Early Growth Stage Root-Associated Endophytes Isolated From Ulex Europaeus L. (Fabaceae) Colonizing Rural Areas in South-Central Chile — Source link


 Background and aimsUlex europaeus L. (Fabaceae), commonly known as gorse, is an invasive woody shrub that easily grows in several locations across the world. However, little is known about the interactions of this invasive species with soil microorganisms and how these microbes can promote reaching rapid grow-rates at early stages of development. We aim at characterizing the endophytic fungal and bacterial microbiota associated with roots of early growth stage U. europaeus colonizing native ecosystems in south-central Chile.MethodsRoot-associated microorganisms were isolated and identified using standard molecular techniques. Furthermore, plant growth-promoting traits were studied and biocontrol activity was assessed to characterize the early growth stage root-associated taxa. ResultsFour endophytic fungi belonging to Sordariomycetes and twelve bacteria assigned to Proteobacteria and Actinobacteria were identified as the principal early growth stage root-endophytic taxa. Plant growth-promoting traits were detected in several isolates such as Fusarium acuminatum and Rhodococcus sp. Besides, some of the isolates such as Rhodococcus sp. and Purpureocillium lilacinum showed biocontrol potential against phytopathogenic fungi. ConclusionsOur results demonstrate that early growth stage root endophytic taxa associated with U. europaeus have beneficial plant growth-promoting traits that can contribute with the rapid growth-rates of the shrub. The interaction with a set of beneficial microorganisms is an additional mechanism to explain the ability of U. europaeus for colonizing in various ecosystems.


Introduction
Ulex europaeus (Fabaceae) is a shrub species having remarkable ecological features and having several impacts in human-dominated landscapes. The species reaches about 3 meters in height, is spiny, has yellow owers, and densely cover extended areas (Gränzig et al. 2021;Quiroz et al. 2009). Besides, U. europaeus can x Nitrogen, grow fast, reproduce by seeds that can persist for several years, and live from sea level up to 4000 m in elevation and at several latitudes (Christina et al., 2020). Nonetheless, U. europaeus is classi ed as an invasive species in several regions of the world (Bowman et al., 2008). Once U. europaeus can establish in a location, it can alter the native species diversity, causing loss of threatened species and altering economic activities (Christina et al., 2020;Hornoy et al., 2013). Norambuena et al. (2001) pointed out that U. europaeus was introduced in Chile for using it as a hedge plant in agricultural lands to keep out livestock and as a fodder source.
The capacity of U. europaeus to handle stressing conditions, successful germination and rapid biomass production is often linked to the ability of the plants from the Fabaceae family to establish symbiotic associations with arbuscular mycorrhizal fungi and symbiosis with nitrogen (N)-xing bacteria ( (Khare et al., 2018). Therefore, due to its rigorous invasiveness, it is expected that speci c interactions with soil microorganisms at early developmental stages can contribute to the rapid growth and development of this invasive shrub.
The aim of this study was to isolate root-associated microorganisms from U. europaeus plantlets colonizing native ecosystems in south-central Chile and to screen for their plant growth promoting capabilities. As far as we know, this is the rst study of early growth stage endophytic interactions of the invasive shrub U. europaeus colonizing ecosystems in the southern hemisphere.

Materials And Methods
Early growth stage U. europaeus plantlets were sampled in agricultural rural areas in a segment of the Nahuelbuta Mountain Range, south-central Chile (38°34'S 72°56'W). The plantlets (n = 20; 3 cm tall) were found in 4 sampling points 2 months after a cleaning and burning treatment of an agricultural soil. The soil samples were collected at a depth of 20 cm for chemical analysis (Fuentes et al., 2020).
The aerial parts of the plantlets were removed and the roots were intensively washed under running tap water to remove the rhizosphere soil. Then, the roots were deposited in 50 ml Falcon tubes and washed ve times with deionized water in a laminar ow cabinet. The cleaned roots were subsequently deposited in sterile 50 ml Falcon tubes and surface disinfected according to Herrera et al. (2019b), where the roots were immersed in 50 ml of a disinfection solution (30 ml of sterile distilled water, 10 ml of sodium hypochlorite and 10 ml of 100 % alcohol) for 5 min, followed by ten washes in sterile deionized water. An aliquot of the last wash was plated in potato dextrose agar (PDA) and Luria Bertani agar (LBA) to rule out the presence of rhizospheric microorganisms in root surface. Six surface-sterilized root segments were placed in Petri dishes containing PDA media supplemented with streptomycin (100 mg l − 1 ), Murashige and Skoog basal medium, oatmeal agar (4 g of oats l − 1 , 10 g of agar, pH 5.6) supplemented with benomyl (4 mg l − 1 ) to reduce growth of ascomycetes (Bruzone et al., 2015), and LBA supplemented with cycloheximide (100 mg l − 1 ). The Petri dishes were incubated in darkness at room temperature until no new fungal and bacterial colonies were detected. Individual bacterial and fungal strains were puri ed in LBA and PDA, respectively, and classi ed according their phenotypic characteristics (i.e., growth rate, color, texture, colony border). Puri ed colonies were stored in individual plates at 4°C and periodically subcultured.
