Introduction
Biogeographic patterns and the underlying mechanisms are essential in biodiversity research. The understanding of biogeographic patterns is crucial for predicting how global change affects biodiversity, designing effective conservation strategies, and achieving sustainable development (Gaston 2000). Studies on biogeographic patterns can be challenging due to the interactions among multiple biotic and abiotic factors, and to complex interrelationships among animals, plants, and microorganisms (Dolan 2006; Delavaux et al., 2019). Previous research focusing solely on macrobiota biodiversity may therefore be inadequate to clarify the full view of biogeography (Mascarenhas et al. , 2020). Despite microbes being the most important biotic driver of global material and energy flows, biogeographic research on microbiota has long lagged behind that of macrobiota, and there are debates on whether heterogeneous biogeographic patterns exists in microbes (Martiny et al. , 2006; Meyer et al.,2018). The followings may be the main obstacles of microbial biogeography.
The primary obstacle for the lagged research on microbial biogeography is evaluating microbial diversity accurately. Microbial communities contain extremely high microbial biodiversities and nonculturable taxa. So, the traditional pure culture methods cannot deal with the obstacle. Recent advances in metagenomics have finally made fast evaluations of microbial diversity possible (Hanson et al. 2012), while concerns about the new methodologies remain. First, metagenomic approaches use operational taxonomic units (OTU) clustered based on arbitrary identity thresholds (usually 97%), which reduces the resolution of the taxonomic data set and generates confusing patterns (Storch et al. 2008). Second, metagenomic approaches are not concern on the strain level and unable to match OTUs with actual taxa or identify dormant microorganisms (Dolan 2006; Land et al.,2015; Locey et al., 2020). Third, metagenomic approaches controversially neglect the rare taxa which are often more important than common taxa in analyses of microbial diversity (Gobet et al., 2012). Although current reaserch have begun to use amplicon sequence variants (ASV) with higher resolution to replace OTU, ASV is still an operational taxonomic unit in essence, which has not been completely solved those problems.
At last, the core obstable is the choice of research units. Many studies about microbial biodiversity lack systematic sampling designs, clear research boundaries, and appropriate research scales (Ladau & Eloe-Fadrosh 2019). The worse is that biogeographic patterns are particularly sensitive to above factors (Bay et al., 2020). Those caused confused results, and the confusions make it impossible for researchers to distinguish variation in microbial biodiversity caused by contemporary or historical factors (Castle et al., 2019).
We propose three solutions to overcoming above obstacles and clarify the confusions. First, microbial biogeographic studies should depend on natural units with clear boundaries. Current studies on microbial biodiversity use either an arbitrary grid as unit of analysis or nothing, which may result in mixing up the effects of various factors. Natural units with clear boundaries should also be the fundamental need for the study about biodiversity, because coupling animal, plant and microbial biodiversity distributions is inevitable for future biogeographic research (Peters et al., 2016). While biogeographic units for animals and plants have long been defined (e.g., ecoregions, biota, biogeographic provinces) (Hausdorf 2002; Smith et al.,2018), they do not match well with each other; great differences were among different regions and taxa (Smith et al., 2020). We propose that natural watersheds are ideal units for biodiversity research, because they are relatively independent or “close” units of material and energy flow, and their boundaries can be natural barriers of transmission (Timur 2013). In addition, the contemporary and historical effects driving species distribution should change gradiently in different scales of watersheds, and watershed units could help us distinguish contemporary and historical effects.
Second, a framework that considers the importance of rare species and removing the coverup of widespread species should be established. Because rare species comprise a large part of the microbial community and contribute greatly to the microbial geographical patterns (Lynch & Neufeld 2015), their narrow distribution cannot be ignored. However, most of the existing research focuses on common species which widely distributed. The widespread species with large biomass are likely to masked the heterogeneous biogeographic patterns of microbial community. This may be due in part to the overconfidence in technology and resource limitations, so inadequate sampling efforts are engaged in current metagenomic approaches. Limitations on the resolution of metagenomic approaches often results in the removal of rare taxa data during data processing (Bay et al., 2020). While the pure culture methods may be more effective at detecting rare species, its high workload makes it difficult to achieve comprehensive monitoring for huge microbial communities. To better account for the issue, we suggest 1) selecting a cultivatable and relatively small microbial taxon as the research object, 2) sampling adequately with stratified sampling strategy, 3) removing the coverup of widespread species at species level, and 4) analysing the widespread species biogeographic patterns at genetic level. These strategies will avoid the common problems plaguing the existing research system and should therefore set the standard for comprehensively understanding the biogeographic patterns of microorganisms.
Third, we suggest that studies about microbial biodiversity should be conducted in highly heterogenous environments. Microorganism dispersal is largely driven by both abiotic and biotic factors, and microbes are readily able to adapt and survive in a variety of environments. Indeed, studies conducted in homogeneous environments and within small spatial scales often reveal that either microorganisms are randomly distributed, or the weakness of environmental effects and dispersal limitations fail to indicate any non-random patterns in these region (Meyer et al., 2018; Liu et al., 2020). We believe it is easier to observe the spatial distribution pattern of microorganisms on large scales with high environmental heterogeneity.
Consequently, we conducted this research in Yunnan Province, China, which holds six major international rivers belonging to the Pacific and Indian Ocean systems. The watersheds were largely shaped by the collision of the Indian Ocean and Eurasian plates, with the formation of the Qinghai-Tibet Plateau. The spatial heterogeneity in Yunnan is very high, given that mountainous areas occupy 88.6% of the province.. Using the stratified sampling method, we collected 2,250 specimens from 228 sites in Yunnan Province (Figure 1, 2). We combined pure culture and molecular methods to investigate how common and rare Nematode-Trapping Fungi (NTF, a cultivable microbial group consisting of 99 known species ) taxa are spatially distributed across Yunnan. By doing so, we aim to verify our three proposed recommendations regarding 1) using watersheds as the research units, 2) removing the coverup of widespread species, and 3) stratified sampling from heterogenous environments, to overcome the challenges of microbial biodiversity research and to test our hypothesis that the microbial biogeographic patterns could be found in the perspective of watersheds.