An array of plants’ secondary metabolites can adversely affect the growth and development of neighboring plants by various physiological and biochemical mechanisms in a process known as allelopathy (Weir et al., 2004; Cheng and Cheng, 2015). Allelopathic interactions predominate in natural and agricultural ecosystems, where they create different ecological and economic implications. Stunted growth of crops and low yields and crop quality due to soil sickness (Huang et al., 2013a), unsuccessful forest regeneration (Souto et al., 2000) and biological invasion of exotic plant species (Bais et al., 2003) are all the consequences of allelopathic activity. On the other hand, allelopathic interactions between crops and weeds have a promising future in strategies for cropping systems improvement (De Albuquerque et al., 2011). The allelochemicals involved in these interactions can be considered as prospective bioherbicides based on new mode-of-actions (MOA) for managing the global problem of rapidly increasing weed resistance to actual synthetic herbicides.

Despite their great diversity, plant allelochemicals can be grouped into three main chemical classes: terpenoids, N-containing compounds, and phenolic compounds (Bachheti et al., 2020). Dihydrochalcones (DHCs) are a special group of phenolic compounds with limited occurrence in the plant kingdom, whose structure is closely related to flavonoid precursors, chalcones. DHCs are defined by the presence of two C6 rings joined by a C3 bridge, but the double bond is reduced in comparison with chalcones (Figure 1A). The unique naturally rich source of DHCs are species of the genus Malus, including the domestic apple (Malus × domestica Borkh.) whose DHCs make up 96–97% of the total phenolics in the leaves and represent approximately 10%–20% of DW leaf tissue (Pontais et al., 2008; Dugé de Bernoville et al., 2011). Phlorizin (phloretin 2´-O-glucoside) is the first discovered and the most common dihydrochalcone in Malus species. The glycosylation of phloretin [3-(4-hydroxyphenyl)-1-(2,4,6-trihydroxyphenyl)-1-propanone; Figure 1A] at position 2′ by specific uridine diphosphate (UDP)-glucose: phloretin 2′-O-glucosyltransferase is the key step in phlorizin biosynthesis that directly determines the concentration of phlorizin in apple tissues (Zhou et al., 2017). Besides phlorizin, there are at least 10 other DHCs described in the genus Malus that differ in sugar group and/or their binding sites, and all are derivatives of aglycone phloretin (Gosch et al., 2010). Due to its typical DHCs structure (Figure 1A), phloretin is a very flexible molecule which can react efficiently with biological macromolecules. The reactivity of phloretin results in its striking health-protecting properties in mammalians and humans such as antimicrobial, anticancer, antioxidant, estrogenic, anti-inflammatory, cytoprotective, and neuroprotective activity (Nithiya and Udayakumar, 2016; Behzad et al., 2017). However, the role of phloretin and other DHCs in plants still remains elusive. Börner (1959, 1960) was the first who suggested autotoxicity of phloretin and phlorizin and related them with “apple replant disease” (ARD), which is characterized by stunted growth, reduced root system and low yields in apple trees replanted on lands that previously supported apple orchards. High concentrations of phlorizin (1–4 mM) were phytotoxic for Malus hupehensis and decreased photosynthesis and respiration rate and increased malondialdehyde content and antioxidative enzymes activities (Jianghong et al., 2007). Yin et al. (2018) suggested that phlorizin and phloretin were harmful to M. hupehensis seedlings at concentrations found in the orchards´ soil, which led to up-regulation of genes encoding defense system-related and free radical scavenging proteins.

To facilitate the examination of phytotoxic attributes of apple root exudates and to exclude possible microbial influence, we recently developed a model system based on the apple hairy root system in vitro (Stanišić et al., 2019). There, we reported a significant inhibitory effect of apple hairy root exudate on the development and growth of the model species Arabidopsis. In the apple hairy root growth medium, in addition to chlorogenic and caffeic acids, dihydrochalcones phlorizin and its aglycone phloretin were detected. Since the determination of structure/activity showed that the presence of a free hydroxyl group at the position C2ʹ and the absence of glycosyl molecules allow phloretin to react more intensively with biological molecules compared to phlorizin (Barreca et al., 2014), we hypothesized that phloretin contributed much more to the observed phytotoxicity of apple root exudate in Arabidopsis than phlorizin.

