Study area and collection site

Nonregenerating individuals (n = 33) of Ophioplocus januarii with similar size (15-20 mm in disk diameter) were collected during the summer 2010 by Scuba diving at Villarino beach (42º23´S - 64º18´W), San José gulf, Argentina, at 6 m depth with a water temperature of 16ºC. Individuals were transported to a wet laboratory and acclimatized for one week. The experiment was conducted by using 501 × 40 × 50 cm (approx. 100 l). The aquarium had a constant supply of filtered seawater with a salinity of 35 ‰. An air-driven box filter containing filter floss and activated charcoal was used to maximize and filtration. The experiment was conducted under a 12 h light: 12 h dark schedule at constant temperature (15 ºC).

Experimental design for morphology observation

For the morphology observations 6 individuals were used. Experimental arm amputation was done following to ^{Dupont and Thorndyke (2006}). The arm into the right of madreporite was used during the experiment. Autotomy was done about 10 mm from the arm tip after relaxation with MgCl₂ 5 % (^{Rubilar, 2010}). After amputation, the individuals were returned to the aquarium. Experimental individuals were kept in the aquaria and were monitored at 2, 4, 6, 8, 11 and 16 days post amputation (dpa). At each time, one individual was randomly selected and 15 mm of the regenerating arm was amputated and used for morphology observations. For SEM (Scanning Electronic Microscope) observations on whole mount were made as follow: the tissue was fixed in paraformaldehyde (PF) 4 % in phosphate buffer solution (PBS) (0.01 M, pH = 7.4) + glutaraldehyde (GA) 2 % and then the tissue was dehydrated in ethanol series and dried to critical point using Hexamethyldisilazne (Sigma). Each arm was extended and mounted with a double-sided adhesive tape on a labelled slide. Arms were observed under a JEOL JSM-6460 LV microscope using the back-scattered electron (BSE) technique at low vacuum (15 - 20 Pa). Digital images were taken and specifications were recorded (magnification, plate inclination and rotation, scale, resolution, etc.).

Experimental design for histological observation

27 individuals were used for histological observations. Amputation procedure was performed as explained above. After amputation, the individuals were returned to the aquarium. Three individuals were used as controls (C₁). Control individuals were not subject to amputation and were kept in the aquarium until completion of the experiment. Experimental individuals were kept in the aquarium and were monitored at 2, 4, 6, 8, 11, 16, 20 and 30 days post amputation (dpa). At each time, three individuals were randomly selected and 15 mm of the regenerating arm was amputated and used for histological observations by fixing the arm tip in PF 4% in PBS (0.01 M, pH = 7.4). After fixation the tissue was rinse with PBS (0.01 M, pH = 7.4) and then decalcified according to Mashanov, Zueva, Heinzeller and Dolmatov (2006) for a week. The tissue was dehydrated in ethanol series and embedded in paraffin. Sections of 7 µm were stained with Masson´s Trichrome. Slides were observed and analyzed using a Zeiss AxioImager Z1 microscope. Images were taken with an AxiocamHRc digital camera attached to the microscope.

Cell count and statistical analysis

Slides were observed and divided by group. Different cell types were counted, measured and identified by using the classification by ^{Smith (1981}), ^{Byrne (1994}), ^{Candia-Carnevali, Bonasoro, Patruno and Thorndyke (1998}) and Candia-Carnevali and Bonasoro (2001). Differential interference contrast was used to improve the cell identification at different days post amputation. The type and number of cells observed were counted in three fields in each section by using a 100x objective. The cell density (cell mm^-2) for each group was calculated throughout the regenerative process. A two-way ANOVA analysis was used to test for differences between the type of cells during the regeneration process and between the different types of cells in each days post amputation. Normality was tested using Shapiro Wilk´s test and homogeneity of variances was tested using Bartlet´s test. The data was transformed with the square root (^{Sokal & Rohlf, 1979}).

Macroscopic observation of the regeneration process

At two days post amputation (dpa) a contraction of the tip of the stump was observed. The wound of the surface of the stump presented a concave shape and was partly covered by the external skeletal shields and a thin epithelium (Fig. 1A). The wound healing was in progress. Fibrous material and tissue debris formed a largely incomplete cicatricle layer. At 4 dpa the wound surface acquired a flattened shape. However, the tissue did not show a defined structure. There were areas with calcium carbonate deposits, characterized by their white coloration and hexagonal shape (Fig. 1B). After 6 dpa the surface of the wound was more regular and continuous. The skeleton was more developed, evidenced by a greater amount of calcium carbonate deposits (Fig. 1C). At 8 dpa the wound surface and the concave shape was lost. The skeleton and surrounding tissues were more organized and the complete re-epithelialization of the wound was observed. The stump surface was covered by a thin new epithelium (Fig. 1D). At 11 and 16 dps the shape of the tip of the arm changed. Without differentiation of any structure (Fig. 1E). This stage of the regenerative process was characterized by the formation of a large number of ossicles (Fig. 1F).

