Class III Peroxidases: Functions, Localization and Redox Regulation of Isoenzymes

Abstract

Class III peroxidases (POXs; EC. 1.11.1.7), are secretory, multifunctional plant enzymes that catalyze the oxidation of a variety of substrates by hydrogen peroxide (H₂O₂). They show a remarkable diversity of isoenzymes, are encoded by a large number of paralogous genes, and are involved in a broad range of metabolic processes throughout plant growth and development. Peroxidases isoenzymes are located in the cell wall, apoplast and vacuole, and may be either soluble or ionically and covalently cell wall bound. They are involved in cell wall cross-linking and loosening, lignification and suberization, auxin catabolism and secondary metabolism. Due to their ability to control the levels of reactive oxygen species (ROS), POXs are efficient components of the antioxidative system induced in response to environmental stress, such as pathogen attack, metal excess, salinity, drought and high light intensity. In addition to the peroxidative function, POXs can catalyze H₂O₂ production in the oxidative cycle. Peroxidases are responsible either for cell elongation or cell wall stiffening, affecting carbon allocation, auxin level and redox homeostasis, which implicates their key role as being in the regulation of growth and defence under stress condition. This chapter will discuss novel insights into the functions of PODs with special emphasis on their localization, substrate specificity and the regulation of redox homeostasis.

1 Introduction

Under optimal physiological conditions, reactive oxygen species (ROS) accumulation and distribution are kept in the controlled range by a complex network of antioxidative enzymes and low molecular weight antioxidants. Intracellular accumulation of ROS is generally lower, while that of antioxidants is higher compared with the apoplast/cell wall compartment, which is, therefore, more oxidized. When accumulated above the certain threshold in the apoplast, H₂O₂ triggers signaling pathways and initiates a complex cascade of downstream events leading to the induction of the antioxidative defence response and transcriptional reprogramming.

The only H₂O₂-scavenging enzymes in the extracellular space are class III peroxidases (POXs; EC 1.11.1.7) classified by Welinder (1992). Peroxidases are glycoproteins synthesized in the endoplasmic reticulum and they are transported via the Golgi apparatus, both to the extracellular space and the vacuoles. Substrates of the endogenous POXs are electron donors for H₂O₂ and most of them belong to secondary metabolites such as phenolics, indoles and amines (Passardi et al. 2007), which are colocalized with POXs in the vacuoles and apoplastic compartment (Cheynier et al. 2013). Oxidation of phenolics by H₂O₂, catalyzed by POXs implies generation of phenoxyl radicals, which are in turn either polymerized or reduced by ascorbate (Asc) in the POXs/Phenolics/Asc (PPA) H₂O₂-scavenging system (Takahama 2004).

Class III plant peroxidases are a ubiquitous multigene family of enzymes with a number of genes identified in various species that are implicated in a broad range of physiological processes (Passardi et al. 2007). In addition to antioxidative defence, POXs are involved in all stages of plant development, from germination to aging, including formation of the secondary cell wall (Lopez-Serrano et al. 2004; Passardi et al. 2006), wound healing (Allison and Schultz 2004), seed germination (Scialabba et al. 2002), pollination (McInnis et al. 2006), fruit ripening (Pandey et al. 2012), senescence (Abarca et al. 2001), and auxin and anthocyanin catabolism (Gazaryan et al. 1996; Movahed et al. 2016). Such diversity of metabolic reactions catalyzed by POX isoforms, and a large number of genes with very divergent promoter sequences, implies a functional specialization of isoenzymes. However, low substrate specificity and a high redundancy of genes make it difficult to assign a specific function to the individual isoenzymes, which is still a considerable challenge in spite of the many attempts made so far (Shigeto and Tsutsumi 2016 and references therein). Molecular tools such as a transcriptome analysis and recombinant DNA in combination with biochemical and physiological approaches could be successful in addressing the specific function to particular POX gene and its protein, at least in Arabidopsis thaliana as a model plant. Recent work on the transcriptional regulation of a set of POXs in A. thaliana root by a transcription factor UPBEAT 1 (UPB1) suggested the important role of POXs in cell differentiation during root development (Tsukagoshi et al. 2010). Similarly, it has been shown that a transcription factor KUODA1 (KUA 1), which repressed the expression of several POXs genes in the leaf of A. thaliana had a role in the redox homeostasis of leaf apoplast (Lu et al. 2014).

In this chapter, we give an overview of the antioxidative and pro-oxidative reactions of the multifunctional POX subfamily, with emphasis on the characterization of isoenzymes and responsible genes, the isoforms’ specialization in stress defence, and role of POXs in the cross-talk between the primary and secondary antioxidants. Ascorbate and glutathione are the primary metabolites that, together with the antioxidative enzymes of ascorbate–glutathione (Asc–GSH) cycle, maintain redox homeostasis in plant cells (Foyer and Noctor 2011). On the other hand, flavonoids and other phenolic compounds, which are among numerous substrates of POXs, can constitute a ‘secondary’ antioxidant system that is activated upon severe stress conditions (Agati et al. 2012).

We incorporate recent findings on the subcellular distribution of POXs, phenolics, and Asc, as well as the intracellular transport of H₂O₂, which all together lead to the hypothesis that POX acts as a central H₂O₂ sink under severe stress and as a key player in the regulation of defence/secondary metabolism pathways.

Despite, POXs’ numerous physiological functions and their abundance in all plant tissues and organs, their contribution in the antioxidative system has been surprisingly overlooked and disregarded in many excellent review papers on antioxidative defence and tolerance. We believe that the evidence from the literature reviewed here will incorporate POX into the complex cellular antioxidative network.

2 Reaction Mechanisms and Structure of POX Isoenzymes

2.1 Three Cycles of POXs

Class III peroxidases are oxidoreductases that catalyze the oxidation of a large number of different substrates (mainly phenolics) to corresponding radicals in the presence of H₂O₂ as an electron acceptor. In the reaction catalyzed by POX, phenoxyl radical (PhO^·) is produced by the subtraction of one electron from the phenolic compound (PhOH) (reaction 1).

