Actinorhizal plants and Frankiaceae : The overlooked future of phytoremediation

Abstract

Bioremediation of degraded soils is increasingly necessary due to rising food demand, reductions in agricultural productivity, and limitations in total available arable area. Several bioremediation strategies could be utilized to combat soil degradation, with phytoremediation emerging as a standout option due to its in situ approach and low implementation and maintenance costs compared to other methods. Phytoremediation is also a sustainable solution, which is increasingly desirable to blunt the progression of global warming. Actinorhizal plants display several desirable traits for application in phytoremediation, including the ability to revegetate saline soil and sequester heavy metals with low foliar translocation. Additionally, when grown in association with Frankiaceae endophytes, these abilities are improved and expanded to include the degradation of anthropogenic pollutants and the restoration of soil fertility. However, despite this significant potential to remediate marginalized land, the actinorhizal‐Frankiaceae symbiosis remains heavily understudied and underutilized. This review aims to collate the scattered studies that demonstrate these bioremediation abilities and explain the mechanics behind such abilities to provide the necessary insight. Finally, this review will conclude with proposed future directions for utilizing this symbiosis and how it can be optimized further to facilitate improved bioremediation outcomes.

Actinorhizal plants, in symbiosis with Frankiaceae bacterial endosymbionts, possess traits beneficial for bioremediation. These plants can thrive in harsh conditions, such as high salinity and metal pollution, with their resilience enhanced by Frankiaceae. Additionally, Frankiaceae improve soil nutrients through nitrogen fixation and can degrade various anthropogenic pollutants.

INTRODUCTION

Bioremediation strategies are a class of “green technologies” that serve to restore degraded land, rendering it habitable and fit for arable use again. Being biologically based, such strategies are generally less intensive and polluting to the environment than physiochemical methods such as electrodialysis and solvent extraction among other examples (Sayqal & Ahmed, 2021). Bioremediation strategies can be divided into two broad classes, consisting of ex situ and in situ methods. Generally, in situ methods are preferable, as they are generally cheaper to implement, requiring less labour and space, as the bioremediation is conducted directly upon the polluted site, whereas ex situ methods consist of transporting the polluted material off site and treating it to remediate it (Azubuike et al., 2016).

In situ methods that introduce exogenous organisms for the purposes of bioremediation (enhanced in situ bioremediation) typically utilize microorganisms, with the exception of phytoremediation which instead utilizes plants (Azubuike et al., 2016; Sayqal & Ahmed, 2021). However, these plants can serve to aid the bioremediation effort alongside endemic soil microorganisms, with the efficacy of this approach also potentially bolstered through inoculation with desirable microorganisms.

Actinorhizal plants are noted pioneer species, well adapted to poor, disturbed soil, in part due to their symbiosis with the nitrogen fixing Frankiaceae. While Frankiaceae are noted as being capable of free‐living nitrogen fixation, through formation of oxygen excluding vesicles (Berry et al., 1993; Mohapatra et al., 2004). During symbiosis, less Frankiaceae resources may need to be devoted towards vesical formation, as the host plant is noted as contributing towards this micro‐oxic environment, in addition to the host providing Frankiaceae with photosynthesis derived carbon (Huss‐Danell, 1997; Persson & Huss‐Danell, 2008). In addition to this, as detailed in this review, this symbiosis, exhibits low foliar recycling of metals, effective revegetation of saline soil, degradation of anthropogenic pollutants and improvements to soil nutrients, making this symbiosis attractive for bioremediation. However, despite these desirable traits, the symbiosis remains heavily underutilized and understudied, with literature on this topic being scattered and difficult to locate.

Thus, this review aims to provide a cohesive overview of the actinorhizal‐Frankiaceae symbiosis to display its bioremediative potential. Such studies were located with relative ease through interrogation of databases such as PubMed searching key words related to the symbiosis and bioremediation, as the pool of literature from which to draw upon is relatively small. As such, the majority of literature detailing the actinorhizal‐Frankiaceae bioremediative capabilities has been considered. Critical review of these sources was warranted to accurately display the bioremediative capabilities of this symbiosis, especially as the limited numbers of studies considered in this review utilized differing methodologies and considered differing plant growth metrics and soil chemistry metrics. In addition, the conclusions reached by some studies differed from each other, therefore critical analysis is warranted to account for such differences. However, prior to discussion of the actinorhizal‐Frankiaceae symbiosis, the sources and effects of soil degradation will be briefly introduced.

Sources and effects of soil degradation

Soil degradation is a natural process to a degree, with volcanic activity, forest fires and rock weathering releasing heavy metals into the soil, with soil salinization driven naturally by evaporation of water from soil and deposition of salts through rock weathering (Hassanien & Shahawy, 2010; Rengasamy, 2006). Nevertheless, anthropogenic activity has greatly increased the rate and scope of these processes (Lal, 2015).

Soil pollution

Heavy metals and other polluting chemicals are released into the environment during energy production, mining, manufacturing and waste disposal, among other activities (Adesokan et al., 2016; Hassanien & Shahawy, 2010; Nicholson & Chambers, 2008; Yang et al., 2018; Zhang, Wu, & Simonnot, 2012), leading to soil contamination. Agricultural activities further soil pollution due to wastewater irrigation and use of pollutant‐ and metal‐containing agrochemicals (Barker & Gimingham, 1911; Mahfooz et al., 2020; Nicholson & Chambers, 2008; Sager, 2007; Sayo et al., 2020). These anthropogenic activities have increased the number of polluted sites across Europe (and presumably the rest of the world), with the majority of these sites remaining unremediated (EEA, 2022; Van Liedekerke et al., 2014).

In terms of effects upon agriculture, heavy metals cause oxidative damage which hinders crop growth (Opdenakker et al., 2012; Pourrut et al., 2008). This growth inhibition results from heavy metal stress leading to reduced photosynthetic pigment concentration, inhibition of enzymes involved in photosynthetic and other key plant production processes (Dhir et al., 2011; Di Salvatore et al., 2008; Li et al., 2005; Rivetta et al., 1997; Sridhar et al., 2014).

In addition to impacts upon agricultural productivity, heavy metal pollution can lead to serious medical conditions such as Minamata and Itai‐itai disease and a range of other health effects, with chronic exposure linked to cancer development (Aoshima, 2016; Gebeyehu & Bayissa, 2019; Harada, 1995; Khan et al., 2008; Sayo et al., 2020; Zhou et al., 2016). Whilst some studies indicate that crops grown upon metal polluted soils pose little risk to consumers (Khan et al., 2008), results presented by Zhuang et al. (2009), Balkhair & Ashraf, 2016, and Gebeyehu and Bayissa (2019) present these metals exceeding safety thresholds in crops grown upon metal polluted sites (including cancer risk thresholds).

Aside from crops, seafood and drinking water may also be polluted by heavy metal run off from contaminated soil (Gupta et al., 2009; Islam et al., 2015; Kobayashi et al., 2009; Nriagu et al., 1979; Taboada‐Castro et al., 2012; von Gunten et al., 1997). Although many studies indicate these food sources are safe for moderate consumption (Djedjibegovic et al., 2020; Maurya et al., 2019; Sobhanardakani, 2017; Sobhanardakani et al., 2018). These studies often only consider metals individually, when considered together using metrics such as the “total health risk index” (and its analogues), the resulting cumulative exposure values indicate risk to human health (El‐Shenawy et al., 2016).

In addition to heavy metals, soil can become polluted from a range of anthropogenic generated compounds. Those pertinent to the actinorhizal‐Frankiaceae symbiosis including atrazine, biphenyl compounds, hydrocarbons and phenolic compounds, will be detailed in this review. In the case of atrazine, this pollutant is released into the environment as it is applied to crops as a broadleaf herbicide. Despite atrazine being banned in Europe in 2004, it is still extensively used, with an estimated 70 million tonnes applied to crops annually in America (US geological survey, 2012). This heavy usage is problematic as the half‐life of atrazine can be up to 742 days, depending upon environmental conditions, potentially allowing accumulation in the environment and food chains (Solomon et al., 1996; Zhang et al., 2014). It is of interest to reduce environmental atrazine levels due to its potential teratogenic, endocrine and immune system disrupting effects (Jablonowski et al., 2011; Rohr & McCoy, 2010; Solomon et al., 1996).

Biphenyl compounds such as polychlorinated biphenyl (PCB), have been used as coolant and insulating fluids in electrical equipment, hydraulic lubricants and applied as plasticisers (Nisbet & Sarofim, 1972). PCBs exhibit a range of negative health effects including carcinogenicity, endocrine disruption, immunosuppression, reproductive and developmental issues, with many of these seen in the Japanese Yusho poisoning (Aoki, 2001; Geusau et al., 2001; Hagmar et al., 2001; Jacobson & Jacobson, 1996; Kramer et al., 2012; Kuratsune et al., 1972; Lauby‐Secretan et al., 2013; Masuda, 2003; Mocarelli et al., 2000; Schell et al., 2014; Stewart et al., 2000; Tsukimori et al., 2008; Yoshimura, 2012). Due to these negative health effects PCBs were banned in 2001, despite this it is estimated that 80% of PCBs are still present within the environment (Othman et al., 2022). Such pollution arises due to volatilization of PCBs disposed of within landfill, leaks from electrical transformers, incineration of PCB containing waste, improper disposal of PCB waste and accidental release (Duke et al., 1970; Kuratsune et al., 1972; Othman et al., 2022; Yoshimura, 2012).

