Plant-Endophyte Partnerships to Assist Petroleum Hydrocarbon Remediation
Petroleum hydrocarbons (PHs) are the most widespread organic contaminants in soil and groundwater worldwide. The financial, environmental, and health impacts of the contaminants are considerable. Regulations require the remediation of contaminated sites and encourage the use of biological methods such as phytoremediation, whereby plants and their associated microorganisms are used. Whilst being cheap and sustainable, there are few elements like the efficiency and predictability that has retarded its implementation into commercial-wide applications. Endophytic bacteria living inside plant tissues are fundamental to plant health and many of them can break-down contaminants taken-up by the host, thereby reducing phytotoxicity. Sequencing of genomic DNA of pure strains and endophytic microbial communities provides critical new opportunities for phytoremediation applications to restore PH contaminated soils. In this chapter, we discuss on a number of beneficial effects of plant-endophyte partnerships, and highlight how new insights from genomics and metagenomics can assist soil remediation to enhance plant growth and ecosystem services of reclamated soil.
The industrialization of modern societies and the ever-increasing demand for energy generation for heat, our transportation networks, and electricity generation has resulted in the extensive exploitation of petroleum hydrocarbons , which are the most widespread class of organic contaminants worldwide (Kästner 2008). Underground leaking of storage tanks are still one of the major sources of toxic hydrocarbon spills. For example, about 435,000 underground tanks with petroleum hydrocarbons have leaked in the United States (Kansas State University, 2005) and nearly 1.5 million underground storage tanks have been closed since 1984, but a large number of sites are still in need of remediation. In Europe, 60% of soil contamination contains mineral oil and trace elements (Panagos et al. 2013). The negative impacts of PH pollution on human health and the environment are of concern for the scientific community, policy makers and the public (Tang et al. 2011). Prolonged exposure to PHs can result in respiratory system disfunction, central nervous system disruption, and increase in the probability of lung, skin, bladder, liver and kidney cancers (Aguilera et al. 2010; Rodríguez-Trigo et al. 2010). Often PH pollution is left untreated, however restored lands are increasingly valuable for the production of edible crops, biofuels and fibers. Generally, conventional physical and chemical in situ and ex situ clean–up technologies for petroleum hydrocarbon remediation such as excavation, air sparging, removal and off-site treatment in biopiles, slurry- and solid phase reactors amongst others (Kulik et al. 2006; Moldes et al. 2011) are environmental unfriendly and invasive, often only result in incomplete removal of the pollutants, and are frequently expensive. In the next decades it is estimated that an average 125,000 dollars will be spent to treat each PH contaminated site by conventional techniques and costs can be upwards to billions when groundwater is polluted (Panagos et al. 2013).
Consequently, a significant amount of research is now focusing on alternative technologies to supplement and/or replace traditional approaches (Gerhardt et al. 2009; Cook and Hesterberg 2013). While the synergistic action of plants and their related microorganisms to remove and degrade petroleum compounds is considered to be advantageous in terms of cost, due to low capital expenditure, sustainability, and flexibility for in situ implementation, there are still numerous aspects about the mechanisms involved and the elements of uncertainty and efficiency that remain the subject of research and debate among members of the scientific community.
Endophytic bacteria living inside plant tissues have attracted major interest for increasing phytoremedation efficiency and predictability (Weyens et al. 2009a; Rylott 2014; Zhu et al. 2014; Ijaz et al. 2015). One of the main reasons is their lifestyle, endophytic bacteria dwelling the internal tissues of plants (roots, stems, leaves) overcome the competition for nutrients and space, and are physically protected from unfavorable environmental conditions which may enhance their survival and catabolic gene expression (Schulz and Boyle 2006). Moreover a number of reports have shown that endophytic bacteria have a better capacity to enhance PH phytoremediation than rhizosphere or soil bacteria (Barac et al. 2004; Andria et al. 2009; Weyens et al. 2010; Yousaf et al. 2011).
The purpose of this chapter is to give a detailed overview of the use endophytes in PH phytoremediation. We will focus on the ecology of endophytic bacteria, the mechanisms of colonisation, and diverse aspects of plant growth promotion (PGP) mechanisms. Furthermore, we will discuss their catabolic behaviour, and examples of endophyte-stimulated phytoremediation. Throughout the discussions we show how developments in genomics has been instrumental for understanding the plant and microbial catabolic mechanisms, directing management and monitoring of phytoremediation of petroleum hydrocarbon sites. Finally, we describe the future applications of endophytes and genomics, and the necessary steps to integrate the data into comprehensive management of PH contaminated sites.
2 Ecology of Endophytic Bacteria
For a long time endophytes were largely ignored or considered contaminants, now it is widely recognised that many endophytes are important plant symbionts without imposing visible sign of infection or negative effect on their host (De Bary 1866; Schulz and Boyle 2006; Reinhold-Hurek and Hurek 2011). After the first definition by De Bary (1866), endophytes were defined as “all organisms occurring within plant tissues,” though later, various researchers used different definitions for endophytes depending on the research context and need (Kogel et al. 2006; Rosenblueth and Martinez-Romero 2006; Schulz and Boyle 2006). Since endophytes can proliferate inside the plant tissue, they can interact very closely with their host, face less competition for nutrients, and are better protected from adverse changes in the environment when compared to bacteria in the rhizosphere and phyllosphere (Andria et al. 2009). These elements increased the attractiveness of using endophytes in many phytoremediation studies (Weyens et al. 2009a, b; Rylott 2014; Ijaz et al. 2015) with often great successes (Barac et al. 2004). Evidence of the occurrence of endophytes has come from culture-dependent and culture-independent analyses, next to fluorescence in situ hybridisation-confocal laser scanning microscopy (FISH-CLSM). In the following paragraphs we give an overview of the diversity and ecology of endophytes as assessed by the culture-dependent and independent techniques.
Hyperaccumulating alpine pennycress (Thlaspi caerulescens) (Lodewyckx et al. 2002a), tall fescue (Malinowski et al. 2000), Arabidopsis seeds (Truyens et al. 2015a, b; 2016) and different grass species (Dalton et al. 2004; Thijs et al. 2014a) to woody tree species such as oak and ash (Weyens et al. 2009c), sycamore (Thijs et al. 2014b), poplar (Porteous Moore et al. 2006; Van der Lelie et al. 2009), Mimosa pudica (Pandey et al. 2005), pine seeds (Cankar et al. 2005), and other forest trees as reviewed (Pirttilä and Frank 2011). These reports illustrate that endophytic bacteria show a tremendously high diversity not only in plant hosts but also in bacterial taxa. Detailed information for host species with their associated endophytic bacterial diversity is available in earlier reviews (Lodewyckx et al. 2002b; Reinhold-Hurek and Hurek 2011; Hardoim et al. 2015). Most isolates belong to the Proteobacteria, Firmicutes, Actinobacteria and Bacteroidetes phyla of the domain bacteria (Bulgarelli et al. 2013). For example, the cultivable endophytic bacterial community associated with English oak was dominated by Actinobacteria (65.1%), with Frigobacterium spp. (45.0%) and Okibacterium spp. (13.0%) forming the majority (Weyens et al. 2009c). Proteobacteria represented 23.1% and were dominated by gamma-Proteobacteria (17.9%) including Pseudomonas spp. (9%), Xanthomonas spp. (4.6%), Enterobacter spp. (3.4%) and Erwinia (0.8%). The remaining part of the endophytes associated with English oak were Firmicutes (11.8%) with 8.8% Bacillaceae and 3.0% Paenibacillaceae (Weyens et al. 2009c). The possible application of these cultivable plant-associated bacteria for improving phytoremediation will be discussed below.
2.1 Critical Factors in the Isolation of Endophytic Bacteria
The isolation procedure is a critical and essential step when investigating endophytic bacterial communities . Commonly used procedures for isolation of endophytes combine surface sterilization of the plant part with subsequently either maceration of the plant tissue and plating onto nutrient agar, or plating of small surface sterilized plant parts onto nutrient agar (Eevers et al. 2015). In general, surface sterilization consists of: (1) thorough washing of the plant tissue, (2) surface sterilization, (3) several aseptic rinses and (4) a sterility check. Considering the high variety in procedures of isolation, comparison between different studies should be made carefully. In our laboratory, sterilization procedures were optimized for the isolation of endophytic bacteria from different parts of poplar (Porteous Moore et al. 2006; Beckers et al. 2016), willow (Weyens et al. 2012), yellow lupine (Weyens et al. 2014), alpine pennycress (Thlaspi caerulescens) (Lodewyckx et al. 2002a), rapeseed (Croes et al. 2013), tobacco (Mastretta et al. 2009) and Arabidopsis (Truyens et al. 2015a). In addition to the sterilization procedure, another very critical step in the isolation procedure is the choice of the growth medium; it directly affects the number and type of strains that can be isolated. Different growth media have been developed and improved over time (Bacon and White 1994; Schulz et al. 2006; Govindasamy et al. 2014). The addition of plant extract to the medium can improve the culturability of endophytes and regrowth of stored endophytes (Eevers et al. 2015). In adition, substantial progress has been made in culturing new groups of organisms through co-culture, and in situ incubation using diffusion systems such as the iChip (Nichols et al. 2010), see for a review here (Stewart 2012).
