How the diversity in digestion in carnivorous plants may have evolved

Summary

Carnivorous plants secrete digestive enzymes for prey degradation. Although carnivorous plants have a polyphyletic origin and evolved several times independently, they surprisingly co‐opted similar digestive enzymes during convergent evolution. However, despite having similar digestive enzymes, the mode of their regulation strongly differs across different phylogenetic lineages. But what factors are responsible for such diversity in their digestion? By combining phylogenetic relationships of digestive fluid proteins and biochemical data, the analyses showed that phylogeny seems to be a significant factor determining the regulation of digestion, but environment (water vs terrestrial) and type of trap do not affect regulation. The oldest carnivorous plant lineage, Caryophyllales, co‐opted phytohormone jasmonic acid (JA) for regulation of digestive enzyme activity. However, the remaining orders of carnivorous plants do not accumulate JA in response to prey capture, and their digestive enzyme activity is not responsive to exogenous JA application. Instead, they use different modes of regulation, for example, development/senescence, osmotically induced and constitutive. These different modes of regulation can be explained by co‐option, albeit of similar genes but different paralogs with different cis regulatory elements that have been fine‐tuned during evolution.

Introduction

Carnivorous plants have the ability to capture, kill, and digest animal prey in specialized traps. They absorb nutrients from the victims and utilize them for plant growth and development (Adamec et al., ²⁰²¹). Carnivorous plants' ability to digest animal prey is most fascinating but has also been shrouded in mystery. It is remarkable that at the beginning of the 21^st century, 30 years after the first man landed on the moon's surface, we had no convincing evidence about the origin of digestive enzymes in carnivorous plants: whether they use endogenous digestive enzymes or rely only on symbiotic microorganisms in digestion. Although there were numerous attempts to chromatographically and electrophoretically purify the digestive enzymes mainly from Nepenthes pitcher plants (Steckelberg et al., ¹⁹⁶⁷; Jentsch, ¹⁹⁷²; Tökes et al., ¹⁹⁷⁴; Heslop‐Harrison, ¹⁹⁷⁵), doubts about the origin of the enzymes in digestive fluids persisted in the last century (for review, see Frazier, ²⁰⁰⁰). But the situation changed in 2004, when a group of Japanese scientists, led by Professor Takahashi, isolated and purified to homogeneity the first endogenous digestive enzyme aspartic protease called nepenthesin from Nepenthes (Athauda et al., ²⁰⁰⁴). They cloned cDNAs for the enzyme from the pitcher tissue of Nepenthes gracilis to deduce the complete amino acid sequence and primary structure determination to confirm that it is indeed a plant‐derived enzyme. Later, a boom in LC‐MS analyses led to the identification of other digestive enzymes in the digestive fluid of carnivorous plants. The analyses found that they are usually related to pathogenesis‐related (PR) proteins, that is, proteins which plants use for defence (Hatano & Hamada, 2008, 2012; Mithöfer, ²⁰¹¹; Lee et al., ²⁰¹⁶). This was the first convincing evidence for the old hypothesis that botanical carnivory has evolved from a plant defence mechanism, as had been suggested before (Juniper et al., ¹⁹⁸⁹). However, after the discoveries of endogenous digestive enzymes, a new question arose: how is the activity and gene expression of the enzymes regulated? And here again the plant defence mechanisms came to the aid of carnivorous plants. Phytohormones from the group of jasmonates (JAs) were discovered to play a significant role in the regulation of digestive enzyme expression in the genera Dionaea, Drosera, and Nepenthes (Libiaková et al., ²⁰¹⁴; Bemm et al., ²⁰¹⁶; Yilamujiang et al., ²⁰¹⁶; Krausko et al., ²⁰¹⁷; Pavlovič et al., ²⁰¹⁷; Pavlovič & Mithöfer, ²⁰¹⁹). JAs are important regulators in plant responses to biotic and abiotic stresses as well as in development processes (Wasternack & Hause, 2013). Mechanical wounding or herbivore attack is one of the most prominent examples where JAs are involved as a signal (Koo & Howe, 2009). The endogenous rise of JAs upon wounding or herbivore and pathogen attack is associated with the induction of synthesis of secondary metabolites and activation of the defence response (Wasternack & Hause, 2013). Particularly, the bioactive isoleucine conjugate of JA (JA‐Ile) triggers an interaction between the CORONATINE INSENSITIVE1 (COI1) receptor and members of the JASMONATE ZIM‐DOMAIN (JAZ) family of repressors. COI1‐mediated degradation of JAZ repressors activates the reprogramming of gene expression leading to the plant defence response (Chini et al., ²⁰⁰⁷; Thines et al., ²⁰⁰⁷; Fonseca et al., ²⁰⁰⁹; Sheard et al., ²⁰¹⁰). In carnivorous plants, JA‐Ile activates the digestive process and thus, carnivorous plants co‐opted JA‐mediated plant defence mechanisms for botanical carnivory. However, despite the fact that JAs are ubiquitous for plant defence mechanisms (Chini et al., ²⁰²³), are they ubiquitous also for botanical carnivory? Recent studies have shown that different carnivorous plant genera employed different modes of regulation of digestion (Nishimura et al., ²⁰¹³; Pavlovič et al., ²⁰²⁴; Procko & Chory, ²⁰²⁴). Here, I will discuss how this mode is dependent on phylogeny rather than on trap types and habitats, and I will outline the molecular mechanisms by which different modes of regulation may have evolved.