Hydrocarbons are released in the environment most obviously through spills, such as the Deepwater Horizon oil spill which released an estimated five million barrels, and oil fires such as those seen in the 1991 Gulf War (McNutt et al., 2012; Yihdego & Al‐Weshah, 2017). Other sources of hydrocarbon pollution are incomplete combustion of organic materials such as coal and oil, and industrial manufacturing (Abdel‐Shafy & Mansour, 2016). Hydrocarbons are associated with damage to the central nervous, gastrointestinal, renal, hepatic, immune and cardiovascular systems, and are potentially carcinogenic and teratogenic (Abdel‐Shafy & Mansour, 2016; ATSDR, 1999; Marris et al., 2020; Perera et al., 2012; Rengarajan et al., 2015; Shiue, 2016; Tong et al., 2018; Zheng et al., 2018).

Phenol pollution arises from a range of sources including pulp and paper mills, petroleum and coal refining, petrochemical manufacture, pharmaceuticals and tanneries (Kumaran & Paruchuri, 1997; Lindström & Nordin, 1976). Phenol compounds are associated with a range of negative health effects in humans, such as nausea, vomiting, diarrhoea, abdominal pain, and a burning sensation in the oral cavity/pharynx, with more severe phenol poisoning being fatal (Boatto et al., 2004; Jarvis et al., 1985; Kim et al., 1994; Philip & Marraffa, 2012). Animal studies suggest phenol exposure may also lead to chronic issues with the nervous, cardiovascular, hepatic, renal, and immune systems; however, human studies regarding phenol are limited, usually with compounding factors (ATSDR, 2008; EPA, 2000; PHE, 2016). Although phenol cannot definitively be said to cause long term adverse medical effects in humans, removal of these pollutants from the environmental is still recommended based upon their acute toxicity and potential to cause chronic effects.

Salinity

Increasing soil salinity presents challenges to land use, with 10 million hectares of land being abandoned annually due to high salinity and 50% of arable land predicted to be salt affected by 2050 (Szabolcs, 1989; Wang et al., 2003). Anthropogenic activity increases soil salinity due to irrigation, which raises water tables bringing salts closer to the soil surface, the use of saline irrigation water and the use of salts as de‐icing agents on roads (Chen et al., 2010; Dobson, 1991; Ma et al., 2008; Oosterbaan, 1988; Pang et al., 2010; Rengasamy, 2006).

Salinity contributes to reductions in agricultural productivity as high salt levels decrease soil water potential, making water and nutrients more difficult for plants to acquire (Cruz et al., 2018; Hu & Schmidhalter, 2005). Furthermore, salinity reduces stomatal and mesophyll conductance, reducing CO₂ diffusion, overall impairing photosynthesis (Delfine et al., 1998; Delfine et al., 1999). Reductions in photosynthesis subsequently lead to generation of reactive oxygen species, causing deleterious oxidative damage, further inhibiting plant growth (Ali et al., 2004; Dobson, 1991; Ozgur et al., 2013; Singh et al., 2012). Reductions in growth are exemplified by many studies which show salt stressed crops performing poorer in a range of metrics compared to unstressed crops (Abbas et al., 2013; Magán et al., 2008).

Soil nutrient depletion

In an agricultural environment soil becomes nutrient depleted over time as crops are removed at harvest, necessitating replacement of the lost macronutrients and micronutrients with chemical fertilizers to maintain productivity. This cannot be avoided, however, crop management practices which leave minimal post‐harvest residual material (such as stover and straw), may lead to greater losses of nutrients than necessary (Blanco‐Canqui & Lal, 2009; Karlen et al., 1994; Salinas‐Garcia et al., 2001).

To alleviate the effects of reduced soil nutrients, economically and energetically expensive fertilizers are applied to soil. Such fertilizers are estimated to contribute 1–2% of global CO₂ emissions, acting as a driver of climate change, in turn hampering agricultural output (Mbow et al., 2019; Walling & Vaneeckhaute, 2020; Xx, 2023). In addition, fertilizers can leach into nearby water bodies causing eutrophication, with the resulting algal blooms depleting the water of oxygen and potentially producing toxins (Anderson, 1994; Carpenter et al., 1998; Flewelling et al., 2005; Gilbert et al., 2006; Jeglitsch et al., 1998; Kotak et al., 1993; Poli et al., 1986; Tong & Chen, 2002; Watkins et al., 2008).

BIOREMEDIATION STRATEGIES

In situ bioremediation

As detailed earlier, in situ methods are preferable, with the most basic of these being known as natural attenuation which involves leaving the site for indigenous organisms to restore through natural biological processes. In some cases this process is enhanced through the use of bioinoculants (bioaugmentation), whereas in other cases may stimulate aerobic degradation by the native soil microflora by introducing oxygen into the soil (bioventing) or by increasing oxygen concentration in the groundwater (biosparging). Such enhanced in situ bioremediation methods are expected to show greater bioremediation rates, particularly in the case of bioaugmentation as the introduced organisms have been specifically selected for their bioremediative properties (Azubuike et al., 2016; Sayqal & Ahmed, 2021).

Phytoremediation and microbially‐assisted phytoremediation

Regarding in situ approaches, phytoremediation is generally viewed as one of the cheaper options in contrast to other methods, which may require acquisition, installation of expensive equipment. Phytoremediation instead utilizes plants to remove pollutants from the soil, presenting a low cost regarding acquisition and planting of the plants, with inputs maintenance as once established plants are generally self‐sufficient (Wang & Delavar, 2023). However, phytoremediation efficacy may be limited by physiological constrains, most notably slow growth which limits turnaround time, small root systems which limit access to deeper pollutants, difficulty in accumulating certain non‐bioavailable metals and poor ability to survive in heavily polluted soils (Adeoye et al., 2021; Sayqal & Ahmed, 2021; Wang & Delavar, 2023).

Some of these constraints such as slow growth are intrinsic to the nature of plants, so phytoremediation may be best applied to sites in which a long‐time horizon is envisaged for the bioremediation process (Wang & Delavar, 2023). Other limitations can be alleviated through careful selection of the plants used, with actinorhizal plants being a particularly attractive example as they exhibit rapid growth upon inhospitable soils while also accumulating metals. In a similar manner to bioaugmentation, microbial inoculants can be deployed alongside plants to further lessen some of the challenges associated with phytoremediation, with this approach known as microbially associated phytoremediation.

Microbes have been shown to aid phytoremediation by forming symbioses with plants, improving growth under stress conditions, enhancing pollutant uptake and degradation of contaminants the host plant is unable to. Successful examples include the supplementation of Alfalfa with Pseudomonas aeruginosa, which increased the rate of petroleum hydrocarbon removal compared to the plants or microbes alone, while also improving plant growth compared to uninoculated plants (Agnello et al., 2016). Similar results have also been observed for other contaminants including 2,4‐dichlorophenoxyacetic acid, phenanthrene, 2‐chlorobenzoic acid, among other examples (Germaine et al., 2006; Sheng & Gong, 2006; Siciliano & Germida, 2009). As such, plants with desirable characteristics and the capability of forming associations with symbionts which can assist in phytoremediation are highly desirable. Actinorhizal plants and their Frankiaceae endosymbionts could represent such a symbiosis, offering significant potential in microbially assisted phytoremediation.

ACTINORHIZAL PLANTS AND THEIR ENDOSYMBIONT

Actinorhizal plants

Actinorhizal plants are a group of rosid plants consisting of eight families of perennial angiosperms, all of which are trees or shrubs (except those in the Datisca genus) (Dawson, 2008; Doyle, 2011). Actinorhizal plants are pioneer species found globally in a range of often ecosystems, often occurring in inhospitable, nutrient poor environments such as deserts and costal dunes (Dawson, 1986; Dawson, 2008).

The ability of actinorhizal plants to grow in inhospitable sites is partly due to their root symbiosis with nitrogen‐fixing Frankiaceae, with 280 of the 420 actinorhizal species displaying this association (Dawson, 2008). This symbiosis is primarily driven by intra‐ or intercellular infection of the host plant roots, with this infection stimulating nodule primordium formation through modification of lateral root cells (Pawlowski, 2008). As detailed earlier, the nodule environment, contributes to creation of a micro‐oxic conditions while also supplying carbon to Frankiaceae, facilitating Frankiaceae growth and nitrogenase function (Huss‐Danell, 1997; Persson & Huss‐Danell, 2008).

Such a symbiosis has been noted as providing 61% of the nitrogen requirement for a stand of Ceanothus velutinus (Dawson, 2008; Zavitkovski & Newton, 1968). Demonstrating why inoculation of actinorhizal plants with Frankiaceae often significantly increases host plant growth. Examples of such improvements to host plant growth include inoculation of Casuarina equisetifolia with Frankiaceae strains CeFr1 and Cefr2 which yielded a positive change in a range of growth metrics after just 90 days, significantly increasing plant height, stem girth and survival after 2 years in field conditions (Karthikeyan, 2016). Likewise, inoculation of Alnus resulted in greater total dry weight and nitrogen content compared to uninoculated trees within 6 months of planting (Wheeler et al., 1991).

It is estimated the actinorhizal‐Frankiaceae symbiosis fixes 240–350 kg ha⁻¹ of nitrogen per year (Wall, 2000). In some instances this may even be exceeded, as it has been shown a 15‐year‐old Ceanothus velutinus stand added 432 kg ha⁻¹ of nitrogen to the soil by deposition of leaf litter alone (Zavitkovski & Newton, 1968). Conversely, this figure may also be drastically lower under unfavourable environmental conditions and poor forestry management, thought this is relatively uncommon (Dommergues, 1997). Such nitrogen rich leaf litter has been noted as improving growth of other plant species, exemplified by conifer forests containing A. rubra exhibited increased soil nitrogen compared to conifer monocultures, due to decomposition of A. rubra leaf litter (Tarrant et al., 1969). This is desirable for bioremediation as this will aid in restoration of soil fertility and reforestation.

