Wnt signaling restores evolutionary loss of regenerative potential in Hydra

Introduction

Why can a flatworm regenerate half or more of its entire body, while we are unable to regrow even a chopped-off fingertip? Indeed, regenerative potential varies widely across animals. Thanks to recent studies, quite a lot of the cellular and molecular processes underlying regeneration have been uncovered. However, the other side of the problem – how regenerative abilities are lost during evolution – remains largely unexplored. Recent studies are beginning to address this gap, bringing up interesting possibilities of evolutionary interplay between regeneration and other traits. For instance, a cross-species study in planarians found that inhibition of canonical Wnt could bypass regeneration defects, and further proposed that Wnt’s role in both regeneration and reproduction might contribute to an evolutionary trade-off.

The cnidarian polyp Hydra vulgaris is a well-established model for studying the molecular mechanisms of regeneration. Interestingly, just within the Hydra genus, regenerative potential varies: when H. vulgaris is cut in half, both halves can form a new hydra by growing a new head or a new foot; however, previous works suggested that this is not the case for some of H. vulgaris‘ close sisters.

In this preprint, Sergio Campos and colleagues found that Oligactis clade hydras cannot regenerate a foot, and also take longer to regenerate heads. Through comparative transcriptomics and pharmacological manipulations, they showed that this loss of regeneration ability may be traced back to attenuated injury-induced Wnt signaling.

Main findings

H.oligactis shows failed foot regeneration and slow head regeneration

The authors first systematically characterized the regeneration ability across the Hydra genus by bisecting animals at different locations along the oral-aboral axis and recording their regeneration outcome across time.

Figure 1 – H. oligactis failed to regenerate foot after bisection at 50%. Figure adapted from the preprint where it is available under a CC-BY-NC 4.0 International license.

For foot regeneration, species of the Oligactis clade showed a low success rate that further decreased when bisected closer to the oral end. RNA-seq in the model species H. oligactis showed that the transcriptomic profile of its aboral wound site tissue did not shift towards the transcriptomic signature of uninjured foot tissue, as few foot-specific genes were upregulated. Moreover, genes specifically expressed during foot regeneration of H. vulgaris were also only weakly upregulated in H. oligactis.

For head regeneration in H. oligactis, while the tissue transcriptome did progressively shift towards the uninjured head, the regeneration process was delayed compared to H. vulgaris. FISH of wnt3 further showed that the establishment of a Wnt signaling center occured slower in H. oligactis, and a tissue grafting assay showed that H. oligactis tissue acquired oral organizer activity later than H. vulgaris.

The authors thus concluded that foot regeneration potential has been lost in H. oligactis, and that head regeneration shows a temporal delay.

Injury-induced transcriptional activation of Wnt pathway genes is reduced in H. oligactis as compared to H. vulgaris

Previous literature suggested that the activation of a regeneration program requires initial Wnt activation, which is a generic response induced by injury. To compare the early injury-induced transcriptional response between H. vulgaris and H. oligactis, the authors conducted RNA-seq on H. oligactis wound site tissue and used the computational method OrthoClust to look for conserved co-expressing gene clusters in the two species.

In doing so, the authors identified one gene cluster that was upregulated during head regeneration in both species and contains several canonical Wnt genes. Compared to H. vulgaris, in H. oligactis the upregulation of this cluster was delayed during head regeneration, and key Wnt pathway genes in this cluster were either not significantly upregulated or expressed at a lower level during foot regeneration.

Figure 2 – Cluster 6, which contains key Wnt pathway genes, showed delayed upregulation in H. oligactis head regeneration, and no significant upregulation in foot regeneration. Figure adapted from the preprint where it is available under a CC-BY-NC 4.0 International license.

Thus, the authors concluded that compared to H. vulgaris, injury-induction of Wnt signaling is either absent or highly reduced in H. oligactis.

