Label-free functional analysis of root-associated microbes with dynamic quantitative oblique back-illumination microscopy

Abstract

The increasing global demand for food, coupled with concerns about the environmental impact of synthetic fertilizers, underscores the urgency of developing sustainable agricultural practices. Nitrogen-fixing bacteria, known as diazotrophs, offer a potential solution by converting atmospheric nitrogen into bioavailable forms, reducing the reliance on synthetic fertilizers. However, a deeper understanding of their interactions with plants and other microbes is needed. In this study, we introduce a recently developed label-free 3D quantitative phase imaging technology called dynamic quantitative oblique back-illumination microscopy (DqOBM) to assess the functional dynamic activity of diazotrophs in vitro and in situ. Our experiments involved three different diazotrophs (Sinorhizobium meliloti, Azotobacter vinelandii, and Rahnella aquatilis) cultured on media with amendments of carbon and nitrogen sources. Over 5 days, we observed increased dynamics in nutrient-amended media. These results suggest that the observed bacterial dynamics correlate with their metabolic activity. Furthermore, we applied qOBM to visualize microbial dynamics within the root cap and elongation zone of Arabidopsis thaliana primary roots. This allowed us to identify distinct areas of microbial infiltration in plant roots without the need for fluorescent markers. Our findings demonstrate that DqOBM can effectively characterize microbial dynamics and provide insights into plant-microbe interactions in situ, offering a valuable tool for advancing our understanding of sustainable agriculture.

Keywords: Label-free imaging; Nitrogen fixation; Quantitative phase imaging.

Figure 1

(A) Overview of the qOBM imaging approach which consists of an inverted brightfield microscope with epi-illumination. The sample (cultured bacteria streaked onto a glass slide) is sequentially illuminated by 4 optical fibers connected to 720 nm LEDs, followed by quantitative phase recovery. (B) Time-lapsed qOBM image stack of cultured Azotobacter vinelandii bacteria. (C) Blue: temporal phase fluctuations of a pixel corresponding to the boxed area of bacteria in (B). Green: log-log representation of the Fourier transform of the temporal phase value. This plot is used to generate the phasor plot (middle) and the colorized DqOBM image (right) demonstrating areas of high and low dynamics. (D) Representative phasor plots and DqOBM images of areas with low dynamics (left) and high dynamics (right).

Figure 2

Representative colorized DqOBM images of the nitrogen-fixing bacteria A. vinelandii. Columns represent culture treatments with varying carbon and nitrogen concentrations. Rows indicate the time expired relative to inoculation.

Figure 3

Representative colorized DqOBM images of the nitrogen-fixing bacteria Rahnella aquatilis. Columns represent culture treatments with varying carbon and nitrogen concentrations. Rows indicate the time expired relative to inoculation.

Figure 4

Representative colorized DqOBM images of the nitrogen-fixing bacteria Sinorhizobium meliloti. Columns represent culture treatments with varying carbon and nitrogen concentrations. Rows indicate the time expired relative to inoculation.

Figure 5

The signal energy over time of A. vinelandii, R. aquatilis and S. meliloti, respectively. (A) Shows a sharp decline to a dormant state of the glucose variants between D3–4 whereas the casamino variants show continued signal energy. (B) Shows some increased dynamics from the casamino acid groups. (C) Shows increased prolonged dynamics for the casamino acid S. meliloti variants. Standard deviation in the plot is demonstrated from the shaded area surrounding the lines and represents the standard deviation of the signal energy from three replicates. (D) Shows the nitrogen fixation values from the ARA study using A. vinelandii microbes. (E) Shows the derivative of the plot in (D) to show the nitrogen fixation rate. (F) Shows a comparison of the DqOBM energy with the nitrogen fixation rate.

Figure 6

Representative qOBM (left) and DqOBM (right) images from uninoculated A. thaliana plants (A–D), A. thaliana inoculated with A. vinelandii (E–H) and A. thaliana inoculated with R. aquatilis (I–N). We look at regions from the elongation zone (A,B, E,F, and I–K) and from the root cap (C,D, G,H, and L–N). In the uninoculated plant (A–D), we see minimal fluctuations in the phase of the images, corresponding to low levels of dynamics. Similarly, (E) and (F) show the elongation zone inoculated by A. vinelandii with minimal dynamics. (I) demonstrates increased dynamics (as indicated by the arrow) in the center of the root, which may be indicative of nutrient flow through the xylem rather than microbial dynamics. (J) and (K) show isolated dynamic areas from individual microbial cells in the elongation zone and higher dynamics and microbe inoculation in the root cap (indicated by arrows in J and K). Further, (K) shows high dynamics from a large number of bacteria growing outside of the root. In the root cap, we see colonization of intercellular spaces as indicated by the arrows in (H) and (M) and the red in (N). All samples were prepared from uncut roots. All scale bars are 100 $\mathrm{\mu }$m.

Figure 7

DAPI-labeled bacteria in plants. (A–H) Show A. vinelandii inoculated plants. (I–P) Show R. aquatilis inoculated plants. (A,E,I,M) Show qOBM phase images, (B,F,J,N) Show DqOBM images, (C,G,K,O) Show DAPI fluorescence images (green) overlaid on qOBM phase images (magenta), and (D,H,L,P) Show DqOBM images (green) overlaid on qOBM phase images (magenta). All scale bars are 100 $\mathrm{\mu }$m.