few as 10 d (Rinkevich et al., 1995, 2007, 2008, 2010; Zondag et al., 2016). Consequently, B. leachii has recently emerged as a model organism for the study of regeneration (Gross, 2007).

BotryUoides leachii is a sessile suspension-feeding ascidian that lives in clonal colonies of adults, termed zooids, that organize in a series of ladder-like parallel rows, known as systems (Fig. 1A, B; Savigny, 1816; Michaelsen, 1921; Brewin, 1946). The colony attaches to its substrate, typically on seagrass or underneath rocks in the shallow subtidal zone, through a gelatinous tunic. While each zooid has an independent heart and an open hemocoelic circulatory system, the entire colony shares a common vascular system that is embedded in the tunic and closed at its periphery by terminal ampullae, the contractile blind ends of marginal vessels (Fig. 1C-F; Mukai et al., 1978). Botrylloides leachii zooids are hermaphrodites and can reproduce both sexually, to colonize new locations through a tadpole larval stage, and asexually, to expand the colony (Berrill, 1947; Mukai et al., 1987). This latter reproduction, also known as palleal budding (Oka and Watanabe, 1959) or blastogenesis (Brunetti, 1976), occurs on the outer epithelial mantle of a zooid, typically on both of its sides, thus producing two offspring. These growing buds start as a thickened disk of cells on the body wall, protrude outward, and invaginate their inner layer to form the required layers of tissue for organ development. Eventually, these buds, located on either side of the older zooid, will mature into new adults, while their parent will be resorbed by the colony (Berrill, 1947). As in the closely related genus Botryllus (Berrill, 1941; Sabbadin, 1969), bud development is synchronous throughout the colony and starts very early in the life cycle of the parent bud (Berrill, 1947; Brunetti, 1976).

In addition to these two methods of reproduction, B. leachii colonies can undergo WBR from a minute fragment of their vascular tissue (~200 cells; Rinkevich et al., 1995). WBR will be triggered only after the loss of all zooids from the colony; otherwise, a more traditional injury healing will occur (Rinkevich et al., 1995). Botrylloides leachii WBR has been reported to start with the healing of the injury sites to prevent further hemolymph loss, followed by the compaction of the marginal ampullae toward the center of the remaining matrix. Within this reorganized tissue, regeneration niches--discrete regeneration loci within the vascular system--will develop and apparently compete through a yet-undetermined process that ultimately leads to the development of a single adult zooid, while all other niches are absorbed by the colony (Fig. 1G). Importantly, WBR ability in B. leachii is retained throughout its developmental cycle (Oka and Watanabe, 1959; Rinkevich et al., 2007), in contrast to the sister species Botryllus schlosseri, where WBR potential is present only during a 1-d period of its asexual reproduction cycle, known as takeover (Voskoboynik et al., 2007; Rinkevich et al., 2008). While the exact origin of the totipotent stemlike cells responsible for WBR remains unknown, it has been reported that even though they will develop a regeneration niche within the vascular system, they are not part of the circulating blood cells (hemocytes) in uninjured colonies but rather appear to migrate from the vascular lining into the vascular system after injury (Rinkevich et al., 2010).

Similar to most other chordates, ascidian hemocytes are continually renewed to compensate for cell death and external sources of loss (Ermak, 1982). The location of hemopoiesis in ascidians is still debated, which might be a consequence of the variety of developmental strategies observed among this clade (Brown and Swalla, 2012). In some solitary ascidians, it principally occurs in small static clusters of hemoblasts located in the lining of the hemocoel, particularly around the pharyngeal basket (Ermak, 1982). In colonial ascidians, clusters of undifferentiated cells with pluripotent properties, called islands, have been detected in the sub-endostylar ventral sinus (Voskoboynik et al., 2008). In addition, ventral islands located near the endostyle have been identified as sites of phagocyte turnover (Lauzon et al., 2013). Furthermore, circulating hemoblasts have been shown to express stemness and proliferation markers around early buds of Botrylloides violaceus, suggesting a circulatory source of hemocytes (Brown et al., 2009). It thus appears that there may be multiple sites for ascidian blood (hemolymph) production, in particular in colonial tunicates. Hemolymph is mostly colorless and composed of a variety of cell types with functions incipient to those of vertebrates (Goodbody, 1975; Wright, 1981; Millar and Ratcliffe, 1989; Cima et al., 2002), and its flow patterns are intimately linked to the health of the whole colony (Dijkstra et al., 2008).

Although there have been a number of studies investigating the morphotypes of ascidian hemocytes (Endean, 1960; Freeman, 1964; Wright and Ermak, 1982; de Leo, 1992; Hirose et al., 2003; Bailarin and Cima, 2005; Ballarin et al., 2011), including of those of B. leachii (Cima et al., 2002), improved resolution and color images would benefit future work on this species. In addition, no common terminology has been accepted to identify these cell types, mainly because of discrepancies between their phenotype and their function in different species as well as a lack of understanding of their differentiation lineages (Endean, 1960; Freeman, 1964; Goodbody, 1975; Wright. 1981; Millar and Ratcliffe, 1989; Cima et al., 2002; Ballarin et al., 2011). Furthermore, accurate identification is crucial for detecting and interpreting variations in the subpopulations of hemocytes during WBR.

Despite the major involvement of the remaining vascular system and the comprised hemocytes during WBR in B. leachii, no study has yet focused on the hematological aspects of regeneration. To obtain a more accurate understanding of the WBR process in B. leachii, we set out to characterize, in a consistent and amended approach, both uninjured and regenerating colonies at the circulatory, tissue, and cellular levels. We used both histological staining and time-lapse recording to dissect this highly dynamic process of regeneration with great resolution.

Materials and Methods

Animal husbandry and manipulation

Botrylloides leachii colonies were collected from two sites in Otago Harbor (45[degrees]52'17" S, 170[degrees]31'43" E; 45[degrees]49'41" S, 170[degrees]38'29" E) and from Nelson Harbor (41[degrees]15'36" S, 173[degrees] 16'48" E), New Zealand, between September and March 2014-2016. Batrylloides leachii grows naturally on submerged structures (e.g., ropes, pontoons, and tires), and colonies were removed from the attached substrata with a singleedge razor blade. Each colony was placed on either a 5.0 x 7.5-cm or 2.6 x 7.6-cm glass slide and left horizontally for 2 d in 200 ml of still seawater, changed daily, to allow the colony to attach to the slide. The slides were then placed vertically and kept at the Portobello Marine Laboratory (University of Otago, Dunedin, New Zealand) in a tank supplied with a constant flow of filtered seawater (FSW) directly from the harbor, at a temperature maintained between 18 and 20 [degrees]C. The glass slides were kept clear of algal growth by using a paintbrush and a single-edge razor blade (after Rinkevich et al., 2007). Colonies were continuously fed. using a peristaltic pump, a 1 : 1 : 1 : 1 by volume mixture of three algae (Chaetoceros muelleri, Pavlova lutheri, and Tetraselmis chuii) and a rotifer (Brachionus plicatilis) culture, complemented with Alltech All-G-Rich algae/yeast powder mix (Alltech, Lexington, KY) (12 g of Alltech All-G-Rich mix was added each week to the rotifer culture).