However, even after optimization of these critical steps in the isolation procedure, cultivation-dependent techniques still strongly underestimate the bacterial numbers, as they do not record the majority of viable but non-cultivable bacteria. Cultivation-independent methods for exploring microbial communities in natural habitats have suggested that cultivable isolates represent less than 1–10% of the bacterial taxa present in a plant (Torsvik et al. 2002). Therefore, in order to obtain a more complete insight in the composition of plant associated microbial communities, cultivation-independent methods based on rRNA gene identification or total environmental DNA or RNA sequencing should be employed. It must be mentioned that we also observed many times that even after their initial isolation and growth in culture medium, many endophytic strains were difficult or could not be further propagated under laboratory conditions, supporting the importance of the analysis of their communities with molecular cultivation-independent techniques. Moreover, mining the yet-to-be-cultured majority provides access to an immense reservoir of unexploited microbial diversity by cultivation.
2.2 Endophytic Communities Identified by Culture-Independent Tools
Insights on the total microbial community level have been obtained using culture-independent 16S rRNA-based methods such as denaturing gradient gel electrophoresis (DGGE), terminal restriction fragment length polymorphism (TRFLP) and 16S rRNA clone libraries, which provided a more detailed overview of the in situ communities than culture-based studies (Su et al. 2012). Since the advent of next-generation sequencing platforms, culture-independent molecular methods have greatly improved to provide high-resolution microbial community profiles (Sloan and Lebeis 2015). Using these molecular sequencing techniques such as pyrosequencing, Illumina and Ion Torrent, the first examples of a high-resolution analysis of endophytic root microbiomes were obtained of the model plant Arabidopsis (Bulgarelli et al. 2012; Lundberg et al. 2012; Bodenhausen et al. 2013; Schlaeppi et al. 2014), and also rice (Sessitsch et al. 2012), willow (Yergeau et al. 2014), poplar (Beckers et al. 2016), and Acer pseudoplatanus (Thijs et al. submitted). These studies showed that soil type was a main determinant of the root endophytic community composition by acting as a “seed bank” of microorganisms that can be recruited into the root-associated community. Despite major differences in soil and microbial community compositions, it has been shown that root endophytic microbiomes are composed of a core set of microbial taxa. Bulgarelli et al. 2013 suggested a two-selection step process to explain this. First, plant rhizodeposits mediate a substrate-driven community shift in the rhizosphere, and second, the host-genotype innate immune system fine-tunes the microbial profile in the selection of root endophyte assemblages (Bulgarelli et al. 2013). Root-endophyte microbial communities generally display an order-of-magnitude reduction in species richness compared to rhizosphere and bulk soil communities and numbers progressively decrease towards the aerial parts (stem, leaves, flowers, seeds) (Bulgarelli et al. 2013; Afzal et al. 2014). Besides soil and air for leaf endophytes, other environmental factors may affect the composition of endophytic communities such as climate variables and interactions with other organisms (Sloan and Lebeis 2015). For leaf endophytic communities for example, the surrounding air and environmental growth conditions were considered important determinants next to host genotype (Kembel et al. 2014).
In most studies, a common trend is that the proportion of Proteobacteria and Actinobacteria are typically increased in the root-endosphere of plants compared to the bulk soil, while Acidobacteria are frequently underrepresented (Bulgarelli et al. 2012; Sessitsch et al. 2012; Beckers et al. 2016). Many Proteobacteria have been characterised for beneficial plant-growth promoting features, while Actinobacteria are famous for their antibiotic producing capacity. Therefore it is not unlikely that some intrinsice plant-microbe signalling can favour the prevalence of these groups. Even between plant species, significant overlap between the core set of microbial taxa have been reported, for example between sugarcane and Arabidopsis root endophytes, and common ivy phyllosphere and Arabidopsis phyllosphere (Stevens et al., submitted). This raised the interesting hypothesis that a number of bacterial families have had long association with plants (Schlaeppi et al. 2014). This warrants a larger survey of edophytic microbial communities from a wider range of plant species growing in different soil types and environments to gain novel insights in the ecology of microbial endophytes.
As most endophytes derive from the rhizosphere, root-endophytic communities are the most diverse, and contain highest densities in the plant. Root colonization by rhizosphere bacteria occurs along the root axis and is a dynamic process with highest cell densities at root hairs and root tips (Fan et al. 2012). Root colonisation can be considered to involve four steps (Compant et al. 2010). The initial step consists of bacteria moving to the plant roots. This movement can either be passive via soil water fluxes, or active, via specific induction of flagellar activity by plant-released compounds (chemotaxis). In the second step, non-specific adsorption of bacteria to the roots is occurring with subsequent anchoring (third step), which represents the firm attachment of the bacterial cell to the root surface. Very often, bacteria form dense biofilms on the root surface and this happens more under stress conditions such as contamination and low nutrient levels (Ramey et al. 2004). Ultimately, a subset of the rhizosphere microbiota can enter the root tissues and establish as endophytes. Bulgarelli et al. (2013) listed different root exudate components that are involved in these processes. Also border-like cell derived proteoglycans have been attributed to be involved in the attachment of bacterial cells, in particular Rhizobium, to root cells of non-leguminous plants (Santaella et al. 2008).
Since many facultative endophytic bacteria are also surviving in the rhizosphere, it is clear that there exists a close relationship between endophytes and bacteria colonizing the rhizosphere. The root is the primary site where endophytes gain entry into plants, with the exception of bacteria transmitted through the seeds, which are already present in the embryo at the time of germination (Porteous Moore et al. 2006; Truyens et al. 2015b). Bacterial entry into plants mainly occurs at locations of epidermal damage, that arise as a result of normal growth of the plant like formation of laterals, or through root hairs and at epidermal conjunctions (Compant et al. 2010). Furthermore, exudates leaking through these wounds are a source of nutrients for the colonizing bacteria and hence create favourable conditions. Several microscopic studies confirmed this route of colonization (Watt et al. 2006; Lagendijk et al. 2010; Fan et al. 2012). This was further supported by the analysis of the genome of Enterobacter sp. 638, a plant growth promoting endophyte from poplar (Taghavi et al. 2010) whose genome was sequenced by the JGI. This analysis revealed the presence of several gene clusters important for cell mobility including four operons for flagellar biosynthesis (FlgNMABCDEFGHIJKL, flhEAB fimA yraIJ lpfD cheZYBR tap tar csuEDCAB int cheWA motBA flhCD, fliYZA fliCDSTEFGHJKLMNOPQR and fliEFHIJKLMNOPQR). The Enterobacter sp. 638 genome further contained a number of genes associated with agglutination and cell adhesion, similar to those found in both animal and plant pathogens (Taghavi et al. 2009; Taghavi et al. 2010). Many of these genes were not found in E. coli K12, and are hypothezised to be important for plant colonization. Wounds and lateral root formation, however, are not absolutely required for the entrance of endophytic bacteria in their host plant. For instance, vector organisms (e.g., Saccharicoccus sacchari, arbuscular mycorrhizae, and insects) are possible candidates to assist potential endophytes to gain entrance to the apoplastic spaces and to colonize the host plant (Frey-Klett et al. 2011).
Several studies reported increased cellulase and pectinase activities during colonization of endophytes suggesting that active penetration is also an option (McCully 2001; Rosenblueth and Martinez-Romero 2006). Although Enterobacter sp. 638 was not able to grow on pectin (poly(1,4-d-galacturonate)) as a sole carbon source, its genome carries the genes involved in pectate degradation, a demethylated backbone of pectin (Taghavi et al. 2010). In addition, other regions on the genome of the same strain encode for carbohydrate uptake and metabolism. However, cellulose hydrolases were not found on the genome of Enterobacter sp. 638 (Taghavi et al. 2010).
Except for particular cases such as Azoarcus spp. and Rhizobium in grasses, or Alcaligenes faecalis in rice, endophytic bacteria primarily colonize intercellularly (Hardoim et al. 2008). Highly specific adaptations are those of endosymbionts like rhizobia which live intracellularly in nodule cells of leguminous plants (Murray 2011). To better understand the modes and sites of entry of endophytic bacteria, gfp and egfp marked strains were inoculated in various plant species and their colonization was investigated by means of confocal microscopy (Bloemberg et al. 2000; Germaine et al. 2004; Gilbertson et al. 2007; Cardinale 2014).