Frankiaceae

Frankiaceae is a poorly studied family of Gram‐positive, difficult to cultivate actinobacteria, which grows primarily through filamentous vegetative hyphae, producing vesicles and intercalary or multilocular sporangia. Frankiaceae are symbionts of actinorhizal plants but have also been found free‐living in soil, devoid of hosts (Benson & Silvester, 1993; Maunuksela et al., 1999; Ridgway et al., 2004).

The family Frankiaceae, within the order Frankiales (Sen et al., 2014) within the phylum Actinomycetota, was proposed in 1970 (Becking, 1970), initially solely containing the Frankia genus (Brunchorst, 1886). All Frankiaceae species were initially grouped into four clusters, with these clusters elevated to genus level in 2022 and named Parafrankia, Protofrankia and Pseudofrankia in addition to Frankia (Gtari, 2022) (Table 1).

TABLE 1.

Validly named and candidatus species within the Frankiaceae family, their infectivity ranges and strains belonging to each species (Gtari et al., 2020; Herrera‐Belaroussi et al., 2020; Nguyen et al., 2019; Normand et al., 2023; Normand & Fernandez, 2019; Nouioui, Ghodhbane‐Gtari, Jando, et al., 2023; Nouioui, Ghodhbane‐Gtari, Pötter, et al., 2023; Nouioui, Neumann‐Schaal, Pujic, et al., 2023; Pozzi et al., 2020), with type species indicated in bold. The “*” symbol indicates this genus is infective to Casuarinaceae except Gymnostoma. While Frankia nepalensis is classified as a Frankia species, it is believed this was done under the old four cluster classification system rather than the new proposed genera (Gtari, 2022). This is based upon F. nepalensis exhibiting traits associated with Pseudofrankia and being most closely related to species within this genus, thus F. nepalensis will be considered a Pseudofrankia for the purposes of this review.

Genus	Species within the genus (strains of each species)	Infectivity range
Frankia	Frankia alni (ACN14a T , AvcI1, MpI1, M16467, M16477)	Alnus
	Frankia canadensis (ARgP5 T )	Casuarinaceae*
	Frankia casuarinae (CcI3 T , ORS020606 (CeD), ORS022602, ORS020607 (CeF), ORS020608, HFP022801 (AllI1), ORS021001 (Cj1‐82), ORS02060, CcI2, BMG5.23, Thr, CeD, TA, Cj1‐82, Cg70.4, Allo2, CcI6, BR, Cg70.3, ORS022602, CeF, ORS020608, AllI1, ORS020609)	Myricaceae
	Frankia gtarii (Agncl‐4 T , Agncl‐10)
	Frankia tisai (Agncl‐8 T , Agncl‐18)
	Frankia torreyi (CpI1 T , CoN24d, ACN1Ag, Ag24‐251, ArI3, ARgN22d, Ar24H3, Ar24O2, A2J)
	Frankia umida (Ag45/Mut15 T )
	Candidatus Frankia alpina (AiOr T , AvVan)
	Candidatus Frankia nodulisporulans (AgTrS T , AgUmASt1 and AgUmASH1)
Parafrankia	Parafrankia colletiae (Cc1.17 T )	Colletieae
	Parafrankia elaeagni (BMG5.12 T )	Elaeagnaceae
	Parafrankia discariae (BCU110501 T )	Gymnostoma
	Parafrankia irregularis (G2 T )	Myricaceae
	Parafrankia soli (Cj T )
Protofrankia	Protofrankia coriariae (BMG5.1 T )	Coriariaceae
	Candidatus Protofrankia californiensis (Dg2 T )	Datiscaceae
	Candidatus Protofrankia datiscae (Dg1 T )	Dryadoideae
	Candidatus Protofrankia meridionalis (Cppng1_Ca_nod T )	Ceanothus
Pseudofrankia	Pseudofrankia asymbiotica (M16386)	Unable to reinfect host plants
	Pseudofrankia inefficax (EuI1c T )
	Pseudofrankia saprophytica (CN3 T )
	Frankia nepalensis (CN4 T , CN6, CN7, CNm7)

Despite Frankia first being described by Woronin (1866), the type species Frankia alni was not validated until 2016, due to difficulty in locating a strain that matched the original description (Nouioui et al., 2016). The proposal of the type strain has since led to a large boom in the classification of Frankiaceae species (Gtari, 2022). Currently 17 validly named Frankiaceae species have been proposed, alongside five candidatus species (Table 1).

FRANKIACEAE‐ACTINORHIZAL BIOREMEDIATIVE ABILITIES

Frankiaceae‐actinorhizal tolerance to salinity and salinity reduction

Revegetation of saline sites can reduce further salinization and potentially reduce existing salinity. This occurs as revegetation reduces wind and water soil erosion and lowers the water table by intercepting rainfall and increasing water uptake by the plants (Schofield & Scott, 1991) (Figure 1). Actinorhizal plants may facilitate revegetation of saline sites, with the Frankiaceae symbiosis improving their survival and in turn, their ability to reduce salinity (Table 2).

FIGURE 1.

Diagrammatic representation of the bioremediative capabilities of the Frankiaceae‐actinorhizal symbiosis. Left side of the image shows the abilities of the Frankiaceae within the nodule and the right side of the image shows the bioremediative abilities of the actinorhizal plants.

TABLE 2.

Studies related to bioremediation that utilize Frankaiceae strains and actinorhizal plants. These studies are grouped in the below regarding which aspect of bioremediation they are concerned with. In cases where only Frankiaceae species or actinorhizal plants species were utilized in the study this has been noted with the phrase not utilized listed under the corresponding column of the table. The names of the Frankiaceae species and strains used in the studies detailed in the below table have been updated to include the species name where possible.