• (1) 2PhOH + H₂O₂ → 2PhO^· + 2H₂O

• (2) 2PhO^· → cross-linking

• (3) PhO^· + Asc → PhOH + MDA^·

• (4) PhO^· + MDA^· → PhOH + DHA

• (5) 2MDA^· → Asc + DHA

In the absence of Asc, the generated PhO^· react with each other, resulting in the formation of cross-linked products (reaction 2); for example, lignin, suberin or quinones (Fry 1986; Bernards et al. 1999). One of the products is a quinhydrone-derived radical, which may be formed from caffeic and chlorogenic acids and their quinones and hydroquinones (Takahama and Oniki 2000). The physiological function of quinhydrone as an electron-charged structure linked to the cell wall has been discussed elsewhere (Kukavica et al. 2008; Morina et al. 2010). Additionally, PhO^· (derived from monophenols, p-coumaric and ferulic acids) may react with Asc (reaction 3), yielding monodehydroascorbyl radical (MDA^·), recovering the corresponding substrate in a non-radical way. Phenoxyl radicals can further react with MDA^· (reaction 4) to form dehydroascorbate (DHA), while Asc recovery is possible through dismutation of 2 MDA^· (reaction 5) (Takahama and Oniki 1992). As mentioned above, PhO^· can dismutate to quinones, such as o-dihydroxyphenolics, caffeic acid and 3,4-dihydroxyphenylalanine quinones (Takahama 1992, 1993).

Besides oxidizing various phenolic compounds during H₂O₂ reduction (peroxidatic cycle), POXs have the capacity to produce H₂O₂ via one-electron reduction of molecular oxygen (oxidative cycle), a mechanism proposed by Chen and Schopfer (1999). The oxidative cycle involves the transfer of one electron from a reducing substrate (NADH or phenolics) to O₂, with ${{\text{O}}_{2}}^{ \cdot - }$ and H₂O₂ as intermediates. In addition, the oxidizing reaction of POX is determined by the trace amounts of dihydroxyphenols and Mn²⁺ (Halliwell 1978).

Horseradish peroxidase (HRP) is the most frequently used POX in in vitro mechanistic and kinetic studies. H₂O₂ coordinatively bonds to the iron atom and initiates the peroxidatic cycle and the oxidation of phenolic substrates to the corresponding radicals. The reduction of Compound I to Compound II, as well as Compound II’s return to the resting state is carried out by reduction substrates (phenolics or aromatic amines).

Superoxide radical converts the ferric form of the enzyme (Fe³⁺-peroxidase) into the labile perferryl form (${\text{Fe}}^{2 + } {-} {{\text{O}}_{2}}^{ - \cdot } {\text{Fe}}^{3 + } {-} {{\text{O}}_{2}}^{ \cdot - }$ POX), otherwise named Compound III. Compound III contains Fe²⁺ in the heme that can be converted to Fe³⁺ to act as a Fenton reagent and reduce H₂O₂ to ^·OH in the hydroxylic cycle. Chen and Schopfer (1999) have demonstrated the ability of different types of POXs (e.g., unfractionated HRP mixture, acidic and alkaline HRP fractions, soybean POX) to catalyze the formation of ^·OH in vitro in the presence of NADH (200 mM) as a reducing substrate. The largest capacity for generating ^·OH was observed for the alkaline HRP fraction and Arthromyces POX, while myeloperoxidase showed the lowest capacity. The authors also showed that, beside NADH, HRP oxidizes NADPH and dihydroxyfumarate, while Asc was completely ineffective as a reducing substrate.

The proposed HRP catalyzed reactions in peroxidatic, oxidative and hydroxylic cycles in vitro with NAD(P)H acting as the exogenous reductant are presented below: Peroxidatic cycle

• (1) POX (Fe³⁺) + H₂O₂ → Comp I + 2H₂O

• (2) Comp I + PhOH → Comp II + PhO^·

• (3) Comp II + PhOH → POX (Fe³⁺) + PhO^·

Oxidative cycle

• (4) NAD^· + O₂ → NAD⁺ + ${{\text{O}}_{2}}^{ \cdot - }$

• (5) NADH + H⁺ + ${{\text{O}}_{2}}^{ \cdot - }$ → NAD⁺ + H₂O₂

Hydroxylic cycle

• (6) POX (Fe³⁺) + ${{\text{O}}_{2}}^{ \cdot - }$ → Comp III

• (7) Comp III + H₂O₂ → POX (Fe³⁺) + ^·OH + OH⁻ + PhOH + O₂

2.2 Structural Characterisation of POX Isoenzymes

Most of POXs are glycoproteins with one polypeptide chain of 300–350 amino acid residues, and their molecular weight (Mw) varies from 33–55 kDa. Though a majority of POXs are monomers, it has been shown that coconut POXs are homotetramers with a Mw of 55 kDa for each subunit (Mujer et al. 1983), POXs from Leucaena leucocephala are heterotrimers (consisting of two subunits of 66 kDa and one subunit of 58 kDa) (Pandey and Dwivedi 2011), while POXs from Brussels sprouts are homodimers (two subunits of 45 kDa) (Regalado et al. 1999).

According to the crystallographic and modeling studies (Ros Barceló et al. 2007), the primary protein structure of POXs consists of 10–12 conserved α-helices in which the prosthetic group is embedded, two short β-strands, and four conserved disulphide bridges (Passardi et al. 2004). There are three highly conserved domains: one is a distal heme-binding domain, the other is unknown, and the third one is a proximal heme-binding domain (Cosio and Dunand 2008). Welinder (1992) showed that all POX proteins, encoded by 73 genes in A. thaliana, were expressed in all organs, and they had up to 98% amino acid sequence identity.

The variation in terms of Mw of POXs isolated from different species may be attributed to the carbohydrate component bound to the polypeptide chain, as it has been reported for POXs extracted from Brasica napus root (Duarte-Vázquez et al. 2000). Variations in the Mw of the same enzyme may be a result of the distinct techniques used for their purification and separation (Deepa and Arumughan 2002). The different Mw of secreted POXs of cell suspension cultures of Cassia didymobotrya were estimated using SDS PAGE (43 kDa) and gel filtration (50 kDa) (Vitali et al. 1998). The other extracellular POX isoenzymes were isolated from the cell suspension of Vaccinium myrtillus (Melo et al. 1995) and Hevea brasiliensis (Chanwun et al. 2013) with Mws of 34 and 38 kDa for V. myrtillus and 70 kDa for H. brasiliensis.

