Isolation and prospection of diazotrophic rhizobacteria associated with sugarcane under organic management

Microorganisms associated with organic management are essential in nutrient transformation and release for plant use. The present study aimed to isolate, identify and characterize plant growth promoting diazotrophic rhizobacteria associated with sugarcane under organic management. Rhizospheres of organic sugarcane varieties IAC 911099 and CTC4 were sampled and inoculated onto nitrogen free NFb and Burk media. The isolated microorganisms were screened in vitro concerning their ability to produce plant growth promoting factors. Eighty-one bacteria were isolated; 45.6% were positive for the nifH gene and produced at least one of the evaluated plant growth promotion factors. The production of indole-3-acetic acid was observed in 46% of the isolates, while phosphate solubilization was observed in 86.5%. No isolates were hydrogen cyanide producers, while 81% were ammonia producers, 19% produced cellulases and 2.7%, chitinases. Microorganisms belonging to the Burkholderia genus were able to inhibit Fusarium moniliforme growth in vitro. Plant growth promoting microorganisms associated with organic sugarcane, especially belonging to Burkholderia, Sphingobium, Rhizobium and Enterobacter genera, can be environmentally friendly alternatives to improve sugarcane production.


INTRODUCTION
The use of petroleum-derived fertilizers and chemical pesticides is a common practice in agriculture.In view of the current scenario, the use of these inputs is indispensable, making cultivation expensive and, associated with the expansion of the agricultural areas, causing a series of environmental impacts (Goldemberg et al. 2008, Severiano et al. 2009, Schultz et al. 2014).In this sense, a growing interest in cheaper and less environmentally ARIANA A. RODRIGUES et al. impacting alternatives that reduce or replace the use of these implements in traditional agriculture has been observed (Crowder andReganold 2015, Kanchiswamy et al. 2015).
The use of plant growth promoting rhizobacteria (PGPR) is one of the available alternatives to reduce the use of synthetic fertilizers and products from non-renewable sources, presenting low production/implantation costs (Sadeghi et al. 2012, Szilagyi-Zecchin et al. 2014).PGPR correspond to bacteria that inhabit the rhizosphere, the soil zone immediately adjacent to the root, capable of promoting plant growth by one or more mechanisms (Noumavo et al. 2016).Their use is commonly adopted in traditional agriculture and empirically in organic agriculture, since this type of agriculture is extremely dependent on nutrient transformations mediated by natural soil microbiota (Pariona-Llanos et al. 2010, Bhardwaj et al. 2014, Järvan and Edesi 2015).
PGPR can promote plant growth by increasing plant nutrient availability through nitrogen fixation, mineral solubilization and iron adsorption (Kukla et al. 2014).They can also produce phytostimulators, such as indole-acetic acid (IAA), gibberilinic acid, cytokinins and 1-aminocyclopropane-1carboxylate (ACC) deaminase, which stimulate root development and plant tissue growth (Sgroy et al. 2009).In addition, they play a fundamental role as biopesticides, through the production of metabolites that exhibit antibiotic action, and can also participate in the degradation of organic pollutants (Ahemad and Kibret 2014).
Due to the biological importance of nitrogen in biomolecule constitution and the high costs and environmental impacts caused by the production of synthetic nitrogen fertilizers, nitrogen-fixing bacteria (diazotrophic bacteria) are targets of interest (Karagöz et al. 2012).These microorganisms are able to convert molecular nitrogen (N 2 ) into ammonia (NH 3 ), a nitrogenous form absorbed by plants.This conversion is mostly mediated by the iron-molybdenum (FeMo) nitrogenase enzyme, which is encoded by the nif complex genes (nifH, nifD, nifK).Because they are highly conserved throughout species, genes from the nif complex can be used as indirect evidence of the presence of nitrogenase and as molecular markers for diazotrophic organism screening (Dai et al. 2014, Ji et al. 2014, Kumar 2014, MacKellar et al. 2016).
Diabotrophic PGPR associated with sugarcane present high taxonomic diversity, with several noteworthy genera, such as Herbaspirillum, Pantoea, Burkholderia, Azospirillum and Glucononacetobacter (Lin et al. 2012, Gopalakrishnan et al. 2015, Tam and Diep 2015).In this scenario, several studies have been conducted to isolate diazotrophic bacteria associated with sugarcane under traditional management and to characterize plant growth promoting factors produced by these organisms (Luvizotto et al. 2010, Lima et al. 2011, Beneduzi et al. 2013, Kruasuwan and Thamchaipenet 2016).
However, research on diazotrophic PGPR associated with sugarcane under organic management is scarce.This group is of great interest, since one of the major challenges of organic agriculture is soil nitrogen limitation, as synthetic nitrogen compounds are abolished (Wongphatcharachai et al. 2015).Thus, PGPR presenting multiple plant growth promotion factors can become a viable alternative in biofertilization, phytostimulation and biocontrol practices in organic sugarcane plantations.
In this context, the present study aimed to isolate and identify diazotrophic rhizobacteria associated with sugarcane under organic management from two sugarcane processing plants located in the state of Goiás, Brazil, and to evaluate plant growth promotion factor production.

