Skip to main content
Download PDF

Tillandsia landbeckii phyllosphere and laimosphere as refugia for bacterial life in a hyperarid desert environment

Microbiome volume 11, Article number: 246 (2023) Cite this article

Abstract

Background

The lack of water is a major constraint for microbial life in hyperarid deserts. Consequently, the abundance and diversity of microorganisms in common habitats such as soil are strongly reduced, and colonization occurs primarily by specifically adapted microorganisms that thrive in particular refugia to escape the harsh conditions that prevail in these deserts. We suggest that plants provide another refugium for microbial life in hyperarid deserts. We studied the bacterial colonization of Tillandsia landbeckii (Bromeliaceae) plants, which occur in the hyperarid regions of the Atacama Desert in Chile, one of the driest and oldest deserts on Earth.

Results

We detected clear differences between the bacterial communities being plant associated to those of the bare soil surface (PERMANOVA, R2 = 0.187, p = 0.001), indicating that Tillandsia plants host a specific bacterial community, not only dust-deposited cells. Moreover, the bacterial communities in the phyllosphere were distinct from those in the laimosphere, i.e., on buried shoots (R2 = 0.108, p = 0.001), indicating further habitat differentiation within plant individuals. The bacterial taxa detected in the phyllosphere are partly well-known phyllosphere colonizers, but in addition, some rather unusual taxa (subgroup2 Acidobacteriae, Acidiphilum) and insect endosymbionts (Wolbachia, “Candidatus Uzinura”) were found. The laimosphere hosted phyllosphere-associated as well as soil-derived taxa. The phyllosphere bacterial communities showed biogeographic patterns across the desert (R2 = 0.331, p = 0.001). These patterns were different and even more pronounced in the laimosphere (R2 = 0.467, p = 0.001), indicating that different factors determine community assembly in the two plant compartments. Furthermore, the phyllosphere microbiota underwent temporal changes (R2 = 0.064, p = 0.001).

Conclusions

Our data demonstrate that T. landbeckii plants host specific bacterial communities in the phyllosphere as well as in the laimosphere. Therewith, these plants provide compartment-specific refugia for microbial life in hyperarid desert environments. The bacterial communities show biogeographic patterns and temporal variation, as known from other plant microbiomes, demonstrating environmental responsiveness and suggesting that bacteria inhabit these plants as viable microorganisms.

Video Abstract

Introduction

Hyperarid deserts such as the Atacama Desert in Chile are largely devoid of macroscopic life, especially in the inner core, where available water has been reported to be below the limit for photosynthetic activity and therewith for plant growth [1]. The lack of water also limits microbial life in such deserts. Besides strong water limitation, microorganisms have to cope with a combination of high-temperature fluctuations, strong UV radiation, and in part saline soil conditions [2]. Consequently, habitats such as soils, which are consistently colonized by microorganisms and often host the most diverse microbial communities on Earth [3], are strongly depleted of microorganisms in hyperarid deserts [4,5,6]. Strong changes in bacterial community composition have been observed particularly in the transition from arid to hyperarid conditions, commonly characterized by an enrichment of Actinobacteria and in part Chloroflexi, whereas common soil bacteria such as Proteobacteria become often depleted, especially in surface soils [4, 7, 8]. Besides, altered community assembly processes have been reported with increasing aridity [9]. Thus, specialized microbial communities of limited diversity and substantially reduced in abundance and activity are found in this habitat [6]. Microbial life rather flourishes in specific refugia, which support highly specialized microbial communities. These can be observed beneath translucent quartz stones, underneath rocks, or within halite crusts (e.g., [1, 2, 10,11,12]). Desert plants might represent another refugium of microbial life hosting specifically adapted microorganisms, but this has so far not been studied for plants growing under hyperarid conditions.

In general, plants are well-known to be colonized by microorganisms, aboveground in the phyllosphere as well as belowground in the rhizosphere [13, 14]. Nonetheless, the bacterial communities of plants have only been analyzed in a few studies in semiarid or arid landscapes, including work on Acacia and Tamarix trees, Agave and Atriplex species, or cacti [15,16,17,18,19,20]. These studies reported differences in the microbial community composition in different plant compartments and plant species, temporal changes, or biogeographic patterns. Plants growing in hyperarid environments such as the Atacama Desert have so far only been studied with focus on the rhizosphere microbiota [21,22,23,24], while the phyllosphere microbiota has not been studied yet. Consequently, it is not known whether the phyllosphere provides another refugium for microbial life in this extreme environment and whether a typical phyllosphere microbiota can establish even under these conditions.

The Atacama Desert in Chile is considered as one of the driest deserts on Earth with plants occurring primarily along the eastern and western margins, where they are supported by sporadic rainfall or fog [25,26,27]. One particular plant genus being well adapted to hyperaridity in this desert is Tillandsia, a member of the Bromeliaceae family. Populations of the most widely distributed species, Tillandsia landbeckii, occur along the Coastal Cordillera, where they form so-called Tillandsia lomas (Fig. 1A) [28]. These vegetation formations occur mostly as spatially isolated generally monospecific populations in the Chilean and Peruvian coastal desert systems, although occasionally few other species of the genus might be intermingled at a low density [29, 30]. In Chile, Tillandsia lomas occur from 3 to 45 km inland at an elevation of 900–1300 m [28, 31], where rainfall has been reported to be < 2 mm y−1 (Fig. 1G), therewith representing the driest part of the desert [32]. Tillandsia plants are primarily supported by fog, which reaches the lomas in corridors from the coast (Fig. 1B) [28, 33]. They often form parallel bands, perpendicular to fog penetration (Fig. 1C) [28, 34]. They obtain fog water by water-absorbing trichomes, which cover their leaves, rather than by functional roots, which are usually lacking [33, 35, 36]. Nutrients such as nitrogen are considered to be also largely fog supplied [37, 38], whereas the lower parts of the shoots are usually buried under sandy substrate, therewith supporting anchorage [29]. Such belowground shoots are not considered as rhizosphere but have been termed laimosphere [39]. This compartment has not gained much attention in microbiome research so far but very likely represents a different habitat compared to the phyllosphere or the rhizosphere of other plants. A very first recent study reports that T. landbeckii supports microbial life in the soil beneath the plants, as this soil showed increased bacterial abundance and diversity compared to the bare soil [40]. However, the plant itself has not been analyzed in that work, neither above- nor belowground.

Fig. 1
figure 1

Tillandsia landbeckii populations in the Atacama Desert. A Plants of the Cerro Chipana population growing on a southwest facing hillslope. B Incoming fog at Cerro Chipana. C Salitrera San Lorenzo 2 population on a southwest facing hillslope and on sand dunes. D Partially dead plants of the Cerro Peninsula population. E and F Shoots and inflorescences of T. landbeckii. G Map of the Atacama Desert with elevation being color coded and sampling sites highlighted by yellow circles. Mean annual rainfall is indicated by black lines

Full size image

The aim of this study was to analyze the microbiota associated with T. landbeckii plants, which are adapted to grow under the driest conditions on Earth. Plant populations sampled in this study were located in regions with an aridity index ≤ 0.0125 [41] or rainfall ≤ 1 mm y−1 [38]. The overarching question was whether T. landbeckii provides specific habitats for bacteria in a hyperarid desert environment and supports microbial life. To evaluate this, we addressed the following more specific questions:

  1. 1)

    Are T. landbeckii plants colonized by a specific bacterial community that is distinct from its surroundings?

  2. 2)

    Do we see habitat differentiation, i.e., does the phyllosphere host a different bacterial community compared to the laimosphere?

  3. 3)

    Do biogeographic patterns exist in the T. landbeckii-associated bacterial communities?

  4. 4)

    Does the plant-associated bacterial community undergo seasonal (temporal) variation?

To address these questions, we collected shoots of T. landbeckii plants from seven different locations in the Atacama Desert in northern Chile. The plant-associated bacterial community was compared to the surface soil community to assess habitat specificity. We compared the bacterial community of the plant phyllosphere to the laimosphere in order to evaluate the capacity of T. landbeckii to offer different habitats for bacteria in a hyperarid desert and to host microbial communities resulting from specific selection processes. To assess seasonality, selected sampling sites were visited twice. In all samples, bacterial community composition was analyzed based on 16S rRNA gene-based amplicon sequencing.

Material and methods

Study sites and sample collection

Samples were collected from seven different T. landbeckii populations [31] in October 2016 (spring) and March 2017 (autumn) in the northern Chilean Atacama Desert (Fig. 1G and Table 1). As phyllosphere samples, shoots were collected from ten individual plants per study site. Collection occurred along a zigzag path within an area with a radius of approx. 15 to 110 m, depending on the population size at the respective site. Five to ten laimosphere samples were collected from the same plant individuals that were selected for taking phyllosphere samples (Table S1). In addition, dead plant material from two to five plants was collected above- or belowground at two sites (Salar Grande, Cerro Pajonal) in 2016. Furthermore, three to four barren soil surface samples (0–1 cm depth) were collected per site. These were taken near the first, middle, and last plant collected at each site and thus coincide with the spatial area of the sampled plant population.

Table 1 Location of study sites from south to north and sampling dates
Full size table

For the phyllosphere samples, several plant shoots per plant individual were manually plucked and transferred into paper bags. These were placed on silica in sealable plastic bags to dry the plant material. Silica was replaced during the drying process when needed. For the laimosphere samples, buried plant shoots were pulled from the ground and shaken to remove loosely attached sand particles before bagging. The buried shoots were up to 15 cm long and were buried in sand to a maximum depth of approx. 10 cm. Surface soil samples were collected with a small shovel and transferred into sealable plastic bags. All samples were shipped to Germany for further processing. Considering that all sample material was completely dry for shipping, we do not expect that substantial changes occurred in the microbiota due to shipment at ambient temperature. However, we cannot fully exclude that some changes were introduced during the drying process of vital plant material.

