The impact of root exudates, volatile organic compounds, and common mycorrhizal networks on root system architecture in root-root interactions

ABSTRACT Plants constantly communicate with coexisting neighbors and adjust their physiological and morphological characteristics, including changes in root system architecture (RSA). Increased or decreased biomass accumulation, horizontal and vertical asymmetric distribution are the main behavioral performances. Some evidence has shown that these performances are associated with plant plasticity such as secretion of root exudates and release of volatile organic compounds (VOCs) and describe the role of common mycorrhizal networks (CMNs) as a communication pathway during belowground interplant interaction. Here, we highlight the direct role of root exudates as cues and signals and the indirect effects via regulating soil nutrients and soil microorganisms of these media in root-root interactions on RSA have been taken into consideration. At last, the existing knowledge gaps and potential research directions have been outlined for a better understanding of plant belowground interactions via RSA.


Introduction
The sessile nature gives plants a passive characteristic that makes them inevitable to interact with coexisting neighbors in natural and agricultural systems (Schandry and Becker 2020;Bilas et al. 2021;Ninkovic et al. 2021). Plants often grow together, and change traits above-and belowground to avoid the worst ecological consequences and challenges (Crepy and Casal 2015;Chen et al. 2021a). For example, plants can eavesdrop on volatile organic compounds (VOCs) signals emitted from herbivore-attacked neighbors and prime defenses (Baldwin et al. 2006). Similarly, the unstressed plants perceived and responded to osmotic and drought stress cues from their stressed neighbors and closed stomata (Falik et al. 2012). Due to the accessibility of aerial tissues facilitating shoot research, the aboveground interactions have been established in detail. In the last two decades, plant belowground interactions gain ever-increasing attention in contexts of root-root interactions (Semchenko et al. 2007b;Belter and Cahill 2015), root-soil interactions (Cahill et al. 2010;Yang et al. 2018), and root-soil-microbe interactions (Hellequin et al. 2021;Yu et al. 2021;Xu et al. 2021a).
Root-root interactions occur belowground, where roots can detect and respond to their neighbors which is conducive to plant performance and fitness (Liu et al. 2015;Rogers and Benfey 2015;Li et al. 2016b;Guo et al. 2020;Wang et al. 2020). Plant roots are plastic, resulting in alterations of root growth and distribution when growing with other plants. Root system architecture (RSA), the spatial configuration with some functional significance to the root system, serves pivotal role in obtaining water and mineral nutrients (Lynch 1995;McKay Fletcher et al. 2020;Liu 2021). There are different classifications of root systems, including tap roots (primary roots), lateral roots, basal roots (seminal roots) and shoot-borne roots (Lynch 2013;Satbhai et al. 2015;Hochholdinger et al. 2018). Root hairs, the tubular structure formed by the outwardly protruding epidermal cells, are also important for the root system (Leavitt 1901;Dolan 2017;Bienert et al. 2021). In addition, root hairs can increase the contents of exudates to enhance the nutrient absorption and rhizosphere interactions (Holz et al. 2018).
The mechanisms involved in the root-root interactions on RSA are well sophisticated. The action of root exudates, VOCs, and common mycorrhizal networks (CMNs) in plant-plant interaction is summarized in detail (Khashi u Rahman et al. 2019). In this review, we aim to shed light on our current knowledge of the role of root exudates, VOCs, and CMNs in root-root interaction on RSA, and to discuss potential directions to increase future research and understanding.

