Plant endophytic microbial communities consist of many latent plant pathogens and, also, many pathogen-related species with reduced virulence. Though with a long history of co-evolution, the diversity and composition of the endophytic mycobiota, especially the pathogen-related fungal groups, has been under-investigated in Citrus (C.). Based on the amplicon sequencing of fungal internal transcribed spacer (ITS), the leaf endophytic mycobiota were profiled on citrus varieties from different citrus-producing regions. The pomelo variety shared significantly distinctive leaf mycobiota when compared to the mandarin and sweet orange; these conform to their host genetic relationships. In addition, a data set of 241 citrus-related fungi, including 171 (71%) pathogens and potential pathogens, was summarized from previous studies. Under the criteria of local BLAST (covered ITS nucleotide ≥ 150 bp, sequence identity ≥ 99%), a total of 935 fungal operational taxonomic units (OTUs) were assigned to 62 pathogen-related fungal groups, representing 14.9% of the relative abundance in the whole community. Of which, the top groups consisted of Colletotrichum gloeosporioides (mean relative abundance, 4.3%), Co. citricola and Co. karstii (2.7%), Zasmidium citri-griseum (2.4%), and Z. fructigenum (1.4%). At the genus level, the ratio of the pathogen-related fungal groups in 64% of fungal genera (16 out of 25) exceeded 50%, which are the solely or mainly occurring fungi of their genus in citrus. Our study suggests that the leaf endophytic compartment may be an important place for the growth of latent pathogens.

For perennial plants, newly emerged organs are fresh hot spots for environmental microbes to occupy and assemble to form mature microbial communities. In the microbial community, some commensal fungi can play important roles in microbial succession, thus significantly improving host plant growth and disease resistance. However, their participating patterns in microbial assembly and succession remain largely unknown. In this study, we profiled the fungal community and found a similar fungal succession pattern of spring-emerged leaves from March to October in two pomelo orchards. Specifically, the fungal species, tracked on the old leaves, dominated the spring leaves after emergence and then decreased in relative abundance. This reduction in priority effects on the spring leaves was then followed by an increase in the number of observed species, Shannon and phylogenetic diversity indices, and the pathogen-associated fungal groups. In addition, we found that the temporal fungal succession on the spring leaves highly correlated with the disease occurrence in the orchards and with the temperature and precipitation variation from spring to summer. Of the pathogen-associated fungal groups, an increase in the relative abundance of Mycosphaerellaceae, hosting the causal agent of citrus greasy spot, correlated with the occurrence of the disease, while the relative abundance of Diaporthaceae, hosting the causal agent of melanose, was extremely low during the fungal succession. These results confirm that the two kinds of pathogen-associated fungal groups share different lifestyles on citrus, and also suggest that the study of temporal fungal succession in microbial communities can add to our understanding of the epidemiology of potential plant pathogens.