Cellular distribution of POXs (apoplastic or vacuolar) can be predicted by the absence or the presence of the extended C-terminal peptides, which are vacuolar sorting signals. However, when the C-terminal sequence was fused to the N-terminus of the protein, no vacuolar localization was obtained (Matsui et al. 2011). According to their N-terminal signal peptides, the majority of POXs are targeted to the endoplasmic reticulum and are considered to be secreted extracellular proteins.

Profiling POX isoforms is an important tool for studying the species-specific isoenzymes and also for the identification of the specific function of individual isoforms of POXs induced by particular stress, despite their substrate redundancy (Bernards et al. 1999; Minibayeva et al. 2015). One of the most common classifications of POXs in the literature is based on their isoelectrophoretic (IEF) mobility, and they are classified as anionic, neutral and cationic isoforms, with quite a wide range of pI values (from 2 to 11.6) (Quiroga et al. 2000; Dicko et al. 2006). A commercial HRP that is widely used for studying POX kinetics and substrate affinity consists of 42 isoforms with a pI range of 2–10 (Hoyle 1977).

Basic isoenzymes were detected in the vacuoles of mesophyll cells, and in cell suspension cultures of tobacco (Schloss et al. 1987), in grape cell suspension cultures (García-Florenciano et al. 1992), and in Catharanthus roseus leaves (Sottomayor et al. 1996). On the other hand, the acidic and basic POX isoforms with different substrate specificities were determined in different plants: oat coleoptile (González et al. 1999), pea root (Kukavica et al. 2012), ginkgo and birch (Kukavica and Veljović-Jovanović 2004), Ramonda serbica (Veljović-Jovanović et al. 2006) and maize root (Šukalović et al. 2015).

Peroxidase isozyme pattern obtained by the separation on either native or semi-native IEF is also a useful indicator of the oxidizing capacity of specific POX isoenzymes (Bolwell et al. 1998; Daudi et al. 2012; Achary 2012). However, it is difficult to relate the isozyme to the corresponding gene, since there is no obvious quantitative relationship between the transcript expression level and the POX activity (Dunand et al. 2003).

Due to different post-translational modifications (PTMs), which are crucial mechanisms of regulation of enzymatic activities in vivo, often more than one protein form originates from a particular gene (Gabaldón et al. 2007; Laugesen et al. 2007). Moreover, it has been reported that particular isoenzymes were involved in two processes, such as, for example, lignification and the oxidative burst induced by a pathogen attack (Young et al. 1995; Morimoto et al. 1999). To our knowledge, there have been relatively few in-depth reports of post-translational regulation other than glycosylation of POXs. Class III peroxidases are mostly glycosylated, differing in the number of sugar moieties and the degree of branching, as well as their arrangement along the polypeptide chain (Kim and Kim 1996; Deepa and Arumughan 2002). It has been shown that an extensive glycosylation of peroxidase isolated from latex of Ficus benghalensis was related to N-glycosylation of seven asparagine residues (Palm et al. 2014). Despite the identification of more potential sites for various PTMs on the primary structure of ZePrxs in Zinnia elegans, only the formation of N-terminal pyroglutamate residues, disulphide bridges and N-glycosylation were documented (Gabaldón et al. 2007).

The role of glycans in the structure and function of POXs has been intensively examined, and it was reported that N-glycosylation affected protein folding, catalytic activity, subcellular localisation, Km value, thermostability, proteolytic sensitivity and trafficking within the cell (Hu and van Huystee ; Sánchez-Romero et al. 1994; Lige et al. 2001; Zhang et al. 2004). For example, N-linked glycans attached to the vacuolar POX make up approximately 20% of its Mw (Welinder 1979), and the number of N-glycosylation signals is about four times higher in vacuolar than in extracellular POX (Matsui et al. 2011). One of the most stable and the most efficient POXs is the isoform of palm tree leaves that contains 21–29% of carbohydrates (Baker et al. 2016). In some cases, glycans did not interfere with substrate binding; for example, HRP retained full activity after deglycosylation (Tams and Welinder 1995). Moreover, the only two isolated plasma membrane-bound POX (PMPOX) isoforms are also glycosylated (Mika and Lüthje 2003). Furthermore, deglycosylation of ionically bound cell wall POXs isolated from pea affected the electrophoretic mobility without any effects on its activity (Kukavica et al. 2012).

The examination of the thermostability of different POX isoenzymes gives an alternative insight into their structural properties as well as to their structure-function ratio. Carbohydrate moiety of POXs generally increases thermostability, which has also been shown for anionic cotton ovule POXs (Mellon 1991), pepper fruit acidic POXs (Pomar et al. 1997), and cationic peanut POXs (Lige et al. 2001). In addition, the variation in thermostability between the individual isoenzymes can be a consequence of the different structural organization of enzymes—namely the folding pattern into the tertiary structure of the native protein—and therefore it may be used for the identification of different isoforms. Higher temperatures (<40 °C) inactivate an enzyme’s activity due to the weakening of the bonds that hold the tertiary structure (e.g., hydrogen bonds), which may also result in the dissociation of the heme prosthetic group from apoprotein (Duarte-Vázquez et al. 2000). Thermostability studies are important for improving the biotechnological application of POXs in food and the pharmaceutical industry, and environmental protection (Bansal and Kanwar 2013; Gurung et al. 2013).

Determination of the thermostability of isoforms after IEF separation, instead of treating in solution, allows for the direct determination of the temperature effect on the individual isoforms. Kukavica and colleagues (2012) reported a quite difference in the thermostability of the ionic versus the covalently bound cell wall POX (CWPOX) in pea roots.

In addition to glycans, Ca²⁺ also plays an important role in the maintenance of the structural stability and activity of POX (Marañón and van Huystee 1994; Manu and Prasada Rao 2009; Fahmy et al. 2012).