Effect of root-root interactions on RSA
Interaction between two roots can be studied as inter-or intraspecific considering the identity of neighbors (Faget et al. 2013). There are increasing evidence that interactions induce a range of changes in RSA (Munoz-Parra et al. 2017a;Colom and Baucom 2020;Bendes and Lincoln 2021). The root system has evolved different behavioral strategies, including adjustment of growth and allocation ( Figure 1; Table 1), which are often characterized by root length, angle, and biomass (Waidmann et al. 2020;Xiong et al. 2020).
The main outcomes of growth are increases, decreases and no changes in biomass accumulation. Plants usually over-proliferate their roots at the expense of reduced seed production when plants share soil resources, which is known as the tragedy of the commons (Gersani et al. 2001). In a split-root experiment, sharing soybeans (with interplant root competition) produced 85% more root mass and 30% less reproductive yield when there was no interplant root competition (Gersani et al. 2001). More phenomena about the tragedy of the commons were reported later (Maina et al. 2002;O'Brien et al. 2005;Zhu et al. 2019). Plants also adjust the lateral root number, which increases when Arabidopsis thaliana plants interact with their kins and strangers (Palmer et al. 2016). Furthermore, plants also decrease the amount of root production in response to the presence of neighbors. Cahill et al. (2010) found that Abutilon theophrasti plants adopted restricted foraging strategies when neighboring plants were present. Other examples are consistent with this result (Mommer et al. 2012;Chen et al. 2015;Broadbent et al. 2018). However, other results show that the growth of the root system depends on the availability of soil nutrients rather than on neighboring plants, that is, the ideal free distribution (Semchenko et al. 2007a;McNickle et al. 2014;Chen et al. 2021a).
All else being equal, roots preferentially grow in areas where there are no other roots (Gersani et al. 2001). Plant species, therefore, have developed a wide variety of traits that allow species to occupy different niches in specific time and specific locations, which is called niche complementarity (Faget et al. 2013). Studies reported that plant avoidance or spatial segregation during growing with neighbors creates a phenomenon of ecological niche complementarity (Schmid et al. 2015;Yang et al. 2018;Cabal et al. 2020). Spatial niche differentiation in root architecture enhances nutrient capture and biomass production in ancient maize/ bean and maize/bean/squash polycultures (Postma and Lynch 2012). Many studies also observed that the root length density of the upper or deeper layer is greater in the intercropping system (Streit et al. 2019;Gong et al. 2020). There are also some examples that roots aggregate beside neighboring plants or do not respond (Fang et al. 2011(Fang et al. , 2013Sattler and Bartelheimer 2018). Different species and genotypes have distinct responses to flexible neighbors. Yang et al. (2018) found that Oryza sativa roots invaded closely related cultivars and avoided distantly related cultivars. Moreover, Zea mays GZ1 roots invaded their roots but avoided Glycine max HX3 roots (Fang et al. 2011). Therefore, more research is needed to elucidate the effects of adjacent plants on focal plants.

Pathways of interactions on RSA
Plants can detect their neighboring plants (Cahill et al. 2010;Veits et al. 2019), and distinguish each other, including self/ non-self (Falik et al. 2003(Falik et al. , 2006, kin/non-kin (Callaway and Mahall 2007;Yang et al. 2018), and conspecific/heterospecific (Fang et al. 2011;Zhang et al. 2020). The alteration of RSA is one of the outcomes of response to plant interaction. Plants do not hear, see, smell or taste, instead, they secrete root exudates and release VOCs which directly affect RSA as cues and signals, establish CMNs for transmission of those cues and signals, or indirectly shape RSA by changing soil nutrients and soil microbial community composition ( Figure 2; Table 2). The complex mechanisms of the effect of interplant interactions on RSA are discussed in detail below.

Role of root exudates in interplant interactions
Roots secrete different substances to their surrounding environment generally described as root exudates (Walker et al. 2003;Bais et al. 2006;Biedrzycki et al. 2010). Root exudates are composed of high molecular weight compounds such as polysaccharides and proteins, and low molecular weight compounds which are further categorized as primary metabolites (e.g. sugars, amino acids and organic acids) and secondary metabolites (e.g. terpenes, flavonoids, glucosinolates and alkaloids) (Bais et al. 2006;Canarini et al. 2019;Ehlers et al. 2020;Chai and Schachtman 2022). These metabolites perform essential functions in different biological processes (Bais et al. 2006;Badri and Vivanco 2009). There is also conclusive supporting evidence that root exudates are a kind of wireless signal in cross-plant communications (Sharifi and Ryu 2021). They exist in the soil for a long time and affect coexisting neighbors and future generations by changing soil properties described as plant-soil feedback (Karlovsky 2008;Hu et al. 2018;Delory et al. 2021).

