Quantification of lateral heterogeneity in carbohydrate permeability of isolated plant leaf cuticles

Abstract

In phyllosphere microbiology, the distribution of resources available to bacterial colonizers of leaf surfaces is generally understood to be very heterogeneous. However, there is little quantitative understanding of the mechanisms that underlie this heterogeneity. Here, we tested the hypothesis that different parts of the cuticle vary in the degree to which they allow diffusion of the leaf sugar fructose to the surface. To this end, individual, isolated cuticles of poplar leaves were each analyzed for two properties: (1) the permeability for fructose, which involved measurement of diffused fructose by gas chromatography and flame ionization detection (GC-FID), and (2) the number and size of fructose-permeable sites on the cuticle, which was achieved using a green-fluorescent protein (GFP)-based bacterial bioreporter for fructose. Bulk flux measurements revealed an average permeance P of 3.39 × 10(-9) ms(-1), while the bioreporter showed that most of the leaching fructose was clustered to sites around the base of shed trichomes, which accounted for only 0.37% of the surface of the cuticles under study. Combined, the GC-FID and GFP measurements allowed us to calculate an apparent rate of fructose diffusion at these preferential leaching sites of 9.15 × 10(-7) ms(-1). To the best of our knowledge, this study represents the first successful attempt to quantify cuticle permeability at a resolution that is most relevant to bacterial colonizers of plant leaves. The estimates for P at different spatial scales will be useful for future models that aim to explain and predict temporal and spatial patterns of bacterial colonization of plant foliage based on lateral heterogeneity in sugar permeability of the leaf cuticle.

Keywords: Erwinia herbicola; Populus x canescens; aqueous pores; fructose; gas chromatography; phyllosphere; poplar.

Figure 1

Setup of the cuticle diffusion chamber. See [Measuring of fructose bulk flux by gas chromatography and flame ionization detection (GC–FID)] for details.

Figure 2

Diffusion of fructose across isolated Populus x canescens leaf cuticles. (A) Accumulation of fructose (in grams) as a function of time at a donor concentration of 0.18 g mL⁻¹. Shown is the linear regression (r² = 0.99) with 95% confidence intervals (broken lines) of the mean (n = 10) diffusion of fructose over Populus x canescens leaf cuticles. Error-bars represent the standard error of measurement. (B) Probability plot of the log-transformed fructose permeability (P, in ms⁻¹) of individual Populus x canescens CMs. The dataset passed the D’Agostino and Pearson omnibus test for normality (P = 0.28).

Figure 3

Green-fluorescent protein-based bioreporting of fructose availability on isolated Populus x canescens cuticles. (A–E) Epifluorescent images of Populus x canescens CMs inoculated with Eh299(pP_fruB-gfp[AAV]). White scale bars represent 20 μm. Images are pseudo-colored merges of DAPI (blue) and GFP (green) channels; when fluorescence occurred in both channels the visible color is yellow. The blue channel mainly shows autofluorescence of the CM and DAPI counterstained, non-fructose reporting bacteria. (A) Area with many bacterial bioreporter cells but none that fluoresce to indicate exposure to fructose; the right-hand panel is a magnification of the photograph shown on the left-hand site, to show individual DAPI-stained bacteria. (B–D) Typical arrangements of bioreporting bacteria on isolated Populus x canescens CMs. (F) Light microscopy image of a Populus x canescens CM. Arrows point to sites with shed-off trichomes.

Figure 4

Probability plot of the log-transformed size of fructose-permeable sites (in μm²) found on enzymatically isolated Populus x canescens CMs. Data points were derived and pooled from three individual CMs.