Diversity of the Rhizosphere Soil Culture-Dependent Fungi of Mature Tobacco

Problem statement: The maturation of tobacco is a very important peri od of tobacco production. Until recently, there are no many repor ts on tobacco rhizospheric culturable fungi, especially at the mature stage. Approach: Five rhizosphere soils of tobacco and five non-rhi zosphere soils of tobacco were collected in Yanbian county, Panzhihua. Sixty-one fungal strains isolated from these soil samples were analyzed by 18S rDNA PCR-RF LP. And 14 representative strains of them were chosen for 18S rDNA sequencing. Results: The results indicated that most of the quantity of fungi of most rhizosphere soils was bigger than rhi zosphere’s. All strains can be clustered together a t similar of 67% in the analyzing of 18S rDNA PCR-RFL P. The strains came from one sample were not always clustered together always, whereas the strai n which came from rhizosphere soils or nonrhizosphere soils often clustered together. Conclusion: The results of 18S rDNA sequencing showed that dominant fungal species of non-rhizosphere soi ls were more abundant than rhizosphere. The culture-dependent fungal quantity, community struct ure and diversity of rhizosphere soil of mature tobacco were affected by the worsening environment of later stage of tobacco.


INTRODUCTION
The Yanbian county of Sichuan, belonging to south subtropical dry river valley climatic region, has typical characteristics of south subtropical drought and monsoon climate. It has sweltering spring but cool summer, strong sun radiation, plenty of sunshine and more heat, an average annual rainfall 1065.6 mm and annual mean temperature of 19.2°C, its advantaged climate resource is very suitable for planting all kinds of flue-cured tobacco.
Plant rhizospheric microorganisms community structure has always been a research hotspot. Many studies showed that the physiological activities of the rhizospheric microorganisms had an important influence on soil properties, nutrient uptake and plant growth and development (He and Li, 1999). The fungus is an important component part of microorganisms and is often closely linked with plant health, even some fungi can directly or indirectly improve the endurance to the poison of heavy metal (Li and Feng, 2001;Wang and Lin, 2007). Zhang et al. (2009) found that fungus of the tobacco field was the minimum in three bacterium groups, which was similar to the soil microbial distribution in general. Zhan et al. (2005) thought that the quantity of tobacco rhizospheric fungus changed with the growth period of tobacco in same fertile soil. For example, the quantity was on a parabola change in the field of purple soil and yellow soil, at least in the rosette stage, then increasing gradually, reaching the top at budding stage and then to decrease. Meanwhile, Shishido and Chanway (1998) found that, as the harmful metabolism products accumulated, the actinomyces and fungi which had strong resistance increased at later growth stage of plant. Penicillium, Trichoderma, Aspergillus and Fusarium were widely seen as dominant fungi in rhizosphere of field crops (Curl and Truelove, 1986;Gadgil, 1965;Parkinson and Clarke, 1964).
In this study, in order to understand the health degree of ecological system in the late period of tobacco and provide certain scientific basis and guidance for the harvest of flue-cured tobacco, the rhizospheric culturable fungi of rhizosphere and nonrhizosphere soils at mature stage had been studied by combining the traditional method with 18S rDNA PCR-RFLP and sequence analysis.

Soil samples:
The soil samples were collected from Yanbian county, Sichuan Province where is very suitable for the growth of tobacco. Rhizosphere soil was collected from plants by first removing all visible bulk soil by hand, then the samples consisted of soils from both loosely adhering to roots and that could be brushed or scraped off the root surface (Smalla et al., 2001). The visible bulk soil was collected for nonrhizosphere soil. Fifteen individual healthy plants were collected from every plot. After sampling, the soils were brought to the laboratory and any obvious plant or animal residues were removed by handpicking. Part of the samples were kept with moist in the dark at 4°C to assess microbial biomass (Tian et al., 2009).
Enumeration of culturable fungi: Ten grams of each of the soil samples were individually dispensed into 90 mL of deionised water containing about 20 g of glass beads (3 mm diameter). Tenfold dilutions were made in sterile deionised water after soil suspensions were centrifuged at 120 r min −1 for 30 min. Then 0.1 mL aliquots of each soil dilution (10 −2 -10 −4 ) were spread on the surface of the different substrates in sterile Petri dishes (9 cm diameter). Three plates were used per dilution. The plates were dried in a laminar flow cabinet for 1 h and then incubated. The CFU of fungi was estimated on Rose Bengal (33 µg mL −1 ) and streptomycin (30 µg mL −1 ) agar on which 100 mL of 10-fold serially diluted soil samples were spread. The CFU was counted after incubation for 7 days for fungi at 28°C (Kong et al., 2008).
18S rDNA sequencing: According to the results of 18S rDNA PCR-RFLP, fourteen representative strains were chosen for 18S rDNA gene sequencing carried by Yingjun Biotechnology Ltd. (Shanghai, China). These sequences and their closest match sequences which from GenBank database were pairwise aligned using Clustal X (Thompson et al., 1997). Phylogenetic trees were constructed using the Neighbor-Joining method in MEGA program version 4.0 (Tamura et al., 2007). These sequences were submitted to the GenBank database and the accession numbers were from JN176199 to JN176212.
Statistical analysis: Analysis Of Variance (ANOVA) was conducted on collected data and the mean values of plant fresh weight and disease index were statistically analyzed using the LSD test. Differences were considered to be significant when the probability was less than 0.05.

Analysis of the 18S rDNA PCR-RFLP:
The PCR products of 18S rDNA were digested by four restriction enzymes of HinfI, TaqI, HaeIII and MspI. The Fig. 2 showed that the bands of Marker and others were clearly visible and all the 4 restriction endonucleases were suitable to digest the PCR products of 18S rDNA. Fingerprints of the strains generated by PCR-RFLP of the ribosomal genes were used to construct dendrogram by using UPGMA analysis (Fig. 3). All strains could be clustered together at similar value of 67% and they would be divided into nine groups at similar value of 83%. Parts of the strains which isolated from one sample were clustered together, such as SAUFC4-2 and SAUFC4-7, SAUFC2-5 and SAUFC2-6. Whereas some were far different, for example, the strains isolated from non-rhizosphere soils in Gude. The strains which isolated from rhizosphere soils often clustered together and the strains isolated from non-rhizosphere soils was same, this showed rhizosphere environment had an influence on the distribution of fungi.
Analysis 18S rDNA sequences: Sixteen representative strains and their closest match sequences which from GenBank database, such as Isaria takamizusanensis, Paecilomyces lilacinus and Gibberella fujikuroi, were selected to make Phylogenetic tree (Fig. 4).

DISCUSSION
In general, tobacco rhizospheric exudations in different growth periods have influence on the quantity of fungi. The quantity of culturable fungi of rhizosphere was more than non-rhizosphere's in most of samples except Fuxing A and Gude. The quantity of fungi of R2 hadn't significant difference with NR2 (p>0.05). The quantity of fungi of non-rhizosphere of Gude site was more than rhizosphere's (NR5>R5).
From the representative strains reflected, the dominant fungal species in tobacco rhizosphere soil were Aspergillus sp., Gibberella fujikuroi, Rhizopus oryzae, Absidia psychrophilia, Penicillium decumbens and Neosartorya fischeri. Whereas, in tobacco nonrhizosphere soil were Aspergillus sp., Chaetomium globosum, Ascomycete sp., Fusarium oxysporum, Eupenicillium javanicum, Paecilomyces lilacinus, Penicillium sp. and Cerrena sp.. English and Mitchell (1988) found that Penicillium, Trichoderma, Aspergillus and Fusarium grew quickly in tobacco rhizosphere soil. However, Trichoderma, Aspergillus and Fusarium hadn't been found in dominant fungal species in tobacco rhizosphere soils in our study. On the contrary, Aspergillus and Fusarium appeared in the nonrhizosphere soils. Gibberella fujikuroi is pathogenic fungi of rice bakanaea disease in rhizosphere soil, although it has little effect on tobacco, it not only can infect non-crop Fistula arundinaceous, Leerier sayanuka Ohwi and Digit aria sanguinely (L.) Scoop. But also can infect field plants, such as wheat, sorghum, maize, etc. Gibberella fujikuroi became a dominant fungal species at later growth stage of tobacco may be related to the current degradation of tobacco rhizospheric environment. Compared with nonrhizosphere, Neosartorya fischeri which has antiinflammatory action was found in dominant fungal species of rhizosphere.

ACKNOWLEDGEMENT
The researchers are grateful to an anonymous reviewer for kindly correcting this manuscript. This work was supported by the Chinese National Natural Science Fund (No. 31070004) and the specialized research found for the doctoral program of higher education 20060626006.

CONCLUSION
The quantity of fungi and population of tobacco rhizosphere soil at later growth stage were different with non-rhizosphere soil, especially the population, which was influenced by the degradation of rhizosphere environment. If the environment aggravates to a definite level, tobacco diseases may occur and lead to yield reduction. It is not suitable to control tobacco diseases using pesticide at maturity stage. In order to obtain high yield at later growth stage, we should improve the field ventilated condition, reduce the field temperature properly, clean diseased leaves and disabled body timely and bake at the right time, except for selecting resistant varieties with local conditions.