Pharmacological manipulation of Wnt signaling during the injury response is sufficient to rescue or block foot regeneration

The authors hypothesized that the lack of injury-induced Wnt expression may contribute to foot regeneration failure in H. oligactis. To test this hypothesis, they transiently treated H. oligactis with the Wnt agonist ALP after amputation to mimic the transient Wnt signaling activation seen in H. vulgaris. The treatment resulted in a partial but significant rescue of foot regeneration, and RNA-seq showed that the treatment moved regenerating tissue towards the transcriptomic signature of homeostatic foot tissue, as more foot-specific genes were expressed.

Figure 3 – H. oligactis showed partially rescued foot regeneration after transient Wnt agonist treatment. Figure adapted from the preprint where it is available under a CC-BY-NC 4.0 International license.

Conversely, the authors transiently treated H. vulgaris with the Wnt inhibitor iCRT14. This delayed head regeneration and inhibited foot regeneration, mimicking H. oligactis.

Thus, the authors concluded that the transient Wnt activation in response to injury is essential for foot regeneration in Hydra, and that the lack of it is responsible for foot regeneration failure in H. oligactis.

Transient activation of Wnt signaling triggers the transcription of foot-specific genes in H. oligactis

To gain insight into the molecular network regulating foot regeneration, the authors compared RNA-seq data of ALP-treated and non-treated H. oligactis during regeneration.

They found that the earliest foot-specific transcription factor activated after treatment was dlx2, which is essential for foot regeneration in H. vulgaris. Using FISH, the authors showed that dlx2 expression is undetectable in H. oligactis, whereas it is clearly expressed at the aboral injury site in H. vulgaris during regeneration. This correlation supports the idea that dlx2 might be a key factor in foot regeneration.

Next, the authors clustered differentially expressed genes by their expression dynamics and conducted GO analysis to determine the biological functions of identified modules. Based on the temporal order of gene module upregulation, the authors proposed that the transient Wnt activation sequentially activates non-tissue specific transcription factors, followed by foot-specific factors—starting with dlx2—and ultimately genes involved in morphological changes such as extracellular matrix remodeling.

What I like about this preprint

I was initially attracted by the study’s unique focus on the loss of regenerative potential. The system that the authors employed here is a simple and effective way to address their question, and the authors made good use of prior mechanistic knowledge and recent genomic data to build and test their hypotheses. Finally, the result that activating a single pathway is sufficient to restore regeneration potential is not only a compelling point supporting the narrative, but also a surprising finding in itself.

Future directions and questions for the authors

1. You showed differences in regeneration kinetics both in different Hydra species, and in the same species when amputated at different oral-aboral positions. Can the correlation between injury-induced Wnt signaling and regeneration be extended to – or account for – these differences?
2. What is the relationship between generic wound healing, regeneration, and positional identity? The current data seems to imply that the strength of the transient injury-induced Wnt activation determines the nature of the downstream response: low or no activation, as in oligactis aboral wound, results in healing without regeneration; an intermediate activation, as in H. vulgaris or in ALP treated H. oligactis, results in foot regeneration; finally, hyper-activation may somehow overwrite prior positional information and result in the formation of a second head. A closer look into the transcriptional regulation network in H. oligactis may provide mechanistic insights on this apparent dosage-dependency.
3. You noted that foot-specific genes are expressed at a lower level in uninjured oligactis foot tissue. The spatial expression pattern of dlx2 is also more restricted to the aboral end in H. oligactis. Are these differences in homeostatic tissues related to regeneration potential?

References:

Vila-Farré M, Rozanski A, Ivanković M, et al. Evolutionary dynamics of whole-body regeneration across planarian flatworms. Nat Ecol Evol. 2023;7(12):2108-2124. doi:10.1038/s41559-023-02221-7

Bely AE, Nyberg KG. Evolution of animal regeneration: re-emergence of a field. Trends Ecol Evol. 2010;25(3):161-170. doi:10.1016/j.tree.2009.08.005

Cazet JF, Cho A, Juliano CE. Generic injuries are sufficient to induce ectopic Wnt organizers in Hydra. Elife. 2021;10:e60562. Published 2021 Mar 29. doi:10.7554/eLife.60562

 

doi: https://doi.org/10.1242/prelights.40244

Read preprint (No Ratings Yet)