Besides plant roots as entry for many endophytes, entry may also occur through the seeds and through natural holes in the phyllosphere such as leaf stomata and lenticels (pores for gas exchange in stems) (Kluepfel 1993), hydrathodes (water pores), and nectarthodes (opening in nectary glands of blossoms). A general model of leaf colonization by phyllobacteria consisting of 8 important steps was presented by Beattie and Lindow (1999): (1) bacterial immigration, (2) habitat modification, (3) bacterial division, (4) microcolony formation, (5) formation of large aggregates, (6) entry into internal spaces, (7) habitat modification and bacterial division, and (8) egression onto the leaf surface (8). However, they acknowledged that these steps will vary in function of different bacterial species (Beattie and Lindow 1999). Recently, a bacterial bioreporter for fructose and sucrose used to quantify the availability of nutrients to individual cells in the phyllosphere, revealed that growth of Erwinia herbicola occurred at the expense of sugars but that there was a highly heterogenous availability of fructose to individual cells (Leveau and Lindow 2001).
3 Beneficial Effects of Plant-Microbe Partnerships
As reviewed by Lugtenberg et al. (2002), Ryu et al. (2005); Glick (2012), Bloemberg and Lugtenberg (2001), plant growth promotion has been intensively investigated for plant growth promoting rhizobacteria (PGPR). We already mentioned before that there exists a close relationship between rhizosphere and endophytic bacteria suggesting that, presumably, they all use similar mechanisms to benefit their host plant. Many mechanisms of plant growth promotion by plant-associated bacteria have been described, and were broadly grouped into two categories: direct and indirect (Spaepen et al. 2009). Many efforts have been made to elucidate both the direct and indirect mechanisms by which plant-associated bacteria improve plant growth (Lugtenberg and Kamilova 2009; Roca et al. 2013; Farrar et al. 2014). Mechanisms for direct plant growth promotion may involve nitrogen fixation especially diazotrophs, the supply of less available nutrients such as phosphorus and other essential nutrients (Lugtenberg and Kamilova 2009; Drogue et al. 2012), the production of plant growth regulators such as auxins, cytokinins and gibberellines (Somers et al. 2004; Gray and Smith 2005), and the inhibition of plant ethylene production due to bacterial 1-aminocyclopropane-1-carboxylate (ACC) deaminase activity (Glick 2005). Indirect mechanisms for plant growth promotion are related mainly to the suppression of pathogenic and deleterious microorganisms through competition for space and nutrients, antibiosis, production of hydrolytic enzymes, inhibition of pathogen-produced enzymes or toxins, and induction of plant defence mechanisms (Compant et al. 2005; Couillerot et al. 2009; Zamioudis and Pieterse 2012). All these mechanisms are attributed more to plant health than direct growth promotion.
Recently, genomic insights of plant growth promoting endophytes (PGPE) has increased our understanding of the molecular mechanisms of plant growth promotion (Bruto et al. 2014). For example, based on a genome sequence analysis of 304 Proteobacteria, it was shown that the number of genes contributing to plant-beneficial functions increased along the continuum from animal pathogens, phytopathogens, saprophytes, endophytes/symbionts to PGPR, suggesting that the accumulation of these PGP-genes might be an intrinsic PGPR feature (Bruto et al. 2014). The best-studied examples of PGPR belong to diverse genera and include, amongst others, Azospirillum, Bacillus, Burkholderia, Enterobacter, Klebsiella, Paenibacillus, Pseudomonas and Rhizobium. A variety of PGPR and PGPE have also been used as inoculants for the remediation of contaminants (Zhuang et al. 2007). In this respect, it is important to understand the mechanisms of plant growth promotion to predict how bacteria interact with plants and whether they can establish themselves in the plant environment after inoculation in the field.
In general, free living bacteria usually do not show one single mechanism of plant growth promotion, but they may involve a combination of the individual mechanisms. In addition, synergistic interactions between PGPR and other microbes such as mycorrhizal fungi have been reported (Artursson et al. 2006). Besides understanding the molecular mecchanisms of PGP of single individuals, it is important to also comprehend the environmental and physiological factors regulating the biosynthesis of growth promoting and antimicrobial compounds produced by plant-associated bacteria. This is an essential step towards improving the level and reliability of their growth promoting activity. Below, some important PGP mechanisms and individual contribution to plant growth promotion are discussed more in detail.
3.1 Direct Promotion of Plant Growth
Although 78% of the earth’s atmosphere is nitrogen, this element often is a limiting factor for plant growth. This because of the fact that atmospheric nitrogen exists as dinitrogen (N2), a form of the element that is inaccessible to all except a few specially adapted prokaryotic organisms including some eubacteria, cyanobacteria and actinomycetes. For all other organisms (also plants), nitrogen should occur under the form of ammonia or nitrate before it can be incorporated into organic molecules. Indeed, higher plants cannot carry out this process in the absence of associated diazotrophic bacteria. These bacteria are equipped with the enzyme nitrogenase, an O2-sensitive enzyme catalyzing the reduction of atmospheric nitrogen to ammonia. Diazotrophic bacteria can be grouped into those that involve close associations with plants, often called symbioses and those that involve only loose associations, often called associative interactions (De Bruijn 2015).
Symbiotic diazotrophic bacteria show highly specific, intimate interactions leading to the induction of new organs or organ-like structures in the host. The intimacy of this type of association maximizes the transfer of fixed nitrogen to the host plant. The plant supplies the prokaryotic symbiont with energy-rich compounds, which is necessary to support the high energy demands of nitrogen fixation. The plant further provides the prokaryote a protected environment. Examples of symbiotic diazotrophs are (1) legume symbionts, which are Gram negative bacteria that form nodules on roots of leguminous plants, (2) members of the genus Frankia, which are Gram positive bacteria that form nodules on roots of woody, dicotyledonous trees and shrubs and (3) cyanobacteria that form mutualisms with many Pteridophytes. Originally, all legume symbionts were classified into the genus Rhizobium; currently these symbionts are classified into several genera with most species belonging to the genera Rhizobium, Sinorhizobium, Mesorhizobium and Bradyrhizobium (Dobbelaere et al. 2003).
Associative diazotrophic bacteria vary in how closely they interact with their host plant. It is tempting to speculate that increasing closeness in this association should correspond to an increasing capacity to transfer fixed nitrogen to the host plant (Ladha et al. 1983). It is reasonable to suppose that nitrogen fixed by free-living diazotrophs on plant surfaces is transferred less efficiently to the plant host and is subject to greater losses than when nitrogen is fixed by endophytic bacteria. Availability of energy-rich carbon compounds and the capability of the heterotrophic N2-fixing bacteria to capture and use it efficiently are the key factors that determine the quantity of fixed nitrogen by associative nitrogen-fixing bacteria (De Bruijn 2015). Associative diazotrophic endophytes have been mainly isolated from grasses such as rice, sorghum, wheat, maize and sugarcane. The most profoundly investigated diazotrophic endophytes are Azoarcus spp. in Kallar grass, Alcaligenes, Azospirillum, Bacillus, Enterobacter, Herbaspirillum, Klebsiella, Pseudomonas and Rhizobium in rice and maize and Gluconacetobacter diazotrophicus and Herbaspirillum spp. in sugarcane (Ladha et al. 1983; Bashan et al. 2004; Iniguez et al. 2004; Alves et al. 2014). Several results, such as from N-balance, 15N isotope dilution and 15N natural abundance, have supplied evidence that some plants can acquire at least part of their nitrogen from associative nitrogen fixation. This transfer of fixed nitrogen to the host plant can occur directly or indirectly due to the death and subsequent mineralization of the bacterial cells (De Bruijn 2015). Many reports proved a direct contribution of endophytic bacteria to plant nitrogen fixation, such as in rice, wheat, and sugarcane (Ladha et al. 2005). Certain Brazilian cultivars of sugarcane get over half their needs for nitrogen from biological nitrogen fixation (Ladha et al. 2005). It has been shown that up to 70% of the nitrogen in sugarcane grown in presence of diazotrophic bacteria can originate from biological nitrogen fixation (Ladha et al. 2005). In rice plants, inoculation with Azospirillum contributed to 66% of the total ammonium present in the plants. As poplar (Populus spp.) and willow (Salix spp.) are very appropriate tree species for phytoremediation purposes, Doty et al. (2009) investigated their associated endophytes and revealed that some of their endophytes appear to be fixing nitrogen (Doty et al. 2009).