Phenolic compounds have been frequently reported to affect plant development by modulating the homeostasis of plant hormone auxin (Brown et al., 2001; Peer and Murphy, 2007; Ahmed et al., 2020). In apple, silencing of chalcone synthase CHS led to the loss of almost all flavonoid and dihydrochalcone compounds, and induced significant phenotypic effects, which included a highly dwarfed growth pattern presumably as a result of significantly increased auxin transport from shoots to root (Dare et al., 2013). Downregulation of phloretin-specific glycosyltransferase UGT88F1 led to decrease in the concentration of phlorizin, phloretin and a number of other polyphenolic compounds in transgenic apple plants. These changes increased auxin flux from the shoot apex and produced a highly dwarfed phenotype that phenocopied previously characterized CHS-silenced plants (Dare et al., 2017).

Auxin plays a fundamental role in plant development and coordinates plant responses to the constantly changing environment through the regulation of cell division, elongation and differentiation of the cells in meristematic tissues. Delicately regulated co-action of auxin biosynthesis, conjugation, degradation and transport through plant tissues generates morphogenic gradients of auxin that govern cell fate decisions and underlies plant phenotypic plasticity. The biologically active native plant auxin indole-3-acetic acid (IAA) is biosynthesized in Arabidopsis mainly via a tryptophan-dependent metabolic pathway containing four biosynthetic routes: the indole-3-pyruvic acid (IPyA), the indole-3-acetaldoxime (IAOx), indole-3-acetamide (IAM), and tryptamine (TRA) route (Normanly et al., 1993; Figure 1B). The majority of IAA is produced via the IPyA metabolic pathway (Mashiguchi et al., 2011; Stepanova et al., 2011; Won et al., 2011) where TRYPTOPHAN AMINOTRANSFERASE OF ARABIDOPSIS 1/TRYPTOPHAN AMINOTRANSFERASE RELATED (TAA1/TAR) catalyzes the conversion of tryptophan to IPyA (Stepanova et al., 2008; Won et al., 2011) and the YUCCA (YUC) family of flavin monooxygenases (YUC1 – YUC11) further converts IPyA to IAA (Brumos et al., 2014). The IAOx route is largely restricted to the genus Brassica (Sugawara et al., 2009) where cytochrome P450 monooxygenases: CYP79B2 and CYP79B3 catalyze the conversion of Trp to IAOx which can be further converted to indole-3-acetamide (IAM) and indole-3-acetonitrile (IAN), both capable to produce IAA (Zhao et al., 2002; Sugawara et al., 2009). Due to the delicacy in the regulation of processes depending on auxin levels, plants have developed several mechanisms to control the levels of active auxin. One of them is the formation of inactive conjugates with amino acids, sugars or peptides that function in IAA storage or as intermediates in successive degrading steps leading to the production of irreversibly degradable oxindole-3-acetic acid (oxIAA; Hayashi et al., 2021).

Polar auxin transport (PAT) plays a central role in auxin maxima generation. PAT is mediated by auxin influx carriers, AUXIN1/LIKE AUXIN1 (AUX1/LAX), auxin efflux proteins PIN-FORMED (PIN1, PIN2, PIN3, PIN4, PIN7; Vieten et al., 2005; Michniewicz et al., 2007), as well as proteins from MULTI-DRUG RESISTANCE/P-GLYCOPROTEIN (MDR/PGP) subfamily, belonging to the ATP-BINDING CASSETTE (ABC) transporter superfamily (ABCB1, ABCB4, ABCB19; Zažímalová et al., 2010; Geisler et al., 2017). AUX1/LAX proteins are responsible for the auxin intake (Swarup et al., 2001), while PINs show asymmetric localization on cell membranes in certain cell types, correlated with known acropetal and basipetal directions of auxin flow (Vieten et al., 2005; Michniewicz et al., 2007; Figure 1C). All auxin carriers act synergistically mediating the plasticity of the root system architecture when they adapt to various environmental stimuli (Sun et al., 2010; Yuan et al., 2013; Li et al., 2015; Yuan and Huang, 2016; Bahmani et al., 2020) or respond to soil allelochemicals (Zhang et al., 2018; Li et al., 2019).

