Phyllosphere microbial associations improve plant reproductive success

The above-ground (phyllosphere) plant microbiome is increasingly recognized as an important component of plant health. We hypothesized that phyllosphere bacterial recruitment may be disrupted in a greenhouse setting, and that adding a bacterial amendment would therefore benefit the health and growth of host plants. Using a newly developed synthetic phyllosphere bacterial microbiome for tomato (Solanum lycopersicum), we tested this hypothesis across multiple trials by manipulating microbial inoculation of leaves and measuring subsequent plant growth and reproductive success, comparing results from plants grown in both greenhouse and field settings. We confirmed that greenhouse-grown plants have a relatively depauperate phyllosphere bacterial microbiome, which both makes them an ideal system for testing the impact of phyllosphere communities on plant health and important targets for microbial amendments as we move towards increased agricultural sustainability. We find that the addition of the synthetic microbial community early in greenhouse growth leads to an increase in fruit production in this setting, implicating the phyllosphere microbiome as a key component of plant fitness and emphasizing the role that these bacterial microbiomes likely play in the ecology and evolution of plant communities.

Methods
For 16s Sequencing related data, paired-end reads were filtered and trimmed to 230(F) and 160(R) base pairs (bps), using DADA2 with default parameters (Callahan et al., 2016). Following denoising, merging reads and removing chimeras, DADA2 was used to infer amplicon sequence variants (ASVs), which are analogous to operational taxonomic units (OTUs), and taxonomy was assigned using the DADA2-trained SILVA database. Using DNA extraction and PCR negative controls from 16S sequencing, the decontam package was implemented using default settings to identify and remove potential contamination from the samples (Davis et al., 2018). The assigned ASVs, read count data, and sample metadata were combined in a phyloseq object (McMurdie & Holmes, 2013) for downstream analyses. The phyloseq package was used to calculate beta diversity (using Bray-Curtis distance), and a permutational analysis (PERMANOVA) was performed on data rarified to 400 reads (Weiss et al., 2017) (to account for extraordinarily low read counts in untreated greenhouse samples) using the adonisfunction in the vegan package (Oksanen et al., 2022).

Tomatoes were harvested and their number and weight were recorded multiple times per plant from onset of fruit production to plant termination in the greenhouse, at weeks 17, 18 and 19 in the second trial, and weeks 18 and 24 in the third trial. These metrics were measured only once after harvest from each individual plant grown in the field, at week 24. Tomatoes were weighed individually in trial 2, and as total harvested weight per plant in trial 3, as described. Data was recorded initially in laboratory notebooks, then transferred to `.csv`s before being processed in R.