Liquid cultures of the puri ed fungal and bacterial strains were performed in potato dextrose broth (PDB) and Luria Bertani broth (LBB) respectively to perform DNA extraction. For fungi, six fungal squares (~ 0.5 cm) were inoculated in 100 ml Erlenmeyer asks containing 40 ml of PDB and incubated for 3 weeks in darkness at room temperature and in an orbital shaker at 150 rpm. An aliquot of 8 ml of medium containing fungal mycelia was taken to store puri ed strains at -80°C in glycerol. The rest of the media were ltered and the mycelia were used for DNA extraction using the E.Z.N.A.® HP Fungal DNA Kit (Omega Bio-tek, Norcross, GA, USA), according to the manufacturer's recommendations. For bacteria, the puri ed strains were cultured in 15 ml sterile Falcon tubes containing 5 ml of LBB and cultured in darkness at room temperature and in an orbital shaker at 150 rpm. An aliquot of 800 µl was taken to store the puri ed strains at -80°C in glycerol. DNA extraction was performed from 1 ml of the liquid culture using the Wizard® Genomic DNA Puri cation Kit (Promega, Madison, WI, USA), according to the manufacturer's recommendations. DNA integrity was checked in a 1 % agarose gel, quanti ed using the Qubit uorometer (Thermo Fisher Scienti c, Waltham, MA, USA) and standardized to 20 ng µl − 1 .
The molecular identi cation of fungal strains was performed based on the nucleotide sequence of the internal transcribed spacers of the 18S rRNA gene, ampli ed by PCR using the ITS1 and ITS4 primers (White et al., 1990) following the PCR conditions reported in Herrera et al. (2020b). Similarly, bacterial strains were identi ed based on the partial 16S rRNA gene sequence, ampli ed by PCR using the 27F and 1942R primers (Miller et al., 2013) according to the PCR conditions detailed in Herrera et al. (2020a). The PCR amplicons were checked in a 1.5 % agarose gel, quanti ed in a Qubit uorometer (Thermo Fisher Scienti c) and sequenced at Macrogen (Seoul, South Korea). The nucleotide sequences were compared with those in the GenBank database of the National Center for Biotechnology using BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi), accepting genus at an identity match greater than 95 % and species at an identity greater than 99 %, as suggested by Chen et al. (2011). The sequences were aligned using the ClustalX software with default conditions for gap opening and gap extension penalty (Larkin et al., 2007) and non-conserved regions were removed using the BioEdit software v7.2 (Hall, 1999). Operative taxonomic units (OTUs) were assigned at 97 % sequence similarity. The nucleotide sequences obtained were submitted to the GenBank database under the codes MW599973 to MW599982 for bacteria and MW604808 to MW604810 for fungi.
Screening of plant growth-promoting traits of the root-associated microorganisms was performed following standard procedures. The capacity to utilize tricalcium phosphate on agar, indole acetic acid (IAA) production and siderophore production were screened as reported in Soto et al. (2019). Brie y, microorganisms were assayed on Pikovskaya agar plates and incubated in darkness at 26 ± 2°C for 7 days. A clear halo around cultures indicated solubilization of tricalcium phosphate (Ca 3 (PO 4 ) 2 ). For IAA, microorganisms were cultured in LB or potato dextrose broth 1/7 strength, supplemented with 0.5 mg ml − 1 of L-tryptophan and then incubated in darkness, under stirring at 150 rpm and 26 ± 2°C for 5 days.