3 Substrates

Peroxidases contain variable domains and substrate access channels, which may explain their great substrate diversity (Cosio and Dunand 2008). The reductant substrates of POXs include phenolics, amines, indoles, alkaloids and sulphonates (Sottomayor et al. 2004; Veitch 2004; Ferreres et al. 2011). The most common are phenolics and their glycosides: hydroxycinnamic acids (e.g., ferulic, chlorogenic, p-coumaric), hydroxybenzoic acids, hydroxyl alcohols (e.g. coniferyl, synapyl), different flavonoids (e.g. anthocyanins, flavonols), coumarins and so on (Vidović et al. 2017 and references therein). Phenolics are involved in numerous processes in plants, from growth and development to plant interaction with the environment.

However, POX activity is often determined using an artificial electron donor, such as guaiacol, yielding colored products after oxidation with H₂O₂. Of course, considering that protein extract is a mixture of isoenzymes, this estimation can be taken only as a relative change in the total POX activity under certain conditions. Thus, the activity of isoforms with specific affinity for a different substrate might be masked by the average total POX activity (Bernards et al. 1999; Kukavica et al. 2012).

Class III peroxidases may oxidize phenolic glycosides too, but at much slower rates compared with aglycones (Hirota et al. 1998; Yamasaki et al. 1997). In addition, some specific isoforms of POXs have indole 3-acetic acid (IAA) oxidase activity but it has not yet been clearly established whether such activity could be exclusively ascribed to POX (Hoyle 1977). In their early studies, Hoyle (1977) showed that all HRP isozymes had the capacity to oxidize IAA. However, Van den Berg and colleagues (1983) showed that the peanut’s cationic POX isoforms had a greater specific IAA oxidase activity compared with the anionic isoforms, while other authors demonstrated the activity of both anionic and cationic isoforms (Chibbar and van Huystee 1984). Kukavica and colleagues (2012) showed that three out of five purified ionic isoenzymes CWPOX showed IAA oxidase activity, while IAA oxidase activity was not detected for covalent isoforms. Mika and Lüthje (2003) showed that maize root PMPOXs also had the IAA activity.

Table 1 gives the Michaelis-Menten constant (Km) for each substrate, which is specific to a given isoenzyme–substrate complex; for example, the lower Km indicates a greater specificity. Km is a numerical value that enables the comparison of different enzymes, whereby the various values of Km may also suggest an isoenzyme’s location within the plant since it is accepted that Km determines cellular substrate concentration. Based on the Km of POX isoenzymes, it can be concluded that specificity of POXs isolated from different plant species for the same substrates varies.

Table 1 Substrate specificity (Km) of various purified and partially purified POX isoforms (e.g. soluble, membrane-bound, ionic and covalent cell wall bound, apoplastic, vacuolar, anionic, cationic)

The final oxidation products of POX reactions with the same phenolic compound depend on the localization of isoforms. Among extracellular POX isoforms, those bound to the cell wall matrix were proposed to participate in lignification (Sato et al. 1993), while soluble POX isoforms function as scavengers of H₂O₂. Regardless of Asc in the apoplast, CWPOX exhibited highly specific activity in the formation of cell wall structure, since the oxidation products were continuously deposited even in the presence of Asc (Takahama 2004).

4 Antioxidative Function

In addition to numerous physiological functions, the antioxidative role of POXs is based on the scavenging of H₂O_2, which accumulates in the apoplast and vacuoles (Tognolli et al. 2002; Welinder et al. 2002). Cellular trafficking of H₂O₂ is enabled and facilitated by specific members of the aquaporin family, the Tonoplast Intrinsic Proteins (TIPs) (Bienert et al. 2006; Maurel et al. 2009). Under stress condition that promote enhanced H₂O₂ accumulation in the chloroplast, which in turn overcomes the scavenging capacity of the Asc–GSH cycle, H₂O₂ may escape to the vacuole where it is scavenged (Karuppanapandian et al. 2011; Ferreres et al. 2011; Bienert and Chaumont 2014; Koffler et al. 2014). Ferreres et al. (2011) estimated that specific activity of the leaf vacuolar POX isoform from C. roseus  represents a very efficient H₂O₂ sink of up to 9 mM s⁻¹, utilizing endogenous phenolic substrates in the vacuole. The concept of delocalized H₂O₂ scavenging (scavenging in a compartment different from its source) was proposed for the first time by Takahama (2004) based on the experiments with paraquat (Pq)-induced oxidative stress in Vicia faba leaves (Takahama and Oniki 1992, 1997). The same authors reported that Pq-induced accumulation of H₂O₂ in the chloroplasts of V. faba diffused into the vacuoles where it was utilized in the POX-dependent oxidation of 3,4-dihydroxyphenylalanine (DOPA) to melanin-like compounds (Takahama and Oniki 2000). The importance of a vacuolar PPA system as a general H₂O₂ sink for plant cells under oxidative stress has been reviewed and evaluated by Ferreres et al. (2011) and Zipor et al. (2015).

4.1 Redox Regulation of Peroxidatic Cycle and Subcellular Compartmentation

While the Asc–GSH cycle and 2-Cys peroxiredoxins (PRXs) are the major scavengers of H₂O₂ in chloroplasts, cytosol, and mitochondria, as well as catalase in peroxisomes, none of these enzymes have been found in the apoplast or the vacuole (Noctor and Foyer 2016). The only antioxidative enzymes found to be located in apoplast/cell wall compartment and vacuoles are superoxide dismutase (SOD) and POXs (Takahama 2004; Ferreres et al. 2011).

It has been reported that POXs make up 6.4% of 500 analyzed cell wall proteins (Jamet et al. 2008). About 10% of all POXs are vacuolar, but in some plants, like in C. roseus, this ratio may even reach 90% (Ferreres et al. 2011). The analysis of subcellular localization of 10 POXs genes from poplar labelled with a C-terminal green fluorescence protein (GFP) and transiently expressed in Nicotiana benthamiana  showed that eight of them were targeted to vacuoles, while two were transported to the cell wall (Ren et al. 2014). In addition, proteomic studies showed that 32 POXs isoforms in A. thaliana were in the extracellular fractions, with 17 embedded in the plasma membrane, while seven were in the vacuole-enriched fractions (Francoz et al. 2015).