3.1.2 Solubilization of Unavailable Nutrients
Second to nitrogen among the mineral nutrients that limit terrestrial plant growth, phosphorus is very important. It often is abundant in soil, but mostly occurs in an insoluble form. Even when soils are rich in phosphorous, most of the P is insoluble and only a very limited amount (≈0.1%) is available to plants (Rodrıguez and Fraga 1999). Phosphorous can only be taken up by plants when it is in its monobasic (H2PO4 −) or dibasic (HPO4 2−) soluble form (Rodrıguez and Fraga 1999). In addition, 75% of the phosphate fertilizers applied to soils is rapidly immobilized and is unavailable to plants (Rodrıguez and Fraga 1999). Phosphate solubilizing and phosphate mineralizing bacteria that can improve plant growth commonly occur in the rhizosphere (Delvasto et al. 2009; Bianco and Defez 2010, Drogue et al. 2012) and include Azotobacter chroococcum, Bacillus spp., Enterobacter agglomerans, Pseudomonas chlororaphis, Pseudomonas putida, and Rhizobium and Bradyrhizobium spp.. Also endophytic bacteria have been reported to solubilize immobilized mineral phosphate. It was suggested that during initial colonization, endophytic bacteria could enhance phosphate availability to soybean plants (Rodrıguez and Fraga 1999). This suggestion was further supported by Kuklinsky-Sobral et al. (2004), showing that 52% of the endophytic bacteria isolated from soybean were able to solubilize mineral phosphate (Kuklinsky‐Sobral et al. 2004).
It is evident that phosphate mobilization will especially be important in the rhizosphere. Therefore it is not surprising that the rhizosphere strain P. putida KT2440, possesses three ABC-type phosphate transport operons compared to only one in the closely related endophytic strain P. putida W619 (Loper et al. 2012). Consistently, other endophytic bacteria like Enterobacter sp. 638 and Serratia proteamaculans 568 were found to possess only one ABC phosphate transport operon (Taghavi et al. 2009).
Iron in soils is, like phosphorous, often occurring in an insoluble form, more specifically the highly insoluble ferric hydroxide form. Many bacteria produce organic molecules, called siderophores, which bind Fe3+ and render it available for conversion to the preferred form, Fe2+. Radzki et al. (2013) reviewed the plant-microbe interactions involved in the regulation of siderophore production and their role in mediating competition for iron in the rhizosphere. Bacterial Fe3+-siderophore complexes may not only facilitate uptake of iron into bacteria. Evidence exists that several plant species can recognize and take up bacterial Fe3+-siderophore complexes, and that, especially in calcareous soils, this process is crucial for the uptake of iron by plants (Kloepper et al. 1980a, b; Schwyn and Neilands 1987; Hider and Kong 2010; Radzki et al. 2013).
Bacteria developed several distinct mechanisms to compete against each other for iron resources, a concept applied by plant growth promoting bacteria in the protection of their host plant against pathogens. They can produce a high number of specific iron uptake transporters, secrete great numbers of diverse siderophores (which is energy costly), or synthesize siderophore receptors to utilize siderophores excreted by other bacteria (Crowley et al. 1988; Hider and Kong 2010).
During its adaptation to survive in the soil prior to colonization of a host plant, or as a plant growth promoting endophyte, Enterobacter sp. 638 has developed an intermediate solution to deal with iron uptake. It possesses two ferrous iron uptake systems (FeoAB, EfeUOB) and nine iron ABC transporters (Taghavi et al. 2010). This number is much higher than the four iron ABC transporters present in E. coli K12 or the three found in P. putida KT2440 (Nelson et al. 2002).
Similarly to E.coli K12, Enterobacter sp. 638 is also able to produce the siderophore enterobactin (EntD, EntF, EntC, EntE, EntB and EntA), to secrete it (EntS), recover the Fe-enterobactin complex making use of a ferric siderophore uptake system tonB-dependant (ExbDB), and to extract the iron using an enterobactin esterase (Fes) after internalization of the Fe-enterobactin complex. In addition, Enterobacter sp. 638 possesses 12 outer membrane ferric and ferric-related siderophore receptors (Taghavi et al. 2010). In contrast, E. coli only possesses 6 siderophore receptors, while P. putida KT2440 is equipped with 18 receptors (Nelson et al. 2002), which is consistent with the concept of a plant growth promoting rhizosphere bacterium that has to compete for the iron resources present in the environment.
3.1.3 Phytohormones and Plant Growth Promoting Compounds
Phytohormones produced by plant-associated bacteria often stimulate plant growth. However, this bacterial phytohormone production does not have a direct benefit for the bacteria itself and can be explained in an indirect way. The stimulation of plant growth that is induced will lead to more nutrients available to the plant-associated bacteria. Auxins, cytokinins and gibberellins can be considered as causal agents for improving plant growth and development (Pieterse et al. 2009; Blinkov et al. 2014; Spaepen 2015).
The most investigated phytohormone produced by plant-associated bacteria is the auxin indole-3-acetic acid (IAA) (Spaepen and Vanderleyden 2011). IAA produced by Azospirillum brasilense, Aeromonas veronii, Agrobacterium spp., Alcaligenes piechaudii, Bradyrhizobium spp., Comamonas acidovorans, Enterobacter spp., and Rhizobium leguminosarum, can contribute to plant growth promotion (López-Bucio et al. 2007). IAA production can improve root growth and root length; it also has been associated with proliferation and elongation of root hairs (Pitts et al. 1998; Remans et al. 2007; Nakayama et al. 2012). Depending on the bacterial strain, the production of IAA seems to follow different pathways. In plant beneficial bacteria, IAA is predominantly synthesized via indolepyruvic acid; however, in phytopathogenic bacteria , IAA is generally produced from tryptophan via the intermediate indoleacetamide (Spaepen and Vanderleyden 2011).
IAA-producing plant beneficial bacteria have been isolated from different plant species: lettuce (Barazani and Friedman 1999), wheat, banana, and cotton (Mohite 2013), rice (Mehnaz et al. 2001), sugarcane (Sajjad Mirza et al. 2001) and poplar (Taghavi et al. 2005; Taghavi et al. 2009).
Contrasting observations were made in different studies that were conducted to determine the specific role of IAA production in plant growth promotion. A mutant strain of Pseudomonas putida showing a fourfold increase in IAA production lost the capacity to induce root elongation in canola seedlings, despite the fact that its growth rate and production of 1-aminocyclopropane-1-carboxulate (ACC) deaminase and siderophores remained the same. A supra-optimal IAA concentration can be hypothesized. In other studies, positive effects of bacterial IAA production on plant growth and development have been reported. In Brassica spp., a positive correlation was found between auxin production by different PGPR strains and their ability to increase the grain yield and numbers of branches and pods per plant (Asghar et al. 2002). This observation was further supported by the positive relation between auxin production by PGPR and the increase in number of branches and oil content in B. napus inoculated with these PGPR (Asghar et al. 2004).
Concerning the role of hormones other than auxins, such as cytokinins and gibberellins, less detailed information is available. Cytokinins play a role in the stimulation of cell division, cell enlargement and tissue expansion in certain plant parts (Skoog and Armstrong 1970; Frébort et al. 2011). Cytokinin-producing PGPR have been isolated from rape and lettuce (Arkhipova et al. 2007), wheat (Kudoyarova et al. 2014), soybean (de Garcia Salamone et al. 2006), and pine (Bent et al. 2001). However, the extent to which the cytokinins produced by these bacteria play a role in plant growth promotion is still not unraveled. Gibberellins (gibberellic acid) are known to be involved in the extension growth of plant tissue, in particular of the stem (Bottini et al. 2004). Although gibberellin production by plant-associated bacteria seems to be unusual, Acetobacter diazotrophicus and Herbaspirillum seropedicae spp., Bacillus pumilus and Bacillus licheniformis spp. (Gutiérrez‐Mañero et al. 2001) able to synthesize this phytohormone, were isolated and were shown to improve plant growth and yield.
Volatile compounds such as 3-hydroxy-2-butanone (acetoin ) and 2,3-butanediol are produced by rhizobacteria to enhance plant growth (Ryu et al. 2003b). The endophytic strains Enterobacter sp. 638 and S. proteamaculans 568 possess the genetic capability for the conversion of pyruvate to acetolactate (acetolactate synthase) (Taghavi et al. 2009; Taghavi et al. 2010). Acetolactate is further spontaneously converted either into diacetyl, or into acetoin (acetolactate decarboxylase). Enterobacter sp. 638 possesses an additional gene to convert diacetyl into acetoin (acetoin dehydrogenase), which can be released and converted into 2,3-butanediol (2,3-butanediol dehydrogenase) by the plant. Interestingly, the poplar genome contains the genes necessary for the conversion of acetoin into 2,3-butanediol, but not for the biosynthesis of acetoin itself. This could mean that poplar is relying on endophytic bacteria like Enterobacter sp. 638 for the production of the plant growth hormones acetoin and 2,3-butanediol (Taghavi et al. 2010).