Sample processing and DNA extraction

Dry plant material was crushed under a laminar flow using sterile mortars and afterwards stored in 15-mL tubes at −20 °C until DNA extraction. For DNA extraction, 0.1 g of the homogenized material was transferred to Lysing Matrix A tubes provided with the FastDNA SPIN Kit for DNA extraction (MP Biomedicals). Extraction was performed following the manufacturer’s instructions with the following modifications. To rehydrate the material, 300 µL of sterile water was added, and the samples were incubated for 2–3 h at 4 °C. The kit-supplied cell lysis solution CLS-VF and protein precipitation solution PPS were added, and bead beating was performed in the FastPrep instrument for 90 s twice for lysis. The matrix-bound DNA was additionally purified by removing the supernatant after brief centrifugation at 14.000 × g and resuspension in 1 mL of 5.5-M guanidine thiocyanate solution. The matrix was again pelleted by centrifugation and then resuspended in 600 µL of guanidine thiocyanate solution to load the matrix onto the kit-supplied spin filter for further processing. DNA elution was performed with two times 50 µL of DES elution solution.

DNA extraction from surface soil was most successful with the NucleoMag DNA Microbiome kit (Macherey–Nagel). Extraction was performed from 0.5 g of soil following the instructions of the kit with the following modifications: (i) the volume of the kit-supplied lysis buffer MI1 was increased to 800 µL; (ii) the lysates were additionally incubated with lysozyme (50 mg/mL) for 30 min at 37 °C, followed by incubation with proteinase K (20 mg/mL) for 30 min at 55 °C; (iii) elution of DNA was done in two steps (2 × 50 µL PCR grade water) resulting in 100 µL total eluate for each sample; and (iv) DNA was concentrated using vacuum drying. As extraction controls in both procedures, sterile tubes with only ceramic beads were used and treated similarly to the plant and soil samples. DNA concentration in all extracts was quantified using the QuantiFluor dsDNA System (Promega Corporation, Fitchburg, WI, USA), and DNA was stored at −20 °C for further use.

16S rRNA gene PCR and sequencing

For bacterial community analyses, 16S rRNA gene amplicons were generated in a nested PCR approach similarly as described by Becker et al. [42]. DNA was first amplified using a LNA PCR protocol with modified primer set 63f/1492r to suppress the amplification of plant organelle-derived 16S rRNA genes [43]. The 25-μL PCR reaction mixture consisted of 5 µL of 5 × polymerase buffer, 0.25 µL of each primer [25 mM], 0.25 μL of BSA [0.8 µg/µL], 0.25 µL of dNTP mixture [25 mM of each dNTP], 1 µL of MgCl2 [50 mM], 0.25 μL of Herculase II Fusion DNA polymerase (Agilent Technologies), and 3 μL of DNA template and PCR-grade water (Qiagen). The PCR reaction was conducted upon initial denaturation at 95 °C for 4 min with 30 cycles (95 °C, 30 s; 70 °C, 30 s for LNA primer annealing; 56 °C, 30 s for 63f/1492r primer annealing; 72 °C, 45 s), followed by a final elongation at 72 °C for 10 min. A subsequent nested PCR was performed using the primers 799f/1193r (V5–V7 region) with sample-specific 8-mer barcodes in 3 × 50-μL assays to obtain enough PCR product for the downstream cleanup steps. Here, 5 μL of PCR product from the first round was used as template and applying a temperature profile consisting of initial denaturation at 95 °C for 4 min, 15 cycles (95 °C, 30 s; 45 °C, 30 s; 72 °C, 30 s) and final elongation as before. The three technical replicates of each sample were pooled prior to DNA quantification using the QuantiFluor dsDNA system (Promega). PCR products of the different samples were pooled at equimolar ratios, and the resulting pool was purified and concentrated using the HighPrep PCR Clean-up System kit (MagBio Genomics, Gaithersburg, MD, USA). The pooled PCR products of the correct size were purified from agarose gel using the QIAEX II gel extraction kit (Qiagen) to eliminate traces of unspecific products. Library preparation and sequencing on a NovaSeq system (Illumina) were done by Novogene (UK), generating paired-end reads (2 × 250 bp).

Sequence data analysis

The raw sequence reads were processed as described by Becker et al. [42]. In brief, the data were demultiplexed, primers removed, and reads further processed with QIIME2 version 2022.2 [44]. The classified reads were quality filtered by removing rare amplicon sequence variants (ASVs) that appeared less than 20 times and in less than five of the 156 successfully sequenced samples (Table S1). Likewise, samples with less than 10,000 reads were excluded. Two of the samples were removed as a result of these filtering parameters. Quality-filtered reads (25,922,551 sequences) were processed using default parameters of DADA2 [45] implemented in QIIME2 and grouped into 1457 ASVs, ranging from 17,981 to 1,083,504 sequences per sample (Table S1). The taxonomic assignment of ASVs was done using a custom classify-sklearn plug-in classifer against the SILVA SSU138 Ref NR99 database by sub-setting to the amplicon region and using the last common ancestor method [46,47,48]. ASVs classified as mitochondria, chloroplasts, and Eukaryota were removed (1.3% of assigned ASVs). This dataset consisted of 1438 ASVs and was further analyzed using the phyloseq R package (version 1.40.0) [49].

As DNA extraction controls and PCR-negative controls of the surface soil samples were slightly positive according to agarose gel electrophoresis, they were included in sequencing. The ASVs resulting from these controls were removed from the soil samples following the prevalence-based approach of the decontam R package (version 1.16.0) [50]. Here, the prevalence (presence/absence across samples) of each ASV in true-positive samples was compared to the prevalence in negative controls to identify the contaminants. Both extraction and negative controls were used to assess the prevalence of the contaminant ASV sequences in the surface soil samples. The threshold was set to 0.5, therewith identifying all sequences as contaminants that were more prevalent in negative/extraction controls than in positive samples. This resulted in the removal of 103 ASVs. Additionally, all reads representing Aquabacterium, which dominated in the DNA extraction control but were not fully removed by decontam, were manually removed in phyloseq (301 ASVs). The remaining 1034 ASVs were used for final data analyses (Table S2). We rate the surface soil samples of sufficient quality for the analysis we present and conclusions we draw, even though the extraction control remained similar to some of the soil samples (Figure S1).

Statistical data analysis and visualization

Statistical analyses were performed in QIIME2 and R (version 4.2.0) [51]. Analyses were first done on the complete dataset including all samples from phyllosphere, laimosphere, and surface soil. More detailed analyses were then performed with data subsets to compare phyllosphere vs. laimosphere, living vs. dead plants (Fig. 1C and D), and spatial and seasonal dynamics. Alpha diversity was estimated by Shannon’s diversity index using a feature table rarefied to 10,000 reads per sample. Additionally, evenness and Faith’s PD indices were analyzed in selected datasets upon significant results from Shannon’s diversity. A Kruskal–Wallis test was applied to test for significant differences, followed by Dunn’s test with Benjamini–Hochberg correction for multiple testing. For data comparisons between phyllosphere vs. laimosphere and seasonal dynamics in the phyllosphere, a Wilcoxon signed-rank exact test was applied for paired sample sets and a Wilcoxon sum exact test for some non-paired sample sets (e.g., Cerro Pajonal 2016 vs. 2017).

Differences in bacterial community composition were evaluated by principal component analysis using the non-rarefied dataset upon robust-centered log-ratio (rCLR) transformation [52]. This analysis was performed with the R package vegan (version 2.6–2) [53]. Statistical differences were calculated by adonis2 function (vegan package) on a robust Aitchison distance matrix, which is a form of one-way permutational multivariate analysis of variance (PERMANOVA). This was followed by pairwise comparisons (pairwiseAdonis version 0.4) [54] using Benjamini–Hochberg correction for multiple testing. For the comparison of phyllosphere to laimosphere, pairwise differential abundance analysis at ASV level was performed using analysis of compositions of microbiomes with bias correction (ANCOM-BC) with detection for structural zeros turned on [55]. Conservative variance estimates of the test statistic were used, and p-values were adjusted using Holm’s correction (padj). All ASVs with padj ≤ 0.05 and log2 fold changes of ≥ 2 or ≤ −2 were considered as significantly differentially abundant.

The visualization of the data was done in R using ggplot2 (version 3.3.6) [56]. A Venn diagram was generated with the ps_venn function of the MicEco package (version 0.9.18) [57]. The heat trees were created within the metacoder package using the heat-tree function [58]. Here, the number of reads in each taxon was calculated, and the samples were grouped by phyllosphere or laimosphere. The abundance value of 0.001 was set as threshold to remove the low-abundance taxa and reduce the complexity of the plots.

Results and discussion

Compartment-specific colonization of T. landbeckii plants

PCR amplification and sequencing were successful for the majority of phyllosphere and laimosphere samples (Table S1), indicating that bacteria were associated with these plants. To evaluate whether these two plant compartments host specific bacterial communities, we analyzed the associated bacterial communities comparatively to the community structure in the surface soil. Both the soil surface and the phyllosphere are assumed to receive bacteria from atmospheric deposition processes and should be similar in case deposition would be the major process leading to an accumulation of bacterial cells in these habitats. Differentiation of bacterial communities in dependence on the two plant compartments and surface soil samples was evaluated by principal component analysis (PCA) using the amplicon data of all samples (Fig. 2A). This revealed a clear separation of samples due to the origin from phyllosphere, laimosphere, and the soil surface, especially along the first axis of the plot, which covered 11.4% of the variation. The distinctiveness by compartment was confirmed by PERMANOVA (R2 = 0.187, p = 0.001). Subsequent pairwise PERMANOVA resulted in significant differences between all three sample types with the soil surface being most distinct to phyllosphere (R2 = 0.13, padj. = 0.001) and laimosphere (R2 = 0.14, padj. = 0.001) (Fig. 2A). Thus, it can be concluded that T. landbeckii provides distinct habitats for bacteria in comparison to the surface soil. The leaf-associated microbiota is thus most likely not the result of mere atmospheric deposition, but instead the consequence of selective processes, leading to the establishment of a characteristic phyllosphere microbiome. This is further supported by the finding that the laimosphere was also clearly distinct from the phyllosphere, indicating that this represents another specific habitat for microorganisms.

Fig. 2
figure 2

Differences in bacterial community composition between phyllosphere, laimosphere, and bare soil surface samples. A Principal component plot (of rCLR transformed data) showing differences between the three sample types (symbol color) and of the seven different T. landbeckii populations (symbol shape). Significant differences between groups of samples were evaluated based on PERMANOVA on a robust Aitchison distance matrix and are listed besides the plot. B Differences in Shannon’s index, evenness, and Faith’s PD. Overall differences were assessed based on Kruskal–Wallis tests, followed by Dunn’s test with Benjamini–Hochberg correction for pairwise comparisons

Full size image

Differences were also observed in alpha diversity with lower Shannon’s index in the phyllosphere than in the laimosphere and soil surface (p < 0.001) (Fig. 2B). While evenness showed the same pattern, Faith’s PD, informing about the phylogenetic diversity by considering phylogenetic distances of the detected taxa, was not significantly different between the three sample types. Thus, the observed differences in Shannon’s index are primarily the result of lower evenness in the phyllosphere, while phylogenetic diversity was similar in all three compartments. This is remarkable, because bacterial diversity in the phyllosphere is usually substantially lower than in soil and also lower than in the rhizosphere [14, 59]. Plants in hyperarid deserts may thus be able to support a relatively diverse microbiota compared to their surroundings, even though it has to be kept in mind that the soil surface may be particularly depleted in diversity [4].

At this point, it should be noted that the use of different DNA extraction procedures for phyllosphere and laimosphere samples compared to soil samples may have contributed to observed differences between these sample types. However, the possible method-related variation can be considered to be much smaller than the clear sample-type-specific differences we observed. The use of different kits was necessary to obtain sufficient DNA of adequate quality from the different sample types.

The differences between phyllosphere and laimosphere were assessed more specifically by focusing on corresponding samples, i. e., pairs of phyllosphere and laimosphere samples that were collected from the same plant individual. PCA and PERMANOVA for this data subset confirmed the differences in the bacterial community composition of these two compartments (R2 = 0.108, p = 0.001) (Fig. 3A). As before, Shannon’s diversity index (p < 0.01) and evenness (p < 0.05) were higher in the laimosphere than the phyllosphere, while Faith’s PD remained at the same level (Fig. 3B). Despite equal Faith’s PD, we observed a higher overall ASV richness when looking at all laimosphere samples in comparison to the overall observed richness in all phyllosphere samples (Fig. 3C). In agreement with the higher overall richness, more unique ASVs (32% of total) were detected exclusively in the laimosphere than in the phyllosphere (24% of total) (Fig. 3C). This underlines that the above- and belowground parts of T. landbeckii plants provide distinct habitats for bacteria, with the laimosphere hosting a more even community with more unique ASVs than the phyllosphere.

Fig. 3
figure 3

Differences in bacterial community composition between phyllosphere and laimosphere with focus on those 27 plant individuals that were sampled for both compartments. A Principal component plot (rCLR transformed) illustrating beta diversity. Differences related to compartment were evaluated by PERMANOVA on a robust Aitchison distance matrix. B Differences in Shannon’s index, evenness, and Faith’s PD. Differences were assessed based on Wilcoxon signed-rank test for each pair of indices. C Venn diagram illustrating shared versus unique ASVs in the phyllosphere and laimosphere. The number of corresponding ASVs is indicated, with corresponding percentages given in parentheses

Full size image

Further habitat differentiation may exist between living and dead plants. To study this aspect, we had included some samples from dead plants, collected from the Salar Grande and Cerro Pajonal populations, where T. landbeckii populations were less vital compared to other study sites, with a remarkable number of dead plant individuals side by side to still living individuals (Fig. 1D). However, DNA extraction and PCR amplification of bacterial 16S rRNA genes from the dead plant material were of limited success. As a consequence, differences between living and dead plants could not be robustly assessed. They were overlaid by site-specific effects (Figure S2), which could not be excluded by data sub-setting due to the limited number of samples from dead plants per site. A further differentiation would have been expected, considering that plant-driven selection processes are relevant for shaping the phyllosphere microbiota. This requests further analyses in the future.

Dominant bacterial classes and genera of T. landbeckii plants are well-known phyllosphere colonizers and include arthropod endosymbionts

Phyllosphere bacterial communities of the individual plants were dominated by Alphaproteobacteria, Gammaproteobacteria, Actinobacteria, or Bacteroidia (Fig. 4), which are known to represent dominant classes in the phyllosphere [14]. Likewise, several of the dominant genera (> 10% relative abundance in a phyllosphere sample; Figure S3) are well-known phyllosphere colonizers. This applies to all dominant Gammaproteobacteria, i.e., Pseudomonas, Massilia, and Ralstonia, as well as Hymenobacter (class Bacteroidia), which have all been reported to occur prominently or as part of a phyllosphere core microbiota on other plants (e.g., [14, 51, 60,61,62,63,64,65,66]). Similarly, dominant genera within the class Actinobacteria, i.e., Modestobacter and Kineococcus, are known as phyllosphere colonizers [51, 61]. Several type strains of these two actinobacterial genera were isolated from halophytic plants [67,68,69,70], desert [71, 72], or saline environments [73]. Thus, members of these genera may be particularly well adapted to water-limiting conditions, likewise as reported for soil-dwelling Actinobacteria [74, 75]. However, we did not observe a general dominance of Actinobacteria in the phyllosphere, as often seen in hyperarid desert soils [4, 7, 8]. Instead, a predominance of Actinobacteria was only seen on some plant individuals from two study sites, Cerro Guanaco and Salar Grande (Fig. 4), indicating that drought-adapted members of this genus are not consistently dominating in the phyllosphere.

Fig. 4
figure 4

Bacterial community composition in phyllosphere and laimosphere samples, collected at seven different study sites. All samples collected per site are included here

Full size image

Similarly, the genera Wolbachia and “Candidatus Uzinura” occurred dominantly in the phyllosphere of some T. landbeckii individuals. These organisms are known as endosymbionts from plant(-sap) feeding arthropods and may use plants as intermediate hosts, being possibly transferred to plant tissue upon insect feeding [60, 76,77,78]. This is supported by a rather inconsistent occurrence of these taxa on plant individuals within and between the study sites, which is a logical consequence of selective (e.g., related to plant fitness) or stochastic insect attacks of plant individuals. The detection of these two endosymbionts, likewise as of endosymbionts of the family Morganellaceae (Figure S3), raises the question of which arthropods might live in these hyperarid environments. During field sampling, no insect infestation of the Tillandsia plants was evident, but the presence of the endophytes suggests that the plants support also insect life under hyperarid conditions.

Characteristic and unusual colonizers in the phyllosphere of T. landbeckii plants in comparison to the laimosphere

To identify phyllosphere-specific ASVs more systematically in comparison to the laimosphere, differential abundance analysis was performed using ANCOM-BC on the reduced dataset of corresponding phyllosphere-laimosphere samples obtained from 27 plant individuals across the seven study sites. In the phyllosphere, 51 ASVs were identified as significantly enriched with padj. ≤ 0.05 and log2 fold change ≥ 2 (Fig. 5 and Table S3). These ASVs represented in total 85% of the mean relative abundance in the phyllosphere samples, but only 24% of the mean relative abundance in the laimosphere samples (Table S3), indicating that several of the phyllosphere-enriched ASVs were dominant members in this community and clearly depleted in the laimosphere. They represented mostly Gammaproteobacteria (18 ASVs, summed mean relative abundance 42.0%), followed by Alphaproteobacteria (13 ASVs, 7.1%), Acidobacteriae (6 ASVs, 8.7%), and Bacteroidia (2 ASVs, 42.1%).

Fig. 5
figure 5

Significantly enriched ASVs in the phyllosphere and laimosphere of T. landbeckii plants identified by ANCOM-BC. Displayed are ASVs with padj. ≤ 0.05 and log2 fold changes of  ≥ 2 (abundant in phyllosphere) and ≤ −2 (abundant in laimosphere), with the color gradient illustrating the fold change according to the legend. Each light-bordered box represents one ASV, with the genus name given in black. The class names are indicated in light color, grouping the ASVs in big boxes with dark borders. The segment sizes reflect the mean relative abundance of the ASVs across all samples from the respective compartment, as included in ANCOM-BC, which included all plants from which a phyllosphere and corresponding laimosphere sample was taken. The names of the low-abundant taxa can be found in Table S3

Full size image

The most abundant ASV (ASV406, 32.9% mean relative abundance) in the phyllosphere was representing Ralstonia and had a log2 fold change of 2.5 in comparison to the laimosphere. Ralstonia has been reported as dominant or consistent phyllosphere colonizer in some other studies [60, 63, 79]. The genus contains some serious soil-born plant pathogens (Ralstonia solanacearum, Ralstonia syzygii, Ralstonia pseudosolanacearum) causing bacterial wilt [80], but is not yet known as pathogen of Bromeliaceae. Considering that plants appeared healthy, these were probably nonpathogenic strains, or the plants are not susceptible. The ASV showing the strongest enrichment (log2 fold change 8.7) was “Candidatus Uzinura” (ASV280, with 21.7% mean relative abundance, Table S3), the endosymbiont of arthropods discussed above. Further ASVs occurring with high relative abundance and enriched in the phyllosphere (log2 fold change between 2.4 and 6.8) were members of Acidiphilium (ASV43), Wolbachia (ASV89), Pseudomonas (ASV108), Burkholderia-Caballeronia-Paraburkholderia (ASV334), and subgroup2 of the Acidobacteriae (ASV316) (Fig. 5 and Table S3). Several of these and further less-abundant, significantly enriched taxa such as Acinetobacter, Sphingomonas, Mesorhizobium, and Rhodococcus (Fig. 5) are well-known phyllosphere colonizers [14], thus underlining that T. landbeckii provides a characteristic phyllosphere habitat. The presence of these taxa may be explained by the fact that the phyllosphere of many plants, not only desert plants, is representing a rather hostile environment where microorganisms are exposed to drought and UV radiation [14]. Phyllosphere colonizers are thus somewhat adapted to these conditions, possibly enabling some taxa to colonize even T. landbeckii plants. Besides, it has to be kept in mind that we did not separate endophytes from epiphytes in this study. Endophytes may need less adaptations to the harsh conditions compared to epiphytes for survival in association with plants growing in hyperarid environments.