MICROORGANISM OR BACTERIA ISOLATION
The microorganism/microbial isolations were conducted from rhizospheric soil obtained from to verify possible bacterial morphotypes isolated simultaneously in the two media.After obtaining pure colonies, the samples were stored in 20% glycerol (m/v) at -20 ºC.

DNA EXTRACTION
DNA extraction was performed from microorganism growth in Luria-Bertani broth (LB) for 24 hours at 30 °C.MoBio's UltraClean®Microbial DNA Isolation Kit was used for DNA extraction, according to the manufacturer's recommendations.

nifH GENE AMPLIFICATION
The gene encoding the nitrogenase reductase subunit of the enzyme nitrogenase complex (nifH) was amplified to confirm diazotrophic potential.The PCR reaction of the nifH gene was conducted using the oligonucleotide primers PolF (TGCGAYCCSAARGCBGACTC) and PolR (ATSGCCATCATYTCRCCGGA), which generate a product with approximately 360 base pairs (Poly et al. 2001).
The reaction was prepared to a final volume of 10 μl, consisting of 1 x Taq polymerase buffer, 2.3 mM MgCl 2 , 0.25 mM of each dNTP, 1.0 μM of each primer, 0.2 U of Taq DNA polymerase and 1.0 μl of template DNA (50 ng).Amplification was conducted in a thermal cycler by the following cycling steps: initial denaturation at 94 °C for 5 minutes; followed by 35 denaturation cycles at 94 °C for 45 seconds, annealing at 57 °C for 30 seconds, extension at 72 °C for 30 seconds and a final extension step at 72 °C for 1 minute.The reaction products had their integrity confirmed on 1.2% (m/v) agarose gels.The nitrogen fixing strain Azospirillum brasilense was used as a positive control.

DIAZOTROPHIC RHIZOBACTERIA IDENTIFICATION
All bacteria whose nifH gene can be amplified were identified from the partial sequencing of the 16S two sugarcane varieties (IAC 911099 and CTC4), grown under organic management at farms belonging to two sugarcane processing plants in the cities of Goiatuba (19º 00' 023" S; 049º 40' 319" W) and Goianésia (15º 20' 241" S; 048º 54' 253" W), Goiás, Brazil.Samplings were carried out from October/2015 to January/2016, and each variety was obtained from four distinct farms, with three replications.Farms with sugarcane specimens planted in different years (2009, 2012, 2013 and 2014) and in different phenological stages were chosen, in order to obtain the greatest possible sample heterogeneity.
Sugarcane roots with adhered rhizospheric soil were conditioned in sterile plastic bags and stored at 4 ºC until processing the following day.The rhizospheric soil was considered as the soil intimately connected to the roots at a distance of up to 5 mm, carefully removed with the aid of a sterilized spatula (Santos et al. 2012).
The media were then incubated at 30 °C for 14 days.After growth, in the form of a diffuse film below the surface of the semi-solid medium, the microorganisms were peeled to new semi-solid media.After the growth in semi-solid media (three rounds), the microorganisms were transferred to the corresponding solid media.Microorganism isolation and purification occurred by depletion in solid media.Microorganisms whose colonies displayed the same morphology isolated from the same farm were discarded.After isolation, all colonies were transferred to nutrient agar plates rRNA gene.The reaction was prepared to a final volume of 50 μl, containing 1 x Taq polymerase buffer, 1.5 mM MgCl 2 , 0.2 mM of each dNTP, 0.2 mM of each primer, 2.5 U of Taq DNA polymerase and 1 μl of template DNA (50 ng).The 16S region was amplified using primers 27F (5'-AGA GTT TGA TCC TGG CTC AG-3') and 1541R (5'-AAG GAG GTG ATC CAG CC-3') (Lane 1991).DNA was amplified in a thermal cycler, under the following conditions: initial denaturation at 95 °C for 3 minutes; followed by 30 denaturation cycles at 94 °C for 1 minute, annealing at 55 °C for 30 seconds, extension at 72 °C for 30 seconds and a final extension step at 72 °C for 10 minutes.
The amplified products had their integrity confirmed on 1.2% agarose gels and purified with isopropanol and ethanol using the method described by Sambrook and Russell (2001).Sequencing was conducted on the ABI 3130xl platform (Applied Biosystems), using the BigDye terminator cycle sequencing kit and the oligonucleotide primers 27 F, 530F (5'-TGA CTG ACT GAG TGC CAG CMG CCG CGG-3') and 519R (5'-GTN TTA CNG CGG CKG CTG -3') (Lane 1991).
The reads obtained from the sequencing of each microorganism were evaluated regarding quality.Subsequently, the sequences (reads) were joined in a single sequence (contig).These steps were performed using the CodonCode Aligner software, version 6.0.2.Then, the sequences were compared to those deposited at the NCBI (National Center for Biotechnology Information) 16S rRNA library using the BLAST tool (Altschul et al. 1990).The identified sequences were deposited at the NCBI 16S rRNA database under accession numbers MG429815 to MG429819; MG459255 to MG459472 and MG459472.
In vitro SCREENING FOR PLANT GROWTH PROMOTING TRAITS Microorganisms positive for the nifH gene were screened, in vitro, regarding their IAA, phosphate solubilization, ammonia, hydrogen cyanide (HCN), celulase and chitinase production capacity and antagonism to Fusarium moniliforme.All evaluations were carried out in triplicate and the results are expressed as the means of the replicates.

AMMONIA AND HCN PRODUCTION
F o r a m m o n i a p r o d u c t i o n e v a l u a t i o n , microorganisms were inoculated in peptone water and incubated at 30 °C for 7 days.After centrifugation at 10,000 xg for 10 minutes, the Nessler reagent (10% HgI 2 , 7.0% KI and 50% aqueous NaOH 32% solution) was added to the supernatants at a 2:1 (v/v) ratio (Cappuccino and Sherman 1996).The development of a brown color was considered indicative of ammonia production (Dey et al. 2004).
HCN production was determined according to the technique described by Cattelan (1999), with modifications.Isolates were initially grown for 24 hours in 10% tripticasein soybean (TSA) agar, supplemented with 0.4% glycine.After microorganism growth, filter papers (Whatman # 1) embedded in a picric acid solution (0.5% picric acid and 2.0% Na 2 CO 3 ) were deposited on the plate lids.The plates were then again incubated for 48 hours.Positive HCN production was considered when the filters became brownish (Walpola and Yoon 2013).

CHITINASE AND CELULASE PRODUCTION
The ability of the isolates to produce chitinases was verified according to Cattelan (1999).Briefly, microorganisms were cultured for 14 days at 30 ºC in nitrogen-free medium (MNL) supplemented with 8.0 g L -1 of colloidal chitin, 0.78 g L -1 NH 4 NO 3 and 15 g L -1 agar.The addition of chitin generates turbidity in the medium.A clear halo around the colony is observed when microorganism display the ability to degrade the chitin.
Cellulase production evaluation was carried out by culturing the microorganisms in minimal medium containing carboxymethylcellulose (CMC) (Stamford et al. 1998).After growth at 30 °C for 48 hours, the plates were covered with a 0.3% Lugol solution.The presence of light halos around the colony were indicative of cellulase production (Kasana et al. 2008).For all evaluated enzymes, the enzymatic index (EI) was determined, obtained by dividing the size of the halo by the size of the colony.

ANTAGONISM AGAINST F. moniliforme
Bacteria were evaluated, in vitro, regarding their ability to inhibit the phytopathogenic fungus F. moniliforme, according to the methodology described by El-Sayed et al. ( 2014), with modifications.The microorganisms were initially cultured in TSA broth at 30 °C for 48 hours.Subsequentyl, they were inoculated into two parallel striae, 1.5 cm distant from the plate border, in potato dextrose agar.A 5 mm diameter disc with fungus growth was placed in the center of the plate, previously cultured for 7 days at 30 °C in potato dextrose agar.After 7 days, the ability of the isolates to inhibit fungal growth was evaluated by measuring the distance between the bacterial colony and the fungal colony.A plaque without the bacterial inoculum was used as a negative control.

STATISTICAL ANALYSES
The data obtained from the in vitro tests were submitted to an analysis of variance (ANOVA) and means were compared applying the Scott-Knott test at a 5.0% significance using the Sisvar software, version 5.6 (Ferreira 2011).When necessary, data ARIANA A. RODRIGUES et al. that did not present homogeneous variances or normal distribution were log corrected.

MICROORGANISM ISOLATION AND IDENTIFICATION
Eighty-one sugarcane rhizosphere bacteria were isolated from sugarcane under organic management in nitrogen-free culture media.For the purposes of isolate nomenclature, the first letter corresponds to the isolation medium (B = Burk and N = NFB), the second, to the sugarcane variety (C = CTC4 and U = IAC 911099), and the numbers refers to farm and striking order, respectively (Table I).Regarding origin, 57.3% of the bacteria were isolated from the CTC4 variety and 42.7% from the IAC 911099 variety.The nifH gene was amplified in 45.6% of the isolates.All bacteria whose nifH gene could be amplified produced more than one mechanism to promote plant growth, characterizing them as rhizobacteria carrying multiple plant growth promoting factors.
The 37 diazotrophic microorganisms identified by the partial sequencing of the 16S rRNA gene were distributed throughout 13 genera (Table I).The size of the obtained fragments varied from 473 to 1467 base pairs, with similarity between 97 and 99% (Table I).
All microorganisms belonging to the Bacillus genus were isolated from the CTC4 variety and found in most of the investigated sugarcane processing farms.The Burkholderia genus, however, was most frequently associated with IAC 911099 variety (Table I).The microorganism Sphingobium yanoikuyae (BU32 and BU92) was isolated from different locations and in the same sugarcane variety, while Rhizobium tropici (BC91 and BU24) was isolated from different localities and varieties (Table I).
Ammonia production was detected in 81% of the isolates, being the second most common plant growth promotion factor, after phosphate solubilization.In addition to nitrogen fixation Cellulase producers accounted for 19% of the isolates, with enzymatic indices ranging from 1.5 and 3.41.Most of the cellulase-producing isolates belonged to the Bacillus genus (BC23, BC26, BC27, BC28 and NC27), but representatives of the Flavobacterium (NC26) and Sphingobium (BU32) genera were also proven as enzyme producers (Table II).Only Dyella sp.(BU96) was chitinaseproducing, with an enzyme index of 2.8 (Table II).
The ability to inhibit the pathogenic fungus F. moniliforme was observed for 21.7% of the isolates.All isolates displaying antagonism to the pathogenic fungus belonged to the Burkholderia genus.Isolates NC97, BU95, NU21, NU31 and NU45 exhibited the greatest distances between the fungal and bacterial colonies (Table II).It is noteworthy that not all microorganisms that presented fungus antagonism were indirect plant growth mechanisms producers (ammonia, HCN, cellulases and chitinases), responsible for pathogen attacks.

DISCUSSION
In the present study, 81 microorganisms from sugarcane under organic management were isolated using NFb and Burk nitrogen-free semi-solid media.The bacteria were isolated based on their morphological diversity, which takes into account characteristics such as color, contour and elevation wich allows to classify the bacteria differentially (Di Franco et al. 2002).This strategy was used to capture the largest number of isolates belonging to different species and to maximize resources.On the other hand, when discarding morphologically equivalent microorganisms, some species may have been underestimated, since there is no genetic evidence that two similar colonies are of the same species (Lebaron et al. 1998).
Although the NFb medium facilitates the isolation of bacteria belonging to the Azospirillum genus (Kuss et al. 2007), several studies using this medium have described the isolation of other genera, such as Gluconacetobacter, Herbaspirillum, Burkholderia, Bacillus, Enterobacter, Klebsiella, Pantoea and Pseudomonas (Ambrosini et al. 2012, Tam andDiep 2015).On the other hand, Burk's medium is not specific, being associated to a range of fixing species (Park et al. 2005).Semisolid nitrogen free media are considered ideal for diazotrophic bacteria isolation and screening.The reduced agar concentration determines an ideal microaerophilic condition for nitrogenase action, reduced in high oxygen concentrations (Baldani et al. 2014).
The isolation and application of diazotrophic bacteria in crops under organic regimes is extremely interesting from an agricultural point of view, since the great limitation of organic crops is precisely the available nitrogen supply.Diazotrophic microorganisms could be applied as substitutes to synthetic fertilizers (Wongphatcharachai et al. 2015) or in association with animal and vegetable manure, usually used as nitrogen sources in organic crops (Hirel et al. 2011).
The nifH gene, being conserved throughout species, is often used as molecular marker for the screening of nitrogen-fixing bacteria (Raymond et al. 2004, Gaby andBuckley 2012) and was amplified in 45.6% of the isolates detected in this study.The fact that the nifH gene has not been amplified, however, does not mean that the microorganism is not fixative.The sequence of the main widely used primers, mostly constituted by degenerate bases, are designed to flank the same region or overlapping sites, restricting their reach (Poly et al. 2001, Zehr et al. 2003).In addition, not all organisms exhibit iron-molybdenum nitrogenase (nif), since other nitrogenases also exist, differentiated by the metal IAA= production of 3-indole-acetic acid; Phosphate = phosphate solubilization; Amo = Ammonia; Cel = Cellulase; Chit = Chitinase; Ant = antagonism. 1. Means followed by the same letters do not differ statistically by Scott-Knott's test at a 5% probability. 2. The expressed values correspond to the enzymatic index, obtained by dividing the diameter of the halo by that of the colony.The "-" label indicates bacterial growth without halo formation and label "+" indicates a positive test. 3. The values presented correspond to the distance between the fungal and bacterial colony.The label "-" indicates that the bacterium was not antagonistic to the fungus.
Other microorganisms are capable of growing in nitrogen-free media and are not fixers.It has been suggested that these organisms may incorporate ammonia or other nitrogen species at very low concentrations, or that they are capable of using fractions fixed by other organisms (Beneduzi et al. 2013, MacKellar et al. 2016).In addition, we emphasize that the amplification of the nifH gene has a confirmatory character, but cannot quantify the amount of fixed nitrogen, and the use of techniques such as acetylene reduction (ARA) and the incorporation of the 15 N 2 isotope is recommended (Gtari et al. 2012).
The genus most detected in this study, Burkholderia, is widely distributed in nature and is often found associated with sugarcane under traditional management (Luvizotto et al. 2010, Castro-Gonzalez et al. 2011).From the point of view of organic agriculture, the Burkholderiaceae family and nifH genes have been reported as being more abundant in agricultural soils abundant in swine manure (Xun et al. 2018).In addition to their ability to fix nitrogen, it is also suggested that plant growth promotion performed by these organisms is associated with the production of IAA, phosphate solubilization and in vitro pathogen inhibition (Paungfoo-Lonhienne et al. 2014).
The genus Bacillus, the second most detected in the present study, has also been described in association with sugarcane under traditional management and presents potential regarding the production of enzymes of agronomic interest (Beneduzi et al. 2008, Madhaiyan et al. 2011, Pisa et al. 2011).The other genera found herein have been mostly associated with sugarcane under traditional management or cultivated soils (Nicolaisen et al. 2008, Burbano et al. 2011, Anand et al. 2012, Bers et al. 2012, Santi Ferrara et al. 2012, Fischer et al. 2012, Lin et al. 2012, Hoang et al. 2015, Solanki et al. 2017).