Mechanism of prey capture

Carnivorous plants employ five different trapping mechanisms for prey capture: adhesive (‘flypaper’) traps with a sticky glandular surface (e.g. Drosera, Drosophyllum, Pinguicula, and Byblis); pitfall (‘pitcher’) traps forming a central cavity or small tanks filled with digestive fluid (e.g. Nepenthes, Sarracenia, and Cephalotus); mobile snap‐traps with rapidly closing lobes (e.g. Aldrovanda and Dionaea); suction (‘bladder’) traps suck the prey by actively forming negative pressure inside (e.g. Utricularia); and specialized eel (‘lobster‐pot’ and ‘cork‐screw’) traps using inward‐pointing hairs to force prey to move towards a digestive organ (e.g. Genlisea; Fig. 1; Pavlovič & Saganová, ²⁰¹⁵; Adamec et al., ²⁰²¹). These different trap types have evolved as a result of convergent as well as divergent evolution (Thorogood et al., ²⁰¹⁸). For example, pitfall traps have evolved several times independently in different evolutionary lineages as a result of convergent evolution (Cephalotus in Oxalidales; Sarracenia, Darlingtonia, and Heliamphora in Ericales; Nepenthes in Caryophyllales; Brocchinia and Catopsis in Poales). Sticky traps have evolved also several times independently (Drosera, Drosophyllum, and Triphyophyllum in Caryophyllales; Byblis, Pinguicula, and Philcoxia in Lamiales; Roridula in Ericales). On the other hand, different trap types have evolved within one evolutionary lineage as a result of divergent evolution (e.g. Caryophyllales; sticky trap in Drosera, Drosophyllum, and Triphyophyllum; pitfall traps in Nepenthes; snap traps in Aldrovanda and Dionaea). The results we currently have indicate that JA signalling is suitable for botanical carnivory in different trap types: in sticky traps of Drosera and Drosophyllum (Krausko et al., ²⁰¹⁷; Pavlovič et al., ²⁰²⁴); in snap traps of Dionaea and Aldrovanda (Libiaková et al., ²⁰¹⁴; Jakšová et al., ²⁰²¹); and in pitfall traps of Nepenthes (Fig. 2, Buch et al., ²⁰¹⁵; Yilamujiang et al., ²⁰¹⁶; Saul et al., ²⁰²³). Although all these traps employ JAs for regulation of digestive enzyme activity, the degree of JA dependence is determined by trap types. For example, the snap traps of Dionaea and Aldrovanda are completely dependent on JAs and they do not produce any digestive enzymes and fluid without prey stimuli (Nishimura et al., ²⁰¹³; Libiaková et al., ²⁰¹⁴; Jakšová et al., ²⁰²¹). Sticky traps of Drosera and Drosophyllum contain a small amount of digestive enzymes already in noninduced sticky fluid without any captured prey (Krausko et al., ²⁰¹⁷; Pavlovič et al., ²⁰²⁴) and at least in Drosera, their expression is tissue specific with predominant expression in tentacles in comparison with other plant organs (Nishimura et al., ²⁰¹³; Arai et al., ²⁰²¹). This tissue specificity is regulated by promoter methylation status; that is, in carnivorous organs and glands the promoter is unmethylated suggesting that the promoters do not form a closed chromatin structure and in the presence of transcription activators (or rather absence of JAZ repressor based on our knowledge about JA‐signalling, Thines et al., ²⁰⁰⁷) can be transcriptionally active (Nishimura et al., ²⁰¹³; Arai et al., ²⁰²¹). This basal promoter activity and transcription of digestive enzyme genes can be further strongly enhanced by prey‐induced JA accumulation (Krausko et al., ²⁰¹⁷; Pavlovič et al., ²⁰²⁴). The pitcher plants Nepenthes, which probably evolved from sticky traps within Caryophyllales (Thorogood et al., ²⁰¹⁸), are probably the least dependent on JAs. The pitcher fluid contains many digestive enzymes even in closed immature pitchers and protein levels increase dramatically during pitcher maturation without any prey stimuli (Eilenberg et al., ²⁰⁰⁶; Hatano & Hamada, ²⁰⁰⁸). The level of digestive enzymes in the fluid is autoregulated as indicated by protein and fluid depletion experiments (Miguel et al., ²⁰¹⁹; Wan Zakaria et al., ²⁰¹⁹; Goh et al., ²⁰²⁰). Nevertheless, prey stimuli can enhance the expression and secretion of many digestive enzymes through JAs in Nepenthes (Buch et al., ²⁰¹⁵; Yilamujiang et al., ²⁰¹⁶; Saul et al., ²⁰²³). This may indicate that the expression of digestive enzymes in Nepenthes is under the control of more regulatory elements and its JA and prey‐induced regulation is more relaxed in comparison with Dionaea. On the other hand, the same trap types in different genera of carnivorous plants regulate digestive enzyme activity independently of JAs (e.g. sticky traps of Byblis and Pinguicula, pitfall traps of Cephalotus and Sarracenia). Some of them show continuous enzyme secretion and the enzymes accumulate over time during pitcher ontogeny without any prey stimuli (e.g. Sarracenia), whereas others activate enzyme secretion in response to prey capture by an unknown mechanism (e.g. Byblis, Pinguicula; Kocáb et al., ²⁰²⁰; Pavlovič et al., ^{2024, 2025}). Although Pinguicula can accumulate JAs in response to prey capture, probably as a result of general trichome physiology (Kocáb et al., ²⁰²⁰; Matsumura et al., ²⁰²²; Sun et al., ²⁰²⁴), its digestive enzymes are probably not JA responsive, as evidenced by experiments with exogenous coronatine (i.e. agonist of JA‐Ile signal) application (Kocáb et al., ²⁰²⁰).

