Parallel repair mechanisms in plants and animals

Wounding has a multitude of causes, from a predator's jaws to a burn in the home or an experimental laser in the laboratory. Tissues adjacent to the wound perceive this damage and then must determine the extent of the danger, for example whether the wound is infected, in order to respond appropriately (Fig. 1).

In both plants and animals, wound activation begins with a flash of calcium signalling that radiates from the site of the mechanical insult. This first signal can be visualised using GCaMP reporter (see Glossary, Box 1) transgenes, enabling live imaging of the calcium wave in, almost, real time (Fig. 2A) (Nguyen et al., 2018; Razzell et al., 2013; Xu and Chisholm, 2011; Toyota et al., 2018). For example, after wounding an Arabidopsis leaf, a calcium wave is initiated within 2 s and spreads at a speed of ∼1 cm/min until it reaches a vein, from where it is systemically transmitted, through the vascular system, to the distal extremities of the plant (Toyota et al., 2018). In animal tissues, the calcium flash is even more instantaneous, but remains more local (Razzell et al., 2013; Xu and Chisholm, 2011).

Damage-triggered calcium waves are transduced through various mechanisms in both plants and animals. Extracellular calcium enters through microtears in the plasma membrane of damaged cells and, in animals, spreads through gap junctions to neighbouring cells (Shannon et al., 2017). Presumably, a similar process occurs via the plasmodesmata channels linking neighbouring plant cells. However, there are also less direct routes for the calcium wave. For example, a study in Drosophila revealed how proteases released from damaged cells cleave and activate extracellular growth-blocking peptides that are detected by G protein-coupled receptors (Box 1), such as Methuselah-like 10, on cells neighbouring the wound edge, leading to calcium release from their endoplasmic reticulum (O'Connor et al., 2021). In plants, a parallel signalling cascade exists with glutamate receptor-like (GLR) channels regulating calcium influx when glutamate leaks from the ruptured phloem (Box 1) and binds to GLRs on the xylem contact cells (Box 1) and phloem sieve elements (Box 1) (Qi et al., 2006; Nguyen et al., 2018; Mousavi et al., 2013; Toyota et al., 2018). Calcium-dependent proteases then cleave pro-plant elicitor peptides (PEPs) into active forms that bind neighbouring cells’ PEP receptors to trigger innate immune responses (Hander et al., 2019).

In animal tissues, the calcium wave initiates release of permissive immune cell attractants, including reactive oxygen species (ROS), in particular H₂O₂, which are produced by NADPH oxidase enzymes (Bedard and Krause, 2007). In Drosophila, Duox, the NADPH oxidase enzyme responsible for generating H₂O₂, is directly responsive to calcium (Razzell et al., 2013). Similarly, in plants, transmembrane NAPDH oxidases, called respiratory burst oxidase homologues (RBOHs), produce H₂O₂ in response to wounding, which is regulated by calcium-dependent phosphorylation or by direct binding of calcium to RBOH domains (Dubiella et al., 2013; Drerup et al., 2013). Is this calcium-dependent ROS release conserved from a common plant and animal ancestor, or the result of convergent evolution? We presume the former because calcium-regulated NAPDH oxidases are present in the most ancient organisms (Kawahara et al., 2007), and all examined so far exhibit a wound activated calcium/ROS axis (Moon et al., 2022).

ROS production in both plants and animals can function directly as a microbicidal but also as a signalling mechanism to activate various elements of the defence response at the wound site (Mhamdi et al., 2010; Chang et al., 2004). Crucial to organism survival, ROS initiate key mechanisms of the wound repair cascade. One such mechanism is apoptosis (Box 1), which is critical for driving proliferation as its inhibition in hydra and Xenopus eliminates their regenerative response (Tseng et al., 2007; Galliot and Chera, 2010). H₂O₂ release at the wound site triggers apoptosis, and inhibition of ROS-producing enzymes also inhibits regeneration (Love et al., 2013; Tobiume et al., 2001). Studies in Xenopus and hydra have shown that ROS-triggered apoptosis, and subsequent proliferation, is mediated by WNT/β-catenin signalling (Galliot and Chera, 2010; Love et al., 2013). During fin regeneration in adult zebrafish, it seems that ROS production leads to a second, later wave of apoptosis that, in turn, is key to formation of the regenerative blastema (Box 1) (Gauron et al., 2013). In some animals, successful regeneration can be potentiated by sacrificing entire stretches of damaged tissue, involving programmed cell death or autotomy (Box 1) along pre-defined fracture planes (Box 2).