Root exudates directly affect RSA as cues and signals
Coexisting plants perceive chemicals released from neighbors and perform an inhibitory or stimulatory response. There is reasonable evidence that root exudates signaling is one of the most common mediators of interactions among The main outcomes of growth are increased or decreased amount of root production. Whereas vertical and horizontal asymmetry is found in the allocation. '+' and '-' represent increase and decrease, respectively. plant species, which stimulates RSA changes (Biedrzycki et al. 2010;Semchenko et al. 2014;Kong et al. 2018;Yang et al. 2018). For example, lateral root numbers in kin samples and stranger populations were significantly different from those in solo controls (single plant), and root exudates as cues directly mediated these differences (Biedrzycki et al. 2010;Palmer et al. 2016). Exposure of rice seedlings to distantly related plants or their root exudates induced larger root systems than when exposed to closely related plants or relative root exudates. Na 3 VO 4 , a root secretion inhibitor, further confirmed the important role of root exudates . The application of activated carbon, another tool that interferes with root exudates, indicated that root exudates can trigger changes in RSA (Nettan et al. 2019). Furthermore, glasshouse experiments showed that root exudates may carry specific information and trigger various RSA responses, including changes in root mass, root length density, specific root length, root surface density and root branching intensity (Semchenko et al. 2014;Delory et al. 2021).
Biochemicals in root exudates play a crucial role in the interaction of RSA (Schandry and Becker 2020). For example, allelopathic plants, including wheat (Triticum aestivum) and rice (O. sativa), release chemicals to inhibit the growth of weeds Yang et al. 2018). It has been reported that the benzoxazinoid (BX) family mediates these processes (Li et al. 2016c;Kong et al. 2018;Xu et al. 2021b). Among them, the paddy weeds exhibited an avoidance response to allelopathic rice and root length, area, width and depth were inhibited. Root segregation with nylon mesh increased inhibitory action due to the increment of rice allelochemicals authentic 5,7,4 ′ -trihydroxy-3 ′ ,5 ′dimethoxyflavone, 3-isopropyl-5-acetoxycyclo-hexene-one-1 and momilactone B compared to control (root contact) (Yang and Kong 2017). At the same time, the application of root exudates and root-secreted signaling chemicals (-)-loliolide and jasmonic acid in weeds induced the production of rice allelochemicals momilactone B and tricin, and wheat allelochemicals benzoxazinoids ( functions of allelochemicals is further emphasized. Furthermore, root secreted phytotoxic substances, such as gallic acid, ferulic acid, phydroxybenzoic acid, vanillic acid, salicylic acid, tannic acid and hydroquinone, from tomato plant inhibited the growth of lettuce and egg seedlings (Kim and Kil 1989). In addition, the e phytotoxin (-)-catechin secreted by Centaurea maculosa is considered a weapon for successful invasion .
Overall, the results of these recent studies demonstrate that both root exudates and biochemicals in root exudates are considerable cues and signals in terms of root-root interaction-induced changes in RSA. However, information on signaling substances in certain biological processes remains to be characterized due to methodological limitations (Pantigoso et al. 2021). New approaches are needed to clarify root-secreted specific signaling compounds.