Vacuolar POXs account for more than 90% of soluble guaiacol peroxidase activity of bulk leaf extract, while the rest is attributed to ionically and covalently bound fractions originating from the cell wall. According to the concept of subcellular compartmentation of the antioxidative system, POXs’ endogenous substrates predominately accumulate in the vacuoles and apoplast (reviewed in Vidović et al. 2017). Indeed, upon biosynthesis, phenolic compounds are transported to the vacuole or apoplast, either conjugated with GSH (by glutathione-S-transferase: GST), esterified with malonate, or glycosylated (Kitamura 2006; Zhao 2015). Depending on the conjugation type, flavonoids and anthocyanins are transferred into the vacuole through specific transporters on the tonoplast, such as, ATP-binding cassette (ABC) transporters, multidrug resistance-associated proteins (MRPs; preferentially glutathione-flavonoid complexes), and multidrug and toxic compound extrusion proteins (MATE, preferentially glycosides) (Petrussa et al. 2013; Vidović et al. 2017 and references therein). However, this compartmentation of POXs’ endogenous substrate is specific. Thus, kaempferol, quercetin, and the indole alkaloids were found in the vacuole (Sottomayor et al. 1998; Harborne and Williams 2000; Marinova et al. 2007), while ferulic acid and coniferyl aldehyde were targeted to the apoplastic cell wall compartment (Ros Barceló et al. 2004).

Although Asc transport across plasma membranes has been described previously (Horemans et al. 2000), the specific transporters that carry Asc have not been identified at the molecular level in plants so far (Maurino et al. 2006). It has been proposed that DHA was exclusively transferred through the plasma membrane via different transporters than the glucose transporters (Fernie and Tóth 2015). It has also been suggested that Asc uptake in some plant cells is an active process that requires the presence of a proton-electrochemical gradient across the plasma membrane. During the last decade, the involvement of the nucleobase-Asc transporter (NAT) family proteins in Asc transport through plasma membrane was proposed and studied (Girke et al. 2014; Pick and Weber 2014). To date, 12 members of the NAT family have been identified and molecularly characterized in A.  thaliana, rice and tomato (Maurino et al. 2006; Cai et al. 2014; Niopek-Witz et al. 2014). Trans-membrane Asc-driven electron transport is catalyzed by cytochromes (Cyts) b561. The Asc reducible Cyt b561 is present in the plasma membrane (Horemans et al. 1994; Asard et al. 2001) and in tonoplasts (Griesen et al. 2004).

The Asc localization in the apoplast and vacuoles, concomitantly with the absence of GSH in apoplast and very low concentration of highly oxidized glutathione pool in the vacuole (<0.03 mM; <10%) (Noctor and Foyer 2016 and references therein), justifies Asc function as the main reductant involved in the PPA system (Takahama 2004). Besides biochemical approaches, Asc in the vacuoles was visualized using transmission electron microscopy after immunocytochemical labeling (reviewed in Zechmann 2017), and concentration around 2 mM was revealed. Compared with vacuolar Asc, where it is mostly in the reduced form (Ferreres et al. 2011), Asc in the apoplast is mostly present in its oxidized form, DHA. In the apoplast, Asc plays a crucial role in plant growth, development and defence against pathogens, drought and high ozone levels (Veljovic Jovanovic et al. 2018). Apoplastic Asc is also involved in the redox regulation of antioxidant enzymes, modifications in plant growth through changes in the hormone synthesis and MAPK activity, as well as changes in calcium signaling (Liso et al. 2004; Karpinska et al. 2017).

Numerous reports show that all three constituents of PPA, phenolics, POXs and Asc, were targeted to vacuoles under high light conditions (Neill and Gould 2003; Pollastri and Tattini 2011; Ferreres et al. 2011; Heyneke et al. 2013; Zipor and Oren-Shamir 2013) as well as during early response to drought (Koffler et al. 2014). Although it is lower than in other compartments (chloroplasts, peroxisomes, mitochondria) (Zechmann et al. 2011), a strong enhancement of vacuolar Asc content has been observed in A. thaliana exposed to high light, drought and after inoculation with Pseudomonas syringae (Großkinsky et al. 2012; Heyneke et al. 2013; Koffler et al. 2014). In addition, the redox state of apoplastic Asc has a significant role in tolerance to high light stress (Karpinska et al. 2017). Recycling of Asc from MDA^· and DHA takes place on the plasma membrane and tonoplast through the cytoplasmic Asc–GSH cycle (Fig. 1). As previously mentioned, MDA^· generated in the vicinity of plasma membrane may accept electrons from membrane-bound Cyt b561 (Asard et al. 2001) and form Asc and DHA independently, or accept electron by the action of membrane-bound MDAR (Drazkiewicz et al. 2003). The DHA is further transported (see the previous section) to the cytoplasm where it is reduced to Asc in Asc–GSH cycle (Horemans et al. 2000).

Fig. 1

Schematic overview of the vacuolar and apoplastic H₂O₂ scavenging mechanism by the POX/PhOH/Asc system. Explanation in the text. POX class III peroxidases, DHAR dehydroascorbate reductase, GR glutathione reductase, SOD superoxide dismutase, AO ascorbate oxidase

Taking into the account the light-induced accumulation of all three constituents of the PPA system in vacuoles and apoplast, Km values for endogenous POX substrates, and the relative volume of these compartments (50–55% of the total mesophyll cell volume; Vidović et al. 2016), the PPA system may represent an important sink for excess H₂O₂ under high light stress.

5 Pro-oxidative Functions

5.1 H₂O₂-Producing System (Oxidative Cycle)

In response to various unfavourable conditions, including high levels of ozone, wounding, and infection by pathogens, apoplastic POXs can act as a source of ROS, contributing to the oxidative burst (Doke 1983; Bolwell et al. 1998; Gill and Tuteja 2010).