The presented study aims to evaluate DHC phloretin of apple tree as a new phytotoxic compound, addressing its basic mechanism of action, which we assume involves the modulation of auxin homeostasis. For that purpose, we investigated the growth and gravitropic response of the model plant Arabidopsis to exogenous phloretin. The obtained morphological data revealed a significant dose-dependent growth retardation and the presence of morphological abnormalities and agravitropic behavior in seedlings, especially roots. Microscopic, molecular and biochemical approaches have confirmed that roots are the primary targets of phloretin phytotoxic action. Significantly disturbed auxin metabolome profile in the root with a particularly increased IAA content and its accumulation in the lateral parts of the root apex, with noticeable changes in auxin biosynthetic and PAT-involved genes expression, confirmed the hypothesis that disturbance in auxin homeostasis is the basic mechanism of phytotoxic action of phloretin. This mechanism makes phloretin a prospective candidate for an eco-friendly bioherbicide and also paves the way for further research into the role of phloretin in ARD.

Phloretin is a plant-based secondary metabolite specific to apple species known for its human health-promoting properties (Behzad et al., 2017). For the first time, we characterized phloretin as an allelochemical that exhibited harmful effects on plant species outside the genus Malus. When tested on Arabidopsis, significant inhibition of the growth of primary and lateral roots and leaves were observed, as well as severe morphological abnormalities and agravitropic behavior of seedlings. Histological examination revealed a serious disturbance in the root architecture, which resulted in meristematic and elongation zone length reduction. Similar consequences of phytotoxic activities have been also found in other plant allelochemicals (Sánchez-Moreiras et al., 2008; Soltys et al., 2012; Zhang et al., 2018). However, phloretin induced changes including stunted growth, reduced root system with brownish root color and epidermal cells separating from the subjacent cell layers, resemble a lot to the phenotype of ARD affected apple trees (Yim et al., 2013; Winkelmann et al., 2019). Although it is suggested that many factors are involved in the etiology of ARD (Winkelmann et al., 2019), it is likely that the role of phloretin in the development of ARD symptoms is not negligible.

The inhibitory effect of phytotoxic allelochemicals on roots has been widely reported as a result of changes in the balance of plant hormones, where auxin is mainly responsible for such action. Thus, the accumulation of auxin in the epidermis of the distal elongation zone inhibited elongation of the root and shortened cells in Arabidopsis roots exposed to salicylic acid (Pasternak et al., 2019). Similarly, increased auxin content reduced both meristematic and elongation zones in benzoic acid-treated seedlings (Zhang et al., 2018). Exogenously applied IAA at concentrations above 100 nM induced dose-dependent inhibition of primary root growth in Arabidopsis seedlings (Yang et al., 2021), while both 1-naphthaleneacetic acid (NAA) and 2,4-dichlorophenoxyacetic acid (2,4-D) reduced root meristem size and primary root growth at concentrations higher than 50 and 20 nM, respectively (Růzǐčka et al., 2009). Taking into account the role of auxin in cell cycle modulation and elongation, we tried to connect the retardation of meristematic and elongation zones of roots in phloretin-treated seedlings with disturbance of auxin levels. The results of auxin metabolome profiling confirmed our hypothesis which indicated a more than threefold increase in the IAA content in Arabidopsis roots shortly after phloretin application.

Some studies have reported that environmental factors, such as salt and osmotic stresses, affect not only IAA levels, but even more the quantities of its precursors, conjugates and degradation products (Pavlović et al., 2018; Smolko et al., 2021). Our results indicated a decrease in the content of all auxin precursors except IAM in the root and particularly pronounced reduction in IAN levels shortly after phloretin treatment. IAN is not exclusively involved in auxin biosynthesis in Arabidopsis, as nitrilases play a prevalent role in glucosinolate metabolism and processes such as cyanide detoxification and camalexin homeostasis (Vorwerk et al., 2001; Su et al., 2011). Since camalexin is the major phytoalexin that accumulates after infections (Tsuji et al., 1992; Thomma et al., 1999; Glawischnig, 2007) or abiotic stresses (Zhao et al., 1998; Bouizgarne et al., 2006; Kishimoto et al., 2006), it is possible that a significant decrease in IAN levels in roots during phloretin treatment is a consequence of its depletion in intensive camalexin biosynthesis rather than in IAA overproduction.

