Light cues induce protective anticipation of environmental water loss in terrestrial bacteria

Abstract

The ecological significance of light perception in nonphotosynthetic bacteria remains largely elusive. In terrestrial environments, diurnal oscillations in light are often temporally coupled to other environmental changes, including increased temperature and evaporation. Here, we report that light functions as an anticipatory cue that triggers protective adaptations to tolerate a future rapid loss of environmental water. We demonstrate this photo-anticipatory stress tolerance in leaf-associated Pseudomonas syringae pv. syringae (Pss) and other plant- and soil-associated pseudomonads. We found that light influences the expression of 30% of the Pss genome, indicating that light is a global regulatory signal, and this signaling occurs almost entirely via a bacteriophytochrome photoreceptor that senses red, far-red, and blue wavelengths. Bacteriophytochrome-mediated light control disproportionally up-regulates water-stress adaptation functions and confers enhanced fitness when cells encounter light prior to water limitation. Given the rapid speed at which water can evaporate from leaf surfaces, such anticipatory activation of a protective response enhances fitness beyond that of a reactive stress response alone, with recurring diurnal wet-dry cycles likely further amplifying the fitness advantage over time. These findings demonstrate that nonphotosynthetic bacteria can use light as a cue to mount an adaptive anticipatory response against a physiologically unrelated but ecologically coupled stress.

Keywords: adaptive prediction; anticipatory regulation; light sensing; photoreceptor; stress response.

Fig. 1.

Multiple light wavelengths stimulate global transcriptional reprogramming in Pss B728a. (A) Principal component analysis (PCA) of normalized read counts for 5,181 genes in wild-type (WT) Pss B728a following maintenance in the dark or a 15-min exposure to blue light (450 nm, 20 μmol m⁻² s⁻¹), red light (660 nm, 30 μmol m⁻² s⁻¹), far-red light (730 nm, 188 μmol m⁻² s⁻¹), or white light (34 μmol m⁻² s⁻¹) (n = 3 for each condition). (B) Pie chart and associated UpSet plot of genes identified as differentially expressed (DE) (q value ≤ 0.00105) in the WT strain upon exposure to at least one light condition as compared to in cells maintained in the dark. The horizontal bars on the lower left represent the total numbers of DE genes (DEGs) in individual light conditions. The vertical bars in the UpSet plot represent overlapping DEGs across light conditions. (C–F) Volcano plots showing the mean gene expression changes (log₂ fold change, n = 3) and associated significance (−log₁₀ q value) in the WT strain exposed to the indicated light compared to maintenance in the dark. Each dot represents a single gene; ≤ 4 non-DE outliers were excluded from each plot for visualization purposes. The q value threshold is marked by a gray horizontal line, with DEGs shown in red. The numbers of up-regulated and down-regulated DEGs in each light condition are indicated. Selected DEGs are uniquely colored and labeled to aid visual comparisons across plots.

Fig. 2.

Light-triggered transcriptional reprogramming is almost entirely orchestrated via a single photoreceptor, the bacteriophytochrome BphP1. (A) Principal component analysis (PCA) of gene expression patterns in wild-type (WT) Pss B728a (filled squares), ΔbphOP1 mutant (open circles), and Δlov mutant (open triangles) following maintenance in the dark or a 15-min exposure to the indicated light condition (n = 3 for each strain x treatment combination). PCA was performed using normalized read counts for the 1,686 light-responsive differentially expressed genes (DEGs) identified in the WT strain. (B and C) Plots indicating the number of WT DEGs identified under a given wavelength that were categorized as being light-responsive in a manner dependent (solid bars) or independent (striped bars) of (B) BphP1 or (C) Lov. DEGs were categorized as dependent on a given photoreceptor if differential expression was lost in the deletion strain and restored upon complementation. DEGs were categorized as independent of a given photoreceptor if they were differentially expressed in both the WT and the deletion mutant. DEGs were categorized as having unclear dependence on a given photoreceptor (gray checkered bars) if differential expression was lost in the deletion mutant but not restored upon complementation. (D and E) Volcano plots showing mean gene expression changes (log₂ fold change, n = 3) and associated significance (−log₁₀ q value) of the 1,307 blue-light-responsive DEGs identified in the WT in the (B) ΔbphOP1 mutant or (E) Δlov mutant under blue light compared to maintenance in the dark. Each dot represents a single gene. The q value threshold is marked by a gray horizontal line, with DEGs shown in red. The numbers of up-regulated and down-regulated DEGs in each strain are indicated. Select DEGs are uniquely colored and labeled to aid visual comparisons across plots. (F) Absorption spectra of purified BphP1 protein showing photoconversion in response to blue, red, or far-red light. The black line shows dark-assembled BphP1. The colored lines show BphP1 following a 60-min exposure to the indicated light wavelengths.

Fig. 3.