Fig. 1.

Different types of traps in carnivorous plants. (a) Adhesive trap of sundew Drosera rotundifolia, (b) pitcher trap of Nepenthes mirabilis, (c) snap trap of Venus flytrap (Dionaea muscipula), (d) bladder suction trap of aquatic bladderwort Utricularia reflexa, (e) eel trap of Genlisea hispidula.

Fig. 2.

Phylogeny of jasmonate (JAs) signalling in different carnivorous plant lineages. Carnivorous plants have evolved at least 11 times independently (arrowheads) in six orders of angiosperms. The genera of carnivorous plants in which the JAs are involved in the regulation of enzyme secretion are depicted in red. The genera in which JAs do not play any role in the regulation of enzyme activities are depicted in blue. The genera that have not yet been investigated in this respect are depicted in green. The genera that use fast electrical signalling are marked by a symbol of action potential. The genera using JA signalling form a clear cluster in the Caryophyllales order without any association with the type of environment (aquatic/terrestrial) or trap type. The numbers depict estimated phylogenetic age in million years ago according to Fleischmann et al. (2018). Adapted from Pavlovič et al. (2024).

Thus, in summary, the mechanism of prey capture does not determine the involvement of JAs in the regulation of enzyme secretion but may slightly fine‐tune it (Fig. 2). Where the prey capture is a dynamic process, like rapid trap movement in Dionaea and Aldrovanda, there was probably a strong positive selection pressure for co‐option and conserving the JA signalling pathway for rapid regulation of the digestive process, which is also involved in rapid JA‐mediated plant defence. Where there is a passive mechanism for prey capture, where the prey slowly accumulate and decompose over time in pitcher traps, there has probably been a relaxation of functional constraints on JAs (e.g. Nepenthes which has probably evolved from sticky traps; Thorogood et al., ²⁰¹⁸) or JA has not been co‐opted at all (e.g. Sarracenia; Pavlovič et al., ²⁰²⁴), and digestive enzymes might additionally be regulated also JA‐independently driven by osmotic changes in traps, pitcher ontogeny, autoregulatory mechanisms, or by unknown mechanisms.