Botrylloides leachii hemolymph was collected as described elsewhere (Cima, 2010). Briefly, whole colonies were first cleaned using FSW and incubated for 5 min in anticoagulant solution (0.38 g of sodium citrate per 100 ml of FSW [pH 7.5]). To prevent loss of hemocytes, all instruments were incubated twice for 1 min in the anticoagulant solution prior to collection. Colonies were then dried using blotting paper, and their marginal vessels were broken using a needle. The hemolymph was then collected using a syringe and transferred into a 1.5-ml microfuge tube containing 100 [micro]l of the anticoagulant solution up to a total volume of 600 [micro]l. The tube was left on ice for 5 min, and the supernatant was transferred to a new tube to remove debris. After centrifugation at 780 x g for 12 min, the supernatant was discarded, and the pellet was resuspended in 100 [micro]l of FSW.

Botrylloides leachii regeneration was carried out following an established protocol (Rinkevich et al., 2007). Dissection of adults and buds away from the marginal ampullae during their midcycle blastogenic stage (i.e., not undergoing takeover) was carried out using a scalpel and a single-edge razor blade. The slides with the remaining vascular fragments were then placed into an aerated seawater tank at 19 [degrees]C and left to regenerate until a certain morphological stage had been reached.

Histological staining

Botrylloides leachii regenerating tissue fragments were fixed overnight in 4% paraformaldehyde in FSW, dehydrated in 70% ethanol, and embedded in paraffin wax. The fragments were then sectioned (5 [micro]m thickness) in the transverse plane. Three stains were used on dewaxed slides: hematoxylin and eosin (H&E), Giemsa, and Martius scarlet blue (MSB) trichrome. H&E staining was performed as follows: stain for 4 min in hematoxylin (Leica Biosystems Surgipath Gill II hematoxylin; Leica, Wetzlar, Germany), wash for 2 min in running tap water, place the slide for 2 min in Scott's tap water (2 g of potassium bicarbonate, 20 g of magnesium sulphate per liter), wash again for 2 min, stain for 30 s in eosin (Leica Biosystems Surgipath eosin), wash for 1 min, dehydrate, and mount. Giemsa staining was performed as follows: stain for 30 min in a coplin jar containing Giemsa (7.36 g of Giemsa powder in 500 ml of 50 [degrees]C glycerol and 500 ml of methanol, mix 1 : 45 in distilled [H.sub.2]O [d[H.sub.2]O]) within a 56 [degrees]C water bath, rinse in d[H.sub.2]O, differentiate in 1: 1500 acetic acid solution, rinse in d[H.sub.2]O. dehydrate, and mount. MSB staining was performed as follows: stain for 5 min in celestine blue (5 g of ferric ammonium sulphate, 0.5 g of celestine blue B in 100 ml of d[H.sub.2]O with 14 ml of glycerol), wash for 30 s in running tap water, stain for 5 min in hematoxylin, wash again for 2 min, place the slide for 2 min in Scott's tap water, wash for 2 min, rinse for 1 min in 95% ethanol, stain for 2 min in Martius yellow (20 g of phosphotungstic acid, 5 g of Martius yellow per liter of 95% ethanol), rinse three times for 1 min in d[H.sub.2]O, stain for 10 min in brilliant scarlet (10 g of brilliant scarlet crystal ponceau 6R, 20 ml of glacial acetic acid per liter of d[H.sub.2]O), rinse again three times, place the slide for 4 min in phosphotungstic acid (10 g per liter of d[H.sub.2]O), rinse for 1 min in d[H.sub.2]O, stain for 45 s in methyl blue (5 g of methyl blue, 10 ml of glacial acetic acid per liter of d[H.sub.2]O), rinse for 1 min in acetic acid (10 ml of glacial acetic acid per liter of d[H.sub.2]O), dehydrate, and mount.

Hemolymph smears were obtained as described elsewhere (Cima. 2010). Briefly, drops of 50 [micro]l of isolated hemocytes (see above) were left for 30 min to attach on SuperFrost Plus microscopy slides (Thermo Fisher Scientific, Waltham. MA). The drop was then discarded by placing the slide vertically, and the remaining attached cells were fixed for 30 min at 4 [degrees]C in ascidian fixative solution (1 g of NaCl and 1 g of sucrose in 1% glutaraldehyde in FSW). Slides were then rinsed for 10 min in 0.1 mol [1.sup.-1] phosphate-buffered saline (PBS), stained using 10% Giemsa, mounted using a 9: 1 glycerol/PBS solution, and sealed with a coverslip using nail polish.


All fixed samples were imaged on an Olympus AX70 microscope (Olympus. Tokyo, Japan) equipped with a QImaging MicroPublisher 5.0 camera (2560 x 1920 pixels) using QCapture software. Magnification used ranged from 4 to 100 x. Time-lapse recordings of hemolymph flow in live colonies were acquired on a Leica M205 FA stereomicroscope equipped with an 8-megapixel CCD DFC490 color camera using the Leica Application Suite. Recordings were acquired using the "Movie" continuous mode at different magnifications (7.42-20 x) and exposure times (27.7-55.7 ms) adapted to the recorded portions of the colony, with a resolution of either 544 x 408 or 1088 x 816 pixels.

Data processing

Acquired images were processed for color balance using the open-source GIMP program (GIMP Development Team) and assembled into figures using the open-source Inkscape software (Inkscape Project). Colors were chosen according to the ColorBrewer palette (C. A. Brewer, GeoVISTA Center, Pennsylvania State University). Movies were montaged using the Lightworks program (EditShare) and compressed using the open-source HandBreak tool (Handbreak Team).

Hemolymph composition during WBR was estimated by manually identifying cell types in microscopy images using five randomly located images of Giemsa-stained ampullae, at 60 x magnification. Similarly, 10 images were used to characterize hemolymph composition in intact colonies fixed during their midcycle blastogenic stage. Only cells located inside the ampullae, centered on the imaged section, and unambiguously identified (2.6% of the total 2339 cells could not be determined) were considered. Between 272 and 447 cells were positively identified for each stage of regeneration, 502 for the intact colony. We focused our counting on ampullae for setting up a meaningful comparison between intact and regenerating tissues, as blood vessels become very sparse after tissue contraction.

The appearance of hemocytes was characterized using light microscopy by smearing ascidian hemolymph and by staining histological sections of whole colonies.

