Cellular Sources and Development of the Blastema

A sharp increase in cell proliferation in tissues adjacent to the wound site indicates the onset of blastema development. Another early indicator of blastema formation is the upregulated expression of several sets of genes strongly associated with the regenerative process. These sets include Hox genes; members of the germline multipotency program (GMP) that include annelid orthologues of piwi, vasa, nanos, PL10, and myc; signaling genes including WNT pathway members; other transcription factors like even-skipped, engrailed, and paired domain proteins; and RNA- binding proteins like elav (Kozin and Kostyuchenko 2015; Ozpolat and Bely 2016; Planques et al. 2018). Similar to cell proliferation, earliest gene expression is usually found within the wound epithelium and/or underlying mesoderm, and is first seen at the ventral side at or near to the end of the ventral nerve cord, supporting the hypothesis that the ventral nerve cord plays a key role in blastema initiation and patterning during annelid regeneration (Ozpolat and Bely 2016; Planques et al. 2018; Boilly, Boilly-Marer, and Bely 2017).

As cell proliferation continues, a mass of morphologically undifferentiated cells continues to accrue beneath the highly proliferative wound epithelium, forming a clear regeneration blastema. Both direct observation and experiments with thymidine analog markers show that this blastema forms by contribution of ectoderm, mesoderm, and endoderm (Bely 2014; Jong and Seaver 2017; Planques et al. 2018).

The blastema shows de novo expression of several GMP genes, strongly suggesting that blastema formation requires dedifferentiation of previously differentiated cells (Ozpolat and Bely 2016; Planques et al. 2018). Alternatively, regeneration might be driven by a pool of reserve stem cells present in each tissue that become activated during wound healing. This later hypothesis has weaker support, as it predicts the presence at most tissues of cells showing stem cell-like expression patterns scattered through the body. Such patterns have so far been observed only for pnW-expressing cells in a few species and are most likely related to regeneration of the germline (Tadokoro 2018). Thus, current data suggests that the regeneration blastema forms from proliferation of differentiated cells from tissues proximal to the wound site.

A related question that follows is what is the fate of blastemal cells, and whether these undifferentiated cells regain pluripotency or are still restricted to differentiate back into the same tissues from which they arose. Although several authors initially proposed various scenarios in which a few cell sources were able to reconstruct most of the missing tissues, current knowledge does not support extensive transdifferentiation across tissue types and origins (Stephan-Dubois 1954; Cornec et al. 1987; Bely 2014). The epidermis of the regenerate seems to derive from proliferation of cells forming the wound epidermis. The proliferative epidermis also produces cells that become internalized and participate in the regenerate’s brain and ventral nerve cord ganglia; at the anteroventral surface, this tissue invaginates to form the mouth opening. The origin of new mesodermal tissues is less well understood, but most data supports that they derive from cells coming from old mesodermal tissues. Finally, gut tissues also appear to derive from proliferation of endodermal cells (Tweeten and Reiner 2012; Planques et al. 2018). Thus, and despite the extensive cell movements reported during annelid regeneration, current evidence paints a scenario where most tissues from the regenerate are derived from nearby stump tissues through dedifferentiation into blastemal cells followed by redifferentiation. In other words, the dedifferentiation process does not seem to confer enough cell pluripotency to allow crossing of embryonic germ layer boundaries.

An interesting observation is the fact that the anteriormost non-segmental region, the prostomium, derives in its entirety from anterodorsal blastemal cells (Zattara 2012). Experiments using iontophoretic labeling of cells with the carbocyanine dye Dil in anteriorly amputated individuals of the clitellate Pristina leidyi show that labeled cells located at an anterodorsal region of the early blastema always end up forming part of the prostomium (Figure 10.4A-E). In turn, cells labeled at the anteroventral region of the early blastema are displaced posteriorly and end up in a ventral region posterior to the mouth (Figure 10.4A, F-H). A dorsal origin of the prostomium can also be seen during paratomic reproduction in this species (Bely and Wray 2001; Zattara and Bely 2011). In annelids with indirect development, the prostomial region is derived from the larval episphere, which is also is characterized by expression of the anterior patterning gene optix/six3 (Steinmetz et al. 2010). Expression of six3 is also seen early during paratomic reproduction in Pristina longiseta in dorsolateral stripes located immediately behind the fission plane (Steinmetz et al. 2010). Thus, both regeneration and fission seem to recapitulate prostomial development by upregulating the same genes that generate this region during embryogenesis; the mouth opening, which forms de novo by epidermal

FIGURE 10.4 Cell tracing experiments showing the dorsal origin of the regenerated peri- stomium in the clitellate Pristina leidyi and the participation of neurites from the old VNC in forming the circumenteric connectives and part of the brain. (A) Fate map diagram of the wound epidermis. (B-K) Confocal images of representative experiments. Merge of reflected (red, peak emission of Dil) and transmitted (gray scale, morphology) light channels. Anterior is to the left in all panels. (B-E) Dil labeled cells (yellow arrowheads) at an anterodorsal epidermal patch (green dashed line): an internal labeled cell (red arrowhead) is found at later stages (inset of boxed areas). (B) Stage 1.7 hpa; (C) Stage 2,26 hpa; (D) Stage 3,48 hpa; (D’) Detail of boxed area in D; (E) Stage 5,76 hpa. (E’) Detail of boxed area in E. (F-H) Dil labeling of epidermal anteroventral patch (blue dashed line) showing posteroventral displacement. (F) Stage 2, 18 hpa; (G) Stage 3, 42 hpa; (H) Stage 5, 70 hpa. (I-K) Dil labeling of neurons within the ventral nerve cord shows incorporation of neuronal elements from the old cord into the regenerated new CNS. (I) Stage 1, 3 hpa; (J) Stage 3,44 hpa; (K) Stage 6, 75 hpa. Green arrowhead shows position of the dye injection. VNC, ventral nerve cord; cec, circumenteric connectives; eg, cerebral ganglion. B-K modified after Zattara (2012).

invagination, takes place at the same point where dorsal and ventral tissues had come into contact as a result of wound closure. Another related observation is that in annelids, a dorsal fate seems to be the default state for the body wall, and ventral fate is imposed by interaction with the ventral nerve cord (Boilly, Boilly-Marer, and Bely 2017). This suggests that the timing and spatial extent of neural regeneration (see next section) must be precisely timed to avoid ventralization of the anterodorsal region of the early blastema.

< Prev   CONTENTS   Source   Next >