The enrichment of two ASVs representing Acidobacteriae subgroup2 in the phyllosphere was rather exceptional. Acidobacteriae have been detected in the phyllosphere, especially of tropical trees, but predominantly subgroup1, not subgroup2 [81,82,83]. Similarly, the presence of Acidiphilium was rather unexpected, as it is usually occurring in acidic, mineral, and oligotrophic environments [84], but not in the phyllosphere. This taxon was primarily detected at Alto Chipana but consistently present with lower relative abundance in the phyllosphere of all other plant individuals. Due to the consistent occurrence, these otherwise fairly uncommon phyllosphere colonizers appear to be part of a T. landbeckii-specific microbiota, therewith contributing to the plant-host specificity of the microbiome, which is well-known to exist between different taxonomic groups of plants, sometimes even down to plant cultivar level [85,86,87]. Their occurrence may result from the extreme conditions the plants and associated bacteria are exposed to in hyperarid environments. However, microbiota studies from plants growing in semiarid or arid deserts do not provide further evidence for this; the occurrence of these genera has not been reported there [18,19,20]. It is thus tempting to speculate that the transition from arid to hyperarid conditions introduces specific changes in the plant-associated microbiota, analogous to observations made in soil [4], or that T. landbeckii plants have some unique features that support these taxa, especially in the phyllosphere.

Taken together, these results suggest that the T. landbeckii phyllosphere bacterial community hosts several taxa commonly known from plants grown under less arid conditions but includes in addition some particular taxa. The detected actinobacterial genera and Bacilli suggest that a selection of particularly drought-tolerant taxa may occur. The relevance of drought adaptation of the T. landbeckii microbiota should be studied in more detail in the future, not only for the Actinobacteria and Bacilli but also for the well-known phyllosphere colonizers. Likewise, the plant traits and processes leading to the specific presence of subgroup2 Acidobacteriae and Acidiphilium deserve further attention.

The T. landbeckii laimosphere hosts soil and rhizosphere-dwelling bacteria in addition to phyllosphere colonizers

Similar to the phyllosphere, Gammaproteobacteria, Actinobacteria, or Bacteroidia dominated in the laimosphere (Fig. 4). In addition, Bacilli were quite prominent here, while Alphaproteobacteria were less abundant. Focusing on dominant genera (> 10% relative abundance in a sample), 11 of the 17 genera that were dominant in the phyllosphere were also found as dominant taxa in the laimosphere (Figure S3). When evaluating all ASVs, the overlap between the phyllosphere and laimosphere was 44% (Fig. 3C). This indicates that the laimosphere shares a substantial part of its microbiota with the phyllosphere. Besides, the laimosphere hosted several additional abundant genera compared to the phyllosphere (Figure S3). These were primarily members of the Actinobacteria, Bacteroidia, and Gammaproteobacteria (Figs. 4 and S3). Moreover, some dominant members of the phylum Acidobacteriota were present. In agreement with the higher number of dominant genera, 32% of the ASVs were uniquely detected in these samples (Fig. 3C). The specific comparison between phyllosphere and laimosphere by ANCOM-BC revealed 52 ASVs to be significantly enriched with a log2 fold change ≤ −2 in the laimosphere (padj. ≤ 0.05) relative to the phyllosphere. In total, these 52 ASVs represented 44% of the mean relative abundance of the bacterial community in the laimosphere samples but less than 1% of the mean relative abundance of the phyllosphere samples (Fig. 5 and Table S3), therewith underlining the distinctiveness of the two compartments. Most of these ASVs belonged to the class Actinobacteria (24 ASVs, 11.3% summed mean relative abundance), followed by Bacilli (10 ASVs, 9.5%), Bacteroidia (10 ASVs, 9.0%), and Gammaproteobacteria (7 ASVs, 11.6%).

The most prominent ASV with significant enrichment in the laimosphere was a member of the genus Pseudomonas (ASV81, 6.5% mean relative abundance, log2 fold change of −3.1), which was different from the abundant Pseudomonas ASV in the phyllosphere (ASV108, 3.4%, log2 fold change of 3.8, Table S3). Likewise, different ASVs representing Acidiphilium and Bryobacter were enriched in the phyllosphere and laimosphere (Fig. 5). This points to niche differentiation at species or even strain level within these genera. Among the most strongly enriched taxa in the laimosphere (log2 fold change between −3.5 and −4.6) were Cellulomonas, Kocuria, Modestobacter (all from the class Actinobacteria), Planomicrobium and Bacillus (Bacilli), Salinisphaera, Kushneria and Noviherbaspirillum (Gammaproteobacteria), and Adhaeribacter (Bacteroidia), which were mostly represented by several differentially abundant ASVs (Table S3). All these genera include soil-dwelling organisms, and most include strains that have been reported to exist under dry or high-salt conditions [88,89,90,91,92]. Moreover, some of those were also detected in our surface soil samples (Table S2), indicating that those could be soilborne taxa, which colonized the buried plant parts in addition. In further agreement with this assumption, several of these taxa (Cellulomonas, Kocuria, Planomicrobium, Bacillus, Pseudomonas, and Kushneria) have been reported to establish in the rhizosphere of desert plants or halophilic plants and have potentially plant beneficial traits [88, 93,94,95,96,97]. Furthermore, some have the capability to degrade plant residue (Cellulomonas, Bacillus, Adhaeribacter, Pseudomonas) [98,99,100,101,102]. This indicates that the laimosphere is colonized in part by soil-derived bacteria, which probably benefit from plant carbon. However, it is not yet known whether organic carbon is leaking from the buried plant shoots or whether the bacteria actively decompose the buried shoots under conditions when sufficient water is available for metabolic activity. Most of the time, soil water contents will be too low to allow decay of these buried shoots [41], therewith guaranteeing the maintenance of the anchorage function of these shoots upon burial. These remain largely intact, sometimes for centuries even after burial of the complete plant [38, 103], due to very slow degradation processes. Upon burial, the microbiota probably undergoes a very slow transition from a typical phyllosphere microbiota towards a more saprotrophic microbiota.

Contrasting biogeographic patterns exist in the phyllosphere and laimosphere of T. landbeckii

Biogeographic patterns are commonly observed in the plant-associated microbiota [16, 85]. To gain insight into spatial community heterogeneity of the plant-associated microbiota of T. landbeckii in the two compartments, the data were subsetted by compartment, and the bacterial community structure was compared between study sites. All three alpha-diversity indices, i.e., Shannon’s index, evenness, and Faith’s PD, differed for the phyllosphere bacterial communities between the different study sites (Figure S4). For the laimosphere, differences were less evident. Regarding beta diversity, PCA and PERMANOVA revealed spatial variation in the phyllosphere (R2 = 0.331, p < 0.001), which was even more pronounced in the laimosphere (R2 = 0.467, p < 0.001) (Fig. 6). In the laimosphere, samples from Alto Chipana appeared most separated in the PCA plot, while this was not evident in the phyllosphere, where Salar Grande samples appeared most distinct (Fig. 6). This underlines that the two plant-associated bacterial communities are controlled by different environmental factors. Distance decay will contribute to spatial patterns, but this was not the main factor for differences in the two compartments of T. landbeckii in the Atacama Desert, as sample clustering by sites in the PCA plots does not reflect geographic distances between the study sites. Consequently, other site-specific environmental factors must be more relevant drivers for community assembly. Assuming that the laimosphere is influenced by the surrounding soil substrate, we evaluated comparatively the biogeographic patterns of the bacterial communities in the surface soil samples, revealing that strong spatial patterns exist also in these samples (R2 = 0.517, p < 0.001), with samples from Alto Chipana and Salitrera San Lorenzo 2 being most distinct from the other samples (Figure S1). As samples from Alto Chipana were also most distinct in the laimosphere from all others and those from Salitrera San Lorenzo 2 were separated from many others along the first axis (Fig. 6), soil-related traits and the recruitment of soil-derived microorganisms in the laimosphere likely contributed to the variation between sites in the laimosphere. Soil traits that may play a role are pH, organic carbon content, electric conductivity, or the concentration of specific salts [4, 104, 105]. The plant itself is probably another factor, being relevant for both the phyllosphere and laimosphere. The studied T. landbeckii plants are known to be genetically distinct [31], resulting in potential phenotypic differences, which might lead to altered community structures [106,107,108,109]. This also deserves a more detailed evaluation in the future. Moreover, climatic conditions, such as fog availability, may have an influence in particular on the phyllosphere microbiota, acting either directly or indirectly via altered plant physiology.