The presence of Bacillus and Burkoholderia as more predominant groups was also observed by Xia et al. (2015), when evaluating the taxonomic distribution of maize, melon, pepper and tomato bacteria in organic systems.Oliveira (2009) using selective media for nitrogen fixers, including NFb, observed a range of microorganisms associated with sugarcane plantations under organic management in São Paulo, Brazil, with emphasis on Enterobacter, Klebsiella, Pseudomonas, Burkholderia and Beijerinckia species.Those authors also detected other species, such as Agrobacterium, Azozpirillum, Bosea, Bradyrhizobium, Brucella, Cohnella, Erwinia, Gluconacetobacter, Rhizobium, Stenotrophomonas, Variovorax and Xanthomonas.The similarities and divergences of this study in relation to the study conducted by Oliveira (2009) reinforce the idea that, even if Enterobacter, Burkholderia and Rhizobium species are ubiquitous, factors such as plant age, variety, root exudates, management and soil chemical and physical attributes can influence the community distribution and species (Beneduzi et al. 2008, 2013, Santi Ferrara et al. 2012, Noumavo et al. 2016).Therefore, for each evaluated region, distinct distribution patterns of diazotrophic microorganisms are observed, which is interesting regarding PGPR prospecting.
Less than half of the evaluated isolates produced IAA, and a broad taxonomic variety was observed among the best producers.In addition, some individuals belonging to the same genus produced distinct IAA values and were classified into different groups, according to Scot-Knott's test at a 5.0% probability.Enterobacter sp.NC25 and NU33 strains produced 80.42 and 77.41 mg mL -1 , respectively, similar to the strains reported by Rodrigues et al. (2016) when evaluating Enterobacter sp.rhizospheres from traditionally managed sugarcane.
Although IAA production is commonly found among rhizobacteria (Sagar et al. 2017), different types of fertilizers impact on the number of bacteria in the soil and the diversity of bacteria producing this compound.In systems where, chemical fertilizers are withdrawn or where organic fertilizers are applied, the number and diversity of IAA-producing bacteria increases.The use of organic fertilizers improves the structure of the microbial community, in addition to selecting beneficial microorganisms (Yuan et al. 2011, Duangpaeng et al. 2012).
In addition to being the most active auxin, it is believed that IAA is physiologically active even at low concentrations, so even organisms that produce small IAA fractions may be able to contribute to plant health (Hayat et al. 2010, Vejan et al. 2016).IAA production leads to increased lateral roots, in addition to increasing root surface area and size, leading to increased water and nutrient supplies to plants (Brandl andLindow 1998, Vessey 2003).
The ability to solubilize phosphate was the most common feature of the diazotrophic bacteria evaluated herein, with emphasis on bacteria belonging to the Burkholderia genus.In the case of sugarcane rizobacteria under traditional management, Inui-Kushi et al. (2012) verified that Burkholderia bacteria displayed with the highest phosphate solubilization index among 10 different bacteria belonging to the same genus.
Although not a mandatory feature of the Burkholderia genus, several species belonging to this genus have been associated with varying phosphate solubilization and rhizosphere acidity levels (Castro-Gonzalez et al. 2011).When studying the impact of organic management on the diversity of pepper phosphate-solubilizing bacteria (Capsicum frutescens L. cv.Hua Rua), Surapat et al. (2013) observed that most of their phosphate-solubilizing isolates were recovered from organic systems, and thet most belonged to the Burkholderia genus.
The most well-known phosphate solubilization mechanism involves medium acidification.Plants display the ability to acidify the rhizosphere, leading to phosphate solubilization, although this capacity is quite limited (Hamdali et al. 2008).It is suggested that even the phosphate solubilized in small amounts by bacteria still corresponds to a large part of the soluble phosphate available to plants, especially in organic crops, that are largely dependent on the natural soil microbiota and that display high amounts of organic phosphates (Pariona-Llanos et al. 2010).
Ammonia production is quite common among rhizobacteria, as observed in this study and in studies conducted by Joseph et al. (2007) and Gayathri et al. (2010).HCN production is rare, a fact corroborated by Kavamura et al. (2013) when evaluating bacteria obtained from cacti.Ammonia and HCN are volatile compounds that act as biocontrol agents.In addition to acting as pathogen responses, it is believed that ammonia can be used as a source of nitrogen supplementation by the host plant (Joseph et al. 2007, Marques et al. 2010, Passari et al. 2015).A study conducted by Marques et al. (2010) observed that ammonia production was positively correlated to nitrogen and phosphorus accumulation, root and stem elongation and biomass increases in maize.
The present study observed that most cellulaseproducing isolates belonged to the Bacillus genus, similarly to the results reported by Zhao et al. (2015).Cellulase-producing microorganisms belonging to the Bacillus genus are widely distributed in the sugarcane rhizosphere.They are believed to participate in the degradation of cellulolytic debris originating from the plant itself and, consequently, in element cycling (Ratón et al. 2012).In addition to potential for pathogen inhibition, cellulase production in rhizobacteria plays an important role in the penetration of these organisms into the host plant during colonization (Pariona-Llanos et al. 2010).ARIANA A. RODRIGUES et al.
The BU96 isolate from the Dyella genus isolated herein was the only isolate able to degrade chitin.Microorganisms belonging to this genus have been isolated from soil and described as chitindegrading agents (Lee and Lee 2009).Although the presence of chitinolytic bacteria is ubiquitous, it is estimated that only 1% of rhizobacteria isolated from wheat, rice and maize present chitinolytic potential (Someya et al. 2011), which reinforces the low number of chitinolytics found in this study.The lytic enzymes cellulases and chitinases were selected for this screening since they are considered limiting factors for the growth of pathogenic fungi, as their lytic action is capable of causing degradation of the structural matrix of the fungal cell wall (El-Sayed et al. 2014).
Sugarcane is attacked by several pathogens such as viruses, bacteria, nematodes, insects and fungi.These organisms cause a number of productivity losses (Zhang et al. 2015).Among the main fungi species that attack sugarcane is F. moniliforme, which causes changes in leaf and stem morphology in non-pest-tolerant varieties (Lin et al. 2014(Lin et al. , 2015)).In this sense, rhizobacteria have been reported as biocontrol agents against pathogenic F. moniliforme strains (Hebbar et al. 1992, Figueroa-López et al. 2016).
Bacteria belonging to the Burkholderia genus are noteworthy with regard to the inhibition of the F. moniliforme pathogen.Several bacteria belonging to this genus isolated from different environments, such as tomato (Omar et al. 2006), barley (Simonetti et al. 2018) and soil in the Amazon region (Silva et al. 2012), are also capable of inhibiting the growth of phytopathogenic Fusarium strains.Some microorganisms belonging to this genus are believed to produce the antibiotic pyrrolnitrine, which is a fungus respiratory chain inhibitor (Parke and Gurian-Sherman 2001).
In vitro antagonistic test is important for the detection of potential biocontrol agents, especially in the presence of numerous isolates.However, theres is no guarantees that microorganisms producing antifungal substances in vitro will produce the same effects when inoculated in plants (Knudsen et al. 1997).In this sense, Shehata et al. (2016), simultaneously evaluated, in vitro and in vivo, the antifungal activity of 190 endophytic bacteria against the pathogen Sclerotinia homoeocarpa.These authors observed that out of 5 positive microorganisms in the in vitro tests, only 3 had in vivo activity.To produce antagonistic effects in vivo, microorganisms must be able to properly colonize the plant, settle in the plant tissue affected by the pathogen and survive the competition with the host's natural microbiota (Pliego et al. 2011, Deketelaere et al. 2017).
It is recommended that the inoculant candidate presents the highest number of in vitro plant growth promoting traits, since this would lead to a higher probability of survival in the field, due to their ability to use several substrates (Rana et al. 2011).In this sense, the isolates: Sphingobium yanoikuyae (BU32), Rhizobium sp.(NC24), Enterobacter sp.(NC25) and Enterobacter sp.(NU33) are noteworthy in this study due to their ability to produce IAA, as well as the ability to solubilize phosphate, and microorganisms belonging to the Burkholderia genus, namely NC97, BU95, NU31, NU21 and NU92, due to their simultaneous ability to solubilize phosphate and promote biocontrol.In addition, it is suggested that these bacteria can be applied in mixed form in greenhouse and field evaluations in subsequent studies.

ACKNOWLEDGMENTS
The authors would like to thank Jalles Machado and Goiasa-Goiatuba Álcool Ltda for allowing and assisting in the collection of rhizospheric soil, to the Centro de Tecnologia Canavieira (CTC) for granting the Fusarium moniliforme strain used in the study and to Professor Maria Carolina Quecine Verdi (Esalq-USP) for kindly granting the Azospirillum brasilense strain used as a positive control.The authors would also like to thank the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for the first author's scholarship and to the Programa de Pós-Graduação em Ciências Ambientais (CIAMB) for financial assistance.

TABLE I Rhizobacteria isolated from sugarcane under organic management identification based on comparisons with 16S rRNA region sequences deposited at the GenBank database.
* Regarding isolate nomenclature: N/B corresponds to the isolation medium (NFB and Burk) and C/U correspond to the variety (C = CTC4 and U = IAC1099).ARIANA A. RODRIGUES et al.