3.1.4 Counteracting Stress-Induced Ethylene
Ethylene is a plant hormone that is known to be increased in plants in response to both abiotic and biotic stress conditions. Among its multiple effects on plant development (from seed germination, morphogenesis, flowering induction up to senescence), frequently investigated effects include the inbition of root elongation, lateral root growth and root hair formation (Gamalero and Glick 2012). Therefore, decreases in ethylene production should lead to an indirect promotion of root elongation. Bacteria can affect ethylene production via two main mechanisms (Mastretta et al. 2006): (1) some bacteria can balance ethylene production levels through auxin production; (2) however, the most commonly described mechanism to reduce ethylene production levels is bacterial ACC-deaminase activity. ACC-deaminase was observed in strains of Alcaligenes spp., Bacillus pumilus, Enterobacter cloacae, Burkholderia cepacia, Pseudomonas putida, Pseudomonas spp. and Variovorax paradoxus (Glick et al. 2007; Saleem et al. 2007; Rashid et al. 2011), but is not produced by the endophytic bacteria Enterobacter sp. 638, S. proteamaculans 568, S. maltophilia R551-3 and P. putida W619 (Taghavi et al. 2009). The working mechanism in plant roots is believed to be via cleavage of ACC, which is the immediate precursor of ethylene during ethylene biosynthesis, resulting in increased root growth (Glick 2014). Bacteria, originating from different soils and containing ACC deaminase activity, stimulated plant growth even in soils containing phytotoxic cadmium concentrations (Belimov et al. 2005).
3.2 Indirect Promotion of Plant Growth
Competition has been claimed as an important mechanism of biocontrol ; both, pathogens and non-pathogenic plant-associated bacteria compete for similar niches and the same nutrients. However, experimental evidence for this claim still hardly exists.
Under iron-limiting conditions, many plant growth promoting bacteria, especially pseudomonads, produce high-affinity Fe3+ binding siderophores (Kloepper et al. 1980a; Miethke and Marahiel 2007). By binding available iron, these bacteria are depriving pathogenic bacteria and fungi of iron, which could limit their growth. Many authors have illustrated the importance of siderophores in the inhibition of both fungal and bacterial pathogens (Miethke and Marahiel 2007).
Antibiosis is the production and release of compounds that kill or inhibit the growth of the target pathogen; it is the best-known mechanism by which microbes can control plant diseases (Raaijmakers and Mazzola 2012). The antibiotics generally produced by diverse antagonistic bacteria consist of ammonia, butyrolactones, 2,4-diacetyl phloroglucinol (DAPG) , kanosamine, oligomycin A, oomycin A, phenazine-1-carboxylic acid, pyoluteorin, pyrrolnitrin, viscosinamide, xanthobaccin, and zwittermycin A (Kim et al. 1999; Raaijmakers et al. 2002; Raaijmakers and Mazzola 2012). Many of these antibiotics have a broad-spectrum activity; DAPG was reported to be the most effective and is the most extensively studied antibiotic (Raaijmakers and Mazzola 2012). Using mutation analysis, molecular genetic tools, and using purified antibiotic compounds, the role of individual antibiotic compounds in suppression of root pathogens has been clearly established. Structures and modes of action of many antimicrobial compounds have been extensively reviewed (Poole 2012; Raaijmakers and Mazzola 2012).
Also biosurfactants have been investigated as antimicrobial compounds (De Souza et al. 2003; Bais et al. 2004). As pathogens frequently form a biofilm on the root surface, it is interesting that some biosurfactants were reported to prevent biofilm formation and even degrade existing biofilms (Kuiper et al. 2002; Bais et al. 2004). For instance, Pseudomonas fluorescens produces cyclic lipopeptides surfactants such as viscosinamide and tensin which have antifungal activity against Rhizoctonia solani and Pythium ultimum (Haas and Keel 2003; Nielsen and Sorensen 2003).
3.2.3 Production of Hydrolytic Enzymes
Another potential mechanism for plant-associated bacteria to control fungal pathogens is cell wall lysis. It is well established in the biocontrol of fungal pathogens in the rhizosphere. Endophytic bacteria isolated from potato roots show high levels of hydrolytic enzymes such as cellulase, chitinase and glucanase (Krechel et al. 2002). However, in the endophytic bacteria Enterobacter sp. 638, S. proteamaculans 568 and P. putida W619 no members of the cellulase/endoglucanase (GH5, GH9, GH44, GH48 and GH74), lichenase (GH16) and xylanase (GH10, GH11) families of glycosyl hydrolases (GH) were found (Taghavi et al. 2009). This observation is consistent with the non phytopathogenic behavior of these bacteria.
Among the hydrolytic enzymes, chitinases are of high importance since chitin is a main cell wall component in the majority of the phytopathogenic fungi (Hoster et al. 2005). The endophytic bacteria S. proteamaculans 568, P. putida W619 and S. maltophilia R551-3 are able to grow on chitin as a carbon source. Chitinases from these strains belong to the glycosyl hydrolase family 18 in S. proteamaculans 568 and S. maltophilia R551-3, and glycosyl hydrolase family 19 in P. putida W619 (Taghavi et al. 2009; Taghavi et al. 2010). Gluconases are another important group of hydrolytic enzymes since they degrade the β-1,3-glucans of the fungal cell walls. Fridlender et al., (1993) reported that the production of β-1,3-glucanases by Burkholderia cepacia (formarly Pseudomonas cepacia) caused an inhibition in rhizosphere proliferation of various phytopathogenic fungi including Rhizoctonia solani, Serratia rolfsii and Phytium ultimum (Fridlender et al. 1993). Furthermore, Tanaka and Watanabe (1995) observed that a synergistic action of a combination of chitinases and β-1,3-glucanases resulted in a more effective inhibition of fungal pathogens than by the individual enzymes (Tanaka and Watanabe 1995).
3.2.4 Inhibition of Pathogen-Produced Enzymes or Toxins
For the degradation of polymers in plant cell walls and so facilitate the fungal infection, pathogenic fungi produce extracellular hydrolytic enzymes including cellulases, pectolytic enzymes (exo- and endo- polygalacturonases, pectin lyases), and cutinase. A suppression of the activity of these enzymes correlates with a reduction in virulence (Beraha et al. 1983). For instance, Bertagnolli et al. (1996) reported that the activities of extracellular enzymes, such as cellulase, pectin lyase and pectinase produced by Rhizoctonia solani, were inhibited by Bacillus megaterium B 153-2-2 producing an extracellular endoproteinase (Bertagnolli et al. 1996).
3.2.5 Induction of Plant Defence Mechanisms
In a plant, contact with a necrotizing pathogen or a non-pathogenic biocontrol bacterium can induce a state of physiological immunity, protecting it against subsequent viral, bacterial or fungal attacks (Bakker et al. 2007). The induced resistance associated with the colonization of plant roots by certain plant growth promoting rhizobacteria (PGPR), has been referred as induced systemic resistance (ISR) (Conrath et al. 2002; Ryu et al 2003a; Domenech et al. 2007). ISR is characterized by remote action, long-lasting resistance, and protection against a large number of pathogens. Most of the resistance-inducing microbes reported hitherto are Gram-negative bacteria, with mainly Pseudomonas and Serratia strains (Bakker et al. 2007). However, also a number of Gram-positive bacteria were reported to induce resistance (Kloepper et al. 2004b).
ISR can activate multiple potential defence mechanisms, including an increase in activity of peroxidases, chitinases, β-1,3-glucanases, and other pathogensis-related proteins, the formation of protective biopolymers, such as lignin, cellulose, and hydroyproline-rich glycoproteins, and accumulation of antimicrobial low-molecular-weight substances, such as phytoalexins (Conrath et al. 2002; Ryu et al. 2003b). In addition, a single inducing agent can control a wide variety of pathogens.
4 Catabolic Potential of Endophytic Bacteria
Not only soil and rhizospheric bacteria but also many endophytes are described to have the catabolic enzymes to degrade or transform petroleum hydrocarbons. Petroleum hydrocarbons are a complex mixture of linear, branched and cyclic alkanes and aromatic compounds that are not all as easy biodegradable. From high to low biodegradability they can be ranked as follows: n-alkanes > branched-chain alkanes > branched alkenes > low-molecular-weight n-alkyl aromatics > monoaromatics > cyclic aromatics and polycyclic aromatic hydrocarbons (PAHs) (Chikere et al. 2011).