Computational analysis

Vessel identification in time-lapse recordings was performed as follows. Raw frames of the recording were smoothed using a Gaussian kernel of radius 0.89 [micro]m, the difference between consecutive frames was computed, and moving cells were identified as having an absolute value greater than six times the estimated standard deviation of the image's white noise (Paul et al., 2010). Cells larger than 5 [micro][m.sup.2] were then consecutively morphologically dilated and eroded, using structuring disks of radii 41 and 21 [micro]m, respectively. The resulting detections were then averaged over the entire recording, and vessels were defined as containing moving cells in at least 30% of the frames. The location, length, and width of the vascular systems were then identified using the morphological skeleton of the vessels,

Hemolymph flow was measured using a particle image velocimetry (PIV) approach with multiple passes and subpixel resolution using Fourier space (Liao and Cowen, 2005) based on the difference between consecutive frames. Raw frames of the recording were smoothed using a Gaussian kernel of radius 0.89 [micro]m. the difference between consecutive frames was computed, and PIV was performed between two consecutive differences, using a modified version of matpiv_nfft from the MatPIV toolbox (J. K. Sveen, University of Oslo), through four consecutive passes with interrogation windows of sizes 83 x 83, 62 x 62, 41 x 41, and 21 x 21 [micro]m, and finalized with a subpixel pass in Fourier space. Only the speeds measured inside the previously detected vessel segments were retained and projected onto the tangent of the segment. The orientation of each segment was then adjusted on the basis of the maximal correlation between average speeds. Speeds from all aligned segments were then binned into a two-dimensional histogram of resolution 350 ms x 6 [micro]m [s.sup.-1], and the most likely hemolymph flow was computed using a shortest-path approach (Dijkstra, 1959), implemented using dynamic programming (Bellman, 1952). Finally, to dampen the effect of the binning required by dynamic programming, the resulting hemolymph flow was smoothed using cubic smoothing splines.

Ampullar contractions were measured as follows. We first identified ampullae in every frame as regions darker than 20 times the estimated standard deviation of the image's white noise (Paul et al., 2010). The resulting binarized image was then consecutively morphologically opened and closed, using a structuring disk of radius 9.5 [micro]m. The area and the centroid of each separated object were computed, and the trajectory of each ampulla was reconstructed by clustering across frame detections whose centroids were closer than 19 [micro]m. Only trajectories that spanned more than 95% of the frames were kept for further analysis. All code written for our analysis was implemented as a set of custom MATLAB R2015b (Math-Works, Natick, MA) functions and is available on request.


Hemolymph circulation

As a first step toward dissecting WBR. we characterized the hematological properties of a healthy and intact adult colony of Botrylloides leachii (Fig. 1A, B). Hemolymph circulation in suspension-feeding ascidians undergoes a periodic reversal of flow direction between advisceral (toward the viscera) and abvisceral (toward the branchial basket). The advisceral flow starts on the anterior side of the heart; moves toward the endostyle before arriving at the pharynx and the middorsal vessel, which supplies hemolymph to the digestive tract; and lastly passes to the dorsal end of the heart (Mukai et al., 1978).

Utilizing the transparency of the tunic, we recorded timelapse stereomicroscopy data from eight colonies at different locations in their vascular system (Fig. 1C-F; video 1, available online). Focusing on the largest vessels of the colony (videos 1, 2, available online), we gathered a set of reproducible flow measurements (n = 9, from five different colonies) that allowed us to calculate, using a custom-made PIV-based tracking software, the flow (peak velocity, ~200 [micro]m [s.sup.-1]) and reversal (abvisceral and advisceral duration ~57 s and ~24 s, respectively) rates of a B. leachii colony (Fig. 2A-C).

When recording vascular junctions, we observed (video 1, available online) and quantified (video 3, available online) an apparently pseudoerratic pattern at the time of flow reversal. By examining terminal vessels (video 1, available online), we extracted valuable information on the rate (11.4 [+ or -] 5.3 [micro][m.sup.-2] [s.sup.-1]; n = 18) and extent (dilated-to-contracted surface ratio. 1.55 [+ or -] 0.26; n = 12) of the ampullar contraction-dilatation cycle concomitant with hemolymph flow in ascidians (Mukai et al., 1978). While the general timing of hemolymph and ampullae alternation is synchronized (average lag, 2.9 [+ or -] 2.5 s; n = 24), small variations in the exact time of contraction for each ampulla could be measured ([+ or -]3.1 s; n =24; Fig. 2D-F).

Hemocyte classification

By combining previous classification schemes for botryllid ascidians (Cima et al., 2002; Hirose et al., 2003; Bailarin and Cima, 2005; Bailarin et al., 2011), we propose a generalized classification of B. leachii circulatory hemocytes into five functional groups: undifferentiated cells, immunocytes, mast cell-like cells, transport cells, and storage cells (Table 1). A further detailed description of the observed cell types is provided in Table A1. In this classification, undifferentiated cells are composed of hemoblasts, also called lymphocytes or lymphocyte-like cells, as well as of differentiating cells. Immunocytes can be further classified between phagocytes and cytotoxic cells. Phagocytes include hyaline amebocytes and macrophage-like cells, while cytotoxic cells include granular amebocytes and morula cells. Mast-like cells are represented only by granular cells. Transport cells are composed of compartment amebocytes and compartment cells. Finally, storage cells include both pigment cells and nephrocytes. A comprehensive guide to hemocyte identification using brightfield Giemsa-stained images is provided (Table 2), with highmagnification pictures exemplifying the most common aspect of each cell type (Fig. 3).

Altogether, these descriptions (Table 2) and images (Fig. 3) allowed us to identify most of the cells in both hemolymph smears and histological sections and to provide an unambiguous identification reference for the analysis of the localization and fate of B. leachii hemocytes in the colony.

Asexual budding

To obtain a precise picture of the prevailing mode of asexual reproduction, we revised various stages of blastogenesis in B. leachii previously described by Berrill (1941, 1947) using histological sections. Palleal reproduction involves the formation of a new bud from the peribranchial epithelium of a zooid. Remarkably, in B. leachii, blastogenesis starts very early in the development of a maturing bud. There are typically three generations coexisting in the colony (Fig. 4A): secondgeneration buds (budlets) that are developing from more differentiated buds (first-generation buds), which themselves are attached to adult filter-feeding zooids (blastozooids).

The first visible sign of blastogenesis is the appearance of a bud disk, a thickened disk of cells on the ventral side of the atrial epithelium of a first-generation bud, just before the stage where its stigmata become perforated. In healthy colonies, one bud disk usually appears on each side of the forming pharyngeal basket (Fig. 4B), thus producing two budlets and increasing the overall size of the colony. In well-fed colonies, one of these buds can even divide to give rise to two buds on one side of the zooid, producing one further zooid (Berrill, 1947). The disk cells then proliferate until the bud reaches about 12 cells in diameter (day 1 of the second-generation development), protrudes outside the first-generation bud through its mantle into the tunic (Fig. 4C), and curves to form a hollow sphere of cells inside a mantle pouch highly reminiscent of the early stages of blastogenesis (days 2-3). This pouch connects the growing bud with the hemocoel of the mother bud, ensuring that hemolymph and oxygen reach the new bud, while the rudiment for the vascular connection to the rest of the colony will be initiated underneath by an outgrowth of the epithelium layer (Berrill, 1947), later producing the radial vessel (Burighel and Brunetti, 1971).

The asexually developing bud continues to increase in size and begins to fold, forming three main compartments, which will form the internal structures of the new adult (days 4-6; Fig. 4D). It is at that stage that the first signs of germ cells can also be observed (Fig. 4D). The folds will ultimately join, creating the basic structure for the pharyngeal basket (days 78). Once the extending pharyngeal basket has joined the stomach and started to produce its first stigmata, the stereotypical body plan of the zooid is completed, thus becoming a first-generation bud on which new budlets start to appear (day 9; Fig. 4E). After this stage, the first-generation bud will differentiate into a fully functional zooid, hence growing in size, finalizing its various organs and vascular system (days 10-22; Fig. 4F), and starting its cardiac activity around day 17.