Root exudates indirectly affect RSA by changing nutrients
Root exudates can mobilize soil nutrients, such as nitrogen (N), phosphorus (P), iron (Fe), that play a crucial role in modulating RSA (Li et al. 2014;Chai and Schachtman 2022). The regulation of soil nutrition by root exudates is hence worthy of attention in root interactions on RSA.
Plants rely on foraging major nutrients from the soil, especially N and P (Oldroyd and Leyser 2020). Root exudates at a certain growth stage have a positive effect on soil nutrient mineralization (Zhao et al. 2021). Both primary and secondary metabolites affect nutrient availability. Liu et al. (2022) demonstrated that carbohydrates and organic acids promote N transformations under low fertility conditions. Liu and Murray (2016) reviewed that the root secretion of flavonoids, including flavone, coumestan, isoflavonoid, flavonol, etc., is very important for nodulation and N fixation in the belowground. Notably, different types of flavonoids have different effects on the mineralization of nutrients. For example, luteolin (flavone) and coumestrol (coumestan, isoflavonoid) act on the nod gene (nodulation gene), but the former is an inducer and the latter is an inhibitor (Cooper 2004). However, isoflavones did not affect rhizobia signaling in red clover in a recent study (Weston and Mathesius 2013), which suggested that species identity may be associated with the effect of flavonoids on N fixation. Root exudates can facilitate P uptake as well (Li et al. 2003). The release of protons, carboxylates, and enzymes are the main pathways that facilitate P capture (Li et al. 2014). For example, the exudation of organic acid and acid phosphatase from faba bean and chickpea roots enhanced the absorption of P in maize plants (Li et al. 2004(Li et al. , 2007Zhang et al. 2020).
Trace elements, such as Fe, are also essential for plant growth (Gallego et al. 2012;Colombo et al. 2014). In Leguminosae/Gramineae intercropping, gramineous plants may be used to absorb Fe nutrition in alkaline soils (Dai et al. 2019). The reason could be that the gramineous plants release Fe carriers and bind Fe (III) to increase the effectiveness of Fe (Xiong et al. 2013). Similar results were obtained in A. thaliana Moreover, the secretion of organic acids can alleviate the toxic effects of Al (Ma et al. 2001).
In short, the substances secreted by the roots can affect the major and trace elements. Furthermore, there is substantial evidence that the availability of nutrients affects RSA, including root biomass van Dijk et al. 2021;Wang et al. 2021), root length (Kumar et al. 2020;Xia et al. 2020), lateral root number (Drew 1975;Pongrac et al. 2020), and root horizontal and vertical distribution (Liu et al. 2015;Zhang et al. 2020). Therefore, we speculate that root exudates can indirectly affect RSA by changing soil nutrients.

Root exudates indirectly affect RSA by altering soil microbial community
Plants can interact with neighboring plants and soil microorganisms by producing different chemical components. Another situation to be considered is that since root exudates are the main source of energy for microorganisms, thus play a key role in the root recruitment and assembly of microorganisms (  Root exudates can stimulate positive interactions with soil microbes, which contributes to plant resource absorption, stress tolerance and pathogen defense (Sasse et al. 2018). It has been proposed in the previous section that flavonoids are useful for N fixation, which is related to the interaction between roots and N-fixing bacteria (Liu and Murray 2016;Li et al. 2016a;Liu et al. 2017). Simultaneously, the symbiotic system of legumes usually requires a large amount of P, and the citrate secreted by the root chelates Ca 2+ then P is released from the Ca-P complex, which improves the N fixation capacity of root nodules (Mei et al. 2012). Moreover, exudates of Michelia macclurei improved root growth and placement of Cunninghamia lanceolata through shaping microorganisms for P acquisition (Xia et al. 2016). Therefore, we conclude that microbes can mediate root foraging during interplant interaction. Furthermore, the composition and content of root exudates can change under biotic and abiotic stress conditions (Henry et al. 2007;Ksouri et al. 2007;Mithöfer and Boland 2012;Hoysted et al. 2018), and altered root exudates can further induce changes in soil microbial composition (Bezemer and van Dam 2005). For instance, plants release specific root exudates to promote the colonization of plant growth-promoting bacteria (PGPR), thereby stimulating plant growth (Vives-Peris et al. 2018). This is consistent with the view of Chen et al. (2019) who found that T. aestivum recruited PGPRs to trade off the growth and development at different stages. In addition, root exudates can also act on the pathogen that interferes with the defense system of the plant (Baetz and Martinoia 2014).
Previously, it has been reported that soil microbes, such as PGPRs, can improve root performance effectively (Alzate Zuluaga et al. 2021). Moreover, microbial community assembly is dependent on root exudates and its chemical structure (Yuan et al. 2018;Zhalnina et al. 2018;Kudjordjie et al. 2019;Kawasaki et al. 2021;Wen et al. 2022). These results indicated that different components of root exudates may recruit different microorganisms to cause different root system responses under different growth conditions.