The involvement of CWPOXs (besides plasma membrane NADPH oxidase) was confirmed using the heme inhibition studies with salicylhydroxamic acid (SHAM), NaN₃, and KCN, and inhibitors of NAD(P)H oxidase: diphenyleneiodonium chloride (DPI) and imidazole (IMZ). In addition, treatments with elicitors and pathogen inoculation, besides H₂O₂, trigger the ^·OH and ${{\text{O}}_{2}}^{ \cdot - }$ production in cultured plant cells, protoplasts and tissues (Bestwick et al. 1998; Choi et al. 2007). The oxidative cycle of POX is considered to be an ${{\text{O}}_{2}}^{ \cdot - }$-generating system and a source of H₂O₂ required for cell wall stiffening, and for the formation of the secondary cell wall by lignification (Halliwell 1978). According to the proposed reaction mechanism, in the absence of H₂O₂ and the presence of a suitable reductant, the intermediate catalytically inactive form of the POX (Compound III, Fig. 2) may oxidize NADH. In this reaction, PhOH returns Compound III to the ground state, making the enzyme functional again, while Mn²⁺, as a second cofactor, catalyzes the non-enzymatic reaction between ${{\text{O}}_{2}}^{ \cdot - }$ and NADH to produce H₂O₂, directing the reaction to peroxidatic cycle (Halliwell 1978).

Fig. 2

Proposed model of ^·OH generation in the pea root cell wall (Adopted from Kukavica et al. 2008). In the presence of metal ions PhOH and MnSOD, cell wall-bound POX catalyze the formation of ^·OH through the following reactions: (1) ${\text{POX}} \, ({\text{Fe}}^{3 + } ) + {\text{H}}_{2} {\text{O}}_{2} \to {\text{Comp}}\,{\text{I}}\,({\text{Fe}}^{4 + } {=} {\text{O}})^{ + } + {\text{H}}_{2} {\text{O}}_{2}$; (2) ${\text{POX}} \, ({\text{Fe}}^{3 + } ) + {{\text{O}}_{2}}^{ \cdot - } \to {\text{Comp}}\,{\text{III}}\,({\text{Fe}}^{2 + } {-} {\text{O}}_{2} )$; (3) ${\text{Comp}}\,{\text{II}} \, ({\text{Fe}}^{4 + } {=} {\text{O}}) + {\text{H}}_{2} {\text{O}}_{2} \to {\text{Comp}}\,{\text{III}} \, ({\text{Fe}}^{2 + } {-} {\text{O}}_{2} ) + {\text{H}}_{2} {\text{O}}$; (4) ${\text{Comp}}\,{\text{III}} \, ({\text{Fe}}^{2 + } {-} {\text{O}}_{2} ) + {\text{H}}_{2} {\text{O}}_{2} \to {\text{POX}} \, ({\text{Fe}}^{3 + } ) +^{ \cdot } {\text{OH}} + {\text{O}}_{2} + {\text{OH}}^{ - }$. M^x+ and M^(x−1)+ oxidized and reduced metal ions, PhOH phenolics compounds, PhO^· phenoxyl radical, –Ph=O quinhydrone. The upper panel represents the electron paramagnetic resonance (EPR) spectra of the quinhydrone structure detected without the presence of spin trap in the pea root cell wall. Basic pH increased the signal intensity of quinhydrone. The bottom panel represents the EPR spectra of the DEMPO/OH adduct produced by POX covalently bound to the cell wall, the ionic POX and soluble intracellular POX. Adopted from Kukavica et al. 2012.

The exclusive involvement of the cell wall in the ${{\text{O}}_{2}}^{ \cdot - }$-production system has been reported for the pea and cowpea during plant-microbe interactions (Kiba et al. 1997) and upon metal toxicity, such as Al excess in the onion root (Achary et al. 2012) or Mn excess in cowpea leaf (Fecht-Christoffers et al. 2006). Kiba and colleagues (1997) also showed that the ${{\text{O}}_{2}}^{ \cdot - }$-production system of the same plant species induced by the elicitor, was inhibited by POX inhibitor, SHAM, and not by NADPH oxidase inhibitors (i.e., imidazole and quinacrine).

5.2 Hydroxyl Radical-Generating System (Hydroxylic Cycle)

The capacity of POXs to generate ^·OH has been documented by a number of in vitro and in vivo studies. Several enzymatic and non-enzymatic mechanisms can be found in the literature explaining the cell wall loosening required for extension and cell growth (Fry 1998; Chen and Schopfer 1999; Liszkay et al. 2003; Kukavica et al. 2007, 2008). Cell wall loosening is important under osmotic, drought or salt stress in order to ensure the possibility of cells and organs to expand (Tenhaken 2014). Furthermore, it has been reported that POX can produce ^·OH in the presence of NADH (Liszkay et al. 2003; Schopfer et al. 2001).

The proposed mechanism for the oxidative cycle of cell wall-bound POXs is based on the oxidation of NADH only in the presence of p-coumaric acid and Mn²⁺, while, in the absence of these two cofactors, oxidation of NADH is negligible. However, there is no evidence for the presence of NAD(P)H in the apoplastic compartment, suggesting that ^·OH generation using NADH is unlikely under natural conditions. Instead, plasma membrane-bound NAD(P)H oxidase is regarded as a crucial enzyme responsible for the generation of ${{\text{O}}_{2}}^{ \cdot - }$ in the apoplast (Murphy and Auh 1996; Van Gestelen et al. 1997). Moreover, various cellular components are capable of generating ${{\text{O}}_{2}}^{ \cdot - }$ in the vicinity of plasma membranes (Mojović et al. 2004). An extreme reactivity of ${{\text{O}}_{2}}^{ \cdot - }$ (half-life in water is 0.2 and 20 ms at 10 and 1 µM, respectively, Bielski et al. 1985) is even more accelerated in the presence of SOD. Therefore, it is doubtful that the membrane-bound NAD(P)H oxidase is a source of ^·OH for cell wall reactions. Kukavica et al. (2008) demonstrated that isolated cell wall free from plasma membrane had the capacity for NADH-independent generation of ^·OH, and suggested a similar mechanism in situ. In isolated cell walls of pea roots, ^·OH was detected using a spin-trapping reagent DEPMPO in the absence of any exogenous compounds (Fig. 2). The alternative mechanism for ${{\text{O}}_{2}}^{ \cdot - }$ formation in the pea cell wall isolates involved the oxidation of hydroxycinnamic acids by redox active metals, Fe³⁺ and Cu²⁺, to phenoxyl radicals, which can reduce O₂ to ${{\text{O}}_{2}}^{ \cdot - }$. Both, caffeic and chlorogenic acid are found in apoplastic fluid and cell wall isolates, and are easily auto-oxidizable yielding either a charge-transfer complex, quinhydrone, or polymers that are also auto-oxidizable. Once generated, ${{\text{O}}_{2}}^{ \cdot - }$ may induce the formation of CWPOX Compound III, which can enter the hydroxylic cycle yielding ^·OH and H₂O₂. The cell wall-bound MnSOD can regulate the concentration of ${{\text{O}}_{2}}^{ \cdot - }$ and ^·OH, while apoplastic H₂O₂ can catalyze the formation of ^·OH via the Fenton reaction (Yim et al. 1993). An overview of the whole mechanism proposed for cell wall generation of ^·OH is given in Fig. 2.