In total, it takes around 22 days at 17 [degrees]C to produce a functioning zooid through blastogenesis (Berrill, 1947). Afterward, the zooid will feed for another 8 d before it gets resorbed through whole-body apoptosis (<3 d), in which numerous large macrophage-like cells infiltrate and remove effete cells from the senescent tissues (Franchi et al., 2016) and the next generation succeeds.

Botrylloides leachii WBR

The vascular and hematological characteristics of WBR in B. leachii are described through comparisons with blastogenesis and intact colonies (Fig. 5). For this analysis, we follow our published classification of WBR into five stages (Fig. 1G; Zondag et al., 2016): stage 0, injury of the colony; stage 1, healing of the injury; stage 2, remodeling of the vascular system; stage 3, condensation of the ampullae; stage 4, establishment and development of regeneration niches; and stage 5, fully regenerated zooid.

Stage 0: Injury (0 h). Botrylloides leachii has a great capacity for preventing hemolymph loss from injuries, with hemolymph loss stopped in less than 30 s by a combination of tissue contraction and hemocyte aggregation (video 4, available online). On ablation of the zooids, the remaining vascular tissue initially loses a small volume of hemolymph but quickly contracts its peripheral vessels to halt the oozing, and hemocytes start to aggregate (Fig. 5A). Flow will then restart after a pause, the duration of which is seemingly related to either the number of ampullae or the amount of vascular tissue but was consistently shorter than 10 min. We observed restart of hemolymph flow in as little as 2 min (video 4, available online), starting in the portion of the tissue farthest from the injury in an erratic yet bidirectional fashion and progressively respanning almost the entire vascular system (video 4, available online). Although driven only by ampullar contractions, this novel hemolymph flow exhibits a reversal frequency similar to that of intact vessels in a whole colony. As suspected, given the short amount of time available for new cellular differentiation, the composition of this hemolymph is almost identical to the one from intact colonies, with a slight increase in the population of granular cells (Fig. 5G).

Stage 1: Wound healing (15 h). At this stage of WBR, a dramatic amount of extravascular cellular activity is taking place. The two most apparent aspects are the infiltration of morula cells into the tunic and the remodeling of the tunic matrix (Fig. 5B). When focusing on circulating hemocytes, we observed an incipient decrease in the population of morula cells, accompanied with an increase in the populations of hemoblasts, hyaline amebocytes, and granular cells (Fig. 5G).

Stage 2: Ampulla remodeling (24 h). Concomitant with the observed wound-healing process, terminal ampullae start changing their elongated shape to a more spheroid form, contracting, and creating novel tunic vessels (Fig. 5C; Rinkevich et al., 1995, 2007). One day after injury, previously reported giant cells (Rinkevich et al., 1995, 2007) could be observed throughout the vascular system (Fig. 5C). In addition, this step exhibits the first signs of the subsequent increase in phagocytic cells (Fig. 5G).

Stage 3: Tissue contraction (2-4 d). By stage 3, the vascular system has fully contracted into a dense network (Rinkevich et al., 1995, 2007), and various regeneration niches are visible within the tissue (Fig. 5D). Regeneration niches are defined as spherical aggregates of cells within the vascular lumen. Probably the most striking change at this stage of WBR is the large increase in phagocytic cells (Fig. 5D, G), while numbers of both hemoblasts and differentiating cells return to preinjury levels (Fig. 5G).

Stage 4: Regeneration niches (5-7 d). During stage 4, several regeneration niches are present, and the foundation of an adult can be observed within one of them (Fig. 5E). Nonetheless, only one zooid will regenerate, and the less developed niches will be resorbed (Rinkevich et al., 1995, 2007). Advanced niches already display well-developed axial and tissue organizations reminiscent of the early first-generation stage of blastogenesis (Fig. 4E), including an endostyle as well as an imperforated pharyngeal basket (Fig. 5E). In comparison to palleal budding, the regenerating buds lack visible gametes (cf. Fig. 4E and Fig. 5E).

At this stage of WBR, hemolymph flow can be observed in novel tunic vessels (video 5, available online), and the composition of the hemolymph strives toward the composition observed in intact colonies. Transport, mast cell-like, and storage cells are at their lowest during this period, while hyaline amebocytes and macrophage-like cells remain at a higher proportion than in uninjured colonies (Fig. 5G).

Stage 5: Regenerated zooid (8-10 d). At the conclusion of WBR, a single fully functional zooid is regenerated, displaying the palleal buds expected from an uninjured adult (Fig. 5F). The hemolymph composition of this new adult is largely similar to that of an intact colony, albeit with a slightly smaller phagocytic cell population (Fig. 5G).


Here, we carefully dissected the hematological and histological properties of both intact and regenerating Botrylloides leachii colonies. We compared hemolymph flow patterns to published data from related ascidian species and revised previous studies of B. leachii asexual development. In addition, we followed the quantity and location of each type of hemocyte during WBR, finding that regeneration progresses through a rapid healing response, an increased activity of hemoblasts, the recruitment of macrophage-like cells, and finally their clearing from the hemolymph.

Synchronization of blood flow in intact and regenerating colonies

Using time-lapse recording, we were able to quantify the complex hemolymph flow patterns observed within a B. leachii colony. While the velocity of hemolymph in ascidians has to our knowledge not yet been estimated, the measured average velocity predicts that hemocytes travel an average of 8 mm during one alternation of the flow. In such a small organism (~2 mm in length), this range could be critical for spreading signaling metabolites and thus potentially coordinating colony responses. The values we obtained for the reversal rates are similar to those measured in other ascidian species, albeit slightly more asymmetric and with a relatively shorter advisceral period (21-36 s in B. leachii, 30-50 s in Botryllus primigenus, and 45-60 s in Symplegma reptans, Mukai et al., 1978). We also observed a quick reestablishment of a regular hemolymph flow after injury and despite the removal of all zooids from the vascular system. This flow is thus sustained solely by synchronized and localized patterns of ampullar contractions (Fig. 2G-I; videos 4, 5, available online). While this synchronization could solely be a consequence of hemolymph pressure and elastic vascular tissue (Mukai

et al., 1978), the absence of any zooid heart within the tunic suggests the existence of a yet-unidentified stimulus coordinating this activity.

Heart activity in B. leachii, similar to that in vertebrates (Miquerol and Kelly, 2013), starts before the zooid is fully mature, when capable neither of filter feeding nor of using most of its other organs. Early heart function is consistent with the necessity to circulate oxygen and nutrients throughout the developing organism, even though the parental heart is connected directly to the hemocoel of the bud and would appear to be sufficient for sustaining such circulation. In fact. the pressure exerted by the parental heart is strong enough to impose its direction and reversal rate onto the daughter's heart (Burighel and Brunetti, 1971).

