The clock of a maize plant does not tick alone. A new study published in International Microbiology shows that the daily rhythms of gene expression in maize leaves are accompanied by measurable shifts in the composition of the bacterial communities living on those leaves, hinting that the plant’s internal timekeeping may help shape its own microbiome. The work, led by Renato Augusto Corrêa dos Santos of the University of São Paulo together with collaborators at the University of Georgia and other institutions, is the first study to pair leaf transcriptome and microbiome data from maize across contrasting diurnal periods.
Maize is the most-produced grain on Earth, and its leaves host a rich bacterial community known as the phyllosphere microbiome. This habitat is notoriously harsh: leaves experience intense solar radiation, large temperature swings, fluctuating humidity, and rapid changes in surface moisture, nutrient availability, and hormone and volatile compound release. Previous work in roots — including studies in rice rhizosphere and in Arabidopsis and Brachypodium — has demonstrated that microbial communities shift between day and night and that disrupting the plant’s circadian clock alters these shifts. But no one had yet examined whether the same relationship holds in the phyllosphere of a major crop.
To tackle this, the team capitalized on an unusually rich data source: a large field experiment conducted in August 2014 in which researchers sampled the upper leaves of roughly 250 maize genotypes approximately two weeks after flowering, at two contrasting time points — midday (11 AM to 1 PM, around 30 °C) and midnight (11 PM to 1 AM, around 24 °C). Two previously published studies from the same experiment had independently generated RNA-sequencing data from the leaf tissue and 16S rRNA amplicon sequencing data from the same RNA samples, capturing the metabolically active bacteria. By matching samples from both datasets, the researchers assembled a paired resource of 176 daytime and 228 nighttime samples spanning 173 and 222 maize genotypes respectively — a uniquely powerful dataset for detecting host–microbe associations across the diurnal cycle.
The first step was to verify that the transcriptomic data genuinely reflected the time of sampling. Principal component analysis of gene expression showed a clear separation between midday and midnight samples, and canonical circadian clock genes behaved as expected, with those known to peak near midday — such as PRR73-LIKE — showing high expression in day samples and those peaking at night, such as RVE7-LIKE genes, showing the opposite. One exception was ELF3-LIKE 1, a component of the plant Evening Complex, which did not show the anticipated night-time peak; however, its close homolog ELF3-LIKE 2 did display the expected pattern, suggesting that this homolog may be functionally compensating in maize.
The researchers then constructed a gene co-expression network using corALS, computing Pearson correlation coefficients between all pairs of the 7,201 genes that passed stringent expression filters. Clustering with the Leiden algorithm produced 33 modules, 18 of which showed significantly different expression between day and night. Functional enrichment analysis with GOATOOLS revealed three particularly notable modules: one of 205 genes enriched for nucleic acid binding and populated with transcription factor families such as AP2-EREBP, MYB-related, and WRKY, expressed more highly at midnight; a 67-gene module likely associated with heat stress response, more active at midday; and a 45-gene module enriched for photosynthesis and energy generation, also peaking during the day. The midday-midnight temperature difference of roughly 6 °C plausibly explains the heat response signature, while the photosynthesis module reflects the obvious dominance of light-driven metabolism during daylight hours.
Turning to the bacteriome, the team re-classified the original operational taxonomic units (OTUs) using the Genome Taxonomy Database (GTDB) via a QIIME 2 classifier, then filtered to 276 OTUs representing nearly 75 percent of total reads. These OTUs were dominated by members of the Sphingomonadales, Rhizobiales, and Actinomycetales. Alpha diversity metrics — Shannon, evenness, and Faith’s phylogenetic diversity — showed no significant differences between midday and midnight. But beta diversity told a different story: Weighted UniFrac ordination revealed a significant, if modest, shift in community composition between time points, with PERMANOVA indicating that sampling time explained roughly 2 percent of the variation (Pseudo-F = 8.56, P = 0.001).
Differential abundance analysis with ANCOMBC identified 87 OTUs whose abundance differed between the two periods — 36 more abundant at midday and 51 at midnight. Strikingly, 82 percent of the night-biased OTUs belonged to Sphingomonas or unclassified Sphingomonadaceae, whereas daytime-enriched OTUs were more taxonomically diverse, including members of Enterobacteriaceae, Burkholderiaceae, Methylobacterium, Pedobacter, Chryseobacterium, Agrobacterium, and Brevundimonas. This asymmetric pattern suggests that night conditions — cooler temperatures, higher humidity, and the absence of radiation — may favor a particular subset of leaf bacteria.
To probe the ecological roles of individual taxa, the researchers built co-occurrence networks separately for day and night samples using Spiec-Easi, a method designed for the compositional and sparse nature of microbiome data. Both networks were highly modular, and most nodes were classified as peripheral specialists. Using within-module and among-module connectivity metrics borrowed from the functional cartography framework of Guimerà and Nunes Amaral, the team identified potential keystone taxa: module hubs, connectors, and network hubs. Notably, a Deinococcus species emerged as a shared module hub in both networks, consistent with prior reports of Deinococcus as a keystone taxon in other microbiomes and with its well-known resilience to radiation and oxidative stress — traits well suited to the sun-drenched leaf surface. A Pseudomonas OTU served as a module hub in the daytime network, while most connectors in both networks were Sphingomonas and Methylobacterium species, the two genera that dominate most healthy plant phyllospheres. Interestingly, taxa with higher network connectivity tended to be less abundant overall, suggesting that rare bacteria may exert disproportionate influence on community structure.
The most striking results came from cross-kingdom analyses. Using SparXCC, a method that computes sparse cross-correlations between compositional microbiome data and gene expression, the researchers linked host genes to bacterial taxa across 174 field plots sampled at both time points. They detected 650 gene–OTU correlations in daytime samples (involving 290 genes and 76 OTUs) and, notably, 1,700 correlations in nighttime samples (732 genes and 83 OTUs) — nearly three times as many host–microbe associations at night. Two canonical circadian clock genes showed significant cross-correlations with microbial taxa: GIGANTEA 2 and the Pseudo-Response Regulator gene PRR73, the latter being part of the molecular machinery that represses CCA1/LHY expression during the day. Members of transcription factor families also appeared among the correlated genes in both periods, and a Gene Ontology term for “photosynthesis, light harvesting” was significantly over-represented among the correlated genes in both day and night networks.
The authors emphasize that their findings represent associations rather than causal demonstrations. Because samples were collected under natural field conditions at only two time points, the observed patterns likely reflect the combined effects of endogenous circadian processes and environmental variation — light intensity, temperature, and humidity all shift simultaneously with the diurnal cycle. Disentangling these effects will require follow-up experiments under controlled conditions, denser temporal sampling, host circadian clock mutants, and ideally synthetic microbial communities combined with metabolomics and meta-omics approaches. Still, the study provides the first integrated picture of host transcriptome–phyllosphere bacteriome associations across contrasting diurnal periods in any crop, and it opens a clear path toward that mechanistic work.
The implications extend beyond basic biology. Interest in engineering the phyllosphere microbiome for sustainable crop production has grown rapidly, and genera such as Sphingomonas and Methylobacterium — both of which appeared as key connectors in these networks — are known to include plant growth-promoting and nitrogen-fixing species. If maize diurnal transcriptional dynamics, including the activity of circadian clock genes such as GIGANTEA and PRR73, genuinely contribute to shaping which bacteria thrive on leaves, then breeding or managing crops with the plant’s internal clock in mind could become a tool for steering beneficial microbial communities. Given that circadian clock activity is already linked to hybrid vigor, photosynthesis, carbon fixation, and stress responses in maize and other grasses, the idea that this same clock might extend its influence to the plant’s bacterial partners adds a compelling new dimension to crop improvement — one where time of day, plant genes, and microbial ecology converge on the surface of a leaf.
Subject of Research: Diurnal associations between maize leaf gene expression and phyllosphere bacterial microbiome composition across midday and midnight sampling periods in diverse maize genotypes.
Subject of Research: Biology
Article Title: Diurnal dynamics of maize gene expression is associated with phyllosphere microbiome composition
Article References: dos Santos, R. A. C., Hidalgo-Martinez, K., Muñoz-Perez, J. M., Laspisa, D. J., Li, C., Mendes, L. W., Riaño-Pachón, D. M., & Wallace, J. G. (2026). Diurnal dynamics of maize gene expression is associated with phyllosphere microbiome composition. International Microbiology. https://doi.org/10.1007/s10123-026-00875-4
Image Credits: AI Generated
DOI: 10.1007/s10123-026-00875-4
Keywords: maize, phyllosphere, microbiome, circadian clock, diurnal gene expression, transcriptomics, 16S rRNA, co-expression networks, keystone taxa, Sphingomonas, Deinococcus, host-microbe interactions
Cite Scienmag News
APA MLA Chicago
Juliet Wilcox. (September 11, 2026). Maize gene expression rhythms shape leaf microbiome composition throughout the day. Scienmag. https://scienmag.com/maize-gene-expression-rhythms-shape-leaf-microbiome-composition-throughout-the-day/
Juliet Wilcox. “Maize gene expression rhythms shape leaf microbiome composition throughout the day.” Scienmag, 11 September 2026, https://scienmag.com/maize-gene-expression-rhythms-shape-leaf-microbiome-composition-throughout-the-day/. Accessed 11 September 2026.
Juliet Wilcox. “Maize gene expression rhythms shape leaf microbiome composition throughout the day.” Scienmag. September 11, 2026. https://scienmag.com/maize-gene-expression-rhythms-shape-leaf-microbiome-composition-throughout-the-day/
Copy citation Download RIS
Tags: daily microbial dynamics in maize leavesdieldiurnal shifts in maize microbiomediurnal shifts in phyllosphere bacterial communitiesdiurnal variation in plant-associated bacteriaeffects of circadian regulation on phyllosphere microbiotaenvironmental stress effects on leaf microbiomeenvironmental stressors and microbiome fluctuations in maize leavesgene expression influence on phyllosphere bacteriaimpact of plant internal clock on leaf microbial compositionimpact of plant internal timing on microbiome compositioninfluence of gene expression on microbial diversitymaize circadian biology and microbiome shapingMaize leaf gene expression rhythmsMaize leaf microbiome diurnal variationplant circadian clock and microbiome interactionsplant circadian rhythms and microbial community dynamicsplant-microbe interactions in the phyllosphereplant-microbe interactions influenced by gene expression rhythmsrole of plant circadian rhythms in microbiome shapingtranscriptome-microbiome interactions in maize leavestranscriptome-microbiome pairing in crop plants