There is also an undoubtedly important role for cell death in the plant wound response. Recent studies in Arabidopsis have shown that cell death leads to expression of the transcription factor ethylene-responsive factor 115 (ERF115) in cells neighbouring the dying epidermal wound-edge cells (Johnson et al., 2018). Subsequently, ERF115 triggers regenerative cell division (Heyman et al., 2016), and thus may represent a parallel axis of regenerative control to the animal WNT/β-catenin pathway because of its induction by cell death (Zhou et al., 2019; Kong et al., 2018). Although cell death is clearly an important activator, it is crucial to limit its spread using ‘dampening’ mechanisms to protect the whole organism (Box 2).

ROS signalling is also fundamental for wound-induced changes in metabolism, in both animals (Love et al., 2014) and plants (Jacobo-Velazquez et al., 2015). The metabolic demands at a site of damage and repair are hugely increased to drive cell proliferation and other repair processes. Just as in rapidly growing cancer where the Warburg effect switches metabolic pathways from the citric acid cycle and oxidative phosphorylation to anaerobic glycolysis, it seems that injury and repair can also lead to a metabolic shift in cells (Sinclair et al., 2021; Scott et al., 2022).

Several decades ago, Thomas Hunt and colleagues revealed high levels of lactate in healing wounds in rabbits and speculated that switching to a glycolytic pathway might be key to some elements of the repair process (Ghani et al., 2003). Gene expression studies in the regenerating Xenopus tadpole limb show that many genes linked to glycolytic metabolism are locally induced here too (Love et al., 2014). More recent single-cell transcriptomic analysis of mouse skin wounding indicates a dramatic alteration in expression of metabolism-associated genes (Haensel et al., 2020). In this study, genes associated with oxidative phosphorylation were downregulated, and glycolysis-associated genes were upregulated, in sub-populations of wound-edge cells.

The plant wound response also leads to alterations in metabolism. A recent proteomics study revealed how wounding leads to an increase in starch and sucrose metabolism, likely providing the carbon for biosynthesis of amino acids and secondary metabolites (Guan et al., 2021).

These adjustments in metabolic strategies in response to wounding appear to be essential, across phyla, to provide the energy and resources required to repair a wound and prevent infection at the site of injury. A better understanding of their activation may lead to development of clinical therapeutics and agricultural treatments, or potential prognostic indicators of the ‘health’ status of healing tissues.

In all organisms, a breach in any barrier layer will result in invasion by pathogens, including bacteria, fungi and viruses. These pathogens must be rapidly quelled, or they will overcome the organism.

In all multicellular organisms, these invaders are recognised by pattern recognition receptors (PRRs), of which the most prominent in animals are the Toll-like receptors (Qian and Cao, 2013). In most animal species, there exists some degree of an immune system that has evolved to recognise and clear pathogens when the barrier layer becomes damaged and infection occurs. This triggers a wound inflammatory response, which, at early stages, is largely dependent on motile innate immune cells, such as neutrophils, monocytes and macrophages. Only if wounds become chronic and/or very infected, will the adaptive immune system be enlisted too (Weavers and Martin, 2020).

Plants have a very different portfolio of PRRs (Noman et al., 2019) and do not possess motile immune cell lineages. However, there is a more primitive component to the inflammatory response that is shared between animals and plants, which involves the upregulation and release of antimicrobial peptides (AMPs; Box 1; Fig. 2C). AMPs are crucial to plant immunity. Some AMPs are constitutively expressed, but others are locally or systemically upregulated in response to jasmonate (Box 1) synthesis, downstream of the immediate wound signals (Browse, 2009; Howe et al., 2018; Lenglet et al., 2017). The largest family of AMPs in plants are the defensins, with over 300 members identified in Arabidopsis (Silverstein et al., 2005). Defensins exhibit an impressive range of functions, from pathogenic cell membrane disruption to inhibition of predatory digestive enzymes (Kovaleva et al., 2020). In humans, over 30 beta-defensins have been identified (Schutte et al., 2002), and these appear to act through similar mechanisms (Pazgier et al., 2006). Plant and invertebrate defensins share a characteristic cysteine-stabilised alpha helix, and all defensins, including mammalian defensins, share a similarly sized triple-stranded antiparallel beta sheet, suggesting that defensins are an evolutionarily conserved class of AMPs across kingdoms (Broekaert et al., 1995). Historically, AMPs have somewhat slipped under the radar in animal repair studies, but, increasingly, they are being investigated as important small-molecule candidates for multiple medical applications associated with tissue repair and tumour progression (Baindara et al., 2017; Sun et al.; Williams et al., 2018; Mangoni et al., 2016).