The authors demonstrated an exclusive role of covalently CWPOX in the DEPMPO/OH adduct formation (Fig. 2). During the generation of ^·OH, the concentration of quinhydrone structures (as detected by EPR spectroscopy) increased, suggesting that H₂O₂ required for the formation of ^·OH in isolated cell walls was produced during the oxidation of hydroxycinnamic acids, which form a quinhydrone dimer.

6 Effects of Environmental Stresses on POXs

Induction of overall peroxidase activity is widely accepted as an indicator of abiotic and biotic stress response in plants. In addition to the measurements of guaiacol peroxidase activity of the whole tissue extract or cell wall isolates, numerous studies have revealed differences in POXs isoform profiles. These variations may correspond either to a stimulation or attenuation of some pre-existing isoforms, or to the appearance of the new isoforms (Kukavica et al. 2012; Liu et al. 2013). Such modulations of POXs profile patterns prove their role in the defence mechanism. The differences in POXs profile patterns are also related to the plant species, the type, intensity and the duration of stress. Sometimes a contrasting response of POX isoforms to stress was demonstrated, such as in the case of ionic and covalently cell wall-bound isoforms in pea roots induced by elicitor chitosan (Kukavica et al. 2012).

The expression of POX genes is regulated in response to biotic and abiotic stresses and the underlying molecular mechanism is related to the nature of the 5′ flanking regions with stress-responsive cis-elements (Sasaki et al. 2007; Kim et al. 2012). However, POX gene expression patterns show great variations, and the stress-induced upregulation depends on the developmental stage and organs. According to Cosio and Dunand (2008), at least 19 AtPrxs genes were involved in the specific abiotic stress mechanisms. However, the stimulating effect on POX activity was not always obvious.

Since it has been shown that POXs are quite sensitive to atmospheric pollution and heavy metals, the measurements of their activities have been widely used for the phytomonitoring of industrial and urban areas (Cho and Park 2000; Klumpp et al. 2000; Wu and von Tiedemann 2002; Geebelen et al. 2002). It has been shown that induction of POX activity was higher in metal-sensitive species/populations compared with tolerant ones (Tamás et al. 2002; Morina et al. 2016). Cadmium treatment increased the accumulation of lignin and apoplastic guaiacol peroxidase activities to a higher extent in the sensitive compared with the resistant cultivars of V. faba. In addition, ten ZmPrxs genes were altered (seven of them were down and three were upregulated) in the stress response to Cd (Yue et al. 2017). Tao and colleagues (2013) suggested a synergistic action between salicylic acid and POXs in the snc1 (suppressor of non-expressor of pathogenesis-related gene 1 (npr1-1) constitutive 1) plant leading to its sensitive phenotype under metal excess. Native PAGE of cell wall proteins of Cassia tora exposed to Al stress showed a strong induction of POXs (Xue et al. 2008). The expression of AtPrx64 enhanced root growth and decreased the accumulation of Al and ROS in the roots of transgenic plants in comparison with WT plants (Wu et al. 2017).

The influence of temperature variation and water status on POX activity has been also demonstrated. Exposure to cold stress gradually increased POX activities in the leaves of several Medicago accessions at the beginning of stress; however, POX activity decreased with prolonged stress duration (Nourredine et al. 2015). Nourredine and coauthors (2015) also showed that the decrease in POD activity was less in tolerant than in sensitive plants. Similar results were obtained for naked oat plants (Liu et al. 2013). The decrease of POX activity under long-term exposure to cold stress indicated that low temperatures might affect RNA transcription, and consequently translation, reducing the POX synthesis (Liu et al. 2013). It was also shown that two POX isoforms (TaPrx04 and TaPrxo3) from an apical root segment were responsible for drought tolerance in two wheat cultivars (Csiszár et al. 2012). In addition, seven isoforms of POXs from Tamarix hispida were either upregulated or downregulated upon NaCl, polyethylene glycol (PEG), NaHCO₃ and Cd stress (Gao et al. 2010). Five ZmPrxs genes (Wang et al. 2015), and six POXs from sweet potato (Ipomoea batatas) responded to different abiotic stresses (H₂O₂, SA, NaCl and PEG treatment) (Kim et al. 2007). The POX isoform, MsPrx16, from alfalfa (Medicago sativa) showed decreased expression under cold stress as a part of a general strategy of root cell walls to maintain flexibility under temperature stress (Behr et al. 2015). In addition, the dominant PMPOX from maize, ZmPrx66, was downregulated and two others were upregulated by H₂O₂ stress (Mika et al. 2009). After a short time of flooding (4 h), the neutral form of soluble POX (pI 7.0) of maize was downregulated while long-time flooding (28 or 52 h) resulted in the upregulation of alkaline POXs (pI 9.2, 8 and 7.8) (Meisrimler et al. 2014).

