Variations in phyllosphere microbial community along with the development of angular leaf-spot of cucumber

Vegetables not only act as energy regulators for human, but also a major part of the human diet with great nutritive values. More than a decade ago, cucumbers were planted in an area of around 156,300 ha around the world with a total production of 26,582 tons (Ram 2002). Cucumber yield and quality are affected by many fungal, bacterial and viral diseases. The angular leaf spot caused by Pseudomonas syringae pv. lachrymans (Young et al. 1978), is distributed worldwide and caused heavy economic losses under favorable climatic conditions by decreasing the yield as well as the quality of the cucumber (Pohronezny et al. 1977).

The phyllosphere is colonized by specific microbial communities as a vital plant-associated habitat (Vorholt 2012; Bringel and Couée 2015). Leaves of plant are colonized by potential microbial communities including bacteria, fungi, protists, and viruses. There are numbers of bacteria colonizing in leaf surface ranging from 10⁶ to 10⁷ per cm² (Vorholt 2012). The phyllosphere may form complex microbial consortia that can be beneficial, pathogenic, or antagonistic for the host plant, and contribute greatly to plant health and yield through complex plant-microbial interactions (Vorholt 2012; Bulgarelli et al. 2013; Brader et al. 2017). The phyllosphere microbial community structure were affected strongly due to changes in the relative abundance of those “key” microbes induced by abiotic or biotic factors. For example, it can be influenced by the plant species, season, geographical location, and different environmental conditions (Whipps et al. 2008; Knief et al. 2010; Wellner et al. 2011; Rastogi et al. 2012; Copeland et al. 2015; Ding and Melchner 2016).

Phyllosphere microbes can affect host fitness through the production of plant hormones and protection from pathogen colonization (Innerebner et al. 2011; Ritpitakphong et al. 2016). In order to adapt to host plants, phyllosphere microorganisms can affect community diversity and plant population (Clay and Holah 1999; Bradley et al. 2008), as well as ecosystem functions (Rodriguez et al. 2009; McGuire and Treseder 2010; Allison and Treseder 2011). The pathogens are expected to compete with native plant-associated microbes when they reach plant surfaces (Brandl et al. 2013). At the same time in the phyllosphere, these microbes will also face harsh environmental conditions including nutrient limitations, UV radiation, as well as lack of free water availability (Colla et al. 2017). All host plants are capable of activating an effective generic defense response against a wide range of microbes (Pieterse et al. 2014). Consequently, microbes usually have specialized structure to help them move towards the plant surfaces. Although the microbial community structure in the leaves has been elucidated, there is still a lack of knowledge about microbial interactions among microbial composition (Hacquard and Schadt 2015). In particularly, it remains unclear that how the establishment of microbial community were affected by competition and mutualism among microbes. It is likely that members of the microbial communities associated with plant have evolved complex strategies for interacting in complex microbial communities to maintain specific host niches.

At present, the research is mainly focused on the culturable part of leaf microflora but it is limited on the understanding of the functional traits of phyllosphere microbes. With the development of next-generation sequencing technology and related computational analysis tools, we can now perform the further investigations (Bulgarelli et al. 2012; Lundberg et al. 2012). The previous studies indicated that the bacterial communities of different plant species are plant-host dependent (Knief et al. 2010; Redford et al. 2010; Vorholt 2012). These findings suggested that the plant host and colonizers might have a selection for each other. However, the underlying processes of microbial population dynamics remain to be elucidated. In this study, our aim was to (1) compare the inter-individual and inter-specific variation of phyllosphere microbial communities; (2) characterize the composition of phyllosphere microbial communities at different disease severities of angular leaf-spot of cucumber; and (3) determine the correlation between disease severity of angular leaf-spot of cucumber and phyllosphere population.

The phyllosphere is colonized by a wide variety of microorganisms including plant pathogenic microbes and other kinds of pathogens (Lindow and Brandl 2003). The colonization of phyllosphere by microbes is controlled by many factors such as factors from plants, microbes and natural environment. And these microbes can deposit on the surfaces of the leaf by means of many pathways. The microbes colonizing the leaves serving as pathogens or growth promoters may have the neutral, negative, or positive impact on their host plants (Kinkel 1997; Whipps et al. 2008). Therefore, it is crucial to study the effect of pathogens from the phyllosphere microbial communities, especially from the dominant microbiome. It is suggested that the abundance of dominant members were associated with plant-associated microbial communities and it is an important factor to determine the healthy community state (Shade and Handelsman 2012). Utilizing the beneficial potential of the plant microbiome to mitigate the hazards of major crop diseases is becoming a sustainable way to improve agricultural production (Berg et al. 2013; Gopal et al. 2013; Bakker et al. 2012). In our study, we conducted a field experiment to study the dynamic change of phyllosphere microbial communities by an Illumina MiSeq-based approach, under the different stage of angular leaf-spot of cucumber.

