Spatiotemporal distribution of reactive oxygen species production, delivery, and use in Arabidopsis root hairs

Abstract

Fluorescent selective probes for reactive oxygen species (ROS) detection in living cells are versatile tools for the documentation of ROS production in plant developmental or stress reactions. We employed high-resolution live-cell imaging and semiquantitative analysis of Arabidopsis (Arabidopsis thaliana) stained with CM-H₂DCFDA, CellROX Deep Red, and Amplex Red for functional characterization of the spatiotemporal mode of ROS production, delivery, and utilization during root hair formation. Cell viability marker fluorescein diacetate served as a positive control for dye loading and undisturbed root hair tip growth after staining. Using a colocalization analysis with subcellular molecular markers and two root hair mutants with similar phenotypes of nonelongating root hairs, but with contrasting reasons for this impairment, we found that: (i) CM-H₂DCFDA is a sensitive probe for ROS generation in the cytoplasm, (ii) CellROX Deep Red labels ROS in mitochondria, (iii) Amplex Red labels apoplastic ROS and mitochondria and shows high selectivity to root hairs, (iv) the root hair defective 2-1 (rhd2-1) mutant with nonfunctional NADPH oxidase RESPIRATORY BURST OXIDASE HOMOLOG PROTEIN C/ROOT HAIR-DEFECTIVE 2 (AtRBOHC/RHD2) has a low level of CM-H₂DCFDA-reactive ROS in cytoplasm and lacks Amplex Red-reactive ROS in apoplast, and (v) the ACTIN2-deficient deformed root hairs1-3 (der1-3) mutant is not altered in these aspects. The sensitivity of CellROX Deep Red was documented by discrimination between larger ROS-containing mitochondria and small, yet ROS-free premature mitochondria in the growing tip of root hairs. We characterized spatial changes in ROS production and compartmentalization induced by external ROS modulators, ethylene precursor 1-aminocyclopropane-1-carboxylic acid, and ionophore valinomycin. This dynamic and high-resolution study of ROS production and utilization opens opportunities for precise speciation of particular ROS involved in root hair formation.

Selective fluorescent probes enable high-resolution live-cell imaging of reactive oxygen species production and subcellular localization in bulges and growing root hairs of Arabidopsis.

Introduction

Root hairs are tubular outgrowths of specialized root epidermal cells, trichoblasts. In the trichoblast expanding by diffuse cell growth, a fundamental change in expansion mode occurs, and the root hair itself is further elongated only by polarized tip growth (Dolan et al. 1994; Baluška et al. 2000), a special mode of extension also shared by pollen tubes (Schoenaers et al. 2017). Root hairs increase the root's total surface, substantially improving water and nutrient uptake from soil. They also support the anchoring of the plant in the soil and enable symbiotic interactions with the beneficial microflora in the rhizosphere (Gilroy and Jones 2000). The establishment and regulation of the tip growth is conditioned by the maintenance of the polar organization of the tip-growing cell. Membrane transport processes, temporal and spatial regulation of exocytosis and endocytosis, as well as selective recycling of vesicles play a central role in this maintenance. Polar tip growth requires regulated directional transport, dynamic properties of the cytoskeleton, maintenance of physiological gradients in the tip and, last but not least, signaling cascades (Baluška et al. 2000; Šamaj et al. 2006). Pollen tubes and root hairs are model objects for the study of tip growth in plants (Schoenaers et al. 2017).

The formation of a bulge on the outer cell wall of the trichoblast represents a transitional phase between the diffuse growth of the epidermal cell and the tip growth of the root hair. From a mechanical and functional point of view, it is a rebuilding of the cell wall, which occurs primarily by local loosening of bonds. There is a local decrease in the pH of the cell wall (Bibikova et al. 1998) and a localized increase in the concentration of protons in the cytoplasm (Bibikova et al. 1997). This is associated with increased xyloglucan endotransglycosylase activity (Vissenberg et al. 2001) and specific accumulation of expansins (EXPANSIN 7 and EXPANSIN 10) in trichoblasts (Cho and Cosgrove 2002). The creation of a proton gradient is also closely associated with the production of reactive oxygen species (ROS; Foreman et al. 2003). The tip growth mechanism of root hairs requires a tip-focused gradient of Ca²⁺ in the cytosol, which regulates the activity of vesicular trafficking, generation of ROS, and organization of the cytoskeleton in the growing tip. Ca²⁺-permeable ion channels are localized at the apical plasma membrane (Miedema et al. 2008), and ROS are generated by Arabidopsis (Arabidopsis thaliana) RESPIRATORY BURST OXIDASE HOMOLOG PROTEIN C/ROOT HAIR DEFECTIVE 2 (AtRBOHC/RHD2). ROS production is regulated by Rho of Plants (ROP) nucleotide guanosine triphosphate (GTP)-ases (Foreman et al. 2003), and the whole process is interconnected by a positive feedback mechanism, which determines root hair polarity and tip growth (Carol et al. 2005; Takeda et al. 2008). Tip-focused accumulation of ROS is also tightly associated with pollen tube tip growth (Schoenaers et al. 2017). Arabidopsis ATP-BINDING CASSETTE G28 (AtABCG28), an ABC transporter, is responsible for ROS accumulation at the tip of the growing Arabidopsis pollen tubes (Do et al. 2019). In tip-growing cells, NADPH oxidases, highly regulated transmembrane proteins that use cytosolic NADPH, produce superoxide (O₂^•−) at the apoplastic side of the plasma membrane, which is then converted to hydrogen peroxide (H₂O₂) by the activity of the superoxide dismutase (SOD) in the apoplast (Demidchik 2018). Hydrogen peroxide can diffuse to the cytoplasm of the tip-growing cells via aquaporins, forming a tip-focused gradient of ROS (Bienert and Chaumont 2014).