After incubation, Salkowski's reagent was added to the cell suspension and measured at 530 nm in a BK-UV1800 spectrophotometer (Biobase, Jinan, China). To determine siderophore production, the isolates were cultured in chrome azurol S (CAS) agar for 5 days, and CAS reaction was determined by color change in the blue CAS agar. Production of exopolysaccharides (EPS) was evaluated following Freeman et al. (1989), where isolates were streaked onto Congo red agar plates and incubated in darkness at 26 ± 2°C for 48 hr. The EPS production was detected by variation in colony color. For ammonia production, microorganisms were grown in 4 % peptone broth and incubated for 7 days in darkness under stirring at 28 ± 2°C. After incubation, Nessler's reagent was added to cell suspension and measured at 450 nm in a BK-UV1800 spectrophotometer (Biobase) (Bhattacharyya et al., 2020). The 1-aminocyclopropane-1carboxylic acid (ACC) deaminase activity was measured according to the free living bacteria method described by Brígido et al. (2015). Brie y, isolates were cultured in tryptic soy broth overnight in darkness, at 26 ± 2°C and 150 rpm, and then collected by centrifugation. The cell pellet was washed twice with Dworkin and Foster (DF) salts minimal medium (without a N source) and re-suspended in DF salt minimal medium with 3 mM ACC for 24 h. Cultures were collected by centrifugation and the cell pellet was used for enzyme activity. The absorbance was measured at 540 nm in a BK-UV1800 spectrophotometer (Biobase). Finally, a biocontrol assay was performed to evaluate the potential of the isolates against phytopathogenic fungal species (Fusarium oxysporum, Botrytis cinerea, Rhizoctonia solani, Phoma herbarum) following Jamali et al. (2020). Brie y, a 5 mm diameter disk of a fresh culture of phytopathogenic fungi was placed in the center of a nutrient agar-potato dextrose agar (1:1) mix plate. Then, bacterial isolates were streaked on both sides of the fungal inoculum at a similar distance of 25 mm and incubated for 7 days at 28 ± 2°C in darkness. For fungi, 5 mm diameter mycelia disks were placed at both sides of the phytopathogenic strains at a similar distance of 25 mm and incubated for 7 days at 28 ± 2°C in darkness. The percentage of inhibition was calculated using the following formula: where C is growth in mm in the control and T is growth in mm in the treatment with the isolates.
The quantitative data were analyzed by one-way ANOVA, establishing signi cant differences at p < 0.05. Post hoc pairwise comparisons were performed using Tukey's multiple range test. All statistical tests were conducted using the R software (R Core Team 2018; https://www.R-project.org).

Results
The soil chemical analysis showed that the sampling sites had similar content of nitrogen, phosphorous, potassium soil organic matter and pH ranging from 5.3 to 5.8. High extractable aluminum levels were detected in the sampling points, with values ranging from 386 to 526 mg kg − 1 (Table 1). A total of 79 bacterial colonies were isolated from the analyzed roots, which were separated into 12 different strains based on the morphological characteristic of the colonies and growth rate in culture media. The molecular identi cation of the isolates revealed 9 OTUs, with a dominance of the phyla Proteobacteria and Actinobacteria (Table 2). Speci cally, the isolates UB7 (Novosphingobium sp.), UB9 (Herbaspirillum sp.), UB2 (Paraburkholderia strydomiana), UB4 (Pseudomonas sp.) and UB10 (Herbaspirillum rhizosphaerae) were assigned to taxa included in the phylum Proteobacteria (Table 2). Similarly, the isolates UB5 (Rhodococcus sp.), UB1 (Terrabacter aerolatus) and UB11 (Jatrophihabitans sp.) were assigned to taxa included in the phylum Actinobacteria ( Table 2). The isolates UB3 (Flavobacterium sp.) and UB6 (Paenibacillus odorifer) were assigned to the phyla Bacteroidetes and Firmicutes, respectively ( Table 2). Most of the bacterial sequences were assigned to different OTUs, excluding isolates UB9 and UB10 (Herbaspirillum spp.). Isolates UB8 and UB12 were classi ed as unidenti ed bacteria (with no signi cant match in the GenBank database) ( Table 2). A total of 18 fungal strains were isolated, which were classi ed into 6 different isolates based on the morphological characteristics of the fungal strains. The molecular identi cation revealed 3 OTUs and showed Ascomycetes as the principal endophytic phylum associated with U. europaeus. Speci cally, the isolates UF1 (Fusarium acuminatum), UF2 (Purpureocillium lilacinum) and UF6 (Acremonium alternatum) were related to taxa belonging to the phylum Sordariomycetes ( Table 2). The isolate UF5 was classi ed as unidenti ed endophytic fungus (without signi cant match in the GenBank database) ( Table 2). Isolates UF3 and UF4 did not grow in synthetic media after initial extraction and puri cation.