Hemolymph composition in Styelidae

We have reported a detailed characterization of hemocytes in B. leachii. The identified cell types are compatible with previous reports in this species (Cima et al., 2002), comparable to those in other ascidians (Cima et al., 2014) and in particular to those of the same family (Styelidae; Ballarin and Kawamura, 2009; de Barros et al., 2009; Gutierrez and Brown, 2017). Notably, while granular cells appear to be absent from other botryllid species (Hirose et al., 2003; Ballarin et al., 2011), they have been identified in Styela plicata (de Barros et al., 2009), in Symplegma brakenhiemli (Gutierrez and Brown, 2017), and even in Thalia democratica (Cima et al., 2014). Their presence in more distant species of ascidians suggests either a loss in botryllids that did not affect B. leachii or that their visual characteristics are not observable in these botryllids. Further analyses with different staining techniques (Gutierrez and Brown, 2017) could help resolve this issue.

In addition, we here provide a quantification of the composition of hemolymph throughout WBR, which is particularly interesting to compare with the quantification of hemocytes made in other Styelidae (Gutierrez and Brown, 2017). First, morula cells and macrophage-like cells consistently compose the majority of hemocytes in this whole family of ascidians, highlighting the importance of the immune system in these organisms. Second, the amount of differentiating cells observed in Symplegma brakenhiemli (19%), a shorter-lived colonial species that undergoes continuous and asynchronous takeover, corresponds to the levels measured during stage 1 of WBR and emphasizes the strong requirement for tissue generation at this early stage. Third, the fraction of hyalin amebocyte is the lowest among all quantified Styelidae, potentially owing to a faster differentiation into macrophage-like cells. In addition, these hyalin amebocytes could be preferentially present outside the hemocoel and thus be underrepresented in our counting. Finally, the extreme peak of macrophage-like cells measured during stage 3 is much greater than what is observed during the normal life cycle of Styelidae. One potential explanation is that the period of takeover, when the colony is unable to feed, is much shorter or even totally absent in S. brakenhiemli; thus, there could be a lesser need for storage of nutrients by these cells.

Morphological and cellular changes during distinct phases of Botrylloides leachii WBR

We additionally characterized the morphological and hematological modulation of injured colonies throughout WBR. Morula cells are known constituents of the immune system involved in inflammatory responses, hemolymph aggregation, homeostasis, and tunic repair (de Leo, 1992; Bailarin et al., 2001, 2011; Menin et al., 2005). The large number of infiltrated morula cells observed during stage 1of WBR highlights a need for both clearance of decaying or foreign debris and reorganization of the injured vascular system during the first 15 h postdissection. In addition, we measured an increase in the population of differentiating cells (Fig. 5G), consistent with the need to reestablish homeostasis within the B. leachii vasculature and to compensate for hemolymph loss. An increased number of mast cell-like cells, a cell type apparently absent from related ascidian species (Cima et al., 2002), was also observed. While the role played by these cells remains unclear, two hypotheses have been proposed: either a source of nourishment or the immunosurveillance of the alimentary tract (Cima et al., 2002). Their presence during WBR, in the absence of any feeding zooid. rather supports a nutritive role.

At stage 2, we observed the first signs of the subsequent increase in phagocytic cells, potentially as a result of the previously observed differentiation activity (Fig. 5G) and of the inflammatory response induced by the morula cells (Fig. 5B). Given the role played by these macrophage-like cells in the removal of foreign elements from the tissue (de Leo, 1992), their presence during stage 3 of WBR points toward a recruitment for undertaking the clearing of bacteria, dead cells, and residues of remodeled tunic. The concomitant reduction of hemoblasts to preinjury levels suggests that the wound-healing and remodeling phase has been completed and that cellular proliferation and differentiation have returned to hemopoietic levels within the vascular system. Finally, during stage 4, mast cell-like, transport, and storage cells were observed at their lowest levels throughout WBR. This could be a direct consequence of the imposed absence of feeding, leading to a lack of nutrients to be transported or stored. Regenerating B. leachii colonies would then need to obtain nutrients in an alternative way. Similar to the reactivation of hibernating colonies (Burighel et al., 1976) and to the takeover of a new generation of buds in Botryllus schlosseri (Lauzon et al., 2002), extraction of reserve nutrients from macrophage-like cells would supplement this need and gradually reduce the number of phagocytic cells, as observed experimentally (Fig. 5G).

When we focus on vascular circulation, the initial rapid interruption in hemolymph flow (video 4, available online) could be a consequence either of the reduced density of hemolymph inside the vascular system or of an actively regulated mechanism to reduce the pressure at the sectioned vessels and aid aggregation. Given that hemolymph flow restarts over a relatively short timescale compared with that necessary for substantial hemocyte production and pressure increase. the latter hypothesis appears more likely. This observation thus further supports the existence of an unidentified stimulus systemic to the entire vascular system, potentially relayed directly by the epithelial cells lining the vasculature (Mukai et al., 1978).

Similarities and divergences between blastogenesis and WBR

The major discrepancy between blastogenesis and WBR was the absence during WBR of visible gametes (Fig. 5E). As WBR restores both the soma and the germ line (Rinkevich et al., 1995, 2007), the absence of gametes in regenerating niches indicates that they will be created during subsequent cycles of blastogenesis. This process may be an energetically economical developmental approach that avoids the production of gametes in niches that could subsequently be resorbed. Overall, once the vascular tissue has fully contracted, the timing of WBR initially relates closely to that of blastogenesis (days 1-8) but ultimately produces a functional adult much faster. One potential cause for such increased speed is the temperature at which the regenerating colonies were kept (~ 19 [degrees]C in the present study), which is known to directly correlate with developmental pace (Berrill. 1941, 1947). Temperature alone may not entirely explain such a difference in timing, however; and one likely cause could be that functional regenerated zooids are smaller than palleal ones, thus having a shortened growth phase that normally spans more than half of blastogenesis. Indeed, such reduction in zooid size was observed in Botryllus schlosseri (Voskoboynik et al., 2007).


Overall, this study underlines the complex interplay of mechanisms required for successful WBR and complements previous morphological studies (Rinkevich et al., 1995, 2007) by providing a higher temporal resolution. Our histological analysis of WBR supports our recent sequencing approach that identified an initial healing phase followed by a regeneration response (Zondag et al., 2016), and we have provided the first detailed account of the hematological properties of Botrylloides leachii colonies throughout WBR.


We thank Francesca Cima (Department of Biology, University of Padova) for advice, protocols, and general help on hemocytes as well as for proofreading this article; Rueben Pooley (Department of Marine Science, University of Otago) for his help in animal husbandry; Noel Jhinku (Department of Pathology, University of Otago) for providing the rotifers; Kendall Gadomski (Department of Marine Science, University of Otago) for providing the All-G-Rich powder; Kristian Sveen (Institute for Energy Technology, University of Oslo) for developing the original code of MatPIV; Cynthia A. Brewer (GeoVISTA Center. Pennsylvania State University) for developing ColorBrewer; and Femando Romero Balestra (Department of Cell Signaling, Centro Andaluz de Biologia Molecular y Medicina Regenerativa [CABIMER]), Aude Blanchoud. and Anna Jazwinska (Department of Biology, University of Fribourg) for proofreading this article. MJW was funded by a Dean's Bequest Grant and by the Department of Anatomy (Otago University), SB was funded by the Swiss National Science Foundation (grant P2ELP3_158873), and LZ was funded by a University of Otago postgraduate scholarship.