The phyllosphere microbiome is crucial for leaf biological processes and ecosystem functions (Ortega et al. 2016). The microbial communities composition and structure were observed to be strongly affected by the plant pathogenic microbe (Rastogi et al. 2012; Bulgari et al. 2011; Trivedi et al. 2012). In this study, the richness and Chao1 of phyllosphere microbiome communities decreased significantly from DM1 to DM2, while increased significantly from DM2 to DM3 (Fig. 1). Microbial communities were observed to have the lowest biodiversity at the DM2 stage, which suggested that the pathogen caused an increase in the microbial community richness when disease pressure was higher. Generally speaking, plant disease is a serious threat to indigenous microbiome biodiversity because of its inhibitory action on other microorganisms (Zhou et al. 2010a, b). The current results are in contrast with previous research (Manching et al. 2014), who found that decreased maize leaf epiphytic bacterial richness was correlated with southern leaf blight disease severity.

Significant variations were found in the microbial community structure of phyllosphere samples at different stages of angular leaf-spot of cucumber. The result of the DCA analysis indicated that the samples from different groups were clearly separated (Fig. 2). In addition, the dissimilarity test also revealed significant differences in the composition and structure of the microbiome assembled between different groups (Additional file 1: Table S1). In previous studies, It had been observed that plant pathogenic microbes shaped the microbial communities in previous researches (Zhou et al. 2010a, b; Manching et al. 2014), but it is still unclear at different disease severities. It might be due to the differences in the disease severity of plant disease, which could select the associated microorganisms to colonize leaf surface because of the symbiotic and competitive stresses among microbial species.

The composition and structure of microbial community changes with the aggravation of the disease, there were more than half of shared OTUs were observed in each disease severities group (Additional file 2: Figure S2). However, the highest number of unique OTUs of bacteria were observed in DM3 while for fungi in DM1 (Additional file 2: Figure S2). Previous studies reported that a healthy ecological environment is usually colonized more unique OTUs (Rosenzweig et al. 2012; Zhang et al. 2018). In our study, the number of unique OTUs increased from DM1 to DM2 in bacteria while decreased significantly in fungi and observed to increase significantly from DM2 to DM3 in both bacterial and fungal communities. These results indicated that plant pathogenic microbes would stimulate the growth of more antagonistic microbes in the phyllosphere. The dominating microbial phyla were the members of Proteobacteria and Ascomycota (Additional file 2: Figure S3). Proteobacteria was the most abundant bacterial population among the three disease severities (Additional file 2: Figure S3). Among fungal communities, Ascomycota were the most abundant phylum in DM1 and DM3 groups, while Basidiomycota was the most abundant phylum in DM2 group (Fig. S3). The lesion coverage rate (LCR) were negatively correlated with Sphingomonas and Methylobacterium but positively correlated with Pseudomonas and Kineococcus (p < 0.05, Table 1). It was reported that Sphingomonas contributes to plant health in many plants (Innerebner et al. 2011). In addition, some strains of Pseudomonas and Sphingomonas can enhance plant growth by producing plant growth hormones (Omer et al. 2004; Tsavkelova et al. 2007) and protect host plants from phytopathogens (Innerebner et al. 2011). Previous studies have also indicated that Methylobacterium can protect the host plants from various harmful pathogens (Ardanov et al. 2012; Madhaiyan et al. 2006). Alternaria is a plant pathogen causing black spots on stems and pods of Brassica napus (Bansal et al. 1990). Relative abundance of Alternaria sp. decreased with increasing lesion coverage rate (LCR), which indicated that growth of this genus may was inhibited by the angular leaf-spot causing pathogen.

The network interactions among three different groups were also analyzed in our study. Networks based on random matrix theory could accurately reflect various complex biological systems because of stronger robustness and consistency (Zhou et al. 2011). In general, the more complex network means a more stable community structure (Liang et al. 2016; Mougi and Kondoh 2012). The results of modularity values indicated that all pMENs appear to be modular (Feng et al. 2017). The modularity values for all groups were higher than those of corresponding randomized networks (Table 3). In this study, the interspecies interaction changed with the development of angular leaf-spot of cucumber. It is obvious that the bacterial network size and the interactions, especially the number of positive links (DM1: 42.16%, DM2: 61.34% and DM3: 74.4%), were significantly increased from DM1 to DM3. The results indicated that DM3 group possessed a more complex network than DM1 and DM2 group, indicating that the disease severity could affects the co-occurrence network patterns of overall bacterial communities. As previous studies described, positive and negative interactions usually mean mutualism and competition under environmental stress (Deng et al. 2016; Faust and Raes 2012). Interestingly, our results suggested that the competitive relationships between species were broken with the development of angular leaf-spot of cucumber. It was indicated that the overall bacterial community tends to mutualism from competition. The disease severity may just affects some modules. Therefore, the development of angular leaf-spot of cucumber could disrupt the stability of the microbial community network, which in turn affects ecosystem functioning.