Role of VOCs in interplant interactions
Root exudates are usually secreted in the form of water. In contrast, VOCs are typically released as airborne signals. Plants emit a large amount of VOCs into the surrounding environment when they encounter herbivorous animals, pathogens or mechanical damage, which mediates the plant-plant interactions (Sugimoto et al. 2014;Coppola et al. 2017;Ninkovic et al. 2019Ninkovic et al. , 2021. VOCs as signals are important for the success of plants. For example, the parasitic plant Cuscuta pentagona makes use of VOCs for host location and host selection to adjust their growth (Runyon et al. 2006). Apart from these, Ninkovic (2003) found that volatile profiles also have an impact on biomass allocation. So, the question is raised whether VOCs can modulate RSA?
Increased interest has been found in the belowground plant-environment interactions. Munoz-Parra et al. (2017b) found that VOCs may be perceived by neighboring plants and regulate root physiology and morphological behavior. Centaurea stoebe root VOCs are beneficial for the germination and growth of different neighboring plants in the same domain (Gfeller et al. 2019). Moreover, the volatile 6pentyl-2H-pyran-2-one promoted lateral root formation (Garnica-Vergara et al. 2016). The VOCs also have an impact on primary root growth. However, different VOCs components and concentrations have different effects. For instance, the volatile 6-pentyl-2H-pyran-2-one restrained primary root growth (Garnica-Vergara et al. 2016). The volatile N,N-dimethyl-hexadecylamine also regulated primary root length, enhancing at low concentrations but inhibiting at high concentrations in A. thaliana (Vázquez-Chimalhua et al. 2021). Meanwhile, the physiological situations aboveground affect the secretion of VOCs belowground (Rostás et al. 2015). Root VOCs further reshape microbial Table 2. The examples of root-root interaction on RSA via root exudates, VOCs and CMNs.

No
Passways of interactions on RSA Description Ref. Benzoic, caffeic, ferulic and salicylic acid decreased fresh weight of Solanum lycopersicum, and chlorogenic, ferulic and salicylic acid discolored roots. Jung et al. 2004 3 Root exudates: meta-Tyrosine Multiple fine fescue cultivars and related root exudates meta-Tyrosine decreased lettuce root length. Bertin et al. 2007 4 Root exudates A. thaliana had more lateral roots in strangers (non-siblings) than in sibling root exudates, yet had shorter roots in the sibling or stranger root exudates than in its own root exudates. Biedrzycki et al. 2010;Biedrzycki et al. 2011 5 Root exudates Deschampsia caespitos had greater root length density: in the treatment with root exudates (comparing to the treatment with control solutions), in the the treatment with root exudates of unrelated individuals from the same population (comparing to the treatment with root exudates of related individuals from the same population), in the the treatment with root exudates of conspecifics from the same community (comparing to the treatment with root exudates of heterospecifics from the same community). 10 CMNs Dicymbe corymbosa had greater growth and survivorship with access to an ECM network than which without such access.
community assembly (Kong et al. 2021). The exposure of tomato leaves to Bacillus amyloliquefaciens strain GB03released β-caryophyllene elicited the release of SA from the root, which reassembled the composition of the rhizosphere microbial community (Kong et al. 2021). These results suggest that VOCs have important implications for subsurface plant-microbe interactions. It can be concluded from these findings that VOCs, both above-and belowground, can affect root architecture and soil microorganisms. However, whether VOCs can affect RSA of the neighboring plants by influencing microorganisms is not well understood.