Actinorhizal plant species used	Frankiaceae species used	Reference
Salinity remediation
Casuarina glauca	Frankia sp. CcI156, Frankia sp. CgIM4	Mansour et al., 2016
Casuarina glauca, Casuarina equisetifolia	Frankia casuarinae CcI3, Frankia casuarinae CeD	Ngom et al., 2016
Elaeagnus macrophylla	Frankia sp. Ema1	Tani & Sasakawa, 2000
Casuarina equisetifolia	Frankia sp. Ceq1	Tani & Sasakawa, 2003
Elaeagnus angustifolia	Not utilized	Qi et al., 2018
Elaeagnus angustifolia	Not utilized	Khamzina et al., 2006
Alnus glutinosa	Not utilized	Deptuła et al. (2020)
Not utilized	Frankia casuarinae CcI6	Oshone et al., 2013
Not utilized	Frankia casuarinae Allo2, Frankia casuarinae CcI6, Frankia casuarinae Thr, Frankia casuarinae CeD, Frankia casuarinae CcI3, Frankia casuarinae BMG5.23, Frankia casuarinae CgI82, Frankia casuarinae BR Frankia sp. DC12, Frankia alni ACN14a Frankia inefficax EuI1C, Frankia sp. EAN1pec	Oshone et al., 2017
Not utilized	Frankia alni ACN14a	Ghedira et al. (2017)
Casuarina glauca, Casuarina equisetifolia	Frankia casuarinae CcI3, Frankia casuarina CeD	Ngom et al., 2016
Casuarina glauca	Frankia sp. Thr	Batista‐Santos et al., 2015
Casuarina glauca	Frankia sp. Thr	Duro et al., 2016
Heavy metal bioremediation
Alnus glutinosa	Not utilized	Mertens et al., 2004
Alnus glutinosa, Alnus incana	Not utilized	Lorenc‐Plucińska et al., 2013
Alnus glutinosa	Frankia alni ACN14a	Bélanger et al., 2015
Alnus glutinosa	Frankia sp. UGL 010708, Frankia sp. UFI 010708, Frankia sp. UFI 13270238	Wheeler et al., 2001
Alnus incana	Not utilized	Rosselli et al., 2003
Alnus nepalensis	Not utilized	Jing et al., 2014
Casuarina glauca	Frankia sp. BMG5.22, Frankia casuarinae BMG5.23	Ghazouani et al., 2020
Alnus crispa, Alnus glutinosa	Frankia alni AvcI1	Callender et al., 2016
Alnus glutinosa	Frankia sp. WgAvcI1	Pawlowski et al., 1997
Not utilized	Not utilized	Gupta et al., 2002
Alnus glutinosa	Not utilized	Vandecasteele et al., 2008
Alnus glutinosa	Not utilized	Desai et al. (2019)
Alnus hirsute, Alnus firma	Not utilized	Lee et al., 2009
Not utilized	Frankia torreyi ACN1AG, Frankia casuarinae CcI3, Parafrankia colletiae Cc1.17, Pseudofrankia saprophytica CN3, Frankia torreyi CpI1‐S, Frankia torreyi CpI1‐P, Frankia sp. DC12, Frankia sp. EI5c, Frankia sp. EAN1pec, Pseudofrankia inefficax EuI1c, Frankia sp. EUN1f, Frankia sp. QA3	Richards et al., 2002
Not utilized	Frankia inefficax EuI1c, Frankia saprophytica CN3, Frankia alni ACN14a, Frankia casuarinae CcI3	Rehan et al., 2019
Not utilized	Frankia casuarinae CcI3, Frankia alni ACN14a, Frankia sp. QA3, Frankia sp. EUN1f, Frankia sp. EAN1pec, Pseudofrankia inefficax EuI1c, Pseudofrankia saprophytica CN3, Frankia sp. DC12	Furnholm & Tisa, 2014
Not utilized	Pseudofrankia saprophytica CN3, Frankia sp. DC12, Pseudofrankia inefficax EuI1c	Rehan, Furnholm, et al., 2014
Not utilized	Frankia sp. ACN10a, Frankia sp. ACN12a, Frankia alni ACN14a, Frankia casuarinae CcI3, Frankia sp. CH37, Frankia torreyi CPI1, Frankia casuarinae Cg70.4, Frankia sp. Cg70.9, Frankia casuarinae Cj1‐82, Frankia sp. Ea1‐12, Frankia sp. DC12, Parafrankia discariae BCU110501, Frankia sp. DSMZ 44251	Deicke et al., 2019
Not utilized	Frankia sp. CH37	Mohr et al., 2021
Casuarina equisetifolia	Not specified	Karthikeyan et al., 2009
Alnus glutinosa	Not utilized	Kuznetsova et al., 2011
Anthropogenic pollutant remediation
Not utilized	Frankia alni ACN14a, Pseudofrankia inefficax EuI1c	Rehan, Kluge, et al., 2014
Not utilized	Pseudofrankia inefficax EuI1c, Frankia sp. EUN1f, Frankia sp. CcI49, Parafrankia irregularis G2, Frankia sp. R43	Mansour et al., 2017
Alnus glutinosa	Not utilized	Kuznetsova et al., 2010
Casuarina equisetifolia	Not utilized	Sun et al., 2004
Alnus crispa	Frankia alni AvcI1	Lefrançois et al., 2010
Alnus viridis ssp. crispa, Alnus incana ssp. rugosa	Frankia alni AvcI1	Bissonnette et al., 2014
Not utilized	Frankia sp. QA3, Frankia alni ACN14a, Frankia sp. EAN1pec, Pseudofrankia inefficax EuI1c, Frankia sp. EUN1f, Parafrankia elaeagni BMG5.12	Rehan et al., 2016
Soil nutrient improvement and plant growth promotion
Ceanothus Velutinus	Not specified	Zavitkovski & Newton, 1968
Casuarina equisetifolia	Frankia sp. CeFr1, Frankia sp. CeFr2	Karthikeyan, 2016
Alnus rubra, Alnus glutinosa	Frankia sp. Ar 1.2.5q, Frankia sp. ArI4, Frankia sp. Ag 1.1.8	Wheeler et al., 1991
Alnus rubra	Not specified	Tarrant et al., 1969
Not utilized	Frankia alni ACN14a, Frankia torreyi CPI1, Frankia torreyi ACN1AG, Frankia sp. QA3, Frankia casuarinae CcI3, Frankia casuarinae Allo2, Frankia casuarinae BMG5.23, Frankia casuarinae CcI6, Frankia casuarinae CeD, Frankia casuarinae Thr, Frankia coriariae BMG5.1, candidatus Frankia datiscae Dg1, Frankia elaeagni BMG5.12, Frankia discariae BCU110501, Frankia sp. EUN1f, Frankia sp. EAN1pec, Frankia irregularis DSM 45899, Frankia sp. R43, Frankia saprophytica CN3, Frankia inefficax EuI1c, Frankia sp. DC12	Nouioui, Cortés‐albayay, et al., 2019
Not utilized	Frankia torreyi CpI1	Nouioui, Ghodhbane‐Gtari, et al., 2019
Not utilized	Frankia canadensis ARgP5	Normand et al., 2018
Not utilized	Frankia sp. RT, Frankia sp. Rif, Frankia casuarinae Thr, Frankia sp. BR, Parafrankia irregularis G2, Parafrankia soli Cj, Frankia sp. URU, Frankia sp. CH, Frankia sp. G82	Arahou et al., 1998
Casuarina equisetifolia	Not utilized	Mailly & Margolis, 1992
Casuarina equisetifolia	Not specified	Parrotta, 1999
Casuarina equisetifolia	Not utilized	Izquierdo et al., 2005
Elaeagnus angustifolia	Not utilized	Khamzina et al., 2006
Elaeagnus angustifolia	Not utilized	Qi et al., 2018
Alnus incana	Not utilized	Taylor et al., 1989
Not utilized	Frankia sp. DDNSF‐01, Frankia casuarinae DDNSF‐02	Marappa et al., 2020
Alnus incana	Not utilized	Uri et al., 2001
Not utilized	Frankia sp. 52065, Frankia sp. HFPCcI3, Frankia sp. HFPArI3, Frankia sp. ArI5, Frankia sp. HFPCPI1, Frankia sp. AvsI3, Frankia sp. Air11, Frankia alni AvcI1	Aronson & Boyer, 1994
Not utilized	Frankia sp. AiPs1	Haansuu et al., 2001
Not utilized	Frankia sp. AiPs1	Klika et al., 2001
Not utilized	Frankia sp. AiPs1	Klika et al., 2003
Not utilized	Frankia alni ACN14, Frankia casuarinae CcI3, Frankia sp. EAN1pec	Udwary et al., 2011
Not utilized	Frankia sp. HFPArI3, Frankia sp. HFPCPI1, Frankia sp. HFPCcI2, Frankia sp. HFPCcI3, Frankia sp. HFPGpI1	Safo‐Sampah & Torrey, 1988
Casuarina equisetifolia	Frankia sp. UMCe12, Frankia sp. UMCe23, Frankia sp. UMCe35, Frankia sp. UMCe55	Gopinathan, 1995
Not utilized	Frankia alni AvCI1	Wheeler et al., 1984
Ochetophila trinervis	Parafrankia discariae BCU110501	Solans et al., 2011
Alnus rubra	Frankiaceae sp. HFPArI3	Berry et al., 1989
Not utilized	Frankiaceae sp. HFPArI3	Stevens & Berry, 1988

The salinity tolerance of actinorhizal plants still generates some controversy, with studies such as Mansour et al. (2016) suggesting that species such as Casuarina glauca are not saline tolerant, as these plants are unable to survive in NaCl concentrations greater than 200 mM. Other studies such as Ngom et al. (2016) instead state that C. glauca is saline tolerant. Such discrepancies between studies may be explained by osmotic shock, as the NaCl was added to the planting media more rapidly in Mansour et al. (2016) compared to others. Therefore, such methodological nuances should be considered when reviewing plants for use in saline environments.

Aside from Casuarinaceae, Elaeagnaceae is noted as being suitable for bioremediation of saline sites but displays a lower saline tolerance than Casuarinaceae. Seed germination of Elaeagnus macrophylla decreased sharply in concentrations of NaCl higher than 50 mM, with no germination occurring at 200 mM (Tani & Sasakawa, 2000). However, E. macrophylla seedlings showed minimal decrease in biomass at increasing NaCl concentration, indicating that growth may be less saline affected than germination. Examples of E. macrophylla use for treatment of saline sites include plantation upon areas of the Yellow River Delta (Qi et al., 2018). Such plantation resulted in reductions in soil salinity while increasing nitrogen, phosphorus and potassium compared to unplanted soil, with salinity reduction and nutrient accumulation increasing each year. In another study, Elaeagnus angustifolia was planted upon saline soil (3.3–4.3 dSm⁻¹) around the Aral sea, with 96–100% of these plants surviving 19 months after planting and exhibiting rapid root and aerial tissue growth (Khamzina et al., 2006). Although not measured, such plant growth would be expected to reduce salinity by limiting water loss from the soil, lower the water table and improve the soil structure.

Alnus glutinosa may also be useful in revegetation of saline sites, particularly in boreal environments in which Casuarina and Elaeagnus are not suited. While A. glutinosa is well known for its ability to thrive in waterlogged sites, there has been some conflict regarding its salinity tolerance (Claessens et al., 2010). For example, studies such as Gómez Mercado et al. (2012) and Dirr (1976) suggest that A. glutinosa does not have good saline tolerance, while Dobson (1991), Mertens et al. (2004), and Deptuła et al. (2020) report moderately good saline tolerance in A. glutinosa. Similarly to Elaeagnaceae, A. glutinosa may be more susceptible to salinity during germination and early growth stages, so transplantation of plants after germination to saline sites may facilitate their revegetation with Alnus (Deptuła et al., 2020).

It is well known that certain endophytic bacteria can improve plant tolerance to abiotic stress, with Frankiaceae capable of eliciting such a response in actinorhizal plants (Kamran et al., 2022). Frankiaceae strains have displayed instances of high salinity tolerance, with strains such as CcI6 and Allo2 displaying a minimum inhibitory concentration of 1000 mM to NaCl (Oshone et al., 2013; Oshone et al., 2017). However, others such as Ceq1 and CcI3 present lower minimum inhibitory concentrations of 500 and 475 mM NaCl, respectively (Oshone et al., 2017; Tani & Sasakawa, 2003). When comparing the genomes of salt tolerant strains to less tolerant strains, it was found the former contained 153 additional genes compared to the latter (Oshone et al., 2017). Seven of these genes were upregulated under saline conditions and were associated with alterations to the cell envelope, changes in membrane fluidity and compatible solute synthesis (Oshone et al., 2017).

Aside from the genes exhibited only by the highly salt tolerant strains, salt tolerance in Frankiaceae is also thought to occur through several other mechanisms including exclusion and removal of salt from the cell through the action of P‐type ATPases, Na+/H+ antiporters and Na+‐ATPases. This is exemplified by the cells of Frankiaceae strains Ceq1 and Ema1, which were observed to have intracellular NaCl concentrations less than 30 and 20 mM, respectively, despite the surrounding media containing 500 and 200 mM NaCl, respectively (Oshone et al., 2017; Tani & Sasakawa, 2000; Tani & Sasakawa, 2003). It was noted that increased NaCl concentrations raised the rate of cellular sodium efflux, indicating that active efflux is engaged to remove salts from the cell (Srivastava et al., 2012).

Furthermore, Ghedira et al. (2017) demonstrated in F. alni that ABC transporters, are upregulated in response to desiccation, which maintains osmotic balance by allowing movement of ions. Interestingly, the mechanosensitive ion channel MscL was upregulated in response to desiccation, which opens in response to mechanical changes in the cell membrane, preventing osmotic damage (Ghedira et al., 2017). Usually, such channels open in response to a hypotonic environment, but this channel may be upregulated to prevent overaccumulation of compatible solutes, which could result in cell lysis (Booth & Louis, 1999).