The screening of plant growth-promoting traits showed diverse capabilities associated with the different isolates ( Table 3). Solubilization of Ca 3 (PO 4 ) 2 in agar was detected only in P. strydomiana, whereas for fungi P. lilacinum and unidenti ed fungi (isolate UF5) showed Ca 3 (PO 4 ) 2 solubilization capability ( Table 3). Production of EPS was detected in P. lilacinum, P. odorifer, P. strydomiana and unidenti ed bacteria isolate UB8, whereas siderophore production was detected in P. strydomiana and P. lilacinum (Table 3). Almost all tested isolates showed ammonia production capability (14 out 16), being signi cantly high in the fungal isolate F. acuminatum (Fig. 1). The IAA production was similar for all bacterial isolates with higher values in Pseudomonas sp., whereas for the fungus F. acuminatum, it showed the highest production values detected in the study (Fig. 1). The ACC deaminase activity was signi cantly higher in the unidenti ed bacteria isolate UB8 (Fig. 1), whereas no ACC deaminase activity was detected for fungi. Biocontrol potential against potential phytopathogenic fungal strains was detected in 6 out of 16 isolates, with inhibition percentages ranging from 65.8 ± 8.7 (Rhodococcus sp.) to 11.1 ± 1.9 (A. alternatum) ( Table 3). Speci cally, the isolates Rhodococcus sp., unidenti ed bacteria (isolate UB8), F. acuminatum and P. lilacinum showed the highest inhibition percentages against the phytopathogenic fungi (Table 3).   Early growth stage microbial interactions can play key roles in the establishment of invasive species in the ecosystems, contributing to stress tolerance, plant nutrition and development of the associated plants (Links et al., 2014;Rezki et al., 2018). In our study, we characterized a set of bene cial endophytes associated with initial developmental stages of U. europaeus plantlets, but the source of those microbial strains is unclear. Several microbial strains can be stimulated from the soil by root exudates or can be a component of the seed-associated microbiome, both in uencing the seedling survival, plant health and productivity (Nelson, 2018). Both, soil-borne and seed-associated microbes can be a potential source of the different taxa identi ed in our study, but further studies are necessary to de ne if U. europaeus can associate with soil microorganisms without speci city or if these microbes are mainly seed-associated endophytes. Despite the source of such bene cial microbial taxa, our results suggest that the presence of multiple microbial endophytes with different plant growth-promoting traits can be one of the mechanisms explaining the high growth rates and effective establishment of U. europaeus. In fact, Pitzschke (2018) reported that microbial endophytes contribute to a rapid seed germination and plantlet development of Chenopodium quinoa even under harsh environmental conditions, which is in line with our results revealing diverse bene cial endophytes colonizing early growth stage U. europaeus plantlets. Indeed, microbial endophytes have commonly been detected in association with invasive species, where a key role of such endophytic taxa has been suggested for plant growth promotion, stress tolerance and herbicide resistance of invasive species ( Plants from the Fabaceae family usually establish symbiosis with arbuscular mycorrhizal fungi and Nxing bacteria, which can confer nutritional and physiological bene ts on mature plants. As our study was based on culture-dependent methods, we only identi ed the N-xing nodulating bacterial genus Paraburkholderia, which has been described as a bene cial taxon associated with Mimosa pudica plants (Paulitsch et al., 2020). Such symbiotic microbial taxa have commonly been described as bene cial microorganisms, but other endophytic strains can also contribute to a successful plant establishment, especially at the beginning of the U. europaeus life cycle. In this sense, we have provided evidence about bene cial attributes of Fusarium endophytic strains associated with U. europaeus in biological control of phytopathogenic fungi and plant-growth promotion. Such fungal genus has commonly been reported as . Such increasing evidence about the biocontrol potential of endophytic strains may represent an opportunity to study the mechanisms underlining the high resistance of invasive species to phytopathogens. Therefore, the biocontrol potential can be an indirect mechanism by which endophytic microbial strains contribute to the successful and rapid growth of U. europaeus plantlets.
Andisol soils, where U. europaeus was sampled, have high levels of available aluminum that interfere with the normal growth and development (Mora et al., 2017). However, such high levels of aluminum seem not to be a problem for U. europaeus plantlets. It is expected that effective tolerance mechanisms can support colonization of acidic Andisols. One of the tolerance mechanisms can be related with speci c interactions with microbial strains which can confer metal tolerance to their associated plants . This is the case of Rhodococcus sp., which has been described as a microbial strain with high tolerance to metal(loid)s ( Our study results provide evidence of novel endophytic interactions of U. europaeus with soil-borne microorganisms, from which their plant growth-promoting traits as well as their biocontrol potential can be an additional mechanism to explain the high growth rates and establishment of U. europaeus. However, it is unclear if the high diversity of endophytic interactions is induced by the ability of U. europaeus to select bene cial soil microorganisms from the bulk soil, or if the microbial strains are part of the seed-associated microbiome. Additionally, microbial strains with multiple bene cial traits can be considered potential bioinoculants for improving the plant growth or screening for novel biocontrol strategies of phytopathogenic fungi.

Declarations Funding
This study was supported by the Fondo Nacional de Desarrollo Cientí co y Tecnológico of Chile [grant numbers 1211857 and 3200134].

Declaration of Competing Interest
The authors declare no con ict of interest. Figure 1 Ammonia production (a), indoleacetic acid (IAA) production (b) and 1-Aminocyclopropane-1-carboxylate (ACC) deaminase activity (c) of endophytic microorganisms isolated from young Ulex europaeus