Fig. 6
figure 6

Differences in bacterial community composition in the phyllosphere and laimosphere related to sampling site. Shown are principal component plots and PERMANOVA results based on robust Aitchison distance matrices

Full size image

Temporal changes provide evidence for environmentally responsive bacterial communities in the phyllosphere of T. landbeckii

Having proven the presence of specific bacterial communities in the different compartments of T. landbeckii and the existence of spatial patterns, we were interested in the environmental responsiveness of the plant-associated microbiota. Due to the limitation of water, microbial metabolic activity and, therewith, responsiveness might be strongly limited, similarly as known from the hyperarid desert soil in the absence of water [6]. To gain evidence, we analyzed the temporal dynamics of the bacterial communities in the phyllosphere. Therefore, we collected samples from four study sites twice, in spring (October 2016), and half a year later in autumn 2017 (March). Overall, all three alpha-diversity indices changed significantly over time (p ≤ 0.002) (Figure S5). A more detailed analysis within each site revealed that such changes were not observed consistently across all sites. A significant decline in bacterial diversity (Shannon’s index and Faith’s PD) was observed at Cerro Pajonal, whereas a less pronounced decline was seen in samples from Alto Chipana-Rio Loa (Faith’s PD only). PCA revealed clear differences related to season in beta diversity, partly reflected by a low though significant R2 value in PERMANOVA (R2 = 0.064, p = 0.001) (Fig. 7). The season-related changes were superimposed by the stronger spatial patterning (R2 = 0.326, p = 0.001) and a combined season/site effect (R2 = 0.081, p = 0.001). When assessing the temporal changes site specifically, significant changes in the phyllosphere bacterial community composition over time were seen at three study sites (PERMANOVA R2 ranging from 0.171 to 0.532) (Fig. 7). This leads to the conclusion that the phyllosphere microbiota does undergo temporal changes. It appears likely that these changes are related to weather conditions, which are known to be a controlling factor for plant-associated microorganisms in general [110, 111]. Water availability via incoming fog may be a major factor here. Fog availability is higher between July and November than in the other months [41, 112, 113] and was therewith higher at the first sampling event than the second. In this context, the observed decline in alpha diversity appears reasonable, assuming that only the better adapted microorganisms and endophytes withstand the dryer period of the year. Besides a direct impact of fog on the microorganisms, the increased availability of fog-derived water has implications on plant productivity. T. landbeckii plants at Salitrera San Lorenzo 2 were reported to have lowest growth rates in May to August and highest rates in August to November, along with the higher fog availability at this time [113]. This may change habitat conditions for the phyllosphere microorganisms, contributing to seasonal community compositional changes. The specific roles of fog and host plant factors request further evaluation in the future. The main conclusion drawn from the observed differences here is the existence of temporal variation, serving as indicator for responsiveness of the T. landbeckii phyllosphere microbiota to environmental cues.

Fig. 7
figure 7

Differences in bacterial community composition in the phyllosphere related to sampling site and season. Shown are principal component plots and PERMANOVA results based on robust Aitchison distance matrices

Full size image

Conclusion

In this study, we demonstrated that T. landbeckii plants, which grow under extreme conditions in hyperarid deserts, provide distinct habitats for microorganisms. The phyllosphere hosts in part well-known phyllosphere colonizers, indicating that T. landbeckii plants provide a refugium of microbial life in hyperarid deserts. In addition, some detected genera are rather unique for the Tillandsia phyllosphere, pointing to a selection of particular host- or habitat-specific taxa. The bacterial community of the laimosphere was observed to be partially distinct from the one in the phyllosphere, in part due to the presence of soil-dwelling taxa. Biogeographic patterns were more prominent in the laimosphere than the phyllosphere, indicating that the community assembly in these two compartments is driven by different deterministic as well as stochastic processes. Climate, especially fog and therewith water availability, appears to be an important factor acting in particular on the phyllosphere microbiota, either directly or indirectly via plant fitness, whereas soil traits such as water storage capacity may modulate the laimosphere microbiota in addition. In particular, the role of fog for the plant-associated microbiota deserves more attention in future studies, as it may explain the temporal variation observed in this study. Both the spatial and even more so the temporal variation in the bacterial community structure, we observed here points to a microbiota that is responsive to environmental cues. Thus, it appears unlikely that the plant-associated microbiota of T. landbeckii, growing under hyperarid conditions, is merely the result of an accumulation of (dead) microorganisms from dust depositional processes. Instead, the plant provides specific species-rich refugia of life for microorganisms above- and belowground in this hostile hyperarid desert. The microorganisms that colonize T. landbeckii should be studied in more detail with regard to their in situ metabolic activity, their adaptation mechanisms to live in this hyperarid environment, and how these differ from those of microorganisms living in the phyllosphere under less arid conditions. Moreover, the possible role of the T. landbeckii microbiome for plant growth support deserves attention.

Availability of data and materials

The 16S rRNA gene sequence reads generated in this study have been deposited in the European Nucleotide Archive (ENA) under accession no. PRJEB57843.

References

  1. Warren-Rhodes KA, Rhodes KL, Pointing SB, Ewing SA, Lacap DC, Gómez-Silva B, et al. Hypolithic cyanobacteria, dry limit of photosynthesis, and microbial ecology in the hyperarid Atacama Desert. Microb Ecol. 2006;52:389–98.

    Article  PubMed  Google Scholar 

  2. Azua-Bustos A, González-Silva C, Fairén AG. The Atacama Desert in northern Chile as an analog model of Mars. Front Astron Space Sci. 2022. https://0-doi-org.brum.beds.ac.uk/10.3389/fspas.2021.810426.

    Article  Google Scholar 

  3. Bahram M, Hildebrand F, Forslund SK, Anderson JL, Soudzilovskaia NA, Bodegom PM, et al. Structure and function of the global topsoil microbiome. Nature. 2018;560:233–7.

    Article  CAS  PubMed  Google Scholar 

  4. Knief C, Bol R, Amelung W, Kusch S, Frindte K, Eckmeier E, et al. Tracing elevational changes in microbial life and organic carbon sources in soils of the Atacama Desert. Glob Planet Change. 2020. https://0-doi-org.brum.beds.ac.uk/10.1016/j.gloplacha.2019.103078.

    Article  Google Scholar 

  5. Navarro-González R, Rainey FA, Molina P, Bagaley DR, Hollen BJ, de la Rosa J, et al. Mars-like soils in the Atacama Desert, Chile, and the dry limit of microbial life. Science. 2003;302:1018–21.

    Article  PubMed  Google Scholar 

  6. Schulze-Makuch D, Wagner D, Kounaves SP, Mangelsdorf K, Devine KG, de Vera J-P, et al. Transitory microbial habitat in the hyperarid Atacama Desert. Proc Natl Acad Sci USA. 2018;115:2670–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Vasquez-Dean J, Maza F, Morel I, Pulgar R, Gonzalez M. Microbial communities from arid environments on a global scale. A Syst Rev Biol Res. 2020;53:29.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Leung PM, Bay SK, Meier DV, Chiri E, Cowan DA, Gillor O, et al. Energetic basis of microbial growth and persistence in desert ecosystems. mSystems. 2020;5:e00495. https://0-doi-org.brum.beds.ac.uk/10.1128/msystems.00495-19.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Pan H, Gao H, Peng Z, Chen B, Chen S, Liu Y, et al. Aridity threshold induces abrupt change of soil abundant and rare bacterial biogeography in dryland ecosystems. mSystems. 2022;7:e0130921. https://0-doi-org.brum.beds.ac.uk/10.1128/msystems.01309-21.

    Article  PubMed  Google Scholar 

  10. Azúa-Bustos A, González-Silva C, Mancilla RA, Salas L, Gómez-Silva B, McKay CP, et al. Hypolithic cyanobacteria supported mainly by fog in the coastal range of the Atacama Desert. Microb Ecol. 2011;61:568–81.

    Article  PubMed  Google Scholar 

  11. DiRuggiero J, Wierzchos J, Robinson CK, Souterre T, Ravel J, Artieda O, et al. Microbial colonisation of chasmoendolithic habitats in the hyper-arid zone of the Atacama Desert. Biogeosciences. 2013;10:2439–50.

    Article  Google Scholar 

  12. Robinson CK, Wierzchos J, Black C, Crits-Christoph A, Ma B, Ravel J, et al. Microbial diversity and the presence of algae in halite endolithic communities are correlated to atmospheric moisture in the hyper-arid zone of the Atacama Desert. Environ Microbiol. 2015;17:299–315.

    Article  PubMed  Google Scholar 

  13. Bulgarelli D, Schlaeppi K, Spaepen S, van Ver Loren Themaat E, Schulze-Lefert P. Structure and functions of the bacterial microbiota of plants. Annu Rev Plant Biol. 2013;64:807–38.

    Article  CAS  PubMed  Google Scholar 

  14. Vorholt JA. Microbial life in the phyllosphere. Nat Rev Microbiol. 2012;10:828–40.

    Article  CAS  PubMed  Google Scholar 

  15. Al Ashhab A, Meshner S, Alexander-Shani R, Dimerets H, Brandwein M, Bar-Lavan Y, et al. Temporal and spatial changes in phyllosphere microbiome of acacia trees growing in arid environments. Front Microbiol. 2021. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2021.656269.

    Article  PubMed  PubMed Central  Google Scholar 

  16. Coleman-Derr D, Desgarennes D, Fonseca-Garcia C, Gross S, Clingenpeel S, Woyke T, et al. Plant compartment and biogeography affect microbiome composition in cultivated and native Agave species. New Phytol. 2016;209:798–811.

    Article  CAS  PubMed  Google Scholar 

  17. Finkel OM, Burch AY, Elad T, Huse SM, Lindow SE, Post AF, et al. Distance-decay relationships partially determine diversity patterns of phyllosphere bacteria on Tamrix trees across the Sonoran Desert. Appl Environ Microbiol. 2012;78:6187–93.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  18. Fonseca-García C, Coleman-Derr D, Garrido E, Visel A, Tringe SG, Partida-Martínez LP. The cacti microbiome: interplay between habitat-filtering and host-specificity. Front Microbiol. 2016;7:150. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2016.00150.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Qvit-Raz N, Finkel OM, Al-Deeb TM, Malkawi HI, Hindiyeh MY, Jurkevitch E, et al. Biogeographical diversity of leaf-associated microbial communities from salt-secreting Tamarix trees of the Dead Sea region. Res Microbiol. 2012;163:142–50.

    Article  PubMed  Google Scholar 

  20. Tahtamouni ME, Khresat SE, Lucero M, Sigala J, Unc A. Diversity of endophytes across the soil-plant continuum for Atriplex spp. in arid environments. J Arid Land. 2016;8:241–53.

    Article  Google Scholar 

  21. Araya JP, González M, Cardinale M, Schnell S, Stoll A. Microbiome dynamics associated with the Atacama flowering Desert. Front Microbiol. 2020. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2019.03160.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Eshel G, Araus V, Undurraga S, Soto DC, Moraga C, Montecinos A, et al. Plant ecological genomics at the limits of life in the Atacama Desert. Proc Natl Acad Sci USA. 2021. https://0-doi-org.brum.beds.ac.uk/10.1073/pnas.2101177118.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Lopez BR, Bashan Y, Bacilio M. Endophytic bacteria of Mammillaria fraileana, an endemic rock-colonizing cactus of the southern Sonoran Desert. Arch Microbiol. 2011;193:527–41.

    Article  CAS  PubMed  Google Scholar 

  24. Maldonado JE, Gaete A, Mandakovic D, Aguado-Norese C, Aguilar M, Gutiérrez RA, et al. Partners to survive: Hoffmannseggia doellii root-associated microbiome at the Atacama Desert. New Phytol. 2022;234:2126–39.

    Article  CAS  PubMed  Google Scholar 

  25. Dunai TJ, Melles M, Quandt D, Knief C, Amelung W. Whitepaper: Earth – evolution at the dry limit. Glob Planet Change. 2020. https://0-doi-org.brum.beds.ac.uk/10.1016/j.gloplacha.2020.103275.

    Article  Google Scholar 

  26. Ruhm J, Böhnert T, Weigend M, Merklinger FF, Stoll A, Quandt D, et al. Plant life at the dry limit - spatial patterns of floristic diversity and composition around the hyperarid core of the Atacama Desert. PLoS ONE. 2020. https://0-doi-org.brum.beds.ac.uk/10.1371/journal.pone.0233729.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Rundel PW, Dillon MO, Palma B, Mooney HA, Gulmon S, Ehleringer J. The phytogeography and ecology of the coastal Atacama and Peruvian deserts. Aliso. 1991;13:1–49.

    Article  Google Scholar 

  28. Pinto R, Barría I, Marquet PA. Geographical distribution of Tillandsia lomas in the Atacama Desert, northern Chile. J Arid Environ. 2006;65:543–52.

    Article  Google Scholar 

  29. Hesse R. Spatial distribution of and topographic controls on Tillandsia fog vegetation in coastal southern Peru: remote sensing and modelling. J Arid Environ. 2012;78:33–40.

    Article  Google Scholar 

  30. Wolf N, Siegmund A, del Río C, Osses P, García JL. Remote sensing-based detection and spatial pattern analysis for geo-ecological niche modeling of Tillandsia spp. in the Atacama, Chile. Int Arch PhotoGramm Remote Sens Spatial Inf Sci. 2016;41:251. https://0-doi-org.brum.beds.ac.uk/10.5194/isprs-archives-XLI-B2-251-2016.

  31. Merklinger FF, Zheng Y, Luebert F, Harpke D, Böhnert T, Stoll A, et al. Population genomics of Tillandsia landbeckii reveals unbalanced genetic diversity and founder effects in the Atacama Desert. Glob Planet Change. 2020. https://0-doi-org.brum.beds.ac.uk/10.1016/j.gloplacha.2019.103076.

    Article  Google Scholar 

  32. Houston J. Variability of precipitation in the Atacama Desert: its causes and hydrological impact. Int J Climatol. 2006;26:2181–98.

    Article  Google Scholar 

  33. Rundel PW, Dillon MO. Ecological patterns in the Bromeliaceae of the lomas formations of coastal Chile and Peru. Plant Syst Evol. 1998;212:261–78.

    Article  Google Scholar 

  34. Rundel PW, Palma B, Dillon MO, Sharifi MR, Nilsen ET, Boonpragob K. Tillandsia landbeckii in the coastal Atacama Desert of northern Chile. Rev Chil de Hist Nat. 1997;70:341–9.

    Google Scholar 

  35. Belmonte E, Arriaza B, Arismendi M, Sepúlveda G. Foliar anatomy of three native species of Tillandsia L. from the Atacama Desert, Chile. Plants. 2022;11:870. https://0-doi-org.brum.beds.ac.uk/10.3390/plants11070870.

  36. Raux PS, Gravelle S, Dumais J. Design of a unidirectional water valve in Tillandsia. Nat Commun. 2020. https://0-doi-org.brum.beds.ac.uk/10.1038/s41467-019-14236-5.

    Article  PubMed  PubMed Central  Google Scholar 

  37. González AL, Fariña JM, Pinto R, Pérez C, Weathers KC, Armesto JJ, et al. Bromeliad growth and stoichiometry: responses to atmospheric nutrient supply in fog-dependent ecosystems of the hyper-arid Atacama Desert. Chile Oecologia. 2011;167:835–45.

    Article  PubMed  Google Scholar 

  38. Jaeschke A, Böhm C, Merklinger FF, Bernasconi SM, Reyers M, Kusch S, et al. Variation in δ15N of fog-dependent Tillandsia ecosystems reflect water availability across climate gradients in the hyperarid Atacama Desert. Glob Planet Change. 2019. https://0-doi-org.brum.beds.ac.uk/10.1016/j.gloplacha.2019.103029.

    Article  Google Scholar 

  39. Magyarosy A, Hancock JG. Microbial population of the laimosphere of squash (Cucurbita maxima). Plant Soil. 1972;37:187–90.

    Article  Google Scholar 

  40. Alfaro FD, Manzano M, Almiray C, García J-L, Osses P, del Rio C, et al. Soil bacterial community structure of fog-dependent Tillandsia landbeckii dunes in the Atacama Desert. Plant Syst Evol. 2021. https://0-doi-org.brum.beds.ac.uk/10.1007/s00606-021-01781-0.

    Article  Google Scholar 

  41. Moat J, Orellana-Garcia A, Tovar C, Arakaki M, Arana C, Cano A, et al. Seeing through the clouds – mapping desert fog oasis ecosystems using 20 years of MODIS imagery over Peru and Chile. Int J Appl Earth Obs Geoinf. 2021. https://0-doi-org.brum.beds.ac.uk/10.1016/j.jag.2021.102468.

    Article  Google Scholar 

  42. Becker MF, Hellmann M, Knief C. Spatio-temporal variation in the root-associated microbiota of orchard-grown apple trees. Environ Microbiome. 2022. https://0-doi-org.brum.beds.ac.uk/10.1186/s40793-022-00427-z.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Ikenaga M, Sakai M. Application of locked nucleic acid (LNA) oligonucleotide–PCR clamping technique to selectively PCR amplify the SSU rRNA genes of bacteria in investigating the plant-associated community structures. Microbes Environ. 2014;29:286–95.

    Article  PubMed  PubMed Central  Google Scholar 

  44. Bolyen E, Rideout JR, Dillon MR, Bokulich NA, Abnet CC, Al-Ghalith GA, et al. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nat Biotechnol. 2019;37:852–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  45. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: high-resolution sample inference from Illumina amplicon data. Nat Methods. 2016;13:581–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2013;41:D590–6.

    Article  CAS  PubMed  Google Scholar 

  47. Yilmaz P, Parfrey LW, Yarza P, Gerken J, Pruesse E, Quast C, et al. The SILVA and “All-species Living Tree Project (LTP)” taxonomic frameworks. Nucleic Acids Res. 2013;42:D643–8.

    Article  PubMed  PubMed Central  Google Scholar 

  48. Pruesse E, Peplies J, Glöckner FO. SINA: accurate high-throughput multiple sequence alignment of ribosomal RNA genes. Bioinformatics. 2012;28:1823–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. McMurdie PJ, Holmes S. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE. 2013. https://0-doi-org.brum.beds.ac.uk/10.1371/journal.pone.0061217.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Davis NM, Proctor DM, Holmes SP, Relman DA, Callahan BJ. Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data. Microbiome. 2018. https://0-doi-org.brum.beds.ac.uk/10.1186/s40168-018-0605-2.

    Article  PubMed  PubMed Central  Google Scholar 

  51. Sahu KP, Kumar A, Sakthivel K, Reddy B, Kumar M, Patel A, et al. Deciphering core phyllomicrobiome assemblage on rice genotypes grown in contrasting agroclimatic zones: implications for phyllomicrobiome engineering against blast disease. Environ Microbiome. 2022. https://0-doi-org.brum.beds.ac.uk/10.1186/s40793-022-00421-5.

    Article  PubMed  PubMed Central  Google Scholar 

  52. Martino C, Morton JT, Marotz CA, Thompson LR, Tripathi A, Knight R, et al. A novel sparse compositional technique reveals microbial perturbations. mSystems. 2019;4:10. https://0-doi-org.brum.beds.ac.uk/10.1128/mSystems.00016-19.

    Article  Google Scholar 

  53. Oksanen J, Simpson G, Blanchet F, Kindt R, Legendre P, Minchin P, et al. Vegan: community ecology package - R package version 2.6–2. 2022. https://github.com/vegandevs/vegan.

  54. Martinez AP. pairwiseAdonis: pairwise multilevel comparison using Adonis - R package version 0.4. 2017. https://github.com/pmartinezarbizu/pairwiseAdonis.

  55. Lin H, Peddada SD. Analysis of compositions of microbiomes with bias correction. Nat Commun. 2020. https://0-doi-org.brum.beds.ac.uk/10.1038/s41467-020-17041-7.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Wickham H. ggplot2: elegant graphics for data analysis. 2nd ed. New York: Springer-Verlag; 2016.

    Book  Google Scholar 

  57. Russel J. MicEco: various functions for microbial community data - R package version 0.9.18. 2022. https://github.com/Russel88/MicEco.

  58. Foster ZSL, Sharpton TJ, Grünwald NJ. Metacoder: an R package for visualization and manipulation of community taxonomic diversity data. PLoS Comput Biol. 2017;13:e1005404. https://0-doi-org.brum.beds.ac.uk/10.1371/journal.pcbi.1005404.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  59. Delmotte N, Knief C, Chaffron S, Innerebner G, Roschitzki B, Schlapbach R, et al. Community proteogenomics reveals insights into the physiology of phyllosphere bacteria. Proc Natl Acad Sci USA. 2009;106:16428–33.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Chun S-J, Cui Y, Yoo S-H, Lee JR. Organic connection of holobiont components and the essential roles of core microbes in the holobiont formation of feral Brassica napus. Front Microbiol. 2022. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2022.920759.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Durand A, Maillard F, Alvarez-Lopez V, Guinchard S, Bertheau C, Valot B, et al. Bacterial diversity associated with poplar trees grown on a Hg-contaminated site: community characterization and isolation of Hg-resistant plant growth-promoting bacteria. Sci Total Environ. 2018;622–623:1165–77.