The oxidation of 2′-7′ dichlorodihydrofluorescein (H₂DCF) to 2′-7′dichlorofluorescein (DCF) is used extensively for the quantitation of H₂O₂ in living cells. An important parameter is the good permeability of the diacetate form, H₂DCFDA, and its acetomethyl ester H₂DCFDA-AM through the plasma membrane of living cells. Probes are taken up by cells, and nonspecific cellular esterases cleave off the lipophilic groups, resulting in trapping of the charged compound inside the cell. Oxidation of H₂DCF by ROS in cells converts the molecule to 2′,7′ dichlorofluorescein (DCF), which is highly fluorescent. Although DCF was believed to be specific for H₂O₂, some evidence indicates that other radicals, such as peroxynitrite and hypochlorous acid, can oxidize H₂DCF (Hoffman et al. 2008). However, even though this probe can react with other ROS, due to its high sensitivity, it has been widely used to monitor H₂O₂ production in plants (Swanson et al. 2011). The CellROX Deep Red is a fluorescent reagent that detects O₂^•− and hydroxyl radicals (Alves et al. 2015). It is a cell-permeant dye weakly fluorescent while in a reduced state. However, it exhibits strong and photostable fluorescence upon oxidation by ROS. Amplex Red is a commonly used fluorogenic substrate, which serves as a hydrogen donor in conjugation with horseradish peroxidase (HRP) enzyme to produce an intensely fluorescent product. Amplex Red serves as a potent and sensitive probe in this case, as increasing amounts of H₂O₂ in cells leads to increasing amounts of specific fluorescent products. Amplex Red is oxidized by H₂O₂ in the presence of HRP and converts to resorufin (Reszka et al. 2005).

In tip-growing cells, fluorescence localization of ROS using CM-H₂DCFDA or DCFH₂-DA is commonly employed. In pollen tubes, this approach was utilized for H₂O₂ detection (Potocký et al. 2007; Do et al. 2019). The same approach was used for the localization of ROS production in Arabidopsis roots upon salt stress-induced signal transduction (Leshem et al. 2007) and in root hairs of Arabidopsis upon treatment with LY294002, a phosphatidylinositol 3-kinase (PI3K)-specific inhibitor (Lee et al. 2008). Alternatively, to test the possibility that the ROS produced by mitochondria might influence the tip growth of pollen tubes, a CellROX Deep Red probe that can detect O₂^•− and hydroxyl radicals was used (Do et al. 2019). In the absence of ROS, CellROX Deep Red remains in its reduced state with no fluorescence, which is rapidly induced upon oxidation during cellular oxidative stress. It has been shown that CellROX Deep Red can be used to detect ROS production in ram sperm by in vitro and in vivo oxidative stress induction (Alves et al. 2015). To our knowledge, Amplex Red has not been used to study ROS production, distribution, and utilization during root hair formation.

Sensitive fluorescent probes have not yet been utilized for subcellular discrimination between ROS sources in growing root hairs. Mapping of the origin and subcellular location of the production and dynamics of ROS utilization during bulge formation and tip growth of root hairs has not been documented. In this study, live-cell imaging using high-resolution spinning disk fluorescence microscopy supplemented with semiquantitative analysis was used for the functional characterization and spatiotemporal analysis of ROS production, delivery, and utilization in Arabidopsis root hairs from bulge formation to tip growth. For staining of the Arabidopsis root apex, we used probes for selective ROS detection in living cells, CM-H₂DCFDA, CellROX Deep Red, and Amplex Red. The subcellular staining patterns of these three ROS-selective probes in growing root hairs differed; therefore, we also analyzed subcellular ROS production in the root hair defective 2-1 (rhd2-1) and deformed root hairs1-3 (der1-3) root hair mutants and corresponding Col-0 and C24 wild-type ecotypes. To prove the sensitivity of the used probes, we also determined spatial changes of ROS production and compartmentalization in root hairs induced by external ROS modulators such as 1-aminocyclopropane-1-carboxylic acid (ACC) and valinomycin. Overall, the approach of detailed subcellular qualitative and quantitative analysis of ROS distribution patterns using sensitive probes in living cells may serve as a platform for a deeper understanding of root hair development.

Results

Fluorescence ROS detection in Arabidopsis root hairs by live-cell imaging

Microscopic visualization of growing root hairs, their nondestructive loading with fluorescent probes, and subsequent fluorescence live-cell imaging for tens of minutes is possible only at physiologically relevant conditions. This is fundamental not only when fluorescently detected ROS will be visualized and measured, but also in control media before the probe loading. In our system, together with checking for normal root hair morphology and typical intracellular polarity, a continuation of tip growth was monitored in all experiments. To this point, as a positive control for fluorescent dye loading and performance of treated plants in the microscope, staining with a cell viability marker, fluorescein diacetate (FDA), was used. The probe labeled all cells in the root tip and root hair formation zone of Col-0 (Fig. 1A), rhd2-1 (Fig. 1B) and der1-3 (Fig. 1C) roots. A prominent and strong signal was detected in growing root hairs of Col-0 plants, accumulated mainly in apical and subapical zones of root hairs (Fig. 1A). Detailed analysis revealed labeling of the cytoplasm in growing root hairs of Col-0 (corresponding wild-type ecotype for rhd2-1; Fig. 1D) and C24 (corresponding wild-type ecotype for der1-3; Fig. 1E), supplemented by a signal located in distinct subcellular spots. Root hairs of rhd2-1 and der1-3 mutants cannot elongate by tip growth (Fig. 1, B and C). