The aim of this study was to evaluate the effect of the genotype and location on the structure of rhizosphere bacterial communities, between two Hass avocado crops. Among the most influential variables in the establishment of rhizosphere bacterial communities are abiotic factors such edaphic conditions and climatic variables; between the biotic factors, plant species, the genotype and the types of root exudates released into the soil are the most relevant (Qu et al., 2020). Here, significant differences are found when comparing evenness indices (Shannon, Pielou) between farms. No differences in the alpha diversity indices were observed when comparing between genotypes and no taxa with a relative differential abundance between genotypes was found. Although it has been reported that the genotype has a marked effect on the rhizosphere bacterial communities (Gallart et al., 2018; Marques et al., 2014; Shenton et al., 2016), there is evidence that this effect is attenuated by other factors like location, environmental conditions and soil type (Bowsher et al., 2020; F. Liu et al., 2019; Weinert et al., 2011).
The different root exudates produced by plants can recruit different bacterial groups in the rhizosphere (Phazna Devi et al., 2020; Zhang et al., 2014). Additionally, variation in root exudation patterns has been reported between different genotypes of the same species (Micallef et al., 2009; Mönchgesang et al., 2016). In this study it was not observed an effect of the genotype on the composition of rhizosphere bacterial communities, which might be an indication that these genotype does not have differences in its root exudation patterns, so probably no differential taxa are being recruited, as has been shown in other investigations, in which bacterial communities are similar under the same root exudates treatment (Bressan et al., 2009; Shi et al., 2011). In turn, between farms, at phylum level no differences were found in the composition of the bacterial communities, which is consistent with the hypothesis that plant species have similar microbiomes associated with the root (Berendsen et al., 2012; de Sousa et al., 2022).
In this study phyla Acidobacteria, Actinobacteria, Bacteriodetes, Cyanobacteria, Firmicutes Gemmatimanadetes, Proteobacteria, and Verrucomicrobia, were found in the avocado rhizosphere, which are common in rhizospheres of other plants (Trivedi et al., 2020). Also, here a phylogeny is proposed as hypothesis of the microbiome associated with the Hass avocado rhizosphere based on a classification of microorganisms according to estimated speciation events to form lineages (Washburne et al., 2018). The proposed phylogeny shows a wide phylogenetic diversity, basal phyla such as Actinobacteria, Chloroflexi, and more derived phyla such as Proteobacteria are observed (Hug et al., 2016). It has been proposed that niche differences increase with the phylogenetic distance of taxa (Stegan et al., 2012), which is consistent with the co-occurrence analysis that suggests multiple niches. Following this, the metagenome prediction also shows several enriched pathways.
The results presented here are similar from those reported in the study of Aguirre-von-Wobeser et al., (2021) where the rhizosphere communities of a Hass avocado intensive crop in Mexico were characterized through metagenomics, in soils managed with chemical and organic fertilization, and were dominated by Proteobacteria, Actinobacteria and Acidobacteria phyla. Solís-García et al., (2021) also characterized the composition of the Phytophthora-infected Persea americana rhizosphere in Mexico and reported in order of greater to lesser relative abundances the phyla Proteobacteria, Acidobacteria, Verrucomicrobia, Actinobacteria and Chloroflexi. Although these studies were conducted under different conditions, these results lead to the thought that for avocado, there are similar rhizosphere bacterial communities that has weak differences in relative abundances within and between cultivars.
In addition to this, in native ecosystems, plant-microbiome interactions responded to different environmental conditions independently between ecosystems making a long-term coevolution of plant-microbe interactions. Thus, the genotypes of each plant have selected different bacterial communities depending on the conditions of each ecosystem. In contrast, in agroecosystems, each plant's genotypes are subjected to similar environmental conditions, so, similar rhizosphere bacterial communities are expected (Philippot et al., 2013). Alongside these considerations, it is important to consider that soil is the primary source of inoculum in the establishment of the rhizosphere microbiome, consequently, the soil is the main microbial reservoir in the rhizosphere, and in consequence, if there are differences between soils, these are probably mirrored in the rhizosphere microbiome, i.e. the rhizosphere community is a portion of the bulk soil community (Bulgarelli et al., 2013; Mendes et al., 2014). Thus, considering that this study was carried out in two different agroecosystems, the soil is potentially the most important driver of the compositional changes of bacterial communities and the differences in diversity observed between farms (Edwards et al., 2015; Han et al., 2020).
The differences observed in soil variables, allow us to categorize LE as a more fertile farm compared to CS, which enables us to relate the differences associated in bacterial communities with differences in the edaphic conditions. In this study, more diversity is observed in the least fertile soil (CS), compared with the most fertile soil (LE). Likely, a nutrient gradient takes place ranging from non-rhizospheric soil with a limited content of nutrients to rhizospheric soil with nutrients derived from rhizodeposits and exudates. So, the rhizosphere acts as a microenvironmental oasis that supports more diversity (Barajas et al., 2020). According to the theory of Copiotrophs-Oligotrophs, bacterial taxa can be classified according to their nutritional requirements. Copiotroph microorganisms have high nutritional requirements and exhibit high growth rates. By contrast, oligotrophic microorganisms sustain slower growth rates and can competitively displace copiotrophs in low-nutrient conditions (Fierer et al., 2007; Yao et al., 2017). Here, they were found ASVs exclusively present in LE, which is the farm with the best nutritional conditions in the soil. Among these unique ASVs in LE, there are 288 belonging to genera Chujaibacter and 196 belonging to the family Burkholderiaceae, which belong to the classes Gammaproteobacteria and Betaproteobacteria, respectively, both considered in general as copiotrophs. In addition, LE showed a higher proportion of Pseudomonas, and Acidibacter both genres belonging to the class copiotrophic Gammaproteobacteria (Fierer et al., 2007; Yao et al., 2017). It is important to mention the importance of the genus Pseudomonas as PGPR, considering the ability of different species of the genus to control pathogens or improve production in crops of interest (Dorjey, 2017). Here Pseudomonas is one of the main representative genera in both farms (Fig. 1B). Regarding CS, LE has a higher proportion of Pseudomonas, which could be benefiting the crop. For example, in the control of pathogens such as Phytophthora cinnamomi (Mamani & Aragón, 2018).
Co-occurrence networks shows relevant changes between treatments associated with the modules, nodes, edges, and centrality of the taxa that are not visible with traditional diversity metrics (Karimi et al., 2017). CS, the most diverse farm and the least fertile, presents a co-occurrence network with fewer nodes and edges but with more modules concerning the LE co-occurrence network. Given the fertility conditions of LE, it is probable that there is a weak nutrient availability gradient between rhizosphere and bulk soil, allowing wider colonization in the rhizosphere because the plant is less selective to recruit its microbiome (Barajas et al., 2020), and it is reflected in a network of co-occurrences with a higher number of nodes and edges as compared to CS. Thus, in the most fertile soil, there are more co-occurrences and probably an intense activity, which can make this microbiome more resilient to disturbances as well as better stability over time (Röttjers & Faust, 2018).
Modules can represent different niches within the microbiome (Röttjers & Faust, 2018), suggesting that more processes that structure the bacterial communities concerning LE could be occurring. Here we observed that some modules are constituted mainly by members of the same phylum, showing an ecological coherence in phylum rank, evidence also shows ecological coherence between the Phyla and the niches they occupy (Philippot et al., 2010). Likewise, here we observe that modules are constituted by representatives of different phyla, suggesting variability of the coherence between different taxonomic levels (Koeppel & Wu, 2012). A high number of co-occurrences indicates that the taxa share similar niches, while simpler networks like the one found in CS suggest that the taxa have variable ecological niches (Shen et al., 2021). Similarly, more complex co-occurrence networks have been observed in bulk soil which is a complex environment with more niches compared to rhizosphere soil which is a subset of the bulk soil (Mendes et al., 2014; Zhang et al., 2021).
