Analysis Of Pathotypic And Genotypic Diversity Of Xanthomonas Oryzae PV.Oryzae In China And Study On The Homologs Of Avrbs3/Ptha Gene Family Mediated Phenotype

Virulence diversity based on near-isogenic rice linesThe strains of Xanthomonas oryzae pv. oryzae (Xoo) were polymorphic for virulence on the13near-isogenic lines using IR24as the recurrent parent. No strain was virulence to all these R genes tested, while no R gene could overcome all these strains tested. Over fifty percent of Chinese strains were virulent to the lines IRBB1, IRBB3, IRBB4. The gene, xa5, Xa7, and Xa10, were resistant against over eighty percent of tested Chinese strains of Xoo. About eight percent of tested strains were virulent to IRBB21. So the resistant gene Xa21is the best candidate R gene for breeding against Xoo in China. The strain KS6-6from Jiangsu province is the most virulent to all major varieties tested except IRBB7and IRBB21. Nine near-isogenic lines were susceptible to the strains HuB03-4, HeN03-18, YN30and YN41. Among near-isogenic lines using Toyonishiki as the recurrent parent, only IRBB214exhibited the best broad resistance against seventy-one percent of the Chinese strains,In addition, IRBB50(Xa4+xa5) and IRBB54(xa5+Xa21) have the best combination of R gene against BB. Among traditional varieties, DV85was resistant to all strains tested. Other two lines, IR26and BJ1are resistant to over ninety percent of tested strains. Java14and Asminori were attacked by over fifty percent of the strains tested.Among the103strains tested,61pathotypes were identified based on the13near-isogenic lines using IR24as the recurrent parent. Some pathotypes were shared by four or less strain, such as pathotype28comprised four strains from middle-southern and southern China. The majority of pathotypes included one or two strains. Pathotypes58and61from southern China attacked one or two R genes. Further analysis of virulence data using the consensus of three clustering statistics and UPGMA revealed4clusters. Cluster2was the most heterogeneous and contained64Chinese strains composed of37pathotypes distributed in11provinces. Clusters1,3and4contained18,4and2pathotypes, respectively. Tested Xoo strains from YN province were found in all clusters.Interestingly, we observed some Xoo strains tested had lost their capacity to induce a nonhost hypersensitivity reaction (HR) when infiltrated with a needleless syringe inoculation in the leaves of Nicotiana tabacum L. cv. Samsun. This phenomenon was also observed in other Xoo strains from Philippines. We also found that several rice cultivars showed very short lesions (<0.5cm) with chlorotic or dark necrotic margin, like hypersensitive response, emerged after leaves were inoculated by certain Xoo strains.Genotypic diversity with a repetitive fragment of avrXa352Chinese strains, which were used for virulence analysis from10provinces, were also selected as representative strains for RFLP research. The DNA fingerprint pattern generated by the probe revealed the high genetic diversity in Xoo strains. The size of DNA fragments that hybridized mainly ranged from1to5kbs, therefore, there are about9to50repeat units inferred with102bps in central region between the conservative Sph Ⅰ site among avrBs3/PathA family. The results indicated the presence of multiple homologous copies of the avrXa3/PthA gene family among the pathogens, ranging from12-25signal bands with an average of16bands, whereas the control Japanese strains and Philippine strains possess12-15and15signal bands respectively. This is also different from the Xoo from west Africa, which have a specific and intriguing features of a reduction in avrBs3/pthA effector genes (Gonzalez et al.2007). For example, strains YN7and AH15showed12and25bands respectively. Alignment of detected signal bands indicated that three bands with the size of1.8,2.2,2.6kbs are shared by all of tested strains including two Xanthomonas oryzae pv. oryzicola (Xoc) strains JX1and JX2from Jiangxi province, and another one with1.3kbs is common for all Xoo strains.The DNA fingerprint pattern was referred to as molecular haplotype. At least35bands positions were scored for the collections. In general, no single haplotype dominated the Xoo population in YN and JL provinces. However, most haplotypes were only found in one or two strains from one or two provinces. Four haplotypes12,20,26, and31were found in two strains and only the haplotype12included two strains from different provinces. Only two haplotypes,6and17, were shared by4strains in HuN province and YN province, respectively. The remained haplotypes occurred at only one site in different provinces of China. Forty-two molecular haplotypes were shared by these52strains. On the basis of consensus of three clustering statistics, four lineages were found among these strains. Genetic diversity was high in all lineages and No significant (P<0.05) differences were revealed by t test. The distribution of the strains from each lineage is independent to geographic original, but the hybridized band patterns ranging from2.5to3.5kbs