    Article  PubMed  Google Scholar 

  62. Janakiev T, Berić T, Stević T, Stanković S, Bačić J, Majstorović H, et al. The microbiome of the ‘Williams’ pear variety grown in the organic orchard and antifungal activity by the autochthonous bacterial and yeast isolates. Microorganisms. 2022;10:1282. https://0-doi-org.brum.beds.ac.uk/10.3390/microorganisms10071282.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  63. Mason CJ, Pfammatter JA, Holeski LM, Raffa KF. Foliar bacterial communities of trembling aspen in a common garden. Can J Microbiol. 2015;61:143–9.

    Article  CAS  PubMed  Google Scholar 

  64. Rastogi G, Sbodio A, Tech JJ, Suslow TV, Coaker GL, Leveau JHJ. Leaf microbiota in an agroecosystem: spatiotemporal variation in bacterial community composition on field-grown lettuce. ISME J. 2012;6:1812–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  65. Wu H, Zhang Z, Wang J, Qin X, Chen J, Wu L, et al. Bio-fertilizer amendment alleviates the replanting disease under consecutive monoculture regimes by reshaping leaf and root microbiome. Microb Ecol. 2022;84:452–64.

    Article  CAS  PubMed  Google Scholar 

  66. Yashiro E, McManus PS. Effect of streptomycin treatment on bacterial community structure in the apple phyllosphere. PLoS ONE. 2012;7:e37131. https://0-doi-org.brum.beds.ac.uk/10.1371/journal.pone.0037131.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  67. Bian G-K, Feng Z-Z, Qin S, Xing K, Wang Z, Cao C-L, et al. Kineococcus endophytica sp. nov., a novel endophytic actinomycete isolated from a coastal halophyte in Jiangsu China. Antonie Leeuwenhoek. 2012;102:621–8.

    Article  CAS  PubMed  Google Scholar 

  68. Duangmal K, Thamchaipenet A, Ara I, Matsumoto A, Takahashi Y. Kineococcus gynurae sp. nov., isolated from a Thai medicinal plant. Int J Syst Evol Microbiol. 2008;58:2439–42.

    Article  CAS  PubMed  Google Scholar 

  69. Lee SD. Kineococcus rhizosphaerae sp. nov., isolated from rhizosphere soil. Int J Syst Evol Microbiol. 2009;59:2204–7.

    Article  CAS  PubMed  Google Scholar 

  70. Qin S, Bian G-K, Zhang Y-J, Xing K, Cao C-L, Liu C-H, et al. Modestobacter roseus sp. nov., an endophytic actinomycete isolated from the coastal halophyte Salicornia europaea Linn., and emended description of the genus Modestobacter. Int J Syst Evol Microbiol. 2013;63:2197–202.

    Article  CAS  PubMed  Google Scholar 

  71. Busarakam K, Bull AT, Trujillo ME, Riesco R, Sangal V, van Wezel GP, et al. Modestobacter caceresii sp. nov., novel actinobacteria with an insight into their adaptive mechanisms for survival in extreme hyper-arid Atacama Desert soils. Syst Appl Microbiol. 2016;39:243–51.

    Article  CAS  PubMed  Google Scholar 

  72. Molina-Menor E, Gimeno-Valero H, Pascual J, Peretó J, Porcar M. Kineococcus vitellinus sp. nov., Kineococcus indalonis sp nov and Kineococcus siccus sp. nov., isolated nearby the Tabernas Desert (Almería, Spain). Microorganisms. 2020;8:1547. https://0-doi-org.brum.beds.ac.uk/10.3390/microorganisms8101547.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  73. Xu F-J, Li Q-Y, Li G-D, Chen X, Jiang Y, Jiang C-L. Kineococcus terrestris sp. nov. and Kineococcus aureolus sp. nov., isolated from saline sediment. Int J Syst Evol Microbiol. 2017;67:4801–7.

    Article  CAS  PubMed  Google Scholar 

  74. Makhalanyane TP, Valverde A, Gunnigle E, Frossard A, Ramond J-B, Cowan DA. Microbial ecology of hot desert edaphic systems. FEMS Microbiol Rev. 2015;39:203–21.

    Article  CAS  PubMed  Google Scholar 

  75. Mohammadipanah F, Wink J. Actinobacteria from arid and desert habitats: diversity and biological activity. Front Microbiol. 2016;6:1541. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2015.01541.

    Article  PubMed  PubMed Central  Google Scholar 

  76. Hansen TE, Enders LS. Host plant species influences the composition of milkweed and monarch microbiomes. Front Microbiol. 2022;13:840078. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2022.840078.

    Article  PubMed  PubMed Central  Google Scholar 

  77. Li S-J, Ahmed MZ, Lv N, Shi P-Q, Wang X-M, Huang J-L, et al. Plant-mediated horizontal transmission of Wolbachia between whiteflies. ISME J. 2017;11:1019–28.

    Article  CAS  PubMed  Google Scholar 

  78. Sabree ZL, Huang CY, Okusu A, Moran NA, Normark BB. The nutrient supplying capabilities of Uzinura, an endosymbiont of armoured scale insects. Environ Microbiol. 2013;15:1988–99.

    Article  CAS  PubMed  Google Scholar 

  79. Besaury L, Rémond C. Culturable and metagenomic approaches of wheat bran and wheat straw phyllosphere’s highlight new lignocellulolytic microorganisms. Lett Appl Microbiol. 2022;74:840–50.

    Article  CAS  PubMed  Google Scholar 

  80. Safni I, Subandiyah S, Fegan M. Ecology, epidemiology and disease management of Ralstonia syzygii in Indonesia. Front Microbiol. 2018. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2018.00419.

    Article  PubMed  PubMed Central  Google Scholar 

  81. Jackson CR, Denney WC. Annual and seasonal variation in the phyllosphere bacterial community associated with leaves of the southern magnolia (Magnolia grandiflora). Microb Ecol. 2011;61:113–22.

    Article  PubMed  Google Scholar 

  82. Kim M, Singh D, Lai-Hoe A, Go R, Abdul Rahim R, Ainuddin A, et al. Distinctive phyllosphere bacterial communities in tropical trees. Microb Ecol. 2012;63:674–81.

    Article  PubMed  Google Scholar 

  83. Lambais MR, Lucheta AR, Crowley DE. Bacterial community assemblages associated with the phyllosphere, dermosphere, and rhizosphere of tree species of the Atlantic Forest are host taxon dependent. Microb Ecol. 2014;68:567–74.

    Article  PubMed  Google Scholar 

  84. Hiraishi A, Imhoff JF. Acidiphilium. In: Whitman WB, Rainey F, Kämpfer P, Trujillo M, Chun J, DeVos P, Hedlund B, Dedysh S, editors. Bergey’s Manual of Systematics of Archaea and Bacteria. Hoboken, NJ: John Wiley & Sons; 2015. p. 1–14.

    Google Scholar 

  85. Knief C, Ramette A, Frances L, Alonso-Blanco C, Vorholt JA. Site and plant species are important determinants of the Methylobacterium community composition in the plant phyllosphere. ISME J. 2010;4:719–28.

    Article  CAS  PubMed  Google Scholar 

  86. Mina D, Pereira JA, Lino-Neto T, Baptista P. Epiphytic and endophytic bacteria on olive tree phyllosphere: exploring tissue and cultivar effect. Microb Ecol. 2020;80:145–57.

    Article  PubMed  Google Scholar 

  87. Schlechter RO, Miebach M, Remus-Emsermann MNP. Driving factors of epiphytic bacterial communities: a review. J Adv Res. 2019;19:57–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  88. Kearl J, McNary C, Lowman JS, Mei C, Aanderud ZT, Smith ST, et al. Salt-tolerant halophyte rhizosphere bacteria stimulate growth of alfalfa in salty soil. Front Microbiol. 2019;10:1849. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2019.01849.

    Article  PubMed  PubMed Central  Google Scholar 

  89. Manni A, Filali-Maltouf A. Diversity and bioprospecting for industrial hydrolytic enzymes of microbial communities isolated from deserted areas of South-East Morocco. AIMS Microbiol. 2022;8:5. https://0-doi-org.brum.beds.ac.uk/10.3934/microbiol.2022002.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  90. Osman JR, Wang Y, Jaubert C, Nguyen T-N, Fernandes GR, DuBow MS. The bacterial communities of surface soils from desert sites in the eastern Utah (USA) portion of the Colorado Plateau. Microbiol Res. 2021;244:126664. https://0-doi-org.brum.beds.ac.uk/10.1016/j.micres.2020.126664.

    Article  CAS  PubMed  Google Scholar 

  91. Palmisano MM, Nakamura LK, Duncan KE, Istock CA, Cohan FM. Bacillus sonorensis sp. nov., a close relative of Bacillus licheniformis, isolated from soil in the Sonoran Desert Arizona. Int J Syst Evol Microbiol. 2001;51:1671–9.

    Article  CAS  PubMed  Google Scholar 

  92. Xie F, Pathom-aree W. Actinobacteria from desert: diversity and biotechnological applications. Front Microbiol. 2021;12:765331. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2021.765531.

    Article  Google Scholar 

  93. Khan N, Bano A, Babar MDA. The root growth of wheat plants, the water conservation and fertility status of sandy soils influenced by plant growth promoting rhizobacteria. Symbiosis. 2017;72:195–205.

    Article  CAS  Google Scholar 

  94. Mukhtar S, Mehnaz S, Malik KA. Comparative study of the rhizosphere and root endosphere microbiomes of Cholistan Desert plants. Front Microbiol. 2021;12:618742. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2021.618742.

    Article  PubMed  PubMed Central  Google Scholar 

  95. Radwan SS, Al-Awadhi H, Sorkhoh NA, El-Nemr IM. Rhizospheric hydrocarbon-utilizing microorganisms as potential contributors to phytoremediation for the oil Kuwaiti Desert. Microbiol Res. 1998;153:247–51.