Soil analyses showed that there are significant differences in phosphorus levels between LE and CS. This has already been reported as an explaining factor of the bacterial communities’ structure (Hu et al., 2020). Considering that phosphorus is a limiting factor in the soil, it represents a determining component in the establishment of bacterial groups since it selects the taxa capable of using the different phosphorus sources (Shen et al., 2016). Here, the farm with more phosphorus content (LE) presents a co-occurrence network with a greater number of nodes and edges, as compared to CS. Similarly, Gumiere et al. (2019) reported that the soil with more available phosphorus has bacterial co-occurrence networks with more nodes and edges in its rhizosphere, compared with soils with less available phosphorus. Here, it was found that the CS soil has a higher amount of Iron and also presents greater diversity concerning the LE farm. It has been reported that soil with higher iron content can be associated with greater bacterial diversity in the rhizosphere (Pii et al., 2016). According to this, Iron could be influential in observed diversity patterns.
Canonical Correspondence Analysis (CCA) shows that three soil factors partially explain the variation found in rhizosphere bacterial communities. These are: Soil protein, which is related to soil health because it represents a source of nitrogen (Adamczyk et al., 2010; Pii et al., 2016). It has already been reported that the amount of protein in the soil is a determining factor in the structuring of rhizosphere bacterial communities (Hurisso et al., 2018). Soil respiration is another variable that explains the variation found in bacterial communities. The measurement of soil respiration reflects the use of organic matter mediated by microbial activity so, it is a method to quantify soil fertility and nutrient availability (An et al., 2020). Thus, in this study, P, soil respiration, and soil protein are variables that correlate directly with the observed community structure. Also, LE presents a better production of avocados in terms of the number and weight of the avocados harvested. Apart from the soil conditions, the temperature may be influencing the results obtained, given that the optimal ranges of temperature for the Hass avocado are 17.9–19.7 ºC (Elfiati et al., 2019; Haney et al., 2008), and in CS, the average temperature was lower (P < 0.05), so this crop would not be within the optimal temperature range.
According to the metagenome prediction performed with PICRUSt2, in the microbiome of the Hass avocado root, the degradation of aromatic compounds is the most represented metabolic pathway. This indicates that in the microbiome, there are bacteria which are capable of using non-conventional sources of carbon and that can degrade complex molecules and consequently makes other carbon sources available (Ledger et al., 2012). The second most represented category of metabolic pathways is carbohydrate degradation, which may be directly associated with consuming carbohydrate-rich rhizodeposits (Zhou et al., 2020). These conventional carbon sources may allow the establishment of generalist taxa in the avocado rhizosphere. This, in turn, relates to the found few pathways associated with carbohydrate synthesis, which may reflect the use of rhizodeposits as a source of carbohydrates downplaying its synthesis by bacterial taxa (Bai et al., 2015; Xu et al., 2018). In this work, the pathways associated with the amino acid metabolism had a low representation (Fig. 7). Here, few pathways associated with amino acid synthesis were predicted. It has been reported that the absence of genes related to the synthesis of amino acids can confer an evolutionary advantage in environments with high amino acid and carbohydrates contents (Cole et al., 2017). Possibly, the avocado rhizosphere constitutes a similar environment that favors the absence of pathways associated with the synthesis of amino acids. In general, the prediction of the metagenome shows that there are more degradation pathways than synthesis. This could indicate that in the avocado rhizosphere, bacterial populations are actively consuming the rhizodeposits of plants.
In this work, the effect of the scion was not considered as a driver in the structuring of the rhizosphere bacterial communities, since the rootstock genotype has a main role in the molding of the rhizosphere bacterial communities, this possibly due to the exudation patterns associated with each genotype (Bernal et al., 2020). Nevertheless, some research shows that to use a resistant plant (rootstock) to replace the root system of a plant of economic interest (scion) but more susceptible, known as grafting practices, can have an effect on the structuring of the rhizosphere bacterial microbiome (Berlanas et al., 2019; D’Amico et al., 2018; Dries et al., 2021). Finally, whereas it has already been reported that rootstock has an effect on some parameters of interest for Hass avocado cultivation e.g. Number of fruits produced (Vink et al., 2021; Wang et al., 2020), in this study no differences associated with rhizosphere bacterial communities were found, so these seem to be a common trait to native rootstocks.