Aquatic vs terrestrial habitats

Approximately 60 out of 810 species of carnivorous plants are submerged aquatic plants, comprising the monotypic Aldrovanda vesiculosa and dozens of Utricularia species (Adamec et al., ²⁰²¹). Although the remaining carnivorous plants are terrestrial, they often grow in wet environments, often rooted in waterlogged soil, and can even be temporarily seasonally submerged. Some carnivorous plants can tolerate seasonally dry environments (e.g. Drosophyllum lusitanicum) but they often give up carnivory temporarily (e.g. Mexican Pinguicula, Australian pygmy Drosera; Paniw et al., ²⁰¹⁷; Pavlovič, ²⁰²²). A recent study has shown that the submergence of Arabidopsis thaliana plants deactivates wound‐induced plant defence against herbivores mediated by JAs (Lee et al., ²⁰²⁰). This finding motivated us to perform experiments with aquatic carnivorous plants, and we found that aquatic A. vesiculosa, which is sister to terrestrial D. muscipula, is fully able to activate the JA signalling pathway for the regulation of digestive enzyme activity. It clearly accumulated JAs in response to feeding, and digestive enzymes responded to their exogenous application (Jakšová et al., ²⁰²¹). It has also strongly expanded genes encoding JA signalling in its genome (Palfalvi et al., ²⁰²⁰). By contrast, aquatic Utricularia reflexa does not use JAs for the regulation of enzyme activity (Jakšová et al., ²⁰²¹), and the expression of digestive enzymes in Utricularia is probably constitutive (Sirová et al., ²⁰⁰³) and tissue specific, with high expression in traps in comparison with shoots and inflorescence (Lan et al., ²⁰¹⁷). Following the Lee et al. (2020) protocol, we artificially flooded the terrestrial sister of Aldrovanda, Dionaea plants, and investigated their ability to trigger the digestive process in an aquatic environment. Dionaea plants can often grow in submerged environments in natural habitats. Repeated mechanical stimulation under the water triggered hermetical trap sealing, accumulation of JA and JA‐Ile, and expression of digestive enzymes, but to a lesser extent (Jílková, 2023). This suggests, combined with the fact that its sister genera Aldrovanda retained its use of JA signalling for carnivory, that an aquatic environment is not an obstacle for using JA signalling. Aldrovanda vesiculosa probably entered the aquatic environment already as a carnivorous plant (Cameron et al., ²⁰⁰²), and JA signalling was retained during this transition. Even on the opposite moisture scale, the most drought‐tolerant carnivorous plant, Drosophyllum lusitanicum (Paniw et al., ²⁰¹⁷), co‐opted JAs as its aquatic relative Aldrovanda (Pavlovič et al., ²⁰²⁴). Thus, the water environment was probably not an important selection factor determining JA signalling for the regulation of botanical carnivory (Fig. 2).

Phylogeny

Carnivorous plants have a polyphyletic origin and have evolved at least 11 times independently in a time span of 95.1–1.9 Mya (Fleischmann et al., ²⁰¹⁸) in 6 orders, 13 families, and 20 genera (Adamec et al., ²⁰²¹ and updated by the new genera Triantha, Lin et al., ²⁰²¹). Studies on the regulation of digestive enzyme activity in carnivorous plants during the last decade indicate that JAs have been co‐opted probably only once in the oldest order of carnivorous plants, Caryophyllales (Fig. 2, Pavlovič et al., ²⁰²⁴). These plants include terrestrial (e.g. Dionaea) as well as aquatic carnivorous plants (e.g. Aldrovanda) with different trapping mechanisms (e.g. sticky traps in Drosera and Drosophyllum; snap traps in Dionaea; and pitcher traps in Nepenthes) as mentioned above. Although similar trapping strategies and the same digestive enzymes also occur in the remaining evolutionary lineages of carnivorous plants (Fukushima et al., ²⁰¹⁷), it seems that JAs have never been recruited again. The pitcher traps of Sarracenia (order: Ericales) and Cephalotus (order: Oxalidales) rely more on developmental combined with water‐induced regulation (i.e. osmotic regulation) and constitutive enzyme secretion, respectively (Gallie & Chang, 1997; Nishimura et al., ²⁰¹³; Pavlovič et al., ²⁰²⁴). So, how is it possible that the same digestive enzymes are regulated so differentially in different evolutionary lineages of carnivorous plants?