To further facilitate salinity tolerance in Frankiaceae several genes, such as glycosyl transferases, nucleoside polysaccharide deacetylases and sugar epimerases, are upregulated to reduce salt influx through cell envelope modification (Oshone et al., 2017). This was corroborated by Ghedira et al. (2017), as the most upregulated F. alni gene in response to desiccation was a putative autotransporter adhesin, which may play a role in cell envelope remodelling. Under salt stress the Frankiaceae cell membrane may also be modified to become more fluid, to better resist salinity imposed osmotic stress. This modification is driven by upregulation of enzymes such as acyl‐acyl carrier protein desaturases, which converts saturated fatty acids to unsaturated fatty acids, and upregulation of ubiquinone biosynthesis to increase mechanical stability and fluidity of the membrane (Oshone et al., 2017).

Another method employed by Frankiaceae to improve salinity tolerance is upregulated synthesis of compatible solutes to reduce osmotic water loss from the cell (Oshone et al., 2017) (Figure 1). This occurs through several mechanisms, with genes facilitating synthesis of N‐acetylglutaminylglutamine amide, upregulated synthesis of trehalose and upregulated production of glutamate (Ghedira et al., 2017; Oshone et al., 2017). In addition, the action of unregulated peptidases and proteases indirectly result in the release of amino acids, which act as compatible solutes (Ghedira et al., 2017).

To further protect the cell from osmotic damage, several reactive oxygen species scavenging genes and DNA repair genes are also upregulated (Ghedira et al., 2017). In tandem, three major groups of proteins are downregulated in response to salinity, with these genes involved in nitrogen fixation, respiration and homologous recombination. These are likely downregulated as they facilitate highly energy demanding processes that are not essential to immediate cellular survival (Ghedira et al., 2017).

Drawing upon the few examples within the literature, inoculation of actinorhizal plants with saline tolerant Frankiaceae leads to increases in actinorhizal salinity tolerance and seems a feasible method to improve performance in this regard. Ngom et al. (2016) showed that inoculation of C. glauca with Frankiaceae strains CcI3 and CeD increased the growth, biomass and chlorophyll content of shoots across all tested salinity ranges (0–500 mM) compared to uninoculated plants. Additionally, strain CcI3 increased root biomass compared to uninoculated C. glauca plants in all saline concentrations and inoculation with strain CeD increased root biomass in NaCl concentrations greater than 100 mM. This increased salinity tolerance may be due to accumulation of the osmoprotectant proline, with proline concentration increasing at all salinity concentration in C. glauca plants inoculated with strains CcI3 and CeD compared to uninoculated plants (Ngom et al., 2016) (figure 2.1).

However, there are also examples illustrating more modest improvements in actinorhizal salinity tolerance when inoculated with Frankiaceae. For example, C. equisetifolia inoculated with Frankiaceae strain CeD increased shoot growth, total biomass and chlorophyll content only in NaCl concentrations lower than 200 mM (Ngom et al., 2016). Likewise, C. glauca Sieb. ex Spreng inoculated with Frankiaceae sp. Thr did not show increased survival in saline conditions compared to uninoculated plants supplied with KNO₃ (Batista‐Santos et al., 2015; Duro et al., 2016). Such differences may arise due to intraspecific variation between Frankiaceae associated with differing actinorhizal species, demonstrating why it is useful to investigate a range of Frankiaceae for differing plants.

Frankiaceae‐actinorhizal tolerance to heavy metals and reduction of soil heavy metals

Actinorhizal plants possess several characteristics that make them suited to bioremediation of heavy metal polluted soils (Table 2), with the ability of A. glutinosa to tolerate high concentration of high metals being well documented (Figure 1). For example, 92% of these plants survived after two growing seasons when planted upon heavy metal contaminated saline river dredgings, with this being 21% higher than other pioneer species such as White Poplar (Populus alba L.) (Mertens et al., 2004).

A further example of such tolerance was observed when growing A. glutinosa and Alnus incana upon copper and lead polluted soil derived from a copper smelter. While under metal stress, leaf dry mass was significantly reduced in both species, however, there was no significant reduction in stem, root and nodule dry biomass or total seedling biomass (except in the case of Krzyż derived A. incana) (Lorenc‐Plucińska et al., 2013). In a similar manner to actinorhizal plant performance, Frankiaceae symbiosis and nitrogen fixation is reduced under heavy metal stress, but not completely inhibited (Bélanger et al., 2015; Lorenc‐Plucińska et al., 2013). This demonstrates that while growth and Frankiaceae symbiosis are somewhat decreased under metal contaminated soils, A. glutinosa and A. incana are still capable of growth under these conditions in association with Frankiaceae. It is probably that such tolerance is not exclusive to Alnus, as many actinorhizal species are found upon metal polluted sites.

Studies such as Wheeler et al. (2001) challenge the notion of actinorhizal plants being metal tolerant, whereas A. glutinosa was negatively impacted by nickel concentrations as low as 0.225 mM. However, it should be noted that this investigation was conducted under hydroponic conditions, with a highly soluble form of nickel (NiSO₄.6H₂O). Thus, this may result in greater nickel bioavailability than is typical within soil, increasing the toxicity of nickel to plants by allowing more nickel to be taken up than usual. This may explain why the findings of this study differ from the majority of the other literature regarding actinorhizal plants metal tolerance.

In addition to their tolerance to heavy metals, actinorhizal plants are well documented as metal excluders (Ghazouani et al., 2020; Jing et al., 2014; Rosselli et al., 2003). Metals taken up by the plants are stored in root tissue of these species, with some metals such as zinc and nickel being sequestered in the nodules specifically (Callender et al., 2016; Lorenc‐Plucińska et al., 2013; Wheeler et al., 2001). This nodule metal sequestration may be due to the production of metallohistins by A. glutinosa which are proteins that sequester a range of metals (zinc, nickel, cobalt, copper, cadmium, and mercury) in the nodule, possibly for use in Frankiaceae metabolism (Gupta et al., 2002; Pawlowski et al., 1997).

Examples of such metal exclusion is evident in multiple studies investigating is A. glutinosa grown upon heavy metal polluted soils. With these plants presenting foliar heavy metal concentrations within the normal range for cadmium, copper, lead, zinc, magnesium, and iron despite the metal polluted conditions (Callender et al., 2016; Desai et al., 2019; Mertens et al., 2004; Vandecasteele et al., 2008). Nevertheless, there is some discrepancy between these studies, as some certain metals show elevated foliar concentrations compared to control plants. However, such differences likely arise from variability in the amounts of metals present and the particular chemistry of the soil at the differing study sties and thus do not detract from the notion of these plants being metal excluders. Also in some instances, the level of foliar translocation of these metals decreased over time, as seen in the case of lead in Desai et al. (2019), likely due to growth in root biomass allowing greater metal storage.

This ability of certain plants to exclude metals from their aerial tissue is desirable for bioremediation, ensuring metals are not taken up into the leaves and recycled back into the soil in high amounts during leaf senescence (Figure 1). Aside from preventing metal accumulation in the soil, this also prevents metals becoming more bioavailable and mobile as they will not be released alongside the metal complexing dissolved organic matter released during leaf litter decomposition (Antoniadis & Alloway, 2002; Kügler et al., 2019; Küsel & Drake, 1998; Mertens et al., 2004; Mohr et al., 2022; Robinson et al., 2000; Shahid et al., 2013; Strobel et al., 2001).

Regarding application to bioremediation of heavy metal polluted sites, A. glutinosa was planted upon the former coal mine in Varteg, Wales. Over the course of 14 years, soil metal levels were significantly reduced compared to unplanted sites, with these reductions ranging from 35 to 52% across a range of metals (Desai et al., 2019). An additional example was the plantation of A. glutinosa and A. incana grown upon copper and lead polluted soil from a copper smelter (Lorenc‐Plucińska et al., 2013). Likewise, Alnus hirsute and Alnus firma have been shown to decrease the concentration of a range of metals when planted at high or low density (Lee et al., 2009). However, it was noted in this study that leaching due to rain contributed to more metals being removed from the soil than the plants, but it is likely that this would decrease over time as larger root systems develop to facilitate metal uptake (Lee et al., 2009).

Frankiaceae is a well‐documented family of heavy metal tolerant bacteria (Table 2). The majority of Frankiaceae species have been described as possessing some degree of resistance to metals including lead, chromium and selenium, with slightly lower resistance reported to silver, cadmium, antimony and nickel, with copper tolerance varying between 2 and 20 mM dependent upon the species (Richards et al., 2002). Tolerance to other metals may depend upon their chemical form, as seen in the case of arsenic, where tolerance to AsO₄ ³⁻ was greater than AsO₂ ¹⁻ (Richards et al., 2002).

Heavy metal tolerance in Frankiaceae may occur through several mechanisms, with chemical modification of the metal to a less toxic form and removal of metals from the cells being the primary methods. This is illustrated by the reduction of selenite to selenium via an NADH dehydrogenase after selenite is exported from the cell (Rehan et al., 2019), the reduction of arsenate to arsenite by a thioredoxin‐dependent phosphotyrosine‐phosphotase (ArsC2) and the presence of a cadmium‐inducible glyoxylase (Furnholm & Tisa, 2014). In addition to reduction, precipitation is used as a means of chemical modification of metals, for example, undecaprenyl phosphatase and DUF347 (unknown function) causes the precipitation of lead on the Frankiaceae cell surface after exportation from the cell by CopA (Furnholm & Tisa, 2014). Additional examples of metal removal from the Frankiaceae cell include the CzcD transporter, which is involved in export of zinc, cadmium, copper, and arsenic (Furnholm & Tisa, 2014) and the P‐type ATPases, CopA, which serves to remove copper (Rehan, Furnholm, et al., 2014) (Figure 1).