Role of CMNs in interplant interactions
Researchers showed that the connection established by fungal hyphae through the root cells is known as CMNs (Genre et al. 2005;Bever et al. 2010). CMNs can promote plant growth, facilitate effective nutrient exchange and affect the composition of plants and soil microbial communities (Wipf et al. 2019;Sharifi and Ryu 2021).
Arbuscular mycorrhizal (AM) fungi, a class of microorganisms widely found in terrestrial ecosystems, can form symbiosis with most higher plants (Smith and Read 2008), which strongly regulates RSA and thereby promotes plant growth and development (Gutjahr and Paszkowski 2013). According to Sun et al. (2015), AM fungus Gigaspora margarita spores release volatiles and exudates that significantly affects lateral root formation in Lotus japonicus and A. thaliana. RNA-sequencing analysis revealed that AM regulates lateral root formation of Poncirus trifoliata and Camellia sinensis, associated with the hormone pathways and the nutrient pathways (Chen et al. 2017(Chen et al. , 2021b. However, after AM fungal treatment, different hormones had different effects on lateral root formation, with auxin being positive and ethylene being negative (Chen et al. 2017(Chen et al. , 2021b. Moreover, studies also showed that inoculation with AM fungi significantly increased the number of root hairs of Poncirus trifoliata, which was also associated with hormones and nutrient signaling (Wu et al. 2016). Some studies confirmed that AM fungal mycelium can obtain nutrients (such as N, P, etc.) from the soil and transmit them to the host plant (Ryan et al. 2012;Walder et al. 2012). Other reports also indicated the importance of hormone and nutrient signals for RSA (Giehl et al. 2014;McCleery et al. 2017;Shahzad and Amtmann 2017). Our existing knowledge also suggests that AM fungi can mobilize bacteria through fungal hyphae (Simon et al. 2015;Otto et al. 2017). A study found that AM fungal hyphae has dispersal networks that can transport phosphate solubilizing bacteria for the mineralization of organic phosphorus (Jiang et al. 2021). In addition, ectomycorrhizal (EM) fungi, another class of microorganisms that can form mycorrhizal networks belowground. Mycorrhizal network formed by EM fungi mediates interspecific communication between Betula papyrifera and Pseudotsuga menziesii (Simard et al. 1997). Another report showed that EM fungi stimulated lateral root production by synthesizing sesquiterpenes (Ditengou et al. 2015).
Combine mentioned information earlier, mineral nutritions and microorganisms affect RSA. Thence it has been noted that CMNs benefit from the reprograming of RSA through releasing chemicals or reshaping nutrients and microorganisms in the soil.

Conclusion and future directions
As the work reviewed here shows that root-root interactions, to a large extent, dominate the fate of the plant's success as a result of the central role of roots in foraging water and nutrients. Root-root interactions can be positive, neutral or negative and roots show outcomes of over-proliferation, decreased amount of root production, and vertical and horizontal asymmetry, etc. Root exudates, VOCs, and CMNs contribute important roles in the strategic processes of interactions on RSA directly or perhaps indirectly. However, some questions have not been answered yet and future studies are needed to investigate the detailed interactions among root exudates, VOCs, CMNs and RSA.
Social interactions usually exist in agriculture and natural ecosystems. So far, most of the studies of cross-plant interactions are carried out in greenhouses or controlled conditions. Therefore, a general universal understanding of interactions on RSA in the field remains to be found. Moreover, the composition of root exudates is complex and exudates are difficult to extract and characterize due to method and technical limitations (Canarini et al. 2019;Ehlers et al. 2020;Williams et al. 2021). Thus, what signal molecules and what functions in root exudates need to be further confirmed in the interactions on RSA. Furthermore, it is well understood that root exudates, VOCs and CMNs interact with each other Tian et al. 2021). Root structure and function affect plant exudation and nutrient absorption (McKay Fletcher et al. 2020;Sun et al. 2021). However, whether these interactions affect RSA requires further investigation.

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work was supported by the National Natural Science Foundation of China [grant number 31872156].

Notes on contributors
Xiu Zhang is a PhD student at the College of Horticulture and Landscape Architecture, Northeast Agricultural University, Harbin, P. R. China. She researches effects of root interactions on root architecture and and the mediation of root exudates within it.
Jingfan Yan is a postgraduate student at the College of Horticulture and Landscape Architecture, Northeast Agricultural University, Harbin, P. R. China. She researches the effects of root interactions on root system architecture under different nutrient conditions. Muhammad Khashi u Rahman is a PhD student at the College of Horticulture and Landscape Architecture, Northeast Agricultural University, Harbin, P. R. China. He researches mediums of plant interactions, particularly chemistry and functioning of root exudates.
Fengzhi Wu, an Associate Professor at the College of Horticulture and Landscape Architecture, Northeast Agricultural University, Harbin, P. R. China, is specialized in the prevention and control of soil-borne diseases and efficient utilization of nutrients. Some key technologies for safe production have been well applied.