    Article  CAS  Google Scholar 

  96. Safdarian M, Askari H, Nematzadeh G, Sofo A. Halophile plant growth-promoting rhizobacteria induce salt tolerance traits in wheat seedlings (Triticum aestivum L.). Pedosphere. 2020;30:684–93.

    Article  CAS  Google Scholar 

  97. Shahid M, Akram MS, Khan MA, Zubair M, Shah SM, Ismail M, et al. A phytobeneficial strain Planomicrobium sp. MSSA-10 triggered oxidative stress responsive mechanisms and regulated the growth of pea plants under induced saline environment. J Appl Microbiol. 2018;124:1566–79.

    Article  CAS  PubMed  Google Scholar 

  98. Beary TP, Boopathy R, Templet P. Accelerated decomposition of sugarcane crop residue using a fungal–bacterial consortium. Int Biodet Biodegrad. 2002;50:41–6.

    Article  Google Scholar 

  99. Calleja-Cervantes ME, Menéndez S, Fernández-González AJ, Irigoyen I, Cibriáin-Sabalza JF, Toro N, et al. Changes in soil nutrient content and bacterial community after 12 years of organic amendment application to a vineyard. Eur J Soil Sci. 2015;66:802–12.

    Article  CAS  Google Scholar 

  100. Fan F, Yin C, Tang Y, Li Z, Song A, Wakelin SA, et al. Probing potential microbial coupling of carbon and nitrogen cycling during decomposition of maize residue by 13C-DNA-SIP. Soil Biol Biochem. 2014;70:12–21.

    Article  CAS  Google Scholar 

  101. Kramer S, Dibbern D, Moll J, Huenninghaus M, Koller R, Krueger D, et al. Resource partitioning between bacteria, fungi, and protists in the detritusphere of an agricultural soil. Front Microbiol. 2016;7:1524. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2016.01524.

    Article  PubMed  PubMed Central  Google Scholar 

  102. Maarastawi SA, Frindte K, Geer R, Kröber E, Knief C. Temporal dynamics and compartment specific rice straw degradation in bulk soil and the rhizosphere of maize. Soil Biol Biochem. 2018;127:200–12.

    Article  CAS  Google Scholar 

  103. Latorre C, González AL, Quade J, Fariña JM, Pinto R, Marquet PA. Establishment and formation of fog-dependent Tillandsia landbeckii dunes in the Atacama Desert: evidence from radiocarbon and stable isotopes. J Geophys Res Biogeosci. 2011. https://0-doi-org.brum.beds.ac.uk/10.1029/2010jg001521.

    Article  Google Scholar 

  104. Fuentes B, Choque A, Gomez F, Alarcon J, Castro-Nallar E, Arenas F, et al. Influence of physical-chemical soil parameters on microbiota composition and diversity in a deep hyperarid core of the Atacama Desert. Front Microbiol. 2022. https://0-doi-org.brum.beds.ac.uk/10.3389/fmicb.2021.794743.

    Article  PubMed  PubMed Central  Google Scholar 

  105. Scola V, Ramond JB, Frossard A, Zablocki O, Adriaenssens EM, Johnson RM, et al. Namib Desert soil microbial community diversity, assembly, and function along a natural xeric gradient. Microb Ecol. 2018;75:193–203.

    Article  CAS  PubMed  Google Scholar 

  106. Bodenhausen N, Bortfeld-Miller M, Ackermann M, Vorholt JA. A synthetic community approach reveals plant genotypes affecting the phyllosphere microbiota. PLoS Genet. 2014;10:e1004283. https://0-doi-org.brum.beds.ac.uk/10.1371/journal.pgen.1004283.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  107. Morella NM, Weng FC-H, Joubert PM, Metcalf CJE, Lindow S, Koskella B. Successive passaging of a plant-associated microbiome reveals robust habitat and host genotype-dependent selection. Proc Natl Acad Sci USA. 2020;117:1148–59.

    Article  CAS  PubMed  Google Scholar 

  108. Wagner MR, Lundberg DS, del Rio TG, Tringe SG, Dangl JL, Mitchell-Olds T. Host genotype and age shape the leaf and root microbiomes of a wild perennial plant. Nat Commun. 2016;7:12151. https://0-doi-org.brum.beds.ac.uk/10.1038/ncomms12151.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  109. Wallace JG, Kremling KA, Kovar LL, Buckler ES. Quantitative genetics of the maize leaf microbiome. Phytobiomes J. 2019;2:208–24.

    Article  Google Scholar 

  110. Durán P, Ellis TJ, Thiergart T, Ågren J, Hacquard S. Climate drives rhizosphere microbiome variation and divergent selection between geographically distant Arabidopsis populations. New Phytol. 2022;236:608–21.

    Article  PubMed  Google Scholar 

  111. Laforest-Lapointe I, Messier C, Kembel SW. Host species identity, site and time drive temperate tree phyllosphere bacterial community structure. Microbiome. 2016;4:1. https://0-doi-org.brum.beds.ac.uk/10.1186/s40168-016-0174-1.

    Article  Google Scholar 

  112. García J-L, Lobos-Roco F, Schween JH, del Río C, Osses P, Vives R, et al. Climate and coastal low-cloud dynamic in the hyperarid Atacama fog Desert and the geographic distribution of Tillandsia landbeckii (Bromeliaceae) dune ecosystems. Plant Syst Evol. 2021;307:1. https://0-doi-org.brum.beds.ac.uk/10.1007/s00606-021-01775-y.

    Article  Google Scholar 

  113. Koch MA, Stock C, Kleinpeter D, del Río C, Osses P, Merklinger FF, et al. Vegetation growth and landscape genetics of Tillandsia lomas at their dry limits in the Atacama Desert show fine-scale response to environmental parameters. Ecol Evol. 2020;10:13260–74.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

Katharina Werner and Tanja Schütte are gratefully acknowledged for support during the field sampling campaigns. Marius Röder and Merle Noschinski-Reetz are acknowledged for excellent technical assistance and Max Becker for his support in data analysis. Benedikt Ritter (University of Cologne) is acknowledged for preparing the map displaying the sample sites. Felix Merklinger (Sukkulenten-Sammlung Zürich, Switzerland) is gratefully acknowledged for critically reading the manuscript.

Funding

Open Access funding enabled and organized by Projekt DEAL. This work was funded by the German Research Foundation (DFG) under project number 268236062 as part of the CRC 1211.

Author information

Authors and Affiliations

  1. Molecular Biology of the Rhizosphere, Institute for Crop Science and Resource Conservation (INRES), University of Bonn, 53115, Bonn, Germany

    Anna Hakobyan, Stefanie Velte, Wiebke Sickel & Claudia Knief

  2. Institute of Biodiversity, Johann Heinrich Von Thünen Institute, Brunswick, Germany

    Wiebke Sickel

  3. Nees Institute for Biodiversity of Plants, University of Bonn, Bonn, Germany

    Dietmar Quandt

  4. Centro de Estudios Avanzados en Zonas Áridas Ceaza, La Serena, Chile

    Alexandra Stoll

  5. Instituto de Investigación Multidisciplinar en Ciencia y Tecnología, Universidad de La Serena, La Serena, Chile

    Alexandra Stoll

Authors
  1. Anna Hakobyan
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Stefanie Velte
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Wiebke Sickel
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Dietmar Quandt
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Alexandra Stoll
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Claudia Knief
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

C.K. designed the study. C.K., W.S. and A.S. performed the field sampling. Data analysis was done by C.K., A.H., and S.V., and the manuscript was drafted by C.K. and A.H. All authors read and approved the final version of the manuscript.

Corresponding author

Correspondence to Claudia Knief.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Supplementray figures: Fig. S1.

Principal component plot of surface soil samples showing the variation in bacterial community composition between samples from different study sites. DNA extraction negative control (EC) and PCR negative control (EC) are included. They both remained slightly positive after applying a decontamination algorithm and cluster with surface soil samples from some, though not all sampling sites. Fig. S2. Principal component plots showing the variation in bacterial community composition in phyllosphere and laimosphere samples from living and dead plant material, collected at Salar Grande and Cerro Pajonal. The tables below the figures summarize PERMANOVA results. Fig. S3. Heat trees showing the bacterial community structure up to genus level in the complete datasets of the phyllosphere and laimosphere. The color and size of nodes and edges are correlated with the sum of the mean abundance and number of ASVs, respectively. The most abundant genera (> 10% abundance in individual samples) are highlighted. The abundance value of 0.001 was set as a threshold to remove low-abundance taxa and reduce the complexity of the plots. Fig. S4. Bacterial diversity in the phyllosphere and laimosphere of T. landbeckii. Shown are Shannon’s index (upper row), evenness (middle row) and Faith’s PD (lower row). Significant differences in dependence on study site were evaluated based on Kruskal-Wallis tests. Please note the different y-axis scales for phyllosphere and laimosphere. Fig. S5. Bacterial diversity in the phyllosphere in dependence on the season for the samples from four different study sites. Shown are Shannon’s index, evenness, and Faith’s PD. Significant differences in dependence on the season were evaluated based on Kruskal-Wallis tests and are reported above the respective panel. Pairwise comparisons within each study site were done using Wilcoxon signed rank exact test for the paired samples and the Wilcoxon sum exact test for non-paired samples (e.g. Cerro Pajonal 2016 vs. 2017). Significance values: *p < 0.05, ***p < 0.001.

Additional file 2:

Supplementary tables: Table S1. Information about sample number and sequencing results. Table S2. The complete ASV table with raw read numbers and taxonomic ranking after decontamination. Table S3. The differentially enriched ASVs in phyllosphere and laimosphere assessed by ANCOM-BC analysis.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Hakobyan, A., Velte, S., Sickel, W. et al. Tillandsia landbeckii phyllosphere and laimosphere as refugia for bacterial life in a hyperarid desert environment. Microbiome 11, 246 (2023). https://0-doi-org.brum.beds.ac.uk/10.1186/s40168-023-01684-x

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://0-doi-org.brum.beds.ac.uk/10.1186/s40168-023-01684-x

Keywords

Microbiome

ISSN: 2049-2618

Contact us