have obviously dependent to spatial source, such as the bands patterns in the strains YN1, YN3, YN5, YN8, YN11,YN13, YN16,YN18, YN20, YN39were identical and collected from the same place YN Province. The strains in each lineage come from at least two different provinces. All nine rice-producing provinces in China contained one or more lineages. Such as lineage1comprises strains originated from8of the12provinces, together with the Japanese strains and Philippin strains.The number of signal bands was not directly associated with the virulence of a given strain. For example, strains AH15and YN12possess25and24bands, respectively, and they were virulent to7and3tested near isogenic rice lines, respectively. At least10bands are different between AH15and YN12. Strain KS6-6only contains15bands, but is pathogenic to11tested near isogenic rice lines. The band pattern of KS6-6is obviously different from that of AH15or YN12. In addition, the two strains YN18and YN31without nonhost HR have15and17hybrid bands, respectively, and the band patterns are also quit different each other.The relationship between haplotypes and pathotypesTwo patterns were observed between the molecular haplotypes and pathotypes:(ⅰ) the strains with identical pathotypes exhibited multiple haplotypes, For example, the strains YNl and HuN61share the same pathotype58, but they belong to the different haplotype18and32, respectively, and at least4bands are different between them (ⅱ) groups of strains that had the same haplotype shared several pathotypes. For instance, the strains YN11and YN39share the same haplotype26, but they belong to pathotype1and3, respectively, and at least2bands are different between them. Furthermore, the two Xoc strains, JX1and JX2, used as controls, possessed21and20bands, respectively. The bands patterns of these strains are more genetically distinct than that of Xoo strains. When the similarity matrix of virulence data was compared with that generated by the RFLP data, the correlation coefficient was0.42(data not shown), which indicated a weak relationship between the two types.The clone of the homology of avBs3/pthA gene family from XooFour DNA fragments from a genomic library of a Japanese race5JXOV hybridized with the Sph I fragment of avrXa3, an avirulence gene from Xoo in Southern blot. The analysis of nucleotide sequence homology show that they have95%identical with avBs3/pthA gene family, conserved BamH I and Sph I restriction site, one leucine zipper (LZ), three nuclear localization signals (NLS), an acidic transcriptional activation domain (AAD)) at C-terminal, different numbers of102-bp repeats in central repeat region.The length of four fragments B17, B19, D82and D41is2131,3346,3345and4482bp, with5.6,16.4,16.4and26.3102-bp repeats in central repeat region, respectivelyThese homologies of avBs3/pthA gene family are not defined by corresponding resistance genes of which a relatively large number have now been cloned. Because these transformant KB1, KB19, KD82and KD41with B17, B19, D82, D41inserted in comsmid pMH1lost pathogenecity on the following13near iso-genetic lines, each carrying a specific R gene:IRBB1(Xal), IRBB2(Xa2), IRBB3(Xa3), IRBB4(Xa4), IRBB5(xa5), IRBB7(Xa7), IRBB8(xa8), IRBB10(Xa10), IRBB11(Xall), IRBB13(xa13), IRBB14(Xal4), IRBB21(Xa21), IR24(Xa18). IR24was used as a susceptible check. The transformant KD41only lost virulence on IRBB4compared with CK wild strain KS6-6A and KS6-6A with cosmid pMH1The phenotype mediated by the homology of avBs3/pthA gene family from XooThe spatial-temporal expression patterns of cell wall reinforcement and oxidative burst were analyzed in leaves of rice IRBB4in response to CK strain KA and KP (KA with cosmid pHM1), treatment K2(KA with cosmic pHMl inserted the homologs of avBs3/pthA gene family from strain JXOV), and K8(KA with cosmic pHMl inserted the homologs of avBs3/pthA gene family from strain JXOV) after inoculation3days. The accumulation patterns of callose and lignin deposits were histochemically analyzed by paraffin section based on the basic Safranin-FastgreenFcF double staining technique, allowed us to visible monitor callose and lignin accumulation in plant cells. The phenomenon of oxidative burst was detected by employing the in situ DAB-uptake assay after clip-inoculation for four days running on rice IRBB4. The growth curve after inoculation was observed by taking count of the number of colony. Lignin was accumulated frequently in vascular bundle cells by visiblered color within the inoculation part, while the lignin accumulation was restricted to the portion of inoculation. Differences in the timing and amount of oxidative burst were observed. Oxidative burst was not detected in the first day after inoculation, but mainly to the accumulation brown in cells surrounding inoculation part was observed from the second day to the fourth day after inoculation. The treatment CK and K2have the same growth trend, but this·growth of K2was delayed and weaker, no significant variations in the rate of colony accumulating took place until2day...