Literature Cited

Alvarado, A. S. 2000. Regeneration in the metazoans: Why does it happen? BioEssays 22: 578-590.

Bailarin, L., and F. Cima. 2005. Cytochemical properties of Botryllus schlosseri hemocytes: indications for morpho-functional characterization. Eur. J. Hislochem. 49: 255-264.

Bailarin, L., and K. Kawamura. 2009. The hemocytes of Polyandrocarpu misakiensis. morphology and immune-related activities. Invertebr. Surviv. J. 6: 154-161.

Bailarin, L., A. Franchini, E. Ottaviani, and A. Sabbadin. 2001. Morula cells as the major immunomodulatory hemocytes in ascidians: evidences from the colonial species Botryllus schlosseri. Biol. Bull. 201: 59-64.

Bailarin, L., M. del Favero, and L. Manni. 2011. Relationships among hemocytes, tunic cells, germ cells, and accessory cells in the colonial ascidian Botryllus schlosseri. J. Exp. Zool. B Mol. Dev. Evol. 316: 284-295.

Bellman, R. 1952. On the theory of dynamic programming. Proc. Natl. Acad. Sci. U.S.A. 38: 716-719.

Berrill, N. J. 1941. The development of the bud in Botryllus. Biol, Bull. 80: 169-184.

Berrill, N. J. 1947. The developmental cycle of Botrylloides. Q. J. Microsc. Sci. 88: 393-407.

Brewin, B. I. 1946. Ascidians in the vicinity of the Portobello Marine Biological Station, Otago Harbour. Trans. R. Soc. N. Z. 76: 87-131.

Brown, F. D., and B. J. Swalla. 2012. Evolution and development of budding by stem cells: ascidian coloniality as a case study. Dew Biol 369: 151-162.

Brown, F. D., E. L. Keeling, A. D. Le, and B. J. Swalla. 2009. Whole body regeneration in a colonial ascidian, Botrylloides violaceus. J. Exp. Zool B. Mol Dev. Evol 312: 885-900.

Brunetti, R. 1976. Biological cycle of Botrylloides leachi (Savigny) (Ascidiacea) in the Venetian lagoon. Vie Milieu 26: 105-122.

Burighel, P., and R. Brunetti. 1971. The circulatory system in the blastozooid of the colonial ascidian Botryllus schlosseri (Pallas). Boll. Zool. 38: 273-289.

Burighel, P., R. Brunetti, and G. Zaniolo. 1976. Hibernation of the colonial ascidian Botrylloides leachi (Savigny): histological observations. Ital J. Zool 43: 293-301.

Cima, F. 2010. Microscopy methods for morpho-functional characterisation of marine invertebrate haemocytes. Pp. 1100-1107 in Microscopy: Science, Technology. Applications and Education. Formatex Microscopy Series 4, A. Mendez-Vilas and J. Diaz, eds. Formatex, Badajoz, Spain.

Cima, F., A. Perin, P. Burighel, and L. Ballarin. 2002. Morpho-functional characterization of haemocytes of the compound ascidian Botrylloides leachi (Tunicata, Ascidiacea). Acta Zool 82: 261-274.

Cima, F., F. Caicci, and P. Sordino. 2014. The haemocytes of the salp Thalia democratica (Tunicata, Thaliacea): an ultrastructural and histochemical study in the oozoid. Acta Zool. 95: 375-391.

de Barros, C. M., D. R. de Carvalho, L. R. Andrade, M. S. G. Pavao, and S. Allodi. 2009. Nitric oxide production by hemocytes of the ascidian Styela plicata. Cell Tissue Res. 338: 117-128.

de Leo, G. 1992. Ascidian hemocytes and their involvement in defence reactions. Boll Zool. 59: 195-214.

Delsuc, F., H. Brinkmann, D. Chourrout, and H. Philippe. 2006. Tunicates and not cephalochordates are the closest living relatives of vertebrates. Nature 439: 965-968.

Dijkstra, E. W. 1959. A note on two problems in connexion with graphs. Numer. Math. (Heidelb.) 1: 269-271.

Dijkstra, J., A. Dutton, E. Westerman, and L. G. Harris. 2008. Heart rate reflects osmostic stress levels in two introduced colonial ascidians Botryllus schlosseri and Botrylloides violaceus. Mar. Biol. 154: 805-811.

Endean, R. 1960. The blood-cells of the ascidian, Phallusia mammillata. J. Cell Sci. s3-101: 177-197.

Ermak, T. H. 1982. The renewing cell populations of ascidians. Am. Zool 22: 795-805.

Franchi, N., F. Ballin, L. Manni, F. Schiavon, G. Basso, and L. Ballarin. 2016. Recurrent phagocytosis-induced apoptosis in the cyclical generation change of the compound ascidian Botryllus schlosseri. Dev. Comp. Immunol 62: 8-16.

Freeman, G. 1964. The role of blood cells in the process of asexual reproduction in the tunicate Perophora viridis. J. Exp. Zool 156: 157-183.

Goodbody, I. 1975. The physiology of ascidians. Adv. Mar. Biol. 12: 1-149.

Gross, L. 2007. From one to many and back again: A systemic signal triggers tunicate regeneration. PLoS Biol 5: e98.

Gutierrez, S., and F. D. Brown. 2017. Vascular budding in Symplegma brakenhielmi and the evolution of coloniality in styelid ascidians. Dev. Biol 423: 152-169.

Hirose, E., M. Shirae, and Y. Saito. 2003. Ultrastructures and classification of circulating hemocytes in 9 botryllid ascidians (Chordata: Ascidiacea). Zool Sci. 20: 647-656.

Jazwinska, A., and P. Sallin. 2016. Regeneration versus scarring in vertebrate appendages and heart. J. Pathol. 238: 233-246.

Lauzon, R. J., K. J. Ishizuka, and I. L. Weissman. 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.

Lauzon, R. J., C. Brown, L. Kerr, and S. Tiozzo. 2013. Phagocyte dynamics in a highly regenerative urochordate: insights into development and host defense. Dev. Biol 374: 357-373.

Liao, Q., and E. A. Cowen. 2005. An efficient anti-aliasing spectral continuous window shifting technique for PIV. E.xp. Eluids 38: 197-208.

Menin, A., M. Del Favero, F. Cima, and L. Ballarin. 2005. Release of phagocytosis-stimulating factor(s) by morula cells in a colonial ascidian. Mar. Biol 148: 225-230.

Michaelsen, W. 1921. Die Botrylliden und Didemniden der Nordsee und der zur Ostsee fuhrenden Meeresgebiete. Wiss. Meeresunters. (Abt. Helgol) 14: 99-124.

Milkman, R. 1967. Genetic and developmental studies on Botryllus schlosseri. Biol. Bull. 132: 229-243.

Millar, D. A., and N. A. Ratcliffe. 1989. The evolution of blood cells: facts and enigmas. Endeavour 13: 72-77.

Miquerol, L., and R. G. Kelly. 2013. Organogenesis of the vertebrate heart. Wiley Interdiscip. Rev. Dev. Biol. 2: 17-29.