Now based on the studies on the Caryophyllales order, there is a consensus that carnivorous plants co‐opted plant defence mechanisms for botanical carnivory because many digestive enzymes found in the secretome of carnivorous plants are involved in plant defence reactions in noncarnivorous plants (e.g. aspartic and cysteine proteases, chitinases, class III peroxidases, and β‐1,3‐glucanases; Mithöfer, ²⁰¹¹; Bemm et al., ²⁰¹⁶). However, this view on botanical carnivory can be oversimplified. Whereas some genes are indeed involved in plant defence mechanisms and belong to the PR‐proteins (e.g. chitinases, Loon et al., ²⁰⁰⁶) and are strongly induced by pathogen/herbivore attack and defence‐related hormones (i.e. JA, Dávila‐Lara et al., ²⁰²¹), others, for example aspartic proteases, are a group of very diverse proteolytic enzymes with a broad spectrum of biological roles, with, for example, 78 genes encoding putative aspartic proteases in Arabidopsis (Yu & Feng, 2025). Plant aspartic proteases are suggested to undergo functional specialization and to be crucial in developmental processes, as well as in biotic and abiotic stress responses (Simões & Faro, 2004; Figueiredo et al., ²⁰²¹; Yu & Feng, ²⁰²⁵). In carnivorous plants, aspartic proteases have been co‐opted for botanical carnivory several times independently from paralogous gene lineages, mainly from atypical, nucellin‐like aspartic proteases (Athauda et al., ²⁰⁰⁴; Fukushima et al., ²⁰¹⁷; Soares et al., ²⁰¹⁹) and also from saposin‐like aspartyl protease (e.g. Cephalotus; Fukushima et al., ²⁰¹⁷). Because cis‐regulatory elements are usually conserved in orthologous gene promoter sequences across diverging species, it means that whereas some lineages might indeed have co‐opted homologs of aspartic proteases that are induced by pathogen/herbivore attack and were further improved for more stringent JA control (e.g. Caryophyllales), others might have co‐opted homologs of aspartic proteases involved in development/senescence (or developmentally‐regulated defence) and thus with different regulation (Fig. 3). For example, the closest orthologs to the aspartic protease found in the digestive fluid of S. purpurea in A. thaliana are aspartic protease CHLOROPLAST NUCLEOID DNA BINDING PROTEIN 41 (CND41, AT5G10770) and APOPLASTIC, EDS1‐DEPENDENT 1 (AED1, AT5G10760), which clearly show developmental, osmotic regulation and a role in senescence (Kato et al., ²⁰⁰⁴; Fukushima et al., ²⁰¹⁷). Similarly, the closest ortholog to the aspartic protease found in the digestive fluid of C. follicularis in A. thaliana is CONSTITUTIVE DISEASE RESISTANCE (CDR1, AT5G33340, Fukushima et al., ²⁰¹⁷) which shows rather constitutive expression and salicylic acid (SA)‐induced response (Xia et al., ²⁰⁰⁴). These orthologs also do not show significant insect‐stimulated expression in Dionaea (see in Rentsch et al., ²⁰²⁴, and for expression profile of Arabidopsis orthologs see The Arabidopsis Information Resource (TAIR) or Fukushima et al. (2017)). On the other hand, the closest relative ortholog of the aspartic protease to Dionaea and Nepenthes in A. thaliana (AT2G03200) exhibits mild JA regulation (Fukushima et al., ²⁰¹⁷), which was probably further improved during evolution. This may explain why the expression and secretion of aspartic proteases are regulated developmentally and osmotically in S. purpurea, are rather constitutive in C. follicularis, and are regulated by JA in the Caryophyllales order (Pavlovič et al., ^{2017, 2024}). The same situation is also clear for β‐1,3‐glucanases. Although research on β‐1,3‐glucanases to date has focused primarily on their PR function, β‐1,3‐glucanases also play critical roles in normal developmental plant processes (Doxey et al., ²⁰⁰⁷). The closest relative orthologs of β‐1,3‐glucanase to four species of carnivorous plants in A. thaliana exhibit functions not only in plant defence (At3g57260) but also in pollen maturation (At3g57270) (Doxey et al., ²⁰⁰⁷; Michalko et al., ²⁰¹⁷). The next example is class III peroxidases which are also involved not only in the defence response but also in lignification, cell elongation, and seed germination (Shigeto & Tsutsumi, 2016). The closest relative orthologs of class III peroxidase to the carnivorous plant Cephalotus in A. thaliana (At5g58400, At5g58390) do not exhibit strong induction in response to pathogen attack in contrast to other paralogs (e.g. At5g05340; Fukushima et al., ²⁰¹⁷). The closest relative ortholog of type I chitinase to carnivorous Drosera and Dionaea in A. thaliana (AT3g12500) is involved in the ethylene/JA‐mediated signalling pathway during systemic acquired resistance, in contrast to Cephalotus (AT3g54420). From these examples, it is obvious that some of the genes encoding digestive enzymes in carnivorous plants are regulated similarly to their orthologs in noncarnivorous plants. But there are also indications that the regulation of some of the genes was adjusted later in evolution. Because the promoters of genes encoding digestive enzymes contain several regulatory cis‐acting elements (Arai et al., ²⁰²¹; Libantová et al., ²⁰²¹; Mikitová et al., ²⁰²⁵), it is highly probable that they were adjusted lineage specifically. Nutrient transporters are usually constitutively expressed in roots; however, in carnivorous Droseraceae, they are induced by prey and JAs (Palfalvi et al., ²⁰²⁰). Indeed, co‐option may involve the addition of novel cis‐regulatory elements (True & Carroll, 2002). Palfalvi et al. (2020) found that genes associated with carnivory in Droseraceae are under the control of WRKY transcription factors, which exhibit sub‐functionalization following the duplication event. WRKY transcription factors often act as a positive regulator in JA‐mediated plant defence in noncarnivorous plants and could play an important role in the recruitment of genes to carnivory‐related functions in Caryophyllales (Palfalvi et al., ²⁰²⁰). Indeed, the promoter region of genes encoding digestive enzymes in Droseraceae is enriched in WBOX sequences (Arai et al., ²⁰²¹, Libantová et al., ²⁰²¹, Mikitová et al., ²⁰²⁵). For example, the number of WBOX sequences (WRKY binding site) in the promoter region of S‐like RNAse in Dionaea is four times higher than in Cephalotus (Nishimura et al., ²⁰¹³).

