| Fusarium wilt is a soil-borne disease of watermelon and an economically important problem in the watermelon industry.Grafting of watermelon shoots on resistant root stocks is an effective technical measure to prevent and control soil-borne diseases,and is widely used in watermelon production.However,it is unclear how root exudates of watermelon change after grafting,nor is it known how this change affects the rhizosphere microbiota.Therefore,rhizosphere microbial characteristics and their assembly mechanisms need to be studied in grafted watermelon to reveal how watermelon microbiomes contribute to disease resistance.In this thesis,the effects of grafting on root exudates and rhizosphere microorganisms were studied.We explored different aspects of watermelon grafting in five experimental chapters:1.Root exudates are considered the main signals from the roots.It consists of two classes of compounds,low-molecular weight compounds and high-molecular weight compounds.First,we explored high-molecular weight compounds(proteins)secreted by plant roots.To gain insight into the root-secreted protein profiles,root exudates of three types of seedlings:ungrafted watermelon,watermelon grafted on bottle gourd and bottle gourd were collected under hydroponic conditions,desalted and concentrated using amicon ultracentrifugal filter devices,and separated by SDS-PAGE.Principal component analysis revealed that the protein profile was distinctly altered after grafting,and the diversity of root-secreted proteins of grafted watermelon was significantly higher than that of ungrafted watermelon and bottle gourd.Moreover,using LC-QTOF/MS/MS,proteins associated with biotic and abiotic stress resistance were detected in root exudates of the grafted watermelon only,such as disease resistance protein At4g27190,a callose synthase,an ABA-induced protein HVA22,and a Clp protease.This indicates that grafted watermelon produces and secretes additional stress-related proteins that might contribute to the disease resistance and growth promotion achieved through grafting.For low-molecular weight compounds in the root exudates,freeze dried root exudates were extracted and derivatized,then transferred into GC/TOF-MS for untargeted identification.The types and relative contents of small molecular substances in root exudates of grafted watermelon were significantly different from those of ungrafted watermelon.Grafting significantly increased the Shannon diversity and richness of root exudates,as well as the relative contents of oxoproline,N-Acetyl-5-hydroxytryptamine 1 and 6-hydroxy caproic acid dimer.We conclude that grafting could change the kinds and relative contents of chemical substances in root exudates of watermelon,which could change the allelopathic effect of root exudates of watermelon and improve the resistance of watermelon to soil-borne diseases.2.Pathogens often compete with many other rhizosphere microorganisms for nutrients and space before they infect plants.In this battle,beneficial microorganisms limit the invasion of pathogens by producing antibiotics,consuming nutrients or stimulating the immune system of plants.Through observation and counting of fluorescently labeled pathogen F.oxysporum f.sp.niveum(FON-gfp)in a hydroponic system,we found that the watermelon-specific Fusarium oxysporum could infect all parts of the roots of ungrafted watermelon,but also of grafted watermelon including the vascular bundles.This suggests that the resistance of grafted watermelon to Fusarium wilt is not entirely based on FON's inability to infect bottle gourd,but also on other factors.We then hypothesized that grafted watermelon roots are better able to recruit beneficial bacteria than ungrafted watermelon roots,thus increasing resistance to FON.However,in a similar set-up,we found that ungrafted watermelon roots are better able to recruit the two fluorescently-labelled beneficial bacteria routinely used by our laboratory,Paenibacillus polymyxa SQR-21-gfp and Bacillus amyloliquefaciens SQR-9-gfp,which are both antagonistic to FON.For SQR-9-gfp,at 3 dpi the density of SQR-9-gfp on watermelon roots is 200 times more than that on grafted watermelon.For SQR-21-gfp,the difference is less but still significant.Moreover,the FON infection of ungrafted watermelon root increased over time,FON infection of grafted root decreased over that same period and was always lower than that of the root of ungrafted watermelon.This indicated that the resistance of grafted watermelons was not due to more antagonistic bacteria recruited by grafted watermelon roots,and that other disease resistance mechanisms still exist.This conclusion is further validated by high-throughput sequencing data from in situ soil culture experiments in the next chapter.3.In chapter ?,the root-associated bacterial community of grafted and ungrafted watermelon was compared using 454 pyrosequencing of 16S rRNA gene amplicons under field conditions.Clear distinctions in the rhizosphere bacterial community between the different treatments were observed.Compared with grafted watermelon,ungrafted watermelon recruited significantly higher densities