Mukai, H., K. Sugimoto, and Y. Taneda. 1978. Comparative studies on the circulatory system of the compound ascidians, Botryllus, Botrylloides and Symplegma. J. Morphol 157: 49-78.

Mukai, H., Y. Saito, and H. Watanabe. 1987. Viviparous development in Botrylloides (compound ascidians). J. Morphol. 193: 263-276.

Oka, H., and H. Watanabe. 1957. Vascular budding, a new type of budding in Botryllus. Biol. Bull. 112: 225-240.

Oka, H., and H. Watanabe. 1959. Vascular budding in Botrylloides. Biol. Bull. 117: 340-346.

Paul, P., H. Duessmann, T. Bernas, H. Huber, and D. Kalamatianos. 2010. Automatic noise quantification for confocal fluorescence microscopy images. Comput. Med. Imaging Graph. 34: 426-434.

Rinkevich, B., Z. Shlemberg, and L. Fishelson. 1995. Whole-body prolochordate regeneration from totipotent blood cells. Proc. Natl. Acad. Sci. U.S.A. 92: 7695-7699.

Rinkevich, Y., G. Paz, B. Rinkevich, and R. Reshef. 2007. Systemic bud induction and retinoic acid signaling underlie whole body regeneration in the urochordate Botrylloides leachi. PLoS Biol. 5: e71.

Rinkevich, Y., B. Rinkevich, and R. Reshef. 2008. Cell signaling and transcription factor genes expressed during whole body regeneration in a colonial chordate. BMC Dev. Biol. 8: 100.

Rinkevich, Y., A. Rosner, C. Rabinowitz, Z. Lapidot, E. Moiseeva, and B. Rinkevich. 2010. Piwi positive cells that line the vasculature epithelium, underlie whole body regeneration in a basal chordate. Dev. Biol. 345: 94-104.

Sabbadin, A. 1969. The compound ascidian Botryllus schlosseri in the field and in the laboratory. Pubbl. Staz. Zool. Napoli 37: 62-72.

Sanchez Alvarado, A., and P. A. Tsonis. 2006. Bridging the regeneration gap: genetic insights from diverse animal models. Nat. Rev. Genet. 7: 873-884.

Savigny, J.-C. 1816. Memoires sur les Animaux sans Vertebres. G. Dufour, Paris.

Seifert, A. W., S. G. Kiama, M. G. Seifert, J. R. Goheen, T. M. Palmer, and M. Maden. 2012. Skin shedding and tissue regeneration in African spiny mice (Acomys). Nature 489: 561-565.

Tanaka, E. M., and P. W. Reddien. 2011. The cellular basis for animal regeneration. Dev. Cell 21: 172-185.

Voskoboynik, A., N. Simon-Blecher, Y. Soen, B. Rinkevich, A. W. De Tomaso, K. J. Ishizuka, and I. L. Weissman. 2007. Striving for normality: whole body regeneration through a series of abnormal generations. FASEB J. 21: 1335-1344.

Voskoboynik, A., Y. Soen, Y. Rinkevich, A. Rosner, H. Ueno, R. Reshef, K. J. Ishizuka, K. J. Palmeri, E. Moiseeva, B. Rinkevich, et al. 2008. Identification of the endostyle as a stem cell niche in a colonial chordate. Cell Stem Cell 3: 456-464.

Wright, R. K. 1981. Urochordates. Pp. 565-626 in Invertebrate Blood Cells, Vol. 2, Arthropods to Irochordates, Invertebrates and Vertebrates Compared, N. A. Ratcliffe and A. F. Rowley, eds. Academic Press, London.

Wright, R. K., and T. H. Ermak. 1982. Cellular defense systems of the Protochordata. Pp. 283-320 in Phylogeny and Ontogeny, N. Cohen and M. M. Sigel, eds. Springer, Boston.

Zaniolo, G., A. Sabbadin, and C. Resola. 1976. Dynamics of the colonial cycle in the ascidian, Botryllus schlosseri: the fate of isolated buds, Acta Embryol. Exp. 2: 205-213.

Zondag, L. E., K. Rutherford, N. J, Gemmell, and M. J. Wilson. 2016. Uncovering the pathways underlying whole body regeneration in a chordate model. Botrylloides leachi using de novo transcriptome analysis. BMC Genomics 17: 114.

Table A1
Detailed description of hemolymph cell types present in the
Botrylloides ieachii vascular system based on histological observations

Cell type (% of all hemocytes)  Description

Undifferentiated cells
Hemoblast (4%)                  This small (5-8-[micro]m), almost
                                 perfectly round progenitor cell has a
                                 high nuclear-cytoplasmic ratio. It has
                                 a consistent dark blue nuclear stain
                                 and most commonly a blue cytoplasmic
                                 stain. However, unstained or pink
                                 cytoplasmic staining was occasionally
Differentiating cell (11%)      A number of transitory cells can be
                                 identified in B. leachii hemolymph.
                                 Although it is difficult to determine
                                 which differentiation program these
                                 cells are following, the morphology of
                                 the majority of cells suggests that
                                 they originate from hemoblasts. These
                                 medium-sized (8-12-[micro]m) cells
                                 display a large light blue-stained
                                 nucleus, sometimes a nucleolus, and an
                                 unstained, often fibrous gray
                                 cytoplasm with various morphological
                                 and staining features reminiscent of
                                 their future cell type.
 Hyaline amoebocyte (3%)        As with all other amoebocytes, this
                                 small (6-9-[micro]m) cell has an
                                 irregular shape, typically displaying
                                 a large variety of pseudopodia.
                                 Hyaline amoebocytes have a dual
                                 purple-blue cytoplasmic staining,
                                 where the purple appears to originate
                                 from a variety of clear vesicles. In
                                 histological sections, both entirely
                                 purple cells and purple cells with a
                                 blue crescent containing the nucleus
                                 were observed.
 Macrophage-like cell (19%)     This extremely variable (6-20-[micro]m)
                                 cell shows a wide range of
                                 sizes and shapes depending on the
                                 material it engulfed. Therefore,
                                 although most of the cells were fully
                                 stained dark blue, some color
                                 variations (including shades of
                                 yellow, brown, green, and gray) were
                                 observed depending on the ingested
                                 material. Empty macrophage-like cells
                                 display a red crested cytoplasm (Cima
                                 et al., 2002). Particularly in large
                                 cells, the nucleus is displaced toward
                                 the periphery of the cell by the
                                 phagocytic vacuoles. While they can
                                 resemble compartment cells,
                                 macrophage-like cells have a greater
                                 heterogeneity in the size, shape, and
                                 content of their vesicles.
 Granular amoebocyte (1%)       A small (6-10-[micro]m) amoebocyte
                                 characterized by a number of scattered
                                 dark green cytoplasmic vesicles. While
                                 they resemble granular cells, granular
                                 amoebocytes have a much lower density
                                 of granules, always displaying some
                                 portion of their cytoplasm. The
                                 cytoplasm is stained a light blue-gray,
                                 although dark blue and red have
                                 also been occasionally observed.
 Morula cell (49%)              This typically large (10-16-[micro]m)
                                 cell is mostly characterized in smears
                                 by its orange-green cytoplasmic color
                                 and in histological sections by the
                                 red color of the thin cytoplasm among
                                 the large vacuoles. It displays low
                                 numbers of large ([equivalent]
                                 2-[micro]m) pale-yellow and round
                                 vacuoles, thus resulting in its
                                 characteristic berrylike shape.
Mast cell-like cells
 Granular cell (1%)             A medium-sized (8-12-[micro]m) cell,
                                  mostly circular, with a very
                                  irregular edge owing to the numerous
                                  small vesicles it contains. A variety
                                  of vesicles have been observed, but
                                  the majority are stained dark
                                  blue-green and packed at such a high
                                  density that the whole cell appears
                                  black and opaque.
Transport cells
 Compartment amoebocyte (5%)    A small (7-10-[micro]m) amoebocyte
                                  characterized by numerous clear
                                  protruding vesicles, resulting in a
                                  berrylike-shaped cell when observed
                                  unfixed. Its cytoplasm stains blue,
                                  with a dark blue nucleus easily
                                  visible in sections but harder to
                                  find in smears.
 Compartment cell (4%)          A large (12-18-[micro]m) cell that
                                  displays few (typically fewer than
                                  12) but large (~3-[micro]m) opaque
                                  vesicles that remain mostly unstained
                                  (gray) although varying shades of
                                  blue were observed.
Storage cells
 Pigment cell (2%)              A medium-sized (10-16-[micro]m) cell
                                  that displays a number of large
                                  vacuoles filled with characteristic
                                  minute dark granules, whose color
                                  relates to that of the colony (dark
                                  orange in this particular example).
Nephrocyte (1%)                 This medium-sized (10-16-[micro]m) cell
                                 closely resembles pigment cells but
                                 lacks any pigment color. These cells
                                 do not exhibit a circular shape filled
                                 with vesicles but rather a berrylike
                                 shape, as an aggregate of vesicles.