One of the most urgent tasks at any site of tissue damage is to rapidly seal the gap to prevent loss of tissue fluids and influx of infectious agents. When a vertebrate animal is wounded, platelets or their equivalents leak from ruptured blood vessels adjacent to the wound site, become activated and degranulate upon exposure to extravascular factors, including collagen (Ivanciu and Stalker, 2015). Simultaneously, thrombin cleaves fibrin to fibrinogen to form a cross-linked mesh at the wound site, and the platelets become embedded in this mesh, forming a clot (Martin, 1997; Jansen and Hartmann, 2021; Li et al., 2010). In insects, a similar localised release of vesicular granules and formation of a coagulatory fibrillar network occurs, although there are no platelet equivalents and few, if any, molecular parallels to vertebrate clotting (Schmid et al., 2019).

An analogous process occurs at the site of injury in plants. As plant vessels are ruptured, sap exudes from their cut ends, and phloem proteins reversibly form disulphide cross-links to generate a polymer analogous to the fibrin mesh in a vertebrate animal clot (Ernst et al., 2012; Kehr, 2006). This initial ‘clot’ can then be reinforced by callose (Box 1) deposition to prevent pathogen entry. In plants, this ‘clot’ is a semi-permanent fixture, in contrast to the temporary protective scab/clot formed in certain animals, which is subsequently degraded and shed as the epidermis repairs.

In an example of parallel co-evolution, sap-sucking aphids have evolved to circumvent plant ‘coagulatory’ defences. Much like mosquitos injecting anticoagulants when sucking blood, aphids inject proteases, thereby preventing plugging of the sieve element allowing them to feed for extended periods (Walling, 2000; Furch et al., 2015). However, interestingly, aphids can also aid plant healing, as soldier aphids have been observed depositing body fluids that harden to form protective ‘scabs’ on leaves wherever they are feeding, in order to prevent their host's inconvenient early death (Kutsukake et al., 2009).

The plugging of a clot to seal the wound site, alongside an immune response in animals and release of AMPs in plants and animals, provides the time required to restore a barrier layer and rebuild missing tissues beneath the seal (Fig. 2B).

Among the animal kingdom, there is great disparity in the regenerative capacity of different species after injury. This ranges from healing with a scar in adult mammalian tissues (Fig. 2D), healing without a scar in mammalian embryos (Ferguson and O'Kane, 2004), and ultimately to being able to regenerate whole organs in planaria flatworms, and many fish and amphibian species (Tanaka and Reddien, 2011).

Studies in zebrafish and axolotl show how cells proximal to the injury site form a regenerative mass of dedifferentiated, multipotent stem-like cells, collectively called a blastema (Blum and Begemann, 2012; McCusker et al., 2015). Blastemal formation is regulated by retinoic acid and growth factor signals, key amongst these being FGF and WNT/β-catenin signalling, which drive cell cycle re-entry (Wehner and Weidinger, 2015). Blastemal cells will then re-differentiate, recapitulating developmental pathways to replace the missing cells and tissues of the damaged organ. However, regeneration-competent animals are limited by cellular memory. Within the blastema of an axolotl limb, cells remain lineage and tissue specific, and re-differentiate into their original cell type (Kragl et al., 2009). This means a tissue cannot be regenerated if it has been entirely lost in the wound. Plants, by contrast, are much more widely competent at regenerating their lost tissues. Their equivalent of a blastema is a callus that exhibits much greater plasticity during regeneration (Fig. 2E), and their capacity for trans-differentiation enables regeneration of any tissue after predation, which explains why plants can be propagated from cuttings (Ikeuchi et al., 2016). However, although plants have the capacity to regenerate a leaf, or even a whole plant from a cutting, they tend not to replace small patches of missing tissue lost – for example, within a wounded leaf – which is perhaps due to inhibitory feedback from cell wall biogenesis at the wound site.  The mechanisms behind these different context-dependent responses to wounding are still not fully understood (Ikeuchi et al., 2013).

Epigenetic modification lies at the heart of regeneration competency in animals, with histone modifications suppressing or activating developmental pathways required for regeneration, alongside retention of lineage identity to prevent aberrant homeotic transformation (Hayashi et al., 2020). Studies on zebrafish fin regeneration have revealed rapid changes in histone modification, in particular loss of polycomb-deposited histone H3K27me3 repressive marks on the promoters of key growth and proliferation genes, which are pivotal at early stages in the regenerative process (Stewart et al., 2009). It seems that, as well as clearing repressive histone marks, blastemal cells need to reduce levels of DNA methylation in order to re-express some developmental genes that enable proliferation and other regenerative activities (Takayama et al., 2014). To a lesser degree, some polycomb-mediated epigenetic changes to histone marks also occur in less regenerative organisms, such as mice. This enables wound-responsive changes in cell migration and proliferation, as well as partial epithelial–mesenchymal transition that is needed to enable wound re-epithelialisation (Shaw and Martin, 2009).