of beneficial bacterial genera,such as Bacillus spp.and Paenibacillus spp.,which is in line with our observation of increased colonization by SQR-21-gfp and SQR-9-gfp on these plants in chapter III.This again indicated that the disease resistance of grafted watermelon is not based on an improved ability to harbor highly beneficial bacteria.However,the rhizosphere of grafted watermelon consistently harbored a more diverse microbial community,expressed as a significantly higher Shannon-Wiener index,compared to ungrafted watermelon.We hypothesize that the more diverse community of root-associated bacteria on grafted watermelon occupy a broader range of ecological niches in the rhizosphere from which the pathogen is then excluded.We proposed that,in addition to secreted antagonistic substances,the grafted watermelon might also exert soil-borne disease resistance by maximizing the niche occupancy of rhizosphere rather than by recruiting more beneficial bacteria.4.In the fifth chapter,we extended our investigations to the disease resistance and microbiome assembly to five plant types(ungrafted watermelon,bottle gourd and pumpkin and watermelon grafted on bottle gourd or on pumpkin),grown under field conditions.Rhizospheres were collected to determine the variations in the rhizodeposits and the microbial community structures by GC-MS and Illumina MiSeq sequencing,respectively.The FON population was quantified by qPCR and the disease incidence was calculated during flowering stage.Clear distinctions in the rhizodeposition and the microbial composition between the five different treatments were observed.Compared to ungrafted watermelons,grafted watermelons formed more diverse rhizosphere chemical niches and consequently both bacterial and fungal communities of the grafted watermelons were more diverse than those of ungrafted plants.We observed significant negative correlations between the bacterial alpha diversity and disease incidence(r=-0.832,p=0.005)as well as between alpha diversity and pathogen population(r=-0.786,p=0.012).This strongly suggests that an enriched bacterial diversity in rhizosphere has the potential to constrain invasion by pathogens.Moreover,significant positive correlations of bacterial rather than fungal alpha diversity with the corresponding chemical alpha diversity was observed,and both bacterial and fungal beta diversities were significantly correlated with their corresponding chemical beta diversities,suggesting that the chemical diversity can not only predict bacterial alpha-diversity but also can impact overall microbial community structure in the rhizosphere.Furthermore,we found that certain chemicals,such as methylnaphthalene,heptacosane and 1,2-Dimethylnaphthalene,were significantly affecting the rhizosphere micro flora.These chemical predictors were significantly associated with specific microbial phyla and functional genes.5.In fields where watermelon is continuously cropped,residues from decomposing watermelon plants are a primary source of autotoxins that can inhibit watermelon growth.Moreover,these residues can facilitate invasion of soil-borne pathogens.In this chapter,we therefore sought to identify both the autotoxins produced during the breakdown of watermelon residues as well as the primary microbial decomposers of watermelon residues.Using gas chromatography-mass spectrometry(GC-MS)and 13C-DNA stable isotope probing-denaturing gradient gel electrophoresis(13C-DNA-SIP-DGGE),we analyzed the volatile organic compounds(VOCs,majority of which are believed to act as autotoxins)as well as the bacterial community of watermelon decomposers in a microcosm experiment at several time points during 45 days.Our results indicate that at the beginning of decomposition,there was an increase in the formation of complex phenolic,alcoholic and ester compounds(e.g.,2,6-Bis(1,1-dimethylethyl)-4-(1-oxopropyl)phenol),whereas after 45 days the main VOCs were alkenes and alkanes with simple structure(e.g.,dodecane,caryophyllene,pentadecane).The bacterial community that assimilated the 13C from watermelon residue was dominated by the phyla Alphaproteobacteria,Betaproteobacteria and Firmicutes at 7 dpi.After that,Alphaproteobacteria remained a dominant phylum,but also Gammaproteobacteria were strongly represented among the residue decomposers.In total,there were 27 bacterial species from 9 different phyla that utilized the 13C-residues in the microcosms.Although the VOC profiles and bacterial decomposers might vary from soils and conditions,these results increase our understanding of residue decomposition of autotoxins and may contribute to mitigate soil-borne disease via crop residue management in field practice.We conclud that a more diverse and distinct rhizodeposition of grafted watermelon results in a more diverse and distinct soil microbial community,which serves as a biological barrier that hinders pathogen invasion.These results will not only broaden our view of disease resistance achieved through grafting,but also provide clues to manipulate the microbial composition by controlling the rhizosphere chemical composition. |
PDF Full Text Request