The most complete degradation occurs under aerobic conditions. Alkane-degradation by bacteria for example begins with an oxidative attack at the terminal methyl group with the formation of a fatty alcohol, aldehyde and fatty acid (Van Hamme et al. 2003; Kanaly and Harayama 2010). The carboxylic acid can then be combined with CoA that via acetyl-CoA can enter the tricarboxylic acid (TCA) cycle. Short-chain alkane monooxygenases, including methane monooxygenases (MMO), are the first enzymes involved in oxidising C1–C4 alkanes (Martin et al. 2014). Gaseous alkanes can be metabolized by strains expressing propane or butane monooxygenases that are related to methane monooxygenases, and these have been found for example in Gordonia sp. TY-5 (Kotani et al. 2003). Medium-chain length (C5–C16) alkanes may be oxidized by two main classes of enzymes: integral membrane non-heme iron alkane hydroxylases (alk system) and soluble cytochrome P450s . The most thoroughly characterized alkane degradation pathway is encoded by the OCT plasmid carried by Pseudomonas putida Gpo1 (van Beilen et al. 2001). In this system, the alkBFGHJKL operon encodes the enzymes necessary for converting alkanes into acetyl-coenzyme A (CoA), while alkST encode a rubredoxin reductase (AlkT) and the positive regulator for the alkBFGHJKL operon (AlkS). The genetic characterization in P. putida GPo1 of the alkane degradation pathway boosted the research on the field and since then more than 60 homologues of alkane hydroxylase gene (alkB) have been cloned and sequenced in both Gram-positive and Gram-negative bacteria such as Acinetobacter sp., Mycobacterium sp., Rhodococcus sp., Pseudomonas putida P1, P. aureofaciens and P. fluorescens (Palleroni et al. 2010; Wang et al. 2010a, b). Sequencing of the genomic DNA of pure bacterial cultures isolated from contaminated sites allows the detailed analyses of PH-degradation pathways. Currently, there are well-established and growing genomic databases for the genomes and pure cultures, such as the National Centre for Biotechnology Information (GenBank, RefSeq), the National Institute of Genetics (DNA Data Bank of Japan), and the European Bioinformatics Institute (EMBL). The advantage is that these genomic sequences can now be obtained in a day, rather than years, and for a fraction of the cost. This has opened access to the so-called “uncultured majority.” Advances in this field has already revealed a striking diversity in nucleotide and amino acid sequences of the AlkB gene (van Beilen et al. 2001; Wasmund et al. 2009; Wang et al. 2010a, b) and suggests horizontal gene transfer in many cases of duplicated gene clusters.
5 Phytoremediation of Petroleum Hydrocarbons
Phytoremediation requires the selection of plants with increased pollutant tolerance, production of sufficient root and shoot biomass, suitability for various soil types, effective pollutant uptake mechanisms, appropriate metabolic capabilities to degrade organic pollutants, and association with active degradative microorganisms (Vangronsveld et al. 2009; Wenzel 2009; Fig. 2).
Initially, the response of plants to PHs present in soil includes uptake and translocation followed by accumulation or evapotranspiration. It is known that root uptake of PHs is strongly affected by PHs lipophilicity, a parameter which is expressed as the octanol-water partition coefficient (Kow ) (Gerhardt et al. 2009). PHs with a log Kow < 1 are characterized by high water-solubility, and plant roots do not generally accumulate them at a rate surpassing passive efflux into the transpiration stream with subsequent translocation to the shoot, therefore, that practically means impassability to be taken up by the plant roots, whereas PHs with a log Kow > 3.5 cannot be taken up and translocated into the plant due to tight sorption onto the soil or root surfaces (Weyens et al. 2009b). After being transported inside the plant, PHs can be either sequestered in root tissue, or transported into shoots and then to leaves, where they can be stored in the vacuole or volatized into the atmosphere (Reichenauer and Germida 2008; Fig. 2).
Over the last years, evidence has accumulated that many plant species are suitable for PH remdiation such as Italian ryegrass (Lolium perenne), sorghum (Sorghum bicolour), maize (Zea mays), tall fescue (Festuca arundinacea), alfalfa (Medicago sativa var. Harpe), elephant grass (Penninsetum purpureum), bermuda grass (Cynodon dactylon), birdsfoot trefoil (Lotus corniculatus var. Leo), sunflower (Helianthus annuus), southern crabgrass (Digitaria sanguinalis), red clover (Trifolium pretense), beggar ticks (Bidens cernua), sedge species (Cyperus rotundus), leguminous plants, and willow (Salix spp.) (Kamath et al. 2004; Kaimi et al. 2007; Bell et al. 2014b; Yergeau et al. 2014). Although the plant may often metabolize or sequester environmental toxins, plants are at a significant disadvantage compared to microorganisms in two ways. Firstly, being photoautotrophic , plants do not rely on organic compounds as a source of energy or carbon. Secondly, plant metabolism of organic molecules (other than photosynthates) follows general transformations to more water-soluble forms, and sequestration processes to avoid build-up and potential toxicity to sensitive organelles (green-liver model) (Burken 2003), rather than mineralisation. Therefore, in order to develop a more efficient degradation of organic xenobiotics, plants depend significantly on their associated microorganisms.
5.1 Plant-Bacteria Partnerships
It is clear that plants have a positive effect on the microbial degradation of organic contaminants (Weyens et al. 2009a, b; Fig. 2). This positive effect can be explained by a higher microbial density and metabolic activity in the rhizosphere as a result of bacterial growth on carbon substrates (root exudates and death cells) provided by the plant roots (Segura et al. 2009). Moreover, diverse species of heterotrophic microorganisms are living together at high population densities in the rhizosphere, the phyllosphere and inside the plant (endophytes), which increases the possibilities for stepwise transformation of xenobiotics by consortia, or provides habitats that are conducive to genetic exchange and gene rearrangement (Fig. 2). Additionally, the plant’s water evapotranspiration influences the transport of water soluble compounds by increasing their mass flow to the root surface where they can be acted upon by the rhizosphere microflora. After being taken up by the plant, endophytes can continue the degradation of organic xenobiotics (Weyens et al. 2009a, b). The importance of these plant-microbe partnerships in the remediation of organic pollutants was confirmed in studies at the level of the rhizosphere (Van Aken et al. 2010; Khan et al. 2013; Segura and Ramos 2013; Arslan et al. 2015), the phyllosphere (Scheublin et al. 2014) and inside the plant (Siciliano et al. 2001; Barac et al. 2004; Syranidou et al. 2016; Fig. 2).
5.2 Plant-Endophyte Interactions for Petroleum Hydrocarbon Degradation
In a pioneering study, it was shown that the enrichment of bacteria with the appropriate catabolic genes in the endophytic root compartment was correlated with the type and amount of contaminant and also on the genotype of the plant (Siciliano et al. 2001). Since then, a number of reports have confirmed that endophytic bacteria, rather than rhizosphere or soil bacteria have a better capacity to enhance PH phytoremediation (Newman and Reynolds 2005; Andria et al. 2009; Weyens et al. 2009a; Afzal et al. 2014; Fig. 2).
Bacteria dwelling the internal tissues of plants (roots, stems, leaves) overcome the competition for nutrients and space, and are physically protected from unfavorable environmental conditions (Schulz and Boyle 2006). Some soil bacteria can penetrate into roots and move into shoots, indicating that these soil bacteria are a source for endophytic bacteria (Germaine et al. 2004; Germaine et al. 2009). In plant-endophyte associations , the plant has to confront the toxic nature of the hydrocarbons; therefore, an endophyte with the ability to mitigate the toxicity of the pollutant seems of utmost importance. Indeed some endophytic bacteria have the potential at first to tolerate and then mineralize hydrocarbons (Andria et al. 2009; Germaine et al. 2009; Yousaf et al. 2010; Gkorezis et al. 2015). Hence, this hydrocarbon degrading capacity of endophytic bacteria has been investigated aiming to enhance the remediation potential of trees, herbaceous plants and grasses (Afzal et al. 2014). In addition, other studies have demonstrated that endophytic bacteria with appropriate degradation pathways are metabolically active in the root and shoot of plants vegetated in diesel contaminated soils (Andria et al. 2009; Khan et al. 2013).
5.3 Potential of Meta-Omics Approaches to Elucidate Microbial Functions
Given the rich literature on PH-microbe interactions (Khan et al. 2013), there is also a tremendous scope for further integrating genomics and meta-omics to endophyte-stimulated site remediation (Fig. 2). Using metagenomics coupled to stable isotope probing (SIP), considerable progress has already been made to understand the mode of action by which soil microorganisms degrade organic contaminants, activate enzymes and whole metabolic pathways, interact with each-other and their environment, and for unraveling novel regulatory pathways.