Maintenance of auxin homeostasis is of the utmost importance for plants and is regulated by the balance among three processes: auxin storage, reactivation and irreversible deactivation. Recent study on auxin metabolism suggests that these processes are part of GH3-ILR1-DAO (amidosynthetases-IAA-Leu-Resistant - DIOXYGENASE FOR AUXIN OXIDATION 1)-mediated metabolic pathway. IAA conjugates IAA-Asp and IAA-Glu are the key metabolites and can serve both as reversible storage forms and as degradative forms that are irreversibly oxidized by DAO to form oxIAA-Asp and oxIAA-Glu, which are then hydrolyzed to oxIAA. The balance between recycling and irreversible deactivation of auxin could be turned towards the second process by IAA over-production in planta or its exogenous application. Our auxin metabolome profiling results indicated that the content of oxIAA in the roots of Arabidopsis was significantly elevated and that it closely followed the increase in IAA levels shortly after phloretin treatment. Apparently, the increase in IAA levels in roots led to activation of the oxIAA catabolic pathway as an attempt to eliminate excessive auxin and maintain auxin homeostasis. However, quantity of IAA-Asp and IAA-Glu was not significantly changed or was under control levels in phloretin-treated seedlings. The results of Hayashi et al. (2021) indicated that AtDAO1 enzyme prefers IAA-amino acids to IAA since its oxidation activity for IAA-Glu was c. 6 × 10⁵ times higher than for IAA in an in vitro assay. Thus, excessive accumulation of IAA presumably led to the production of oxIAA through the rapid conversion of IAA-Asp and IAA-Glu in their oxidized forms shortly after phloretin treatment. In addition, the higher Km value of AtDAO1 for IAA-Asp than for IAA-Glu reported in Hayashi et al. (2021) could explain why IAA-Glu content was more reduced than IAA-Asp under phloretin treatment.

As local biosynthesis and transport of auxin act in concert to establish and maintain auxin morphogenic gradients in roots (Brumos et al., 2018), we evaluated the expression of key genes involved in the IPyA biosynthetic route and PAT in Arabidopsis. Despite the general tendency of phloretin to up-regulate auxin biosynthetic genes in roots, the level of TAA1 transcript was significantly reduced in roots after 12 h of phloretin treatment, at the same time point when transcription of TAR2 was elevated. This is presumably the result of the shared role of TAA1 and TAR2 in the regulation of auxin maximum formation in meristematic root tissue due to the known functional redundancy of TAA1 and TAR2 and overlapping expression patterns in young seedlings (Stepanova et al., 2008). Similarly, the expression of YUC3 in roots was antagonistic to YUC4, while YUC6 and YUC8 remained unchanged. Exploring expression profiles of cucumbers’ YUC genes under different stress treatments, Yan et al. (2016) reported that expression of CsYUC10b was dramatically increased, while CsYUC4 was repressed in response to low temperature. Similarly, CsYUC10a and CsYUC11 acted against up-regulation of CsYUC10b under salinity stress, suggesting that distinct YUC members antagonize each other under the same type of stress to maintain appropriate auxin levels. Actually, the fact that enzymes catalyzing both steps of the IPyA biosynthetic route are encoded by multigenic gene families offers a simple mechanism for generating an array of different expression patterns. Thus, the phloretin-induced elevation of endogenous auxin levels is likely to be achieved by the selective expression of particular TAA1/TAR and YUC gene family member at different time points after phloretin application.

In contrast, phloretin coordinately affected all PINs and both AUX1 and LAX3 genes inducing the same time-point pattern of expression, but significantly up−/down-regulated only PIN1, PIN3 or PIN7 in roots and shoots, respectively. PIN1 is found postembryonically in the root stele (Gälweiler et al., 1998) and it is responsible for acropetal auxin flow towards the root tip, together with another two members of PIN family of auxin transporters: PIN3 and PIN7. Both PIN3 and PIN7 genes show a similar pattern of expression in root pericycle and columella cells, redirecting the auxin flow downwards to the root tip and laterally towards the flanks of the root apex (Friml et al., 2002; Billou et al., 2005). Therefore, it is likely that phloretin-induced up-regulation of PIN1, PIN3 and PIN7 genes complements local auxin biosynthesis generating more robust auxin maxima in root tips and induces increased auxin accumulation in lateral parts of the root tips, as visualized in Arabidopsis transgenic line DR5rev::GFP. The phytotoxicity of bisphenol A (Bahmani et al., 2020) and weisiensin B (Li et al., 2019) was also based on the increased expression of PIN1, PIN3 or PIN7 in roots.

Treatment with phloretin also induced a strong agravitropic phenotype of Arabidopsis seedlings. As the lack of starch in the starchless mutants (Caspar et al., 1989; Kiss et al., 1989) or physical elimination of columella cells (Blancaflor et al., 1998; Tsugeki and Fedoroff, 1999) led to severely attenuated gravity responses, redistribution of starch-filled amyloplasts can be considered essential for plant gravity sensing (Baldwin et al., 2013). Light microscopy examination of the root tips of phloretin-treated Arabidopsis seedlings confirmed the assumption that reduction of starch content in columella cells was in the bases of agravitropic phenotype of treated seedlings. Similarly, the allelochemicals narciclasine and artemisinin inhibited the root gravity sensing as a result of reduced starch levels in the root tips of Arabidopsis (Na et al., 2011; Yan et al., 2018).