In plants, cellular identity is, in part, also maintained in differentiated cells by polycomb-deposited H3K27me3 marks, and loss of these marks is critical to the dedifferentiation required for callus formation following wounding (Mozgova et al., 2017; Ikeuchi et al., 2015). The leaf to callus transition requires repression of leaf regulatory genes by histone trimethylation, in parallel with the demethylation of the auxin (Box 1) hormone pathway and root regulatory genes (He et al., 2012). Another key epigenetic marker of wound-induced transcriptional activation of plant genes is histone 3 acetylation, as inhibition of histone acyltransferases abolishes wound callus formation (Rymen et al., 2019). Interestingly, many rapidly wound-induced genes are pre-primed for transcription with acetylation marks, poising them for induction by wound-activated transcriptional regulators (Rymen et al., 2019). It may be that the plant genome is more effectively cleansed of inhibitory epigenetic marks upon wounding, making it more plastic in the regenerative phase. Therefore, we need to be looking to plants – just as much as to axolotls – to identify the best epigenetic clues to enable better regenerative capacity in our own tissues and organs following damage.

Although all organisms have developed strategies for repair, preventing damage in the first place by shielding or protecting tissue is clearly advantageous and will also have been evolutionarily selected for wherever the cost is not too great. Permanent examples of this include camouflage or mimicry, or obnoxious taste, but there are also more transient resilience mechanisms that prime tissues to minimise damage. Recent studies in flies and mice have shown a series of signalling pathways activated by damage itself, and by the associated inflammatory response, which triggers local upregulation of enzymes that sequester ROS, as well as upregulation of DNA and protein repair machineries (Weavers et al., 2019; Haertel et al., 2014; Telorack et al., 2016). These mechanisms that transiently enhance tissue repair capability would be extremely useful to harness in the clinic, for example prior to elective surgery, and might also be beneficial to activate in crops prior to times of potential infestation. However, whether plants have equivalent ‘resilience’ mechanisms remains to be addressed.

Plants clearly have mechanical strategies for detecting impending attack, utilising their sensory trichomes (Box 1), which, when disturbed, can lead to upregulation of antipredator toxins before the first bite (Zhou et al., 2017; Peiffer et al., 2009). Moreover, plants exhibit priming of defence pathways in response to sequential stimuli, with damage or even soft mechanical stress leading to localised disease resistance (Benikhlef et al., 2013; Chassot et al., 2008). Prior exposure to pathogens has powerful priming effects on several key defence proteins, leading to systemic acquired resistance to infection (Conrath et al., 2002). As we better understand these natural resilience mechanisms, more advanced treatments might be developed to mimic or activate these systemic resistance pathways to improve crop survival (Kohler et al., 2002), and some of these strategies may also be ‘borrowed’ for use in the clinic.

What this brief survey of repair mechanisms in plants and animals reveals is that there are indeed several strikingly analogous mechanisms, including ROS and calcium signalling, and metabolic alterations (Fig. 1). However, we highlight clear gaps in areas that are well studied in one kingdom but underexplored, and thus somewhat ‘under the radar’, in the other. For example, much is known about the role of AMPs in plant repair, but this has been studied to a much lesser extent in animal models besides Drosophila (Hanson and Lemaitre, 2020). The rise of antimicrobial resistance has created a demand for antimicrobials and pesticides in the clinic and in agriculture, and the cross-species application of AMPs could hold potential solutions for this crisis. An example of where cross-kingdom application is already happening is the plant defensin NmDef02, which has been recombinantly expressed in Escherichia coli and was shown to exhibit antimicrobial activity against human pathogens in vitro, as well as against plant pathogens (Ceballo et al., 2022). Similarly, the analogous resistance and resilience mechanisms, currently being uncovered in both plants and animals, can offer insights for research direction in diverse phyla. For example, ‘dampening’ mechanisms that limit cell death in plants may hold promise for novel therapeutics aiming to limit fibrosis in patients (Box 2). Now might be an ideal time for more systematic comparative studies on how plants and animals heal their wounds, and an opportunity for the two fields to learn from each other in ways that expand our options for novel treatments to enhance repair in the clinic and to replace pesticides and other agricultural treatments in crop production.

We thank Ashley Pridgeon and Keara Franklin (University of Bristol) for inspiring and supporting us through our exploration of the plant wound response; Stephanie Foster for providing us with images of her wounded hand; and the members of the Martin laboratory for helpful discussion throughout.