Microbial communities have been characterized from hydrocarbon contaminated soils in the Arctic (Thomassin-Lacroix et al. 2002; Bell et al. 2011; Yergeau et al. 2012; Bell et al. 2013a, b), deepwater oil spills (Mason et al. 2014), in crude oil wells in Poland, and underground tank storage leakage sites (Gkorezis et al. 2015). Using carbon isotope (U-13C) labeled naphthalene, phenanthrene, pyrene, fluoranthene and benz[a]anthracene added to polycyclic aromatic hydrocarbon (PAH) contaminated soil, diverse genera were discovered of which some were previously associated with the degradation of those compounds (Pseudomonas, Burkholderia) (Jones et al. 2011), but also newly associated sequences related to Pigmentiphaga and a group of yet-to-be cultured γ-Proteobacteria, identified as “Pyrene group 2” which attacked pyrene and benz[a]anthracene. A similar study using 13C-labeled metagenomes recovered from samples in a long-term biphenyl- and polyaromatic hydrocarbon contaminated soil, showed a dominance of mainly Proteobacteria including Burkholderia, Pandoraea, Dyella and Pseudomonas and some yet-to-be cultured taxa. Some of these strains derived carbon from naphthalene as well as biphenyl/benzoate, pointing out broader biodegradation abilities of some soil microbiota than previously known (Uhlik et al. 2012).
Metatranscriptomics studies have revealed complex responses to hydrocarbon pollution, providing evidence how bacteria and fungi catabolically adapt to changes in the chemical environment (Bell et al. 2011, 2014b, Yergeau et al. 2014). In addition, in some cases it appeared that the natural established plant-microbiome associations were not the most productive in terms of biodegradation efficiency (Bell et al. 2013b). When researchers added antibiotics (gentamicin and vancomycin) to a hydrocarbon-contaminated Arctic soil to inhibit distinct subgroups of the microbial community, the hydrocarbon degradation rates increased compared to the no-antibiotic control, while bacterial and fungal abundance were reduced (Bell et al. 2013). These findings let the authors suggest that a large part of the microbial population is not actively involved in hydrocarbon degradation and rather compete with hydrocarbon degradative microorganisms for nutrients and space. The effect was again different when nutrients were added (Bell et al. 2013b). In this case, nutrient addition promoted a larger fungal population accounting for higher degradation rates (Bell et al. 2014b). Looking in the near future, data on gene, transcript, protein, metabolome, and interactome of plant-microbiota must be combined into holistic models in order to better appreciate the holobiont that is essential for phytoremediation (Fig. 2).
5.4 Critical Factors, Success Factors, and Future Outlook
Although phytoremediation of PHs and organics in general is very promising, it cannot be ignored that there are still some obstacles that need to be overcome, such as (1) the levels of pollutants tolerated by the plant, (2) the bioavailable fraction of the contaminants that often is limited and, (3) in some cases, the evapotranspiration of volatile organic pollutants to the atmosphere. A possible solution to overcome these limitations is the construction of plants specifically tailored for phytoremediation purposes by genetic manipulation (Maestri and Marmiroli 2011). However, since bacteria are much easier to manipulate than plants and natural gene transfer is possible (avoiding the limitations of the use of GMO), many studies focussed on the use of engineered plant-associated bacteria. Dzantor 2007 reviewed the state of the art of rhizosphere “engineering ” for accelerated rhizodegradation of xenobiotic contaminants (Dzantor 2007). Even in case of an efficient rhizodegradation, compounds with a lipophilicity in the optimum range seem to enter the xylem before the soil and rhizosphere microflora can degrade them. Since the retention time of contaminants in the xylem ranges from several hours up to 2 days (Barac et al. 2004), (engineered) degrader endophytes colonizing the xylem are ideal candidates to prevent evapotranspiration of pollutants or intermediates through the leaves into the environment and to reduce phytotoxicity. Endophytic bacteria can be isolated, equipped with interesting characteristics and subsequently re-inoculated in the host plant (Fig. 2). Proof of this concept was provided by inoculating lupine plants (Barac et al. 2004) and poplar (Taghavi et al. 2005) with endophytic bacteria able to degrade toluene, which resulted in both, decreased toluene phytotoxicity and a significant reduction of toluene evapotranspiration. As many catabolic pathways for hydrocarbons are found in soil bacteria where they are encoded on self-transferable plasmids or transposons , natural gene transfer offers great potential for the construction of endophytic strains with new catabolic functions. Moreover, heterologous expression of these catabolic functions might not be a major problem, especially when the donor strain and the recipient endophytic strain are closely related, as frequently is the case.
Although it is obvious that the application of engineered endophytic bacteria to improve phytoremediation of organic contaminants has several advantages, there are still some other factors that need optimisation (Newman and Reynolds 2005). Since phytoremediation projects can conceivably last decades, a major point of concern is the persistence and the stability of the engineered organisms and their degradation capabilities in association with plants growing in the field. As long as there is a selection pressure present, those community members possessing the appropriate degradation characteristics will have an advantage. Nevertheless, this is no guarantee that inoculated strains will become an integrated part of the natural endophytic community. However, instead of integrating a new strain in a stable community, the original community can also get adapted through horizontal gene transfer. Horizontal gene transfer has been shown to play an important role in the adaptation of a microbial community to a new environmental stress factor, including rhizosphere communities (van Elsas et al. 2003), and endophytic communities (Taghavi et al. 2005). Alltogether, the exploitation of endophytic bacteria, horizontal gene transfer, and environmental genomics should all be incorporated into processes at petroleum hydrocarbon contaminated sites to better predict and monitor on site processes.
Hydrocarbon contamination in soil and (ground)water is one of the major challenges we are facing today. The use of plants and microbes can drastically reduce the impact that contaminated sites have on the environment by improving the degradation at a low cost and sustainable manner. We have discussed that plants associate in diverse and intricate partnerships with microorganisms. Plants create specific favourable niches for their associated bacteria by providing them nutrients, and the plant-associated (endophytic) bacteria can improve growth and development of their host plant directly or indirectly. During phytoremediation of contaminated soils and (ground)water, endophytic bacteria possessing the appropriate degradative enzymes may reduce PH phytotoxcicity and enhance PH degradation thereby reducing the chance of evapotranspiration of volatile contaminants and/or degradation intermediates to the atmosphere. The integration of environmental genomics into plant-microbe sitmulated phytoremediaton may further improve the reliability and efficiency of the technique. Phytoremediation will require the use of molecular barcoding and sequencing for rapidly quantifying microbial communities to predict changes in degradation rates and future predictions. Besides using natural occuring endophytes , successful examples in the past have shown the potential of using engineered endophytes, which have a high degradation capacity and ability to transfer their degradative genes to te indigenous microbial community by natural gene transfer . Because the new sequencing technologies are still in development, some risks to the environment may still exist related to the element of uncertainty and long-term performance. Nevertheless, great avenues and discoveries are anticipated in this exciting field in the years to come.
7 Research Needs
To address these challenges, cheaper and more sustainable solutions are needed to remove harmful pollutants from the environment. In this light, phytoremediation, the use of plants and associated microorganisms to degrade aromatic hydrocarbons, has gained increased interest by providing a green and economically beneficial solution.