Other detoxification mechanisms present within Frankiaceae involve the use of chaperones to ensure metals are correctly transported to the sites of metabolism, export and chemical modification. The cobalt chaperones CobN and CbiX, are examples which serve to protect against the toxic effects of cobalt (Furnholm & Tisa, 2014) (Figure 1).

The above detoxification methods will render the metals having reduced mobility and toxicity, increasing the survivability of the host plant and the Frankiaceae, with metal uptake reducing soil metals over time. Aside from the aforementioned detoxification mechanisms, some Frankiaceae (such as sp. CH37) are capable of producing metallophores which allow selective exclusion of metals (Deicke et al., 2019; Kraemer et al., 2015; Mohr et al., 2021). This is demonstrated by Frankobactin A1, which binds copper with high affinity preventing its entry into the cell and therefore increasing Frankiaceae resilience to copper (Mohr et al., 2021). This has also been observed in other species such as Azotobacter vinelandii, which use a similar metallophore system to selectively take up molybdenum while blocking tungsten uptake (Wichard et al., 2008).

In contrast to the beneficial attributes of Frankiaceae regarding metal pollution, it has been proposed that Frankiaceae nitrogen fixation may lead to a decrease in soil pH, increasing metal mobility (Bolan et al., 1991; Mertens et al., 2004; Pavlů et al., 2021; Tang et al., 1999; Wang et al., 2015). However, over the course of 7 years the pH of soil under an A. glutinosa stand did not substantially decrease in pH compared to the soil beneath the non‐nitrogen fixing Silver Birch (Betula pendula Roth.) and Scots Pine (Pinus sylvestris L) (Kuznetsova et al., 2011). Additionally, this rate of acidification (if occurring at all) would likely decrease over time, as nitrogen fixation would slow as soil nitrogen increases (Mertens et al., 2004; Tang et al., 1999).

As Frankiaceae display the ability to detoxify metals, inoculation of actinorhizal plants with Frankiaceae has been linked to improvements in plant performance upon contaminated sites (Table 2). An example of this is the inoculation of C. equisetifolia with Frankiaceae, which, in comparison to uninoculated plants, led to significantly increased plant height, collar diameter, cladophyll biomass and number of branches, both 3 months and 2 years after planting upon bauxite mines spoil (Karthikeyan et al., 2009). In addition, uninoculated plants showed only 35% survival after 2 years, whereas 90% of the Frankiaceae inoculated plants survived. Likewise, inoculation of A. glutinosa with F. alni resulted in increases in total dry weight, root dry weight and shoot dry weight, compared to uninoculated plant, in soils contaminated with Cu, Ni, Zn, Pb, or Cd (Bélanger et al., 2015).

Despite the benefits conferred by Frankiaceae inoculation, only a small number of studies specifically utilizing Frankiaceae inoculated plants for bioremediation of heavy metal polluted sites have been reported. These are exemplified by the bioremediation of mine tailings from Val‐d'Or, Quebec, Canada using Frankia inoculated A. glutinosa and Alnus crispa (Callender et al., 2016). Here, these Frankiaceae inoculated plants were able to significantly lower nickel, chromium, barium and cobalt in both bulk and rhizospheric soil compared to unplanted soil, with the largest decrease being an 84.8% reduction in chromium in A. glutinosa bulk soil (Callender et al., 2016). It was also noted that the levels of copper and sodium were higher in the rhizospheric and bulk soil of planted sites compared to unplanted sites (Callender et al., 2016).

Frankiaceae‐actinorhizal pollution degradation

A number of differing pollutants are degraded by Frankiaceae (Figure 1) (Table 2), In the case of atrazine, Frankia alni has demonstrated capability to grow upon media containing atrazine as the sole carbon and nitrogen source, indicating an ability to degrade atrazine (Rehan, Kluge, et al., 2014). Atrazine metabolism is proposed to occur due to trzN, atzB, and atzR genes which encode an amidohydrolase, an adenosine aminohydrolase and a LysR‐type transcriptional regulator, respectively (Rehan et al., 2016; Rehan, Kluge, et al., 2014). The TrzN enzyme dechlorinates atrazine by hydrolysing the C‐Cl bond, producing hydroxylatrazine, which is dealkylated by AtzB to produce N‐isopropylammelide (Martinez et al., 2001). This N‐isopropylammelide is then metabolized through four hydrolysis reactions, conducted by AtzCDEF, producing 2CO₂ and 2NH₃ (Martinez et al., 2001). Some Frankiaceae such as Pseudofrankia inefficax EuI1c, contain the trzN‐atzBR and atzCDEF clusters, indicating potential to carry out complete degradation of atrazine. Other species such as F. alni, while capable of atrazine degradation in vitro, lack the atzC and atzD genes (Martinez et al., 2001; Rehan, Kluge, et al., 2014). This indicates that varying degrees of biodegradation of atrazine could be present within Frankiaceae, with some species completely mineralising atrazine and others producing intermediates. Alternatively, this could also suggest that there are potentially unknown genes that could conduct an analogous function to atzCD.

According to genomic studies, in Frankiaceae has the potential to degrade biphenyl compounds such as PCB and dioxins. It is proposed that biphenyls are degraded to 2‐hydroxypenta‐2,4‐dienoate and benzoate using the bphABCD operon through the biphenyl upper meta‐cleavage pathway. The resulting 2‐hydroxypenta‐2,4‐dienoate is then further degraded through the lower pathway of aromatic ring degradation using the genes bphEFG (also known as bphHIJ) to produce acetyl‐CoA (Pieper & Seeger, 2008; Rehan et al., 2016). In Frankiaceae genomes, annotation of the bphA1A2A3BCDEFG and bphBCDEFG gene clusters is relatively frequent, having been previously reported in strains EuI1c, EUN1f, CcI49, G2, and R43 (EuI1c lacks bphA3 and bphH) (Mansour et al., 2017; Pieper & Seeger, 2008; Rehan et al., 2016). This would indicate ability to metabolize these pollutants and reduce the exposure risks associated with them; however, the functionality of these has not yet been confirmed in vitro.

Additionally actinorhizal plants have been shown to effectively grow upon hydrocarbon polluted sites, with species such as A. glutinosa demonstrating a 93% survival rate when planted on open cast oil shale mining areas (Kuznetsova et al., 2010). Furthermore, when planted upon hydrocarbon polluted sites, actinorhizal plants including Alnus viridis, A. incana, and C. equisetifolia have reduced concentrations of soil hydrocarbons (Bissonnette et al., 2014; Kuznetsova et al., 2010; Lefrançois et al., 2010; Sun et al., 2004). These decreases are largely attributed to microbial activity, as Frankiaceae demonstrates the ability to degrade hydrocarbons.

It is proposed that Frankiaceae uses the protocatechuate pathway to degrade the polyaromatic hydrocarbon naphthalene (and derivatives) into acetyl‐CoA and succinyl‐CoA, via ortho‐cleavage (Baker et al., 2015; Rehan et al., 2016). The protocatechuate pathway used in such degradation in Frankiaceae QA3 is composed of eight genes encoding: a protocatechuate 3,4 dioxygenase alpha and beta subunits, fumarate lyase (3‐carboxy‐cis,cis muconate cycloisomerase), 4‐carboxymuconolactone decarboxylase/3‐oxoadipate enol lactonase, 4‐hydroxybenzoate 3‐monooxygenase, oxoacid‐CoA transferase alpha and beta subunits and phthalate 4,5 dioxygenase (Baker et al., 2015; Rehan et al., 2016). This is similar to the gene cluster found in Rhodococcus opacus 1CP and Rhodococcus ruber OA1, which also degrade naphthalene (Eulberg et al., 1998; Li et al., 2016; Rehan et al., 2016). The first five genes of this cluster are also found in Frankiaceae EuI1c and EUN1f, indicating these strains can also potentially degrade naphthalene. Furthermore, Frankiaceae may also be able to degrade alkanes found in hydrocarbon fuels, as both Frankiaceae ACN14a and EAN1pec are known to contain the alkane‐1 monooxygenase (alkB) gene (Rehan et al., 2016). With this gene also present in Rhodococcus Q15 and Rhodococcus NRRL B‐16531, both of which have been shown to utilize alkB to degrade C12‐C16 alkanes (Whyte et al., 2002).

Phenol compounds are another class of pollutants which Frankiaceae may be able to degrade. It is thought phenols are degraded by catechol‐2,3‐dioxygenase, which is used in meta‐degradation of phenols in other bacterial species, including the closely related Rhodococcus (Ali et al., 1998; Arif et al., 2011; Hughes et al., 1984; Rehan et al., 2016). Furthermore, Frankiaceae alters production of host plant root exudates to increase production of phenol compounds, as these may be metabolized by Frankiaceae (Popovici et al., 2011). Conversely phenolic compounds have been shown to inhibit growth of some Frankiaceae, although strains such as AvcI1 were less inhibited by certain phenols such as caffeic acid, with strains EuI1b and Pt410 seeming to display increased growth as the concentrations of o‐hydroxyphenylacetic acids increased (Vogel & Dawson, 1986). This suggests that some Frankiaceae can degrade certain phenols, with this possibly being due to the host actinorhizal plant they associate with, as suggested in Vogel and Dawson (1986). Further investigation is warranted to determine the ability of the Frankiaceae‐actinorhizal symbiosis to degrade phenols, as no major studies have been conducted in this area.