In contrast to abiotic stress, a complex role of POXs in biotic stress is based on peroxidatic and pro-oxidative catalytic action (Sects. 4.1 and 5) and possible generation and detoxification of ROS, though a reductant for the formation of the ‘oxidative burst’ has not yet been identified. They are induced in host plant tissues by pathogen infection and belong to the pathogen-related protein 9 subfamily (van Loon et al. 2006). POXs are crucial for the establishment of structural barriers to limit pathogen invasion or the generation of extremely toxic ROS and RNS (Passardi et al. 2005). It has been shown that POX activity or POX gene expression in higher plants was induced by fungi, bacteria, viruses and viroids (references in Almagro et al. 2009 ). The activity of NADH-peroxidases was involved in the generation of ${{\text{O}}_{2}}^{ \cdot - }$ and H₂O₂, and ^·OH in response to biotic stress as was shown in many plants: in pea and cowpea in response to a pathogen (Kiba et al. 1997); in cultured cells of rose and French bean in response to an elicitor derived from the cell walls of Colletotrichum lindemuthianum (Bolwell et al. 1998); in cotton cotyledons in response to a hypersensitive reaction to Xanthomonas campestris (Martinez et al. 1998); and in lettuce leaves during the non-host hypersensitive reaction induced by P. syringae (Bestwick et al. 1998). Lycopersicum esculentum POX (Prxs06) was induced at infection site by accumulating H₂O₂ (Coego et al. 2005). The extracellular wheat isoforms were upregulated by wounding stress (Minibayeva et al. 2015).

As previously mentioned, assigning a function to a particular POX isoform is a rather complex task and requires the integration of physiological, biochemical and genetic studies, especially regarding the possible use for generation of useful transgenic plants (Sasaki et al. 2007).

7 Genetic Manipulation of POX Isoenzymes Related to Plant Defence Against Environmental Stress Conditions

A number of studies on transgenic plants with altered expression of genes of POXs were carried out with the aim to understand the specific role of individual isoenzymes under unfavourable environmental conditions. Overexpression of sweet potato swpa4 POX gene in tobacco plants (Nicotiana tabacum) (Kim et al. 2008) and overexpression of AtPrx3 in A. thaliana significantly increased salinity and drought tolerance (Llorente et al. 2002). Moreover, it has been shown that the overexpression of AtPrx22, AtPrx39, and AtPrx69 increased cold tolerance in the brassinosteroid-insensitive mutants (Kim et al. 2012). The same results—increased germination rate under salt and dehydration, and decreased sensitivity to cold stress—were shown in the experiments with heterologous expression of two POX genes from C. roseus in N. tabacum (Kumar et al. 2012). The knockdown of AtPrx33 and AtPrx34 transcripts by transduction of antisense FBP1 from Phaseolus vulgaris into A.  thaliana resulted in a failed oxidative burst and an increased sensitivity to fungal and bacterial pathogens (Daudi et al. 2012). In addition, transgenic expression of POX2 from Capsicum annuum (CaPrx2) in A. thaliana (Choi et al. 2007) enhanced a broad spectrum resistance (pathogenesis-related gene induction) and H₂O₂ accumulation, as well as tolerance to drought and salt. However, there are no data about changing the phenotype by knockout of simple vacuolar POXs genes.

The overexpression of AtPrx in transgenic tobacco plants enhanced root growth under Al excess and decreased the accumulation of Al in the roots (Wu et al. 2017). The role of specific POX isoforms in growth inhibition was revealed in a genetic study in which AtPRX71 expression was suppressed and the phenotype had a bigger rosette and biomass. On the other hand, a retarded growth of the 35S:AnPGII plants was accompanied by a high activity level of POXs (Raggi et al. 2015). Transgenic plants with antisense suppression of Pry60 in tobacco had a significant reduction in lignin content (Blee et al. 2003). Similarly, overexpression of AtPRX37 caused dwarfism, probably by affecting cell expansion and not cell division (Pedreira et al. 2011). The underlying inhibitory mechanism of these isoforms on cell expansion is proposed to be the promotion of H₂O₂ generation. However, the qua2-1 atprx71-1 double mutant showed decreased ROS accumulation that can be explained by the involvement of the 71 and 53 genes in the cross-linking of cell wall in the hydroxylic POX cycle (Raggi et al. 2015).

8 Conclusion

Class III peroxidases are a ubiquitous, multigene family of secretory enzymes, localized in all plant organs, which is in accordance with their well-established role in plant growth and development, as well as in plant–environment interactions. Their role in stress-related processes (e.g., oxidative burst, signaling, cell wall re-arrangement and antioxidative defence) is determined by (i) the cellular distribution of the specific substrates and isoenzymes; (ii) PTMs (especially the glycosylation pattern) of isoenzymes; (iii) the microenvironment such as pH, Ca²⁺, Mn²⁺ concentration, Asc redox state; and (iv) POXs’ bifunctionality—namely, scavenging/generating H₂O₂ activities. The mechanisms underlying these processes have not yet been fully established and remain a challenge for future research. In the light of recent findings related to H₂O₂ intracellular trafficking and the subcellular localization of the constituents of Takahama’s H₂O₂ scavenging system (POX/Phe/Asc), we re-established the antioxidative role of POXs incorporating it in a perfectly orchestrated complex cellular antioxidative system (Fig. 2). We propose that, in this way, vacuolar POXs are the important sink for H₂O₂ in the plant cell, taking into account its size and the presence of millimolar concentrations of phenolics and Asc, which emphasize the key role of the vacuole in the cellular antioxidant network. Taking all of this into account, POXs can be regarded as a crossover point in metabolism involved in growth and defence regulation by redirecting phenolics to the cell wall and, finally, in adaptation to stress. Due to the more oxidized state and the low redox buffering in the apoplast, POXs located in apoplast/cell wall compartment may have different properties compared with those in the vacuole, such as contribution to cell wall loosening and elongation or, on the other hand, to an oxidative burst.

Further studies including molecular approaches, such as a transcriptome analysis and recombinant DNA, should be conducted in order to reveal the specific functions of particular POX isoforms and to assign a specific function to a particular POX gene and its protein. Comprehensive knowledge of the link between genes, primary structure and PTMs will provide new solutions for developing and engineering plants with improved vigor and stress tolerance. A range of biotechnological approaches are being employed for the manipulation of lignin content for increased stress tolerance, as well as for optimal utilization of plant biomass in different branches of industry.