Light cues sensed via BphP1 induce expression of genes broadly involved in water-stress tolerance in Pss B728a, including the sigma factor AlgU that regulates water-stress adaptations. (A and B) Bubble plots showing the top five overrepresented functional categories (q value < 0.05) among the differentially expressed genes (DEGs) that were (A) up-regulated or (B) down-regulated in a BphP1-dependent manner under blue, red, or far-red light. y-axis labels denote functional categories and light conditions. Rich factor (x axis) denotes the fraction of the total genes annotated to a given category that were differentially expressed under a given light condition. Bubble size and color indicate the number of DEGs in the given category and the significance of the enrichment, respectively. Gray bubbles indicate nonsignificance. (C and D) Heatmaps of the mean fold change expression (n = 3) upon exposure to blue, red, or far-red light compared to maintenance in the dark for all genes annotated to the categories of (C) osmotic stress response or (D) polysaccharide synthesis and regulation. Genes are identified on the y axis, with brackets indicating specific functions for select gene sets. Starred genes in C indicate those tested for involvement in light-enhanced osmotolerance in Fig. 4. (E) UpSet plot showing the numbers and intersections of the photoinduced, Bph1-dependent DEGs identified in this work and the osmo-induced, AlgU-regulated DEGs identified previously (31, 32).

Fig. 4.

Light cues stimulate enhanced osmotolerance in Pss and other plant-associated pseudomonads. (A) Representative normalized growth curves of Pss B728a wild-type (WT) upon osmotic challenge with 1 M NaCl following maintenance in the dark or exposure to 188 μmol m⁻² s⁻¹ far-red light for 30 min, 1 h, 2 h, or 4 h. (B–H) Cumulative growth [area under the curve (AUC)] of individual replicates during osmotic challenge with (B–G) 1 M NaCl or (H) 0.75 M NaCl following maintenance in the dark or exposure to far-red light. Each point represents a single replicate; gray lines connect matched samples that originated from the same independent culture. Replicates were pooled from either two independent experiments (n = 10 or 12 after pooling) or three independent experiments (n = 9 or 18 after pooling). Light-treated samples were exposed to 188 μmol m⁻² s⁻¹ far-red light for 30 min except where specified otherwise. (B) AUCs for B728a WT cells exposed to far-red light for various durations (n = 12). Letters indicate statistical differences among treatments [P < 0.05, one-way repeated-measure (RM) ANOVA with Greenhouse–Geisser and Tukey corrections, where the RM is based on matched samples of a single original culture used for multiple treatments]. (C) AUCs for B728a WT cells exposed to three far-red light intensities (n = 18). (D) AUCs for B728a WT, ΔbphOP1 mutant, and ΔbphOP1 (pN-bphOP1) complemented strain (n = 10). (E and F) AUCs for B728a WT and isogenic strains deleted for (E) light response-related genes (n = 10 for WT, n = 12 per deletion strain) or (F) water-stress response-related genes (n = 12). (G and H) AUCs for the following plant-associated pseudomonads: Pss B728a; Pss HS191, Pss B301D-R, P. putida KT2440, P. cichorii 302699, and P. protegens CHA0 (n = 9). Light-treated samples were exposed to far-red light for 2 h prior to osmotic challenge. Statistical significance was determined by (C–F) two-way RM ANOVA with Šídák correction or (G and H) two-tailed paired-sample Student’s t test (light versus dark for individual strains) with Holm–Šídák correction. Bold gray letters in D and F indicate statistical differences among strains for (D) dark-maintained or (F) light-treated samples (P < 0.05, two-way RM ANOVA with Šídák correction).

Fig. 5.

Light serves as an anticipatory cue for environmental water loss in terrestrial microbes. (A and B) Environmental and surface water evaporation data collected from 6 am to 1 pm on September 15, 2020, in Ames, IA, USA. The periods of dew accumulation and surface drying are denoted in the bar above the graphs. (A) Environmental measurements of relative humidity (purple), air temperature (teal), and dew point (orange). Lines represent means with error bars representing SE (n = 3). (B) Surface water content on a hydrophilic surface (filter paper, green) or hydrophobic surface (weigh boat, gray) and intensities of natural sunlight for the wavelength ranges of blue light, red light, and far-red light. The arrow indicates the addition of an aperture to the light sensor to keep within the dynamic range of the sensor as solar radiation levels increased. For surface water content, lines represent means with error bars representing SE (n = 4). Water content values are reported as the percentage of the mean maximum amount of water measured on the given surface. (C) Model of how bacteria use light cues at dawn to anticipate evaporative water loss and prepare for ensuing water stress. Blue spheres represent osmoprotectants. Red cubes represent solutes. The lightning bolt represents light that promotes conversion of a photoreceptor (dark blue) from an inactive state to an activated state, enhancing expression of water stress-related functions including osmoprotectant transporters (green). The illustration was created with BioRender.com.