Diversity of cultivable microorganisms of Davidia involucrate in rhizosphere soil

The purpose of this study was to assay the cultivable microorganisms of Davidia involucrate in rhizosphere soil and to analyze their diversity using colony and individual morphology and genetic profiling methods. A total of 118 strains were obtained after colony characterization and a microscopic examination, including 52 bacteria strains, 27 actinomycetes strains and 39 fungi strains. Fifty-seven (57) of these strains were randomly selected for rDNA analysis and phylogenetic identification, including 23 rhizobaceria isolates, 29 fungi isolates, and five streptomyces isolates. The selected isolates were accurately identified at the genus level with consistent results using a morphology examination and phylogenetic identification. From the study, we concluded the following: (1) Bacillus, Lactobacillus, Azotobacter, and Streptomyces are predominant microorganisms of D. involucrate in rhizophere soil and Bacillus is a dominant rhizobacteria that might have potential in host growth promotion; (2) the physiological activity of dovetree could significantly influence the microorganism biomass of the rhizophere soil; and (3) the dovetree is a highly complex rhizosphere ecosystem, and additional research on this topic should be carried out in the future.


INTRODUCTION
Rhizophere microorganisms play a central role in the regulation of soil structure (Aira et al., 2010), formation of symbiosis (Filippi et al., 1995), controlling plant pathogens (Yang and Cao, 2012) and nutrient cycling (Cambardella and Eliott, 1992;Collins et al., 1992).Helman et al. (2011) summarized that among the microorganisms inhabiting the rhizosphere, several are plant growth promoting rhizobacteria (PGPR), such as genera of Azospirillum, Herbaspirillum, Gluconacetobacter, Burkholderia, Pseudomonas, and Paenibacillus.Arun et al. (2012) reported that some bacteria isolated from the rhizosphere of Cassia occidentalis exhibited significant growth promoting activities that could enhance root length in Vigna radiate and Vigna mungo.
For assaying the genetic microorganism diversity of rhizophere soils, different culture-independent methods have been developed since Pace et al. (1985) proposed the direct cloning of environmental DNA.Most of those methods are based on the technology of polymerase chain reaction (PCR), such as denaturing gradient gel electrophoresis (DGGE) (Muyzer, 1999), single strand conformation polymorphism (SSCP) (Lee et al., 1996), restriction fragment length polymorphisms (RFLP) (Laguerre et al., 1994), terminal restriction fragment length polymorphisms (T-RFLP), (Dunbar et al., 2000), and quantitative PCR (q-PCR) (Takai and Horikoshi, 2000).Those methods have been widely used in assays of the genetic diversity of environment microbial.
Davidia involucrate is a tertiary relic species endemic to China and is reputed to be a "living fossil" in the plant kingdom.It is also an ornamental tree known as Chinese dovetree by virtue of the large pair of white bracts surrounding the small flower, which looks like a dove.In past studies, the dovetree has been well recognized for the diversity of its population, its gene expression and its distribution (Song and Bao, 2006;Wu et al., 2004;Li et al., 2002).However, the rhizophere microorganisms and their relationship with the dovetree have been infrequently studied.Therefore, in this study, we assay the cultivable microorganism biomass and diversity in the rhizosphere soil of D. involucrate by identifying isolations using colony and individual morphology examinations, and rDNA PCR methods.It was expected to isolate some cultivable strains that have potential to increase soil fertility or to promote host growth and that could be used as growth promotion candidate strains in further studies.

Soil collection
Eight plots in the study site were selected and marked as GT1, GT2, GT3, GT4, and FGT1, FGT2, FGT3, FGT4.The GT plots were set as positive controls with D. involucrate grown.Conversely, the FGT plots were set as negative controls without D. involucrate grown.Soil samples were collected at depths of 0 to 20 cm after removing the surface soil (3 mm) at intervals of 1 m from the tree trunk and three soil samples were collected at each plot.The rhizosphere soil was defined as the soil attached to the root after gentle crushing and shaking of the collected roots.A total of 24 soil samples were collected.These samples were stored in freezer boxes at 4°C and transported to the laboratory within 24 h.

Rhizosphere microorganism cultivation and isolation
Rhizosphere cultivable heterotrophic microorganisms were quantified by the plate dilution method.Nutrition agar (NA) was used as selection medium for isolating bacteria (except for actinomycete), inorganic salt starch agar (ISSA) for actinomycete and potato dextrose agar (PDA, Hangzhou Microbial Reagent CO., LTD, China) for fungi, respectively.In brief, 5 g of mixed fresh soil was extracted with 45 ml sterile physiological saline solution (0.85% NaCl) in an Erlenmeyer flask by shaking the mixture for 30 min at about 150 rpm/min.immediately after shaking, the suspension experienced a series of tenfold dilution by pipetting 1 ml aliquots into 9 ml sterilized water.The final dilution was 10 9 -fold, and 0.2 ml of each dilution of the series was placed onto a Petri dish.Three replicate dishes were made for each dilution.NA plates were incubated at 37C for 48 h, ISSA and PDA plates were incubated at 28C for 5-7 days.After incubation, the colony forming units (CFU) of each plate were counted and each typical rhizosphere microbial isolate was sub cultured on appropriate medium to obtain pure colony.Each selected typical colony was characterized according to the properties of shape, surface, color and transparency (Stainer et al., 1987), as well as an individual morphology examination through Gram's staining.All selected strains were stored at -80C in broth with 10% glycerol.

Genomic DNA extraction and PCR amplification of fungi
The genomic DNA of isolated fungi was extracted by a doublesedimentation method (Wu et al., 2003).PCR was performed using TIANGEN 2×Taq PCR MasterMix Kit by two universal primers: 5'-TCC GTA GGT GAA CCT GCG G-3' (ITS1) and 5'-TCC TCC GCT TAT TGA TAT GC-3' (ITS4).PCR reaction was carried out in a volume of 50 μl containing 25 μl TIANGEN 2×Taq PCR MasterMix, 1 μl of 20 mmol/l ITS1 primer, 1 μl of 20 mmol/l ITS4 primer, 1 μl of template DNA, 22 μl sterile water.PCR conditions were as follows: 3 min at 95C, followed by 30 cycles of 1 min at 94C, 1 min at 52C, 1 min at 72C, 72C for 10 min, and 4C for completion.The expected size of PCR products were 800 bp.PCR amplicons were electrophoresed in 1% agarose gels and stained with Gold-View (0.005% v/v).(Altschul et al., 1990).Phylogenetic and molecular evolutionary analyses of the sequences were conducted using MEGA version 4 (Tamura et al., 2007).Multiple sequence alignments were performed and drawing of phylogenetic tree was carried out by Neighbor-Joining method (Saitou and Nei, 1987).
Correction with 1000 replicates to produce bootstrap values (Felsenstein, 1985) and the phylogenetic tree was confirmed by maximum-parsimony method (Kluge and Farris, 1969) and maximum-likelihood method (Cavalli-Sforza and Edwards, 1967).

Colony counting results
The colonies counting results of plates after incubation are shown in Table 1.The counting results revealed that rhizobacteria was the dominant microorganisms in 24 tested soil samples.There were no significant differences in the population of rhizosphere microorganisms except for GT2.However, it is difficult to calculate the fungi colonies of tested soil samples since rare fungi colonies grew on the PDA plates, indicating that the fungi biomass was very low in each plot.

Colony and morphology characterization
A total of 118 strains were obtained after colony characterization and microscope examination, including 52 strains isolated by NA medium (labeled as NA-isolated), 27 strains isolated by ISSA medium (labeled as ISSAisolates) and 39 strains isolated by PDA medium (labeled as PDA-isolates).The 52 NA-isolates were grouped into nine genera as shown in Table 2; genus Bacillus (36.53%, n=19) and Lactobacillus (26.92%, n=14) were dominant microorganisms in tested soil samples.It wass noticeable that there were significantly more Gram-positive strains (69.23%, n=36) than Gram-negative strains (30.78%, n=16).The 27 ISSA-isolates were identified as streptomyces genera by means of an individual morphology examination according to Taddei's reports (Taddei et al., 2006).

rDNA analysis and phylogenetic identification
57 isolates were randomly selected to perform rDNA analysis and phylogenetic identification, including 23 NAisolates and 29 PDA-isolates (two isolates selected from the group which contains less than three isolates and three from those that contained more than three isolates), and 5 ISSA-isolates.Electrophoresis indicated that the expected sizes of PCR products were obtained (Figure 1).Each rDNA sequence was analyzed by BLAST program on NCBI website (http://blast.ncbi.nlm.nih.gov/Blast.cgi)after complete sequencing.Phylogenetic trees were constructed by using the sequences of randomly selected isolates and three typical strain (ten for streptomyces genera) sequences of each closely related genus for phylogenetic identification (Figures 2, 3 and Figure 4).The phylogenetic analysis results revealed that each group of 57 randomly selected isolates was clearly separated into individual branches, which coincided with the morphology examination results as described above.

DISCUSSION
To our knowledge, it is the first time that cultivable microorganisms were assayed in D. involucrate rhizosphere soil.In this study, we found that the biomass of rhizobacteria in GT2 was much higher than the other plots (Table 1).According to previous studies, physiological activity of plant could influence the rhizobacteria biomass.Wang et al. (2010) found that plants are able to regulate the soil microbial community in their immediate vicinity through the secretion from Rehmannia glutinosa roots to the rhizosphere.Arun et al. (2012) reported that the  bacteria diversity of Cassia occidentalis rhizosphere differed at the different stages of the plant.The physiological activity of plant is generally considered to be closely related to the stage of the plant.Therefore, we considered that the reason for the different biomass of rhizobacteria in plot GT2 might have been due to the special physiological activity of the dovetree growing in GT2, although further researches should be carried out to investigate the physiological activity of the dovetree at different stages or using different tissue.
Strain identification is a time-consuming work although several molecular methods have been established.In the present study, both conven- tional methods and molecular methods were used to identify the isolates at the genus level for the assessment of diversity as previously reported (Marie et al., 2010).The results reveal that genus identification can be done using morphology examination and the phylogenetic identification method.
The importance of rhizosphere microbial diversity for maintenance of root health, nutrient uptake, and tolerance of environmental stress has been well recognized (Bowen and Rovira, 1999;Cook, 2002).In this study, a genera of Bacillus (36.54%, n=19) was detected as the dominant rhizobacteria.A number of investigators have reported that the genus Bacillus has good potential in promoting plant growth (Khurram et al., 2012;Sgroy et al., 2009), which is probably because the Bacillus provides a double benefit to the plant: (1) It gradually releases phosphorus (P) from insoluble P complexes though P-solubilization activity, and (2) it improves root growth and root surface area for better uptake of P and other nutrients through ACC deaminase activity.Based on the benefit mechanisms of Bacillus, we have concluded that the huge mass of Bacillus in the dovetree rhizophere soil should be beneficial to the growth of the dovetree.In this study, 21 cultivable genera were isolated, including nine bacteria genera, 11 fungi genera and one streptomyces genus.Those rhizophere microorganisms constitute a complex rhizophere ecosystem that could significant influence the soil fertility and structure (Lynch and Bragg, 1985).Moreover, it is estimated that 99% of microorganisms observed in nature are typically not cultivated using standard techniques (Amann et al., 1995).Consequently, the rhizophere symbiosis is highly complex and its plant growth promotion mechanism is not fully understood.
In summary, we have concluded that (1) the Bacillus, Lactobacillus, Azotobacter, and Streptomyces are the dominant microorganism's genera in the rhizophere soil of the dovetree and that Bacillus is the dominant rhizobacteria that might have potential in host growth promotion; (2) the physiological activity of the dovetree could significantly influence the microorganism biomass of the rhizophere soil; and (3) the dovetree has a highly complex rhizosphere ecosystem and further study should be carried out in this area.The robustness of individual branches was estimated by using bootstrapping with 1000 replications and the phylogenetic tree was confirmed by the maximum-parsimony method and maximum-likelihood method.*, no typical circinella genera strain is searched in Genbank database, circinella muscae E-5261 isolate was selected to construct the pylogentic tree; †, there were only typical strains were searched in Genbank database.

Figure 2 .
Figure2.Phylogenetic tree of rhizobacteria isolates.The phylogenetic tree was constructed by Kimura 2-parameter model and Neighbor-Joining method.The robustness of individual branches was estimated by using bootstrapping with 1000 replications and the phylogenetic tree was conformed by the maximum-parsimony method and maximum-likelihood method.

Figure 3 .
Figure 3. Phylogenetic tree of streptomyces isolates (A) and Pairwise distance of Streptomyces (B).

Figure 4 .
Figure 4. Phylogenetic tree of fungi isolates.The phylogenetic tree was constructed by Kimura 2-parameter model and Neighbor-Joining method.The robustness of individual branches was estimated by using bootstrapping with 1000 replications and the phylogenetic tree was confirmed by the maximum-parsimony method and maximum-likelihood method.*, no typical circinella genera strain is searched in Genbank database, circinella muscae E-5261 isolate was selected to construct the pylogentic tree; †, there were only typical strains were searched in Genbank database.

Table 1 .
Colony forming units counting results of each soil sample.

Table 2 .
Colony and morphology characterization of rhizobacteria isolates.

Table 3
Colony and morphology characterization of fungi isoltes.