Fig. 3.

Hypothesis of how different modes of regulation of digestive enzyme activity in carnivorous plants may have evolved. Some genes (group I) encode pathogenesis‐related (PR) proteins, which were already jasmonate responsive in non‐carnivorous plants, for example type I chitinase (ChitI). They could undergo duplication and subfunctionalization for pathogenic response (subclass Ia) and carnivory (subclass Ib) as is evident in the Caryophyllales order (Renner & Specht, 2012, 2013). On the other hand, type III chitinase was probably retained as a single gene for both functions (not shown as we still have few data about its JA regulation, Rotloff et al., ²⁰¹¹). Other genes (group II) encode enzymes with a broad spectrum of biological roles which evolved by gene duplications and their functional divergence. One copy of the gene retained its function and a new copy of the gene acquired a new function. Different lineages of carnivorous plants co‐opted different paralogs of genes encoding, for example, aspartic protease (AP), class III peroxidase (PerIII) and β‐1,3‐glucanases (Glu) for carnivory with different sets of cis‐regulatory elements; some of them were involved in defence and some of them in development including senescence and/or developmentally or constitutively regulated defence. They could be retained as a single copy gene for both functions or they (neo‐)subfunctionalized. They were probably also adjusted and fine‐tuned lineage specifically later in evolution, which may explain that in genera of the Caryophyllales order, the digestive enzymes are under the control of JA and in the orders Ericales, Lamiales and Oxalidales are regulated JA‐independently (e.g. developmentally, constitutively). Other genes (group III) encode enzymes involved in plant development like α‐amylase (amy) and are examples of genes with JA‐independent regulation. So far we have still very little information about their molecular evolution in carnivorous plants. Plant JA‐regulated defence‐related genes are shown in red, and developmentally regulated genes are shown in blue. The insets show Western blots of secreted aspartic protease into the digestive fluid of Drosophyllum lusitanicum, which is JA‐responsive (top) and Sarracenia purpurea, which is not (bottom), regulating enzyme secretion rather developmentally (according to Pavlovič et al., ²⁰²⁴). The simplified scheme. [Correction added on 7 June 2025, after first online publication: the word ‘Speciation’ was updated to ‘Divergence’ within Fig. 3].