Frankiaceae‐actinorhizal mediated soil fertility improvements and Frankiaceae plant growth promotion abilities

Frankiaceae is a well‐documented family regarding its plant growth promotion ability and capabilities (Table 2). Frankiaceae are noted nitrogen fixing bacteria, with the nitrogenase encoding nif operon found in Frankia, Protofrankia, and Parafrankia, but lacking in Pseudofrankia (Normand et al., 2018; Nouioui, Cortés‐albayay, et al., 2019; Nouioui, Ghodhbane‐Gtari, et al., 2019) (Figure 1). Based upon the result of Koirala and Brözel (2021), Frankiaceae contain a molybdenum‐iron nitrogenase as opposed to a vanadium or iron‐only nitrogenase. Molybdenum within soil is noted as often being a rare element due to its highly soluble nature, however it is noted that tannins, which are produced by actinorhizal plants such as Alnus, can bind molybdenum preventing molybdenum being leached from the soil (Wichard et al., 2009). As noted by Deicke et al. (2019) the Frankiaceae analysed in their work did not seem to contain a molybdenum chelating system to extract tannin bound molybdenum. Thus, Frankiaceae may use an as of yet undiscovered molybdenum chelating agent, rely on host plant molybdenum uptake, or use catechol type siderophores to facilitate molybdenum uptake into the nodules (Arahou et al., 1998; Bellenger et al., 2008; Pourhassan et al., 2015).

Nevertheless, Frankiaceae symbiotic association with actinorhizal plants allows these plants to take up fixed nitrogen, in the form of ammonia, with this added to soil through decomposition of plant material (Izquierdo et al., 2005; Khamzina et al., 2009; Mailly & Margolis, 1992; Parrotta, 1999; Qi et al., 2018; Tarrant et al., 1969; Wall, 2000; Zavitkovski & Newton, 1968). As demonstrated by A. incana leaf litter, this rich material may aid in the decomposition and nutrient release of other, less degradable material from species such as Populus tremuloides, further improving soil fertility (Taylor et al., 1989). This general improvement in decomposition is likely driven by the nitrogen rich actinorhizal leaf litter facilitating faster growth of saprotrophic organisms which break down organic material (Taylor et al., 1989). Alongside nitrogen fixation, Frankiaceae have also been shown to produce ammonia, which can further aid in plant growth as a nitrogen source (Marappa et al., 2020).

In addition to nitrogen fixation, most Frankiaceae genomes are noted as containing enzymes involved in phosphate solubilization, such as alkaline phosphatase and phosphodiesterase/alkaline phosphatase D, allowing for improved phosphorus uptake in host plants (Nouioui, Cortés‐albayay, et al., 2019) (Figure 1). In addition, several other genes possibly potentially involved in phosphate solubilization, such as ptsABCS, phoBHRU, and various phosphate transporters (lat, pho), appear to be present in several Frankiaceae genomes (Nouioui, Cortés‐albayay, et al., 2019). This was corroborated by in vitro investigations that have demonstrated the ability of Frankiaceae to solubilize inorganic phosphates sources in media (Marappa et al., 2020). Unsurprisingly, actinorhizal growth in field conditions has been shown to improve soil phosphate, likely due to association with Frankiaceae (Qi et al., 2018; Uri et al., 2001).

Siderophore production is another useful trait to aid plant growth and improve soil nutrients, as siderophores chelate iron increasing its bioavailability. As such, plants inoculated with siderophore producing bacteria show greater growth than uninoculated plants on soils that are rich or poor in iron (Rungin et al., 2012). Based upon genomic evidence, the majority of Frankiaceae contain siderophore producing clusters, with production of siderophores in some strains confirmed during in vitro screening (Marappa et al., 2020; Nouioui, Cortés‐albayay, et al., 2019) (Figure 1). Nevertheless during in vitro studies, some Frankiaceae do not produce siderophores, despite containing the genes to produce siderophores. This could possibly be due to these strains lacking the necessary substrates or environmental cues to initiate siderophores production, as commonly seen in production of other secondary metabolites such as antibiotics (Sánchez et al., 2010). Thus, siderophore production in these strains should not be discounted entirely and should be investigated further under differing conditions.

As an alternative to siderophores, some Frankiaceae strains such as strains AvsI3 and HFPCpI1, appear to instead secrete oxalic acid to solubilize iron under in vitro conditions (Arahou et al., 1998; Aronson & Boyer, 1994). In comparison to a range of other organic acids tested, oxalic acid has been shown to be highly effective in solubilizing metals, including iron (Ambikadevi & Lalithambika, 2000; Nworie et al., 2017).

Antibiotic production is another useful characteristic of Frankiaceae, potentially aiding in disease suppression to improve plant growth. Frankiaceae strains such as AiPs1 produce the antibiotic Demethyl (C‐11) cezomycin, originally known as Frankiamide (Haansuu et al., 2001; Klika et al., 2001; Klika et al., 2003). Demethyl (C‐11) cezomycin has demonstrated activity against a range of microorganisms, including 14 Gram positive bacteria and six fungal species, with such activity thought to be due to inhibition of cellular calcium flux (Haansuu et al., 2001). Further genome analysis suggests that Frankiaceae ACN14, CcI3 and EAN1pec could exhibit high potential to produce a range of other antibiotic compounds, with 65 biosynthetic clusters identified between these three strains alone, with many of these appearing to encode unique products (Udwary et al., 2011).

Frankiaceae genomes are also rich in genes that potentially encoding lytic enzymes including chitinases, cellulases, endoglucanases, and extracellular endoglucanases (Nouioui, Cortés‐albayay, et al., 2019). The production of such lytic enzymes by Frankiaceae has been demonstrated to confer antimicrobial activity in vitro against fungi and bacteria such as Pseudomonas and Colletotrichum (Marappa et al., 2020; Nouioui, Cortés‐albayay, et al., 2019; Safo‐Sampah & Torrey, 1988). Furthermore, it was reported that C. equisetifolia plants were more resistant to Rhizoctonia solani infection when inoculated with Frankiaceae UMCe12 conferring up to 81.1% overall disease resistance (Gopinathan, 1995).

Frankiaceae may also directly influence host plant growth by the production of phytohormones. The genomes of several Frankiaceae contain the genes encoding anthranilate synthase and aminase component, indole‐3‐glycerol phosphate synthase and anthranilate phosphoribosyltransferase, which allows production of anthranilate, an indole precursor, which is in turn a precursor of indole‐3‐acetic acid (Di et al., 2016; Nouioui, Cortés‐albayay, et al., 2019). However, it is unknown if the Frankiaceae contain any of the other genes in the five bacterial pathways that allow synthesis of indole‐3‐acetic acid (Di et al., 2016). Despite this lack of genome analysis, species such as Frankiaceae sp. DDNSF‐01, Frankia casuarinae, Frankiaceae sp. AvcI1 and Parafrankia discariae BCU110501 have demonstrated ability to synthesize indole‐3‐acetic acid under in vitro conditions (Marappa et al., 2020; Solans et al., 2011; Wheeler et al., 1984).

When the composition of indole compounds produced by Frankiaceae HFPArI3 were analysed, the most abundant were indole‐3‐ethanol (25.4 ng ml⁻¹) and indole‐3‐acetic acid (8.4 ng ml⁻¹) (Berry et al., 1989). However, these levels were only achieved when exogenous tryptophan was supplied (50 μM). The need for exogenous tryptophan may be indicative of these processes only being upregulated under specific conditions which may be provided by a plant host. In terms of function, increasing concentration of indole‐3‐ethanol applied to the root zone have been found to increase the number of lateral roots produced (Berry et al., 1989). Indicating this compound, despite not being the “classical auxin” can still improve plant growth. Interestingly Berry et al. (1989) observed that the ethyl acetate partitioning step used in analysis reduced indole‐3‐ethanol levels, suggesting this may be a more common constituent of indole compounds produced by PGP bacteria than initially noted.

Aside from indole compounds, Frankiaceae genomes are also characterized by the presence of a conserved biosynthetic cluster containing 11 genes related to the synthesis of cytokinins (Nouioui, Cortés‐albayay, et al., 2019). Production of cytokinins was observed by Stevens and Berry (1988), with Frankiaceae sp. HFPArI3 producing N6‐(Δ2‐isopentenyl) adenosine in vitro, with this likely being converted to zeatin cytokinins. Due to contaminating media components, this could not be definitively inferred from the analysis undertaken, it is possible that other cytokinins such as trans‐zeatin riboside and cis‐zeatin riboside were instead produced. More recent study of P. discariae BCU110501 has been reported to produce zeatin in vitro, supporting the notion of zeatin cytokinins being produced by Frankiaceae (Solans et al., 2011).

Frankiaceae genomes also present 1‐aminocyclopropane‐1‐carboxylic acid (ACC) deaminase encoding genes in all species except F. casuarinae strains (Nouioui, Cortés‐albayay, et al., 2019). ACC deaminase is a highly desirable enzyme, as its presence allows the bacteria to reduce ethylene levels in the plant, promoting growth and resistance to a range of abiotic stressors including drought, salinity and metal pollution (Glick, 2014). However, the functionality of this enzyme within Frankiaceae, has not been investigated in planta and in vitro, so it cannot be confirmed definitively if this enzyme promotes the actinorhizal growth. Finally, Frankiaceae may also be able to produce gibberellic acid, with this being evident in P. discariae BCU110501 under in vitro conditions (Solans et al., 2011). Likewise, this also deserves further investigation to determine both the nature of the gibberellin and the ability of this to improve actinorhizal growth.

FUTURE DIRECTIONS

Deeper insights into Frankiaceae and optimization of an endophyte inoculation system

Frankiaceae‐actinorhizal symbiosis offers a fertile area for future study, particularly regarding the bioremediation potential of this association. Work to select optimal Frankiaceae isolates for each bioremediation task would be useful, as different Frankiaceae vary in terms of their ability to degrade pollutants and their tolerance to salinity and metals (Oshone et al., 2013; Oshone et al., 2017; Richards et al., 2002; Tani & Sasakawa, 2003). In addition, other endophytes able to cooperatively aid in bioremediation should be studied alongside Frankiaceae, as these could be used to further bolster their performance. Examples of this include inoculation of A. glutinosa with Glomus intraradices and Frankiaceae, which lead to improved plant performance in a number of metrics than either isolate alone (Oliveira et al., 2005). This has also been demonstrated as co‐inoculation of Alnus cordata with Frankiaceae and either Glomus mosseae or Glomus fasciculatum produced significantly greater plant growth (after 1 year on mine spoils), than either inocula in isolation (Lumini et al., 1994).