Types of the cell found during histological observations of the regeneration process

Different cells types were identified during the experiment: coelomocytes, sclerocytes, granule cells, presumptive blastemic cells, phagocytes, cells in mitosis, cells with cytoplasmic expansions, amoebocytes and cells in differentiation (Fig. 2). The coelomocytes (3.0 ± 0.5 μ) were observed in the coelomic channels, these cells were characterized by a central nucleus and a small cytoplasm (Fig. 2A). Sclerocytes (5.0 ± 1.0 μ) showed a rounded appearance, acidophilic cytoplasm and lot of granules in the cytoplasm (Fig. 2B). Granule cells (2.0 ± 0.5 μ) were characterized by their large chromatophilic bodies (Fig. 2C). Blastemic cells (= undifferentiated, 8.0 ± 1.0 μ) showed a globular appearance, circular nucleus and translucent pink cytoplasm (Fig. 2D). Phagocytes were distinguished by their large size (14.0 ± 4.0 μ) and the residual bodies of different sizes inside their cytoplasm (Fig. 2E). Cells in mitosis (13.0 ± 4.0 μ) were characterized by the presence of two nuclei located at both ends of a large cytoplasm (Fig. 2F). Cells with cytoplasmic expansions (5.0 ± 0.7 μ) were characterized by the expansion of their cytoplasm and an ovalshaped cytoplasm (Fig. 2G). The amebocytes (19.0 ± 3.0 μ) were characterized by their large size (Fig. 2H). Cells in differentiation were distinguished by their small size (1.0 ± 0.3 μ) and their acidophilic circular-shaped cytoplasm (Fig. 2I). The mean cell density did not vary during the wound healing. However, if the types of cells varied significantly during the experiment (ANOVA, F5,8 = 14.25, p < 0.01). The sclerocytes were abundant throughout the study period, even though the density decreased. The amebocytes were the lesser abundant cell type.

Fig. 1 Scanning Electronic Microscope images. Frontal view of the amputated arm tip at different times post amputation (dpa). (A) 2 dpa, denote the strongly concave shape (arrow) covered by the external skeletal shields and a thin epithelium. (B) 4 dpa, denote the flattened shape and the disorganized tissue (asterisk). (C) 6 dpa and (D) 8 dpa, denote the loss of the concave shape and the organized tissue (asterisk) and the calcium deposits (arrowhead). (E) 11 dpa and (F) 16 dpa, denote the ossicles in formation (arrows) and the growth of the arm (asterisk). AAP aboral arm plate, LS lateral shield, OAP oral arm plate, P podia, S spine. Scale bars: 500 µm. 

Fig. 2 Detail of cells involved in the regeneration process. (A) Coelomocytes (arrow). (B) Sclerocytes. (C) Granule cells. (D) Presumptive blastemic cells. Note the two nuclei of the cell (arrow). (E) Phagocytes. (F) Cells in mitosis. (G) Cells with cytoplasmic expansions. (H) Amoebocytes. (I) Cell in differentiation. Scale bars: 1: 5 μm; 2: 10 μm; 3: 20 μm. 

Histological observation of the regeneration process

At 2 dpa (Fig. 3A, B, C) the wound was closed. The injured area acquired a concave shape. The amputation surface was covered by a complete and thin cicatricial layer (Fig. 3A). This tissue was characterized by a heterogeneous appearance, with no apparent structure. The tissue was discontinuous and had unclear boundaries (Fig. 3B). Beneath the cicatricial layer, the connective tissue was loose and without a clear tissue structure. During this period, the sclerocytes were the more abundant cells, followed by granule cells and coelomocytes. Phagocytes and amoebocytes were also observed (Fig. 3B, C).

At 4 dpa the wound surface showed a smooth concave shape (Fig. 3D, E, F). The amputation surface was covered by a welldeveloped cicatricial layer (Fig. 3E). The sclerocytes started to secrete collagen to form the endoskeleton. The ends of the coelomic cavity, epineural oral canal, the hyponeuralcanal and radial nerve cord were sealed by a large number of cells (Fig. 3F). The sclerocytes continued to have the highest density followed by the granule cells. Coelomocytes and phagocytes were observed. Cells with cytoplasmic expansions and amebocytes were also present in a low number.

At 6 dpa the amputation surface was no longer concave and acquired a more regular appearance at both ends and in the center of the arm (Fig. 3G, H, I). There were a large number of cells, primarily cells in differentiation, in the scar surface (Fig. 3H). The epithelium was continuous and thick (Fig. 3I). The coelomic cavity, the radial water canal and the sinuses surrounding the nerve remain sealed by primarily sclerocytes. The connective tissue appeared to have a denser compact appearance. The sclerocytes had the highest cell density, followed by phagocytes, cells in differentiation and granular cells.

At 8 dpa the wound was completely covered (Fig. 4A, B, C). Beneath the wound surface there was an intense cellular activity in the connective tissue (Fig. 4B). There were a large number of cells at the end of the nerve (Fig. 4C). At 16 dpa below the epidermis layer, the connective tissue presented loose fibers; there were a degree of rearrangement (Fig. 4D, E, F). The sclerocytes were close to each other; however, trabeculae were forming (Fig. 4E). The sclerocytes were the most abundant cells. However, cells in differentiation were important in this stage. Phagocytes, granular cells and coelomocytes were also present.

At 20 dpa the shape of the tip of the arm changed (Fig. 4G, H, I). A protrusion was observed on the amputation surface. The regenerative regrowth of the coelomic canals and the brachial nerve is outlined by a marked cellular flux towards the bud region. The outline of the presumptive blastema formation was observed as accumulation of a large number of cells (Fig. 4H). Trabeculae were already formed at both sides of the wound surface (Fig. 4I). At 30 dpa a presumptive blastema appeared as a small button, slightly emerging in the central-oral area of the autotomy surface was observed (Fig. 5A). The stalked presumptive blastema was formed mostly by presumptive blastemic cells. However, there were also cells in differentiation and sclerocytes (Fig. 5B, C, D).