The microbiome lives in an environment and can adapt to it (Elbeltagy et al. 2000; Wilkins et al. 2019; Zhang et al. 2019). Microbiomes have strong connections with their host plant, and they can affect the metabolism of their host plant and harm or benefit the host plant (Vandenkoornhuyse et al. 2015; Joshi et al. 2019; Yang et al. 2019). Furthermore, the culturing of plant varieties always causes some adverse stress effects due to the management of the planting (Hamid et al. 2017; Tian et al. 2020a). It is well known that hybridization can help promote the stress resistance in plants to their planting conditions (Nirmala et al. 2016; Yun et al. 2018). Studies have shown that the beneficial physiological traits of the hybridized plants are better than their original parents (Nirmala et al. 2016; Yun et al. 2018). However, no studies have demonstrated the relationship of the plants and their root endophytic microbiomes with hybridization.
Here, this work provided a preliminary proof of vertical transmission and heritability of specific endodermal microorganisms (fungi and bacteria) in plants. However, the propagation of microorganisms through heritability is different from the clone network because this network is not composed of plant tissues; hence, the parental filter that occurs on microorganisms (i.e., wild type) is also different. Beta diversity has been popularly used for analyses of biological diversity among microbial community compositions along environmental gradients (Maaß et al. 2014; Rivest et al. 2019). Based on the bacterial and fungal community structures, the examined Asian and African cultivated rice species clustered together, and the indica and japonica species were grouped together (Fig. 2a, b); however, African wild rice clustered together with the F1 offsprings following its crossbreeding with African cultivated rice (Fig. 2a, b). Similar to nivara wild rice and common wild rice, the bacterial and fungal community compositions of nivara and common wild rice species were more similar to their respective F1 offsprings than Asian cultivated rice (Fig. 2a, b). Our demonstration of microbiol transmission supports the idea that microbial consortia and their host constitute a combined unit of selection.
Alpha-diversity measurement is particularly challenging for microbial communities (Haegeman et al. 2013; Flores-Rentería et al. 2016). Commonly, microbial diversity has been characterized as the diversity within a given community generally using the total number of OTUs (richness), their relative abundances (Shannon diversity), or indexes that combine these two dimensions (evenness). Studies have generally used microbial alpha diversity to explore the relationships between structure and functioning of microbial communities (Yuste et al. 2011; Flores-Rentería et al. 2016). The results showed that the bacterial and fungal alpha abundances and diversities were not stabilized in the endophytic microbiomes of both the cultivated and wild rice along with their F1 offsprings. Significantly higher Chao1 and Shannon indexes were found for the bacterial community composition of both African and Asian cultivated rice versus their relative wild rice and respective F1 offsprings (Fig. 2b, d), indicating that the bacterial community composition in cultivated rice had more bacterial species and evenness than wild rice. However, there were no stabilized significant differences in the Simpson, Chao1, ACE and Shannon indexes between African cultivated rice and African wild rice and their F1 offsprings (Fig. 2a, b, c, d). The fungal Simpson and Shannon indexes were significantly higher in nivara wild rice and the F1 offsprings than in cultivated rice and the F1 offsprings (Fig. 3a, d), indicating that nivara wild rice and its F1 offsprings retained more fungal diversity and evenness than Asian cultivated rice. Meanwhile, the Chao1 and ACE indexes in common wild rice were higher than in cultivated rice and the F1 offsprings (Fig. 3b, c). These results together indicated that there were more fungal species in common wild rice than in cultivated rice. In support of this finding, Tian et al. (2017) showed that the diversity and abundance indexes of common wild rice were higher than O. sativa for root-associated bacteria (Tian et al. 2017).
Results of PCA showed that African wild rice (Af-W) clustered together with the F1 offsprings after its crossbreeding with African cultivated rice. The bacterial community compositions of the respective F1 offsprings were more similar to that of nivara and common wild rice than to that of the cultivated rice indica and japonica species (Fig. 4a). For fungi, the examined Asian (InC) and African (AfC1, AfC2) cultivated rice species were clustered together, and the two species of African cultivated rice were grouped together (Fig. 4b). The fungal community structures of the nivara wild rice and common wild rice and their F1 offsprings were more similar than that observed between Asian cultivated rice species and their F1 offsprings (Fig. 4b). Likewise, results also showed that community structure of fungi between African wild rice and its F1 offsprings was more similar than that noted between African cultivated rice and its F1 offsprings (Fig. 4b). Furthermore, Euclidean distance of African wild rice and the F1 offsprings (Af-W vs Af-H) for fungal community comparison was found to be lower than that between cultivated rice and the F1 offsprings (Fig. 4d), indicating that the F1 offsprings were more similar to African wild rice than to African cultivated rice in terms of fungal community composition. Similarly, the Euclidean distances of the fungal community comparison between cultivated rice and F1 offsprings were higher than that observed between common wild rice and F1 offsprings, demonstrating that the F1 offsprings showed higher relationship to common and nivara wild rice species than to cultivated rice indica and japonica species with respect to fungal community composition (Fig. 4d).
Network analysis for microbiomes has been used to explore the co-occurrence patterns among microbial taxa or functions (Ling et al. 2016). For plants, the transmission of a microbe along plant clonal networks extends the concept of physiological integration previously demonstrated for information and resources to microorganisms (Vannier et al. 2019). This integrated network structure and blueprint challenge the concept of a meta-holobiont organization in which plants can act as sinks or sources of microorganisms (Vannier et al. 2019). Such a structure can ensure communication between plants, especially between parent and offspring, to improve the adaptability or fitness of clones as a whole (Vannier et al. 2019). No study has performed yet a network analysis on wild and cultivated rice species along with their F1 offsprings. In this study, the network analysis showed that Asian and African cultivated rice clustered together and had more significant correlations than wild rice both in bacteria and fungi (Fig. 4a, b). However, African wild rice and common wild rice had some common significantly correlated bacterial species (Fig. 4b), while African wild rice and common wild rice had some common significantly correlated fungal species (Fig. 4b). Furthermore, there were more significantly correlated bacteria and fungi between both African and Asian wild rice and their F1 offsprings than between cultivated rice and their F1 offsprings (Fig. 4a, b). The bacterial population associated with the rhizosphere of wild rice species displayed differences with those associated with cultivated rice species, suggesting that the root traits selected in domestication could have significant influence on the rhizosphere microbial composition (Shenton et al. 2016). The network analysis showed significantly different correlated species among cultivated and wild rice along with their F1 offsprings (Fig. 5a, b). The core bacterial species that connected wild rice with its F1 offsprings was Acidovorax, wherea the core bacterial species that linked cultivated rice to its F1 offsprings was Bradyrhizobium (Fig. 5a). Acidovorax was detected as a pathogenic genus for plants in watermelon (Citrullus vulgaris), but these species may function in plant immune systems (Xu et al. 2008; Shavit et al. 2016). On the other hand, Bradyrhizobium can benefit plants in nitrogen utilization under more normal and adverse environmental stress conditions (Sulieman et al. 2015; Ambrosini et al. 2019; Kannan et al. 2019; Sulieman et al. 2019). Pleosporales, Myrothecium and Bullera were found to be the core fungal species that connected wild rice and the F1 offsprings, whereas Dendroclathra spp. were noted to be the core fungal species that linked cultivated rice to their F1 offsprings (Fig. 5b). Bullera spp. have been known to serve as biocontrol agents for benefiting plants in disease resistance and growth, explaining the greater relative abundance of Bullera in wild rice and the F1 offsprings (de Tenório et al. 2019).