After a desirable Frankiaceae isolate (possibly alongside other endophytes) has been found, it may be possible to increase the ability of that strain to infect the host through use of a carrier system. It has been demonstrated that nodulation of the host plant can be relatively low, with Markham (2005) demonstrating that only 45.9% of A. incana developed root nodules after 6 weeks when inoculated with Frankiaceae. Carrier systems have been used to enable easier handling of the bioinoculant, promoting their long‐term storage and effectiveness. An example is the encapsulation of Frankiaceae in alginate, which has been shown to be effective even after storage at room temperature for 2 years (Frioni et al., 1994). Plants inoculated with alginate‐entrapped Frankiaceae showed the expected improvements in nodulation and growth compared to uninoculated plants, demonstrating that the carrier system did not impede the plant growth promoting abilities of Frankiaceae (Sougoufara et al., 1989). Further exploration of more sophisticated formulations and application methods, adapted to the actinorhizal‐Frankiaceae symbiosis is necessary to maximally utilize this symbiosis at large scale.

Frankiaceae genetic modification

Aside from studying naturally occurring Frankiaceae, future research efforts could explore genetic modification of a desirable Frankiaceae species or strain to further improve its bioremediative capabilities. As of yet, stable transformation of Frankiaceae has proved difficult, one reason being that Frankiaceae have no identified phages which could be utilized their genetic modification. It is likely that research in this area could prove fruitful as the genomes of Frankiaceae are known to contain integrated phages (Normand et al., 2007). However, as Frankiaceae do not form lawns very well on solid media, this could prove a barrier to locate phage plaques (Normand & Lalonde, 1986).

Actinobacteria also have low rates of homologous recombination, most likely due to use of non‐homologous end joining instead of homologous recombination in DNA repair. However, as demonstrated by Zhang, Chen, et al. (2012) in Streptomyces avermitilis, when the bacterial homologues of ku70 and ku80 are successfully knocked out, the use of non‐homologous end joining is inhibited. This knowledge could potentially be applied to Frankiaceae to increase the success of genetic transformation.

An additional factor that may pose a barrier to Frankiaceae transformation is the presence of restriction enzymes which may degrade introduced plasmids. This degradation is due to plasmids often being methylated by the enzyme Dam after replication (Barras & Marinus, 1989). However, some actinobacteria possess a system which cleaves methylated DNA using type IV restriction enzymes (Janulaitis et al., 1992; Kieser & Hopwood, 1991; MacNeil, 1988; Sutherland et al., 1992). Furthermore, some actinobacteria methylate DNA in a different pattern to bacteria using the Dam enzyme, also possessing a DNA mismatch repair pathway that lacks the Dam associated DNA repair enzymes, MutS and MutL, with this mismatch repair pathway being more akin to those found in Archaea (Castañeda‐García et al., 2017; Ishino et al., 2016).

Based upon the aforementioned, it is possible that use of unmethylated plasmids could increase transformation efficiency in Frankiaceae. This has been previously demonstrated in both Corynebacteria and Streptomyces, where unmethylated plasmids are not cleaved by the type IV restriction enzymes (Ankri et al., 1996; Kieser & Hopwood, 1991). Gifford et al. (2019) transformed F. alni using the unmethylated plasmid pIGSAF, with this stably maintained in F. alni for up to 3 weeks in the absence of any selective pressure. Similarly, Pesce et al. (2019) stably transformed multiple Frankiaceae strains using a conjugation system, inserting a putative salt tolerance gene from the salt tolerant Frankiaceae sp. CcI6 into the salt sensitive F. casuarinae CcI3, increasing its performance under saline conditions. The method utilized by Pesce et al. (2019) involved conjugation with a methylation positive Escherichia coli, where it is likely that use of a methylation negative E. coli would have further increased transformation efficiency (Flett et al., 1997; Gifford et al., 2019; Stein et al., 1988).

Another method of Frankiaceae transformation that warrants further exploration is the use of Frankiaceae derived plasmids. Such plasmids would most likely allow for increased rates of stable transformation, due to these plasmids being native to Frankiaceae. An analogous example demonstrating this is the transformation of Frankiaceae strain EuI1c using genomic DNA from Frankiaceae strain EAN1pec. This introduced antibiotic resistance in Frankiaceae strain EuI1c, that was maintained over several generations, even in the absence of selective pressure (Myers & Tisa, 2003). Thus, demonstrating transformation of Frankiaceae with Frankiaceae derived DNA allows for effective and stable transformation. Several Frankiaceae plasmids have been studied such as pFQ31 and pFQ11, which could serve to allow stable insertion of desirable genes into Frankiaceae (Lavire et al., 2001; Normand et al., 1985; Normand & Lalonde, 1986; Xu et al., 2002).

Overall, the above research represents strides in Frankiaceae genetic modification which may allow for future modification of these beneficial bacteria. Based upon this review, particularly worthy targets of such modification include increasing Frankiaceae growth rate, permit growth in a wider array of media, increase tolerance to abiotic stressors and introduce novel pollutant degradation pathways.

Increased application of Frankiaceae‐actinorhizal symbiosis in other sectors

Aside from its use in bioremediation, it may also be possible to utilize the actinorhizal‐Frankiaceae symbiosis in silvopasture and agroforestry systems to sustainably improve agricultural productivity while reducing the environmental impact of farming. In terms of silvopasture, application of the actinorhizal‐Frankiaceae symbiosis would result in reduction in pollutants and salinity, providing an increase in soil nutrients and potentially high quality, high yielding grazing material. In addition, the improved growth of the grazing material may mean less time is required for each pasture to lay fallow between grazing, improving land use efficiency. When the trees are of a mature height they can be cut down and replanted or coppiced for timber to promote new growth, with good rates of carbon sequestration expected as the trees produce more biomass (Jose & Dollinger, 2019).

Regarding agroforestry, the actinorhizal‐Frankiaceae symbiosis could be applied to cropping systems such as shade cropping, taungya cropping, and alley cropping. In shade cropping, crops are grown under the tree canopy, protecting the soil from rain erosion and increasing nutrient input through leaf litter. However, in such a cropping system only shade‐tolerant crops could be utilized due to the tree canopy excluding light (Muschler, 2001). In contrast, taungya cropping involves the planting of crops as normal with the desired trees positioned among the crops. These trees are allowed to develop to maturity for timber harvesting, with increasingly shade‐tolerant crops being planted as the canopy develops (Menzies, 1988). Similarly to shade cropping, crops under taungya cropping benefit from increase nutrient input, with the soil protected from rain erosion. Finally, alley cropping involves planting alternate rows of crops and trees or hedgerows, the latter usually being cut prior to planting the former, to reduce shading, with this cut woody material used as organic fertilizer (Akinnifesi et al., 2006). In this case, the actinorhizal trees would act as windbreaks reducing wind erosion, protecting crops from lodging. Where pollution is a concern, the addition of actinorhizal plants to these cropping systems would also enable the removal of contaminants and salts from the soil, thereby increasing crop performance and reducing the risks health associated with produce grown upon polluted sites.

Actinorhizal plants could also be planted upon marginalized land to meet tree planting goals for carbon sequestration. Therefore, freeing land that has been used for these purposes which could instead be utilized for agricultural purposes. Finally, actinorhizal plants, especially those in the Casuarina genus which have previously been used in stabilization of coastal regions, could be utilized to protect agricultural areas from coastal wind and water‐based erosion (De Zoysa, 2008; Jayasingam, 2014). Based upon the tolerance of actinorhizal plants to salinity and waterlogging, it is possible that they could also be generally well suited for plantation in riparian and coastal areas, mitigating degradation and erosion pressures common in such areas, while also serving to reduce flooding risks for farmland and coastal/rural communities.

CONCLUSION

Increasing soil degradation is a global issue that can severely impact food production and human health. The Frankiaceae‐actinorhizal symbiosis may be an attractive method to address such soil degradation, as this symbiosis has the potential to reduce soil pollutants and salinity while increasing soil fertility, with low management requirements and minimal economic costs for establishment and maintenance.

Overall, this system should be further studied and optimized to develop inoculants better suited to the context of each bioremediation case than the endemic bacteria of the actinorhizal plant. Additionally, co‐inoculants and carrier systems could be developed to improve bioremediation efficacy and host plant inoculation. Genetic modification of Frankiaceae inoculants is an attractive long‐term goal for this symbiosis, as it could allow for levels of bioremediation far surpassing those found in nature.

Beyond improvements to the symbiosis, it should be leveraged more widely, not only for bioremediation of polluted sites but also for shade cropping, Taungya cropping, alley cropping, silvopasture, coastal stabilization, and flood defence. For these and many other applications, this symbiosis currently remains under‐exploited.

AUTHOR CONTRIBUTIONS

Ryan Michael Thompson: Conceptualization; writing – review and editing; writing – original draft; project administration; funding acquisition. David George: Writing – review and editing; supervision; funding acquisition. Maria del Carmen Montero‐Calasanz: Writing – review and editing; supervision; funding acquisition.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

Thompson, R.M. , George, D. & del Carmen Montero‐Calasanz, M. (2024) Actinorhizal plants and Frankiaceae : The overlooked future of phytoremediation. Environmental Microbiology Reports, 16(6), e70033. Available from: 10.1111/1758-2229.70033

DATA AVAILABILITY STATEMENT

Data sharing is not applicable to this article as no new data were created or analyzed in this study.