- Arslan M, Imran A, Khan QM, Afzal M (2015) Plant-bacteria partnerships for the remediation of persistent organic pollutants. Environ Sci Pollut Res Int. doi:10.1007/s11356-015-4935-3Google Scholar
- Bacon CW, White JF (1994) Stains, media and procedures for analyzing endophytes. In: Biotechnology of endophytic fungi of grasses. CRC PRess, Boca Raton, pp 47–56Google Scholar
- Bertagnolli B, Dal Soglio F, Sinclair J (1996) Extracellular enzyme profiles of the fungal pathogen Rhizoctonia solani isolate 2B-12 and of two antagonists, Bacillus megaterium strain B153-2-2 and Trichoderma harzianum isolate Th008. I. Possible correlations with inhibition of growth and biocontrol. Physiol Mol Plant Pathol 48:145–160CrossRefGoogle Scholar
- Bloemberg GV, Wijfjes AH, Lamers GE, Stuurman N, Lugtenberg BJ (2000) Simultaneous imaging of Pseudomonas fluorescens WCS365 populations expressing three different autofluorescent proteins in the rhizosphere: new perspectives for studying microbial communities. Mol Plant Microbe Interact 13:1170–1176PubMedCrossRefGoogle Scholar
- Croes S, Weyens N, Janssen J, Vercampt H, Colpaert JV, Carleer R, Vangronsveld J (2013) Bacterial communities associated with Brassica napus L. grown on trace element-contaminated and non-contaminated fields: a genotypic and phenotypic comparison. Microb Biotechnol 6:371–384PubMedPubMedCentralCrossRefGoogle Scholar
- De Bruijn F (2015) Biological nitrogen fixation. In: Lugtenberg B (ed) Principles of plant-microbe interactions Microbes for sustainable agriculture. Springer International Publishing, Switzerland, pp 215–225Google Scholar
- de Garcia Salamone IE, Hynes RK, Nelson LM (2006) Role of cytokinins in plant growth promotion by rhizosphere bacteria. In: Siddiqui ZA (ed) PGPR: biocontrol and biofertilization. Springer, Dordrecht, pp 173–195Google Scholar
- Gkorezis P, Rineau F, Van Hamme J, Franzetti A, Daghio M, Thijs S, Weyens N, Vangronsveld J (2015) Draft genome sequence of acinetobacter oleivorans PF1, a diesel-degrading and plant-growth-promoting endophytic strain isolated from poplar trees growing on a diesel-contaminated plume. Genome Announc 3:e01430PubMedPubMedCentralGoogle Scholar
- Glick BR (2012) Plant growth-promoting bacteria: mechanisms and applications. Scientifica (Cairo) 2012:963401Google Scholar
- Govindasamy V, Franco CMM, Gupta VVSR (2014) Endophytic actinobacteria: diversity and ecology. In Advances in endophytic research (pp. 27–59). Springer IndiaGoogle Scholar
- Hardoim PR, van Overbeek LS, Berg G, Pirttila AM, Compant S, Campisano A, Doring M, Sessitsch A (2015) The hidden world within plants: ecological and evolutionary considerations for defining functioning of microbial endophytes. Microbiol Mol Biol Rev 79:293–320PubMedPubMedCentralCrossRefGoogle Scholar
- Kästner M (2008) Degradation of aromatic and polyaromatic compounds. In: Biotechnology set. Wiley-VCH Verlag GmbH, Weinheim, pp 211–239Google Scholar
- Kim BS, Moon SS, Hwang BK (1999) Isolation, identification, and antifungal activity of a macrolide antibiotic, oligomycin A, produced by Streptomyces libani. Can J Bot 77:850–858Google Scholar
- Ladha JK, Pathak H, Krupnik TJ, Six J, van Kessel C (2005) Efficiency of fertilizer nitrogen in cereal production: retrospects and prospects. In: Advances in Agronomy, vol 87. Academic, pp 85–156. ElsevierGoogle Scholar
- López-Bucio J, Campos-Cuevas JC, Hernández-Calderón E, Velásquez-Becerra C, Farías-Rodríguez R, Macías-Rodríguez LI, Valencia-Cantero E (2007) Bacillus megaterium rhizobacteria promote growth and alter root-system architecture through an auxin- and ethylene-independent signaling mechanism in Arabidopsis thaliana. Mol Plant-Microbe Interact 20:207–217PubMedCrossRefGoogle Scholar
- Mohite B (2013) Isolation and characterization of indole acetic acid (IAA) producing bacteria from rhizospheric soil and its effect on plant growth. J Soil Sci Plant Nutr 13:638–649Google Scholar
- Pandey P, Kang S, Maheshwari D (2005) Isolation of endophytic plant growth promoting Burkholderia sp. MSSP from root nodules of Mimosa pudica. Curr Sci 89:177–180Google Scholar
- Pirttilä AM, Frank AC (2011) Endophytes of forest trees, vol 80. Springer NetherlandsGoogle Scholar
- Porteous Moore F, Barac T, Borremans B, Oeyen L, Vangronsveld J, van der Lelie D, Campbell CD, Moore ER (2006) Endophytic bacterial diversity in poplar trees growing on a BTEX-contaminated site: the characterisation of isolates with potential to enhance phytoremediation. Syst Appl Microbiol 29:539–556CrossRefGoogle Scholar
- Rashid S, Charles TC, Glick BR (2011) Isolation and characterization of new plant growth-promoting bacterial endophytes. Appl Soil Ecol 61(2012):217–224Google Scholar
- Roca A, Pizarro-Tobias P, Udaondo Z, Fernandez M, Matilla MA, Molina-Henares MA, Molina L, Segura A, Duque E, Ramos JL (2013) Analysis of the plant growth-promoting properties encoded by the genome of the rhizobacterium Pseudomonas putida BIRD-1. Environ Microbiol 15:780–794PubMedCrossRefGoogle Scholar
- Santaella C, Schue M, Berge O, Heulin T, Achouak W (2008) The exopolysaccharide of Rhizobium sp. YAS34 is not necessary for biofilm formation on Arabidopsis thaliana and Brassica napus roots but contributes to root colonization. Environ Microbiol 10:2150–2163PubMedPubMedCentralCrossRefGoogle Scholar
- Scheublin TR, Deusch S, Moreno-Forero SK, Muller JA, van der Meer JR, Leveau JH (2014) Transcriptional profiling of Gram-positive Arthrobacter in the phyllosphere: induction of pollutant degradation genes by natural plant phenolic compounds. Environ Microbiol 16:2212–2225PubMedCrossRefGoogle Scholar
- Spaepen S (2015) Plant hormones produced by microbes. In: Lugtenberg B (ed) Principles of plant-microbe interactions Microbes for sustainable agriculture. Springer International Publishing, Switzerland, pp 247–256Google Scholar
- Spaepen S, Vanderleyden J, Okon Y (2009) Chapter 7: plant growth-promoting actions of Rhizobacteria. In: Loon LCV (ed) Advances in botanical research, vol 51. Academic, pp 283–320Google Scholar
- Taghavi S, Garafola C, Monchy S, Newman L, Hoffman A, Weyens N, Barac T, Vangronsveld J, van der Lelie D (2009) Genome survey and characterization of endophytic bacteria exhibiting a beneficial effect on growth and development of poplar trees. Appl Environ Microbiol 75:748–757PubMedCrossRefGoogle Scholar
- Thijs S, Weyens N, Sillen W, Gkorezis P, Carleer R, Vangronsveld J (2014a) Potential for plant growth promotion by a consortium of stress-tolerant 2,4-dinitrotoluene-degrading bacteria: isolation and characterization of a military soil. Microb Biotechnol 7:294–306PubMedPubMedCentralCrossRefGoogle Scholar
- Thijs S, Van Dillewijn P, Sillen W et al (2014b) Exploring the rhizospheric and endophytic bacterial communities of Acer pseudoplatanus growing on a TNT-contaminated soil: towards the development of a rhizocompetent TNT-detoxifying plant growth promoting consortium. Plant Soil 385:15–36CrossRefGoogle Scholar
- Truyens S, Beckers B, Thijs S, Weyens N, Cuypers A, Vangronsveld J (2015a) The effects of the growth substrate on cultivable and total endophytic assemblages of Arabidopsis thaliana. Plant Soil 405(1-2):325–336Google Scholar
- Truyens S, Weyens N, Cuypers A, Vangronsveld J (2015b) Bacterial seed endophytes: genera, vertical transmission and interaction with plants. Environ Microbiol Rep 7(1):40–50Google Scholar
- Truyens S, Beckers B, Thijs S, Weyens N, Cuypers A, Vangronsveld J (2016) Cadmium-induced and transgenerational changes in the cultivable and total seed endophytic community of Arabidopsis thaliana. Plant Biol (Stuttg) 18(3):376–381Google Scholar
- Weyens N, Taghavi S, Barac T, van der Lelie D, Boulet J, Artois T, Carleer R, Vangronsveld J (2009c) Bacteria associated with oak and ash on a TCE-contaminated site: characterization of isolates with potential to avoid evapotranspiration of TCE. Environ Sci Pollut Res Int 16:830–843PubMedCrossRefGoogle Scholar
- Weyens N, Gielen M, Beckers B, Boulet J, Lelie D, Taghavi S, Carleer R, Vangronsveld J (2014) Bacteria associated with yellow lupine grown on a metal‐contaminated soil: in vitro screening and in vivo evaluation for their potential to enhance Cd phytoextraction. Plant Biol 16:988–996PubMedCrossRefGoogle Scholar
- Yousaf S, Andria V, Reichenauer TG, Smalla K, Sessitsch A (2010) Phylogenetic and functional diversity of alkane degrading bacteria associated with Italian ryegrass (Lolium multiflorum) and Birdsfoot trefoil (Lotus corniculatus) in a petroleum oil-contaminated environment. J Hazard Mater 184:523–532PubMedCrossRefGoogle Scholar