(1) Department of Anatomy. Otago School of Medical Sciences. University of Otago. P.O. Box 56, Dunedin 9054, New Zealand: and (2) Department of Marine Science. Division of Sciences, University of Otago. P.O. Box 56, Dunedin 9054, New Zealand

Received 2 February 2017; Accepted I May 2017; Published online 28 August 2017.

(*) Present address; Department of Biology, University of Fribourg. Chemin du Musee 10, 1700 Fribourg. Switzerland.

([dagger]) To whom correspondence should be addressed. E-mail:

Abbrevianons: PIV. particle image velocimetry; WBR, whole-body regeneration.
Table 1
Summary of cell types present in the Botrylloides leachii circulatory

Cell type                    Subtype

Undifferentiated cell (a,b)  Hemoblast  (c,d)

Immunocyte (a)               Phagocyte (c)

                             Cytotoxic cell (d)

Transport cell (c)           Granulocyte (c)

Mast cell-like cell (g)      Granular cell (b)
Storage cell (d)             Vacuolated cell (d)

Cell type                    Circulatory cells

Undifferentiated cell (a,b)  Hemoblast (c,d)
                             Differentiating cell
Immunocyte (a)               Hyaline amebocyte (b,d)
                             Macrophage-like cell (b,d)

                             Granular amebocyte (b,d,f)
                             Morula cell (b,c,f)

Transport cell (c)           Compartment amebocyte (b)
                             Compartment cell (b,e,f)
Mast cell-like cell (g)      Granular cell (b)
Storage cell (d)             Pigment cell (a,b,c,e)
                             Nephrocyte (a,b,c)

Cell type                    Older terminology

Undifferentiated cell (a,b)  Lymphocyte (e,f)
                             Stem cell (b)
Immunocyte (a)               Amebocytes with vacuole (e)
                             Signet ring cell (e,f)
                             Vacuolated cell (a)
                             Leukocyte (a)
                             Cells with acidic vacuole (e)
                             Green cell (f)
                             Leukocyte (a)
                             Vanadocyte (e)
Transport cell (c)           Vacuolated cell (a)

Mast cell-like cell (g)      Unique to B. leachii (b)
Storage cell (d)             Cell with reflecting disks (e)
                             Orange cell (f)

(a) Wright and Ermak, 1982.
(b) Cima et al., 2002.
(c) Hirose et al., 2003.
(d) Ballarin et al., 2011.
(e) Endean, 1960.
(f) Freeman, 1964.
(g) Cima et al., 2014.

Table 2
Classification chart of Botrylloides leachii hemocytes based on light
microscopy Giemsa stain

Cell type              Size        Shape

Hemoblast               5-8        Round
Differentiating cell    8-12       Swollen, irregular

Hyaline amebocyte       6-9        Amebocyte; highly irregular,
                                   including a variety of pseudopodia
Macrophage-like cell    6-20       Mostly round with irregular content

Granular amebocyte      6-10       Amebocyte
Morula cell            10-16       Circular, containing a low number
                                   of large (~2-[micro]m) pale yellow
                                   and round vacuoles
Granular cell           8-12       Circular with irregular edge

Compartment amebocyte   7-10       Amebocyte with a berrylike shape
Compartment cell       12-18       Similar to morula

Pigment cell           10-16       Similar to morula

Nephrocyte             10-16       Berrylike

Cell type              Stain

Hemoblast              Blue cytoplasm with a dark blue nucleus
Differentiating cell   Light blue nucleus and gray cytoplasm

Hyaline amebocyte      Dual purple-blue cytoplasm
                       Blue with vesicles of varying shades
Macrophage-like cell   of yellow, brown, green, and gray

Granular amebocyte     Scattered dark green cytoplasmic vesicles
Morula cell            Orange-green cytoplasmic color in
                       smears, red color around the vacuoles in
                       histological sections
Granular cell          Dense dark blue-green vesicles

Compartment amebocyte  Blue with a dark blue nucleus
Compartment cell       Unstained (gray) or occasionally blue opaque
Pigment cell           Filled with minute dark granules, related to the
                       colony color
Nephrocyte             Dark and speckled gray vesicles

Cell type              Characteristic

Hemoblast              High nucleus-to-cytoplasm ratio
Differentiating cell   Fibrous cytoplasm with a proportionally large
Hyaline amebocyte      Purple-stained clear cytoplasmic  vesicles

Macrophage-like cell   Vesicles highly heterogeneous in  size, shape,
                       and content
Granular amebocyte     Sparser vesicles than granular cells
Morula cell            Most common cell type in B. leachii hemolymph

Granular cell          High density of vesicles renders the cell black
                       and opaque
Compartment amebocyte  Numerous clear protruding vesicles
Compartment cell       Large uniform vesicles

Pigment cell           Content of the vesicle exhibits
                       Brownian movement in living cells
Nephrocyte             Aggregate of dark gray vesicles

Only the most common traits are listed in this table; refer to Table
A1 for a full description of the cell types.
COPYRIGHT 2017 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Blanchoud, Simon; Zondag, Lisa; Lamare, Miles D.; Wilson, Megan J.
Publication:The Biological Bulletin
Article Type:Report
Date:Jun 1, 2017
Previous Article:Errata.
Next Article:Transcriptomic Basis of Metamorphic Competence in the Salt-Marsh-Dwelling Polychaete Capitella teleta.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters