Genetic diversity analyses of Lasiodiplodia theobromae on Morus alba and Agave sisalana based on RAPD and ISSR molecular markers

ABSTRACT Genetic diversity of 23 Lasiodiplodia theobromae isolates on Morus alba and 6 isolates on Agave sisalana in Guangxi province, China, was studied by using random amplified polymorphic DNA and inter-simple sequence repeat molecular markers. Results of two molecular markers showed that the average percentage of polymorphic loci of all isolates was more than 93%. Both dendrograms of two molecular markers showed obvious relationship between groups and the geographical locations where those strains were collected, among which, the 23 isolates on M. alba were divided into 4 populations and the 6 isolates on A. sisalana were separated as a independent population. The average genetic identity and genetic distance of 5 populations were 0.7215, 0.3284 and 0.7915, 0.2347, respectively, which indicated that the genetic identity was high and the genetic distance was short in the 5 populations. Average value of the gene diversity index (H) and the Shannon’s information index (I) of 29 isolates were significantly higher than 5 populations which showed that genetic diversity of those isolates was richer than the populations and the degree of genetic differentiation of the isolates was higher. The Gst and Nm of 29 isolates were 0.4411, 0.6335 and 0.4756, 0.5513, respectively, which showed that the genetic diversity was rich in those isolates.


Introduction
The fungus Lasiodiplodia theobromae (Pat.) Griffon & Maubl., anamorph of Botryosphaeria rhodina (Berk. & M. A. Curtis) Arx, is pleomorphic, plurivorous and ubiquitous soilborne pathogen in the tropics and subtropics (Úrbez-Torres et al. 2008). As a plant pathogen, L. theobromae associated with up to 500 plant hosts and caused numerous diseases, including gummosis, leaf necrosis, dieback, collar rot, root rot, fruit rot, leaf spot, decline, leaf blight, cankers and witches' broom and so on (Úrbez-Torres et al. 2008). It also occurs as an endophyte (Mohali et al. 2005;Rubini et al. 2005). Mulberry root rot caused by L. theobromae in Heng County of Guangxi Province, China, was first reported in 2014 (Xie et al. 2014) and diseased area was about 2400 ha, where approximately 300 ha mulberry was seriously diseased with the diseased and mortality rate more than 80%. It was found that this disease also occurred in many other counties of Guangxi Province where mulberry was cultivated.
Colonies of L. theobromae were initially white, with fluffy and aerial mycelia, and overgrew the Petri dish in 4 days. Then, the colonies turn grew to black and spreading with superficial dark-branched septate mycelia. Black colour of mycelia on reverse sides of Petri plates was visible. The conidia of the pathogen were initially unicellular, hyaline ellipsoid to subovoid. Mature conidia were dark brown, bicelled, thick walled and ellipsoid.
The molecular markers such as random amplified polymorphic DNA (RAPD) and inter-simple sequence repeat (ISSR) have been extensively used to investigate the genetic diversity of plant pathogens at strain and population level (Wu et al. 2000;Liu et al. 2014;Chen et al. 2015). Genetic cluster dendrogram based on the results of molecular markers often keeps a relationship with the geographical region, host, pathogenicity, vegetative compatibility group and other aspects of the tested isolates, and the genetic diversity and difference of intra-and inter-species from different sources could be understood by analysing the dendrogram (Shafagh et al. 2008;Baysal et al. 2010). Dendrogram generated based on RAPD and ISSR divided the 158 isolates of Botryosphaeria into three groups and those groups were identical with the morphological characteristics and cultural properties of those isolates (Zhao 2007). The aim of this study was to evaluate genetic diversity of some L. theobromae isolates collected from Morus alba and Agave sisalana by RAPD and ISSR molecular markers and reveal the genetic variation and the correlation between genetic diversity and geographical locations.

Isolates and identification
Twenty-three isolates and six isolates of L. theobromae, respectively, on M. alba and A. sisalana Perr. ex Engelm. collected from geographical locations were used for analysing the genetic diversity (Table 1), which were confirmed to be the pathogen of mulberry root rot and sisal leaf spot disease on A. sisalana according to the Koch's postulate, respectively, and were identified as L. theobromae based on morphological characteristics and sequences of the rDNA internal transcribed spaces (rDNA-ITS) and nuclear translation elongation factor l-α (EF1-α) previously.
Phylogenetic tree based on the rDNA-ITS sequences of the 29 isolates was made ( Figure 1).

DNA extraction
Isolates were cultured on potato dextrose agar for f5 days at 28°C and the genomic DNA was extracted by using the E.Z.N.A. ® Fungal DNA Min Kit (produced by OMEGA Co. Ltd.). DNA concentration was quantified on a spectrophotometer (Eppendorf Biophotometer, Hamburg, Germany) and was diluted according to experimental needs.

Primers selection
RAPD primers with more bands and better repeatability were screened out by using fifty 10-base oligonucleotide random primers to amplify the DNA of 10 isolates. Then, the selected primers were used to amplify the DNA of the 29 isolates. For tested ISSR primers screening, 60 primers published by University of British Columbia, Vancouver, Canada, were used to amplify the DNA of 10 isolates and the selected primers were used to amplified the DNA of all the isolates. All the primers were synthesized by Sangon Biotech, Co., Ltd. Shanghai, China.

PCR products measured
The amplified products of RAPD and ISSR PCR were size separated in 1.8% agarose gel (containing 4S Green Plus Nucleic Stain, Sangon) under 1 × TAE buffer (40 mM Tris, 20 mM acetic acid and 1 mM EDTA) at 80 V for 120 min and 100 V for 180-200 min at room temperature, respectively. All the gels were visualized under UV light and photographed using a gel documentation system (UVItec ® ). The sizes of amplified DNA fragments were estimated by comparison with 10 kb DNA ladder markers (BBI).

Data analysis
Those clear, intense and reproducible amplified fragments of RAPD and ISSR were converted and normalised into a data pattern as binary matrix with "1" for presence and "0" for absence. Polymorphic regions were scored to determine the similarity among the isolates. Hierarchical clusters were analysed to construct dendrograms, using Dice's coefficient and unweighted pair group method with arithmetic mean (UPGMA) methods with the NTSYS-PC software (Version 2.10e) (Rohlf 1998). The dendrogram was reconstructed 1000 times by repeated sampling with replacement and the frequency on which clusters formed indicating the strength of the clusters. The software POPGENE (Version 1.32) was used for statistical analysis of standard population genetics (Yeh et al. 1999). The mean gene diversity index (H) was calculated as H = (1 − ∑P i 2 ), of which P i was the frequency of allele at the locus (Nei 1973). Heterozygosity and percent polymorphic loci were estimated for all populations. Genotypic diversity was calculated by Shannon's information index (I). Differentiation among populations was estimated by indirect estimation of gene flow using Gst with Nm = 0.5(1 − Gst)/Gst (McDermott and McDonald 1993).

The primers selection and amplification
Sixteen RAPD primers and 13 ISSR primers with more bands and better repeatability were screened out, respectively (Table 2). A total of 168 constantly amplified  DNA bands between 450 and 7200 bp in length (500-4000 bp in most instances) were generated from 16 RAPD primers, among which 97.6% were polymorphic. The number of DNA bands amplified by primer OPC08 was the most and totally 13 bands were polymorphic ( Figure 2). Major size of the 135 DNA bands generated with the 13 ISSR primers was from 350 to 3000 bp, and the average percentage of the polymorphic loci was 93.96%. Among those primers, 13 bands were amplified with the primer 885 and all of them were polymorphic (Figure 3).

Cluster analysing of RAPD and ISSR
Dendrograms produced from UPGMA analysis based on Nei's similarity coefficient of RAPD and ISSR molecular markers showed the similarity coefficient ranged from 0.61 to 0.92 and from 0.69 to 0.99, respectively, and divided the 29 isolates into 7 branches (when the similarity coefficient reached 0.76) and 5 branches (when the similarity coefficient reached 0.746), respectively (Figure 4; Figure 5). The UPGMA dendrogram based on RAPD showed that except the isolates from Luzhai (independently clustered as GL), A. sisalana (independently clustered as GS) and Xiangzhou (independently clustered as GX) isolates from Heng (GH) and Yizhou (GY) were clustered into two or three different groups. The UPGMA dendrogram based on ISSR showed that the isolates from Yizhou, Luzhai, A. sisalanal and Heng (six isolates) were independently clustered in a branch, but two isolates from Heng and five isolates from Xiangzhou were clustered together. The dendrograms based on RAPD and ISSR indicated that most isolates clustered groups reflecting relationship with the geographical locations where those isolates were collected.
Genetic diversity of 29 isolates of L. theobromae collected from different hosts and geographical locations showed that those isolates were highly polymorphism. The genetic similarity coefficients and its amplitudes generated by RAPD and ISSR indicated that two molecular markers could reveal the genetic background and the degree of genetic variation among those isolates was very high.
Clusters on dendrogram were first separated all isolates as two parts, a part belonging to isolates on M. alba and another part belonging to isolates on A. sisalana. Twenty-three isolates on M. alba were  divided into four populations, namely, Xiangzhou (GX), Yizhou (GY), Luzhai (GL) and Heng (GH) just corresponding with geographical locations where these isolates were collected.

Genetic variation analysis
The average genetic identity and average genetic distance of the 5 populations based on RAPD and ISSR molecular markers were 0.7215, 0.3284 and 0.7915, 0.2347, respectively, which showed that there were high genetic identity and short genetic distance among those populations. The average gene diversity index (H) and the average Shannon's information index (I) of the 29 isolates were significant higher than the 5 populations based on the RAPD and ISSR molecular markers, which indicated that the genetic diversity and genetic differentiation of isolates were higher than that among the populations.
Of RAPD molecular markers, Nei's genetic identity and genetic distance were different among the five populations (Table 3, showed by upright numbers). The genetic identity ranged from 0.6546 to 0.7943, and the genetic distance ranged from 0.2303 to 0.4237. The maximum genetic identity appeared between the population GS and GH, and the  maximum genetic distance appeared between the populations GL and GX. Parameters of genetic diversity of the five populations were calculated by PopGene software and results showed that average value of observed alleles (Na) of the 5 populations was 1.5476, the average value of effective alleles (Ne) was 1.4009, the average gene diversity index (H) was 0.2224 and the average Shannon's information index (I) was 0.3230 (Table 4, showed by upright numbers). Parameters of genetic diversity of 29 isolates were calculated by PopGene software and results showed that the mean of Na, Ne, H and I was 1.9821, 1.7200, 0.4014 and 0.5830, respectively (Table 4, showed by upright numbers). The total heterozygosity (Ht = 0.3979), intra-specific heterozygosity (Hs = 0.2224), coefficient of gene differentiation (Gst = 0.4411) and gene flow (Nm = 0.6335) were also calculated for the 29 L. theobromae isolates (Table 5, showed by upright numbers). Of ISSR molecular markers, Nei's genetic identity and genetic distance were also different among the five populations (Table 3, showed by italic numbers). The genetic identity ranged from 0.7298 to 0.8233, and the genetic distance ranged from 0.1944 to 0.3150. The maximum genetic identity (0.8233) appeared between the population GS and GY, and the maximum genetic distance (0.3150) appeared between the populations GX and GY. Parameters of genetic diversity of the 5 populations and the 29 isolates were calculated by PopGene software and showed that the average value of observed alleles (Na) of the 5 populations was 1.3585, the average value of effective alleles (Ne) was 1.2903, the average gene diversity index (H) was 0.1566 and the average Shannon's information index (I) was 0.2240 (Table 4, showed by italic numbers). The mean of Na, Ne, H and I of 29 isolates was 1.8444, 1.5173, 0.3032 and 0.4528, respectively (Table 4, showed by italic numbers). The total heterozygosity (Ht = 0.2987), intraspecific heterozygosity (Hs = 0.1566), coefficient of gene differentiation (Gst = 0.4756) and gene flow (Nm = 0.5513) were also calculated for the 29 L. theobromae isolates (Table 5, showed by italic numbers).
According to Nei's gene diversity and Shannon's information index based on RAPD and ISSR  molecular markers, the genetic differentiation among isolates was relatively higher, whereas smaller among populations, and there was a high level of genetic diversity among those isolates. Zhang et al. (2000) found that 30 isolates belonging to the Botrysphaeria collected from Shangxi, Liaoning, Hunan, Beijing, Shangdong and Henan province of China had highly polymorphism (92.81%), some isolates showed obvious geographical and host differentiation, and the others did not, according to the UPGMA dendrogram based on RAPD. Our results were partly similar to it, that is, most isolates showed obvious geographical and host differentiation. Sangeetha et al. (2011) reported that 12 isolates of L. theobromae collected from 12 different commercial banana cultivars showed high degree of genetic variation and rich genetic diversity among isolates from different banana varieties based on RAPD. Nghia et al.

Discussion
(2012) described that the genetic diversity of 20 isolates of Botryodiplodia theobromae (the synonym of L. theobromae) and 214 DNA bands were generated with 16 ISSR primers (76.6% of those bands were polymorphic); the clusters of dendrogram showed the relationship among isolates from different regions and no relationship between genetic groups and the hosts. But, the dendrogram generated based on ISSR in our study showed significant correlation between the clusters and both the geographical locations and host plants, and only two isolates collected from Heng County mixed with five isolates collected from Xiangzhou County, the reason of which was probably that mulberry seedlings from Xiangzhou County were transported to Heng County. The gene differentiation coefficient (Gst) denoted the level of genetic variation among populations, and the genetic differentiation among populations was generally large when Gst value was above 0.15 (Wright 1978). The strength of gene flow (Nm) between populations was the important factors affecting of the population genetic differentiation, and if Nm value was more than 1, the gene flow could prevent the differentiation of population that caused by genetic drift, but limited gene flow could promote the genetic differentiation among populations when the Nm value was less than 1 (Hartl and Clark 1997). Genetic diversity and gene flow of endophyte L. theobromae were studied, and results showed that there was no evidence of host specificity for isolates of L. theobromae and there was very high gene flow between populations from different hosts and gene flow was less restricted between isolates from the population of Mexico (Mohali et al. 2005). In this study, the Gst value (0.4411 and 0.4756) and Nm value (0.6335 and 0.5513) generated from RAPD and ISSR revealed that genetic differentiation was large among isolates and the frequency of gene exchanges between populations was low.
The sample size of L. theobromae isolates was relatively small. More isolates collected from different countries, geographical regions, seasons, hosts, ecological surroundings and soil conditions should be studied for thoroughly revealing the genetic diversity and genetic differentiation of L. theobromae.

Highlights
• Mulberry root rot caused by Lasiodiplodia theobromae was a new disease in China. • Genetic diversity of 29 L. theobromae isolates was studied. • Genetic diversity of L. theobromae was related to the geographic locations.