In contrast to the above‐mentioned enzymes, the enzymes involved in phosphate utilization, RNAses and acid phosphatases, showed unexpected orthologous relationships among nonrelated carnivorous plants with convergent amino acid changes (Nishimura et al., ²⁰¹⁴; Fukushima et al., ²⁰¹⁷). So, even when different evolutionary lineages co‐opted different paralogs, they were under the strong selective pressure of a carnivorous lifestyle and accumulated convergent amino acid changes (Nishimura et al., ²⁰¹⁴; Fukushima et al., ²⁰¹⁷). But despite convergent amino acid changes in the coding region, they are also regulated differentially (Nishimura et al., ²⁰¹³), indicating that promoters were under different selection pressure in comparison with the coding region of the gene in different lineages and/or trap types.

Recently, we identified the unique enzyme α‐amylase found in the digestive fluid of butterwort Pinguicula, which was not induced by JAs (Kocáb et al., ²⁰²⁰). The α‐amylases are enzymes that are mainly involved in initiating starch degradation and plant development (Stanley et al., ²⁰⁰⁵). There are three families of α‐amylases in plants differing in their cellular localization. None of these proteins are more than 48% identical to each other and are thus not likely to have arisen from recent gene duplications. Family 1 is secretory; family 2 has no predicted targeting peptide, and therefore is thought to localize to the cytoplasm; and family 3 contains a predicted chloroplast transit peptide. Family 1 contains well‐characterized cereal grain α‐amylases secreted from the cells of the aleurone layers to the endosperm promoting germination (Stanley et al., ²⁰⁰⁵). The presence of this enzyme in the secretome of carnivorous plants has not been found in any other genera of carnivorous plants before and is unique for Pinguicula. The function of family 1 in dicotyledons is much less understood than in cereals, but in the leaf of A. thaliana, the AT4g25000 ortholog is induced by different biotic and abiotic stresses (induced by abscisic acid, ABA; but not JAs) and senescence (Stanley et al., ²⁰⁰⁵; Doyle et al., ²⁰⁰⁷). The fact that the Arabidopsis ortholog AT4g25000 is strongly induced by osmotic stress is in line with evidence of osmotically driven digestive fluid secretion mediated by Cl⁻ influx to the head cells of digestive glands in Pinguicula, induced by prey (Heslop‐Harrison & Heslop‐Harrison, 1980). The same is also known in D. muscipula (Rea et al., ¹⁹⁸³) and the hyperosmotic salinity response was among the enriched gene ontology (GO) terms in insect and coronatine stimulated traps in RNA‐seq experiments (Bemm et al., ²⁰¹⁶). The possible regulatory effect of osmotic changes in the glands of carnivorous plants needs further experimentation.

Thus, co‐option of different paralogs of hydrolytic enzymes regulated by different hormones (e.g. ABA, SA, and indole‐3‐acetic acid (IAA)), development, osmotic changes, or senescence can explain their different regulation in different lineages of carnivorous plants and are motivating for the search of other regulatory elements involved in botanical carnivory. This however does not rule out the possibility that many of these paralogs are also involved in plant defence, because, for example, plant senescence is the developmental process leading to nutrient remobilization and finally the degradation and death of tissue and can be triggered by many pathogens (Häffner et al., ²⁰¹⁵). Thus, co‐option of defence‐related genes has been probably the main driving force for evolution of botanical carnivory. But why only the oldest lineage of carnivorous plants co‐opted the plant defence hormone JA as a master factor for regulation of carnivory is unclear, as are the environmental factors which were the driving force for such co‐option in Caryophyllales 90 Mya ago. It is quite interesting that the other well‐known plant defence hormone SA has not been found to regulate digestive enzyme expression in Drosera, Nepenthes, or Pinguicula (Matušíková et al., ²⁰⁰⁵; Buch et al., ²⁰¹⁵; Krausko et al., ²⁰¹⁷; Kocáb et al., ²⁰²⁰). Moreover, no other phytohormone has been shown to be able to trigger the expression of digestive enzymes except JA in carnivorous plants and this issue is open to further investigation. Also phytohormone crosstalk, which is crucial for adjusting plant responses to environmental stimuli in different situations, is highly understudied in carnivorous plants. The antagonism between ABA and JA in Dionaea, and SA and JA in Drosera in regulating excitability and enzyme secretion, respectively, has already been described (Escalante‐Pérez et al., ²⁰¹¹; Krausko et al., ²⁰¹⁷), as well as possible synergism between JA and IAA in trap folding reaction in D. capensis plants (La Porta et al., ²⁰¹⁹).

