Rapid metabolism of glucose detected with FRET glucose nanosensors in epidermal cells and intact roots of Arabidopsis RNA-silencing mutants

Abstract

Genetically encoded glucose nanosensors have been used to measure steady state glucose levels in mammalian cytosol, nuclei, and endoplasmic reticulum. Unfortunately, the same nanosensors in Arabidopsis thaliana transformants manifested transgene silencing and undetectable fluorescence resonance energy transfer changes. Expressing nanosensors in sgs3 and rdr6 transgene silencing mutants eliminated silencing and resulted in high fluorescence levels. To measure glucose changes over a wide range (nanomolar to millimolar), nanosensors with higher signal-to-noise ratios were expressed in these mutants. Perfusion of leaf epidermis with glucose led to concentration-dependent ratio changes for nanosensors with in vitro K(d) values of 600 microM (FLIPglu-600 microDelta13) and 3.2 mM (FLIPglu-3.2 mDelta13), but one with 170 nM K(d) (FLIPglu-170 nDelta13) showed no response. In intact roots, FLIPglu-3.2 mDelta13 gave no response, whereas FLIPglu-600 microDelta13, FLIPglu-2 microDelta13, and FLIPglu-170 nDelta13 all responded to glucose. These results demonstrate that cytosolic steady state glucose levels depend on external supply in both leaves and roots, but under the conditions tested they are lower in root versus epidermal and guard cells. Without photosynthesis and external supply, cytosolic glucose can decrease to <90 nM in root cells. Thus, observed gradients are steeper than expected, and steady state levels do not appear subject to tight homeostatic control. Nanosensor-expressing plants can be used to assess glucose flux differences between cells, invertase-mediated sucrose hydrolysis in vivo, delivery of assimilates to roots, and glucose flux in mutants affected in sugar transport, metabolism, and signaling.

Figure 1.

Expression of Nanosensors in Arabidopsis Wild Type and Silencing Mutants. (A) Number of mature, soil-grown transformants showing significant eYFP fluorescence as determined using an epifluorescence stereomicroscope. (B) Representative fluorescence images of leaves from the different transformants. (C) and (D) Fluorescence (C) and bright-field (D) images of T1 seedlings of highly expressing transformants at the seedling stage.

Figure 2.

Construct Maps and in Vitro Saturation Curves for FLIPglu-Δ13. (A) FLIPglu-Δ13 cassette containing linearly fused eCFP-mglB-eYFP genes. The size of each gene, restriction sites, and transcription start and stop are indicated. (B) pPZP 312 binary vector T-DNA containing a FRET glucose nanosensor. L, left border; MglB, E. coli periplasmic glucose binding protein; P_MAS, MAS promoter; P_35S, CaMV 35S promoter; R, right border; T_Rbcs, Rbcs terminator; T_35S, CAMV 35S terminator. Arrows indicate the direction of transcription. The restriction enzymes used for cloning are indicated. (C) Glucose binding isotherms of FLIPglu-170nΔ13, FLIPglu-2μΔ13, FLIPglu-600μΔ13, and the new low-affinity nanosensor FLIPglu-3.2mΔ13. Fractional saturation of the four nanosensors versus glucose concentrations is given for proteins purified from Escherichia coli. Binding was measured by dividing fluorescence intensity at the eYFP emission peak (528 nm) by that of the eCFP emission peak (485 nm) and fitting to a single-site binding isotherm, (r − r_apo)/(r_sat − r_apo).

Figure 3.

Confocal Images of Cytosolic Expression of FLIPglu-600μΔ13. Cytosolic and nuclear localization of FLIPglu-600μΔ13 in the leaf epidermis were determined by spinning disc confocal microscopy. (A) Optical section through a pavement cell. Note cytoplasmic strands. (B) Same cell as in (A), with optical section at bottom of cell showing the nucleus. Note the fluorescent ring around the nucleus. (C) A different leaf area showing cytosolic fluorescence in guard cells. Quantitation shows that nuclei have ∼70% of fluorescence intensity found in the cytosol.

Figure 4.

Glucose-Induced FRET Changes in the Cytosol of Leaf Epidermal Cells. The FRET sensor FLIPglu-600μΔ13 with an affinity of 600 μM for glucose in stably transformed rdr6-11 Arabidopsis plants responds in the epidermis to perfusion with 50 mM glucose. Quantitative data were derived by pixel-by-pixel integration of the ratiometric images. The scale at right gives fluorescence intensity in arbitrary units (A.U.) for the individual eCFP (480/30) and eYFP (535/40) emission channels; the scale at left gives the ratio of eYFP intensity divided by eCFP intensity. For each phase (plus/minus glucose), one ratiometric image with a pavement and a guard cell is shown.

Figure 5.

Glucose-Induced FRET Changes in Leaf Epidermal Cells. Images were acquired and data were analyzed as described for Figure 4. (A) Response of an epidermal region of a leaf from a transformant expressing FLIPglu-170nΔ13 to perfusion with 5 and 10 mM glucose (no response was seen to concentrations up to 20 mM; data not shown). (B) Responses in a plant expressing FLIPglu-2μΔ13. (C) Responses in a plant expressing FLIPglu-600μΔ13. (D) Responses in a plant expressing FLIPglu-3.2mΔ13.

Figure 6.

Glucose-Induced FRET Changes in Intact Roots. Images were acquired and data were analyzed as described for Figure 4. (A) Response of an intact root tip of an intact seedling from a transformant expressing FLIPglu-170nΔ13 to perfusion with 0.1 to 1 mM glucose (higher glucose levels did not lead to higher ratio changes; data not shown). Root tips were analyzed. (B) Responses in a plant expressing FLIPglu-2μΔ13. For each phase (time points indicated by lines), a ratiometric image of a lateral root tip is shown. A lookup table is provided to the right of the images. (C) Responses in a plant expressing FLIPglu-600μΔ13. (D) Responses in a plant expressing FLIPglu-3.2mΔ13.

Figure 7.

Isotherms for Glucose Binding Proteins Fused to the Codon-Modified eCFP and Venus Encoding Ares and Aphrodite Genes. (A) Glucose binding isotherms for FLIPglu-600μΔ11 (eCFP/eYFP), FLIPglu-600μΔ11 (Ares/eYFP), FLIPglu-600μΔ11 (eCFP/Aphrodite), and FLIPglu-600μΔ11 (Ares/Aphrodite). (B) FLIPglu-600μΔ13 (eCFP/eYFP) and FLIPglu-600μΔ13 (eCFP/Aphrodite). Fractional saturation of the four nanosensors versus glucose concentrations is given for proteins purified from E. coli. Binding was measured as described for Figure 2C. (C) Comparison of r_apo − r_sat in the FLIPglu-600μ variants carrying combinations of the different fluorophores.