The normal gravitropic response further depends on the regular flow of auxin in roots mediated by the symmetrical positioned PIN3 which enables uniform auxin flow toward the flanks of the root tip from where it enters the basipetal stream depending on PIN2 (Friml et al., 2002; Kleine-Vehn et al., 2010). Some polyphenols such as ellagic acid and cold/heat stress are able to disturb the gravitropic root response by affecting the expression, distribution or trafficking of PIN2 and/or PIN3 proteins (Shibasaki et al., 2009; Hanzawa et al., 2013; Yan et al., 2015). Presumably, phloretin-induced up-regulation of PIN3 together with reduced starch content in gravity sensing columella cells are the main factors responsible for agravitropic growth and morphological abnormalities of treated Arabidopsis seedlings.

The activity of the proteins from the ATP-BINDING CASSETTE transporter superfamily, especially ABCB1, ABCB4, and ABCB19, contributes to the directed movement of auxins in plants (Geisler et al., 2005; Terasaka et al., 2005; Lewis et al., 2007). However, the strict role of the ABCBs remains controversial (Geisler et al., 2017). ABCB19 is localized in the vascular tissues of the hypocotyl and root, contributing to root-ward transport of auxin, while ABCB1 and ABCB4 are involved in the loading and transport of auxin into the shoot-ward stream from the root apex. Accordingly, mutants deficient in ABCB1 protein display reduced shoot-ward auxin transport (Geisler et al., 2005). Since ABCB transporters stably retain their localization in plasma membranes in response to internal and external cues (Cho et al., 2012), changes in their expression levels, such as the observed increase in expression of ABCBs in shoots and ABCB1 in roots, could be a way of modulating the root-ward and shoot-ward auxin transport, respectively, contributing to the PINs´ driven auxin accumulation in roots and agravitropic phenotype of phloretin-treated seedlings. In addition, increased expression of ABCB1 in phloretin-treated roots at time points coinciding with elevated levels of IAA and oxIAA, suggests that ABCB1 could mediate transport of IAA breakdown products in the regions exposed to high auxin concentrations, as also proposed by Geisler et al. (2005).

Phloretin human health-promoting properties (Behzad et al., 2017) in combination with its phytotoxic effects could be essential in development of new prospective bioherbicidal agent. Since phloretin could be obtained from waste products such as apple tree residues (fallen apple tree leaves) and juice industry by-product (apple pomace) by deglycosylation of phlorizin, it makes phloretin-based bioherbicides a prospective economical and eco-friendly solution for weed control. In addition, grinded apple tree leaves could be also applied by mulching, since phlorizin from apple mulch is degradable by soil actinomycetes (Huang et al., 2013b) to produce phloretin.

The presented study introduces the apple dihydrochalcone phloretin as a new allelopathic compound with harmful effects on species outside the genus Malus and provides the very first insight into the basic mechanism of its phytotoxic action. Phloretin treatment induced a significant dose-dependent growth retardation and agravitropic phenotype in Arabidopsis seedlings. Profoundly disturbed auxin metabolome profile in roots with highly increased content of IAA accumulated in the lateral parts of the root apex, accompanied with changes in the expression of auxin biosynthetic and transport genes, especially PIN1, PIN3, PIN7 and ABCB1, confirms that the disturbance of auxin homeostasis is the basis of phloretin phytotoxicity. This mechanism makes phloretin an attractive candidate for a new prospective bioherbicidal compound. Insight into the physiological basis of phloretin action also paves the way for further research of phloretin role in ARD and its physiological functioning in planta for a better understanding of its role as a potential growth modulator in apple.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

MS, NB, and SN designed the study. MS, DS, and MT performed morphological tests. DĆ and DS conducted the light microscopy experiments. AP performed auxin metabolite profiling analyses. MS, JS, and TĆ performed qRT-PCR analyses. MB, MS, and DS conducted confocal microscopy analyses. MS, DS, and JS analyzed the data and prepared the figures and tables. MS and DS wrote the manuscript. NB and SN provided critical editing of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, contract number 451-03-68/2022-14/200007.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2022.875528/full#supplementary-material