Future perspectives

JAs are important regulators of botanical carnivory in the oldest lineage of carnivorous plants, Caryophyllales, where they were co‐opted from plant defence mechanisms. The speed of response in plant defence mechanisms is mediated by JAs, which can accumulate very quickly in response to herbivore or pathogen attack (Kimberlin et al., ²⁰²²). This speed is also very important in capturing and retaining insect prey in carnivorous plants, so it is not surprising that the fastest carnivorous plants Aldrovanda, Dionaea, and Drosera have co‐opted JAs for fast response to prey. But now it has become obvious that co‐option of JA signalling is the exception rather than the rule, and many carnivorous plants regulate enzyme activity differently. Co‐option of different paralogs with different cis‐regulatory elements might provide genetic background for further fine‐tuning of regulatory properties. Ontogeny and autoregulating mechanisms are important in the regulation of enzyme expression in some groups of carnivorous plants with passive traps, but which signalling molecules are involved in other genera, which strongly induce enzyme activity in response to prey capture, remains unknown. These plants usually possess glandular trichomes (e.g. Byblis and Pinguicula), which in noncarnivorous plants act as mechanosensors through Ca²⁺ signalling (Matsumura et al., ²⁰²²; Sun et al., ²⁰²⁴). Although Ca²⁺ signal is also important for the initiation of JA response in carnivorous plants within the Caryophyllales order (Suda et al., ²⁰²⁰; Procko et al., ²⁰²²), there are many examples where Ca²⁺ acts independently of JAs (e.g. calmodulin‐binding transcription activator (CAMTA)‐dependent pathway; Darwish et al., ²⁰²²; Matsumura et al., ²⁰²²). Whether this general trichome physiology has been co‐opted for botanical carnivory outside the Caryophyllales order is not clear. Also, chemical signals (e.g. protein and chitin) from insect prey may play an important role, as is known in the Caryophyllales order. How these signals are recognized by carnivorous plants is not clear (Bemm et al., ²⁰¹⁶; Jakšová et al., ²⁰²⁰). There is also a possibility that the secretion of digestive enzymes can be driven by osmotic changes in digestive glands, as the expression of many Arabidopsis orthologs is regulated by osmotic changes, and rapid Cl⁻ uptake in digestive glands, followed by the flow of water, occurs in response to prey capture (Heslop‐Harrison & Heslop‐Harrison, 1980; Rea et al., ¹⁹⁸³). Such osmotic regulation has not been investigated in carnivorous plants so far but has been shown to play a role in the release of presynthesized enzymes from gland cell walls (Heslop‐Harrison & Knox, 1971; Kocáb et al., ²⁰²⁰). Recent success in the dissection of Ca²⁺ and Cl⁻ signalling optogenetically (Ding et al., ²⁰²⁴; Hedrich & Gilliham, ²⁰²⁵) may be useful for the investigation of related processes in carnivorous plants, as both ions play a role in the initiation of digestion (Heslop‐Harrison & Heslop‐Harrison, 1980; Rea et al., ¹⁹⁸³; Suda et al., ²⁰²⁰). Transformation of carnivorous plants from different orders outside the Caryophyllales order by calcium sensor indicators and channelrhodopsins for optogenetic control can help answer these questions. The problems in the past were the recalcitrance of carnivorous plants and the difficulty in their transformation and generation of transgenic and mutant plants, which have recently been overcome (Oropeza‐Aburto et al., ²⁰²⁰; Suda et al., ²⁰²⁰; Procko et al., ^{2022, 2023}; Coronado‐Martín et al., ²⁰²⁴). Also, the database of digestive enzyme sequences is growing (Fukushima et al., ²⁰¹⁷). These genetic tools may become very useful in further studies focused on digestion regulation in carnivorous plants. The classic approach using phytohormonics, proteomics, and transcriptomics has become very useful to unravel the involvement of JAs in the regulation of botanical carnivory and can also be helpful in future studies. Mainly, transcriptomic studies outside the Caryophyllales in response to the application of different stimuli (e.g. prey and phytohormones) can be important when looking for a new regulatory mechanism (Fig. 4).

Fig. 4.

Box showing what we know and future questions and steps in the research on the regulation of enzyme expression and secretion in carnivorous plants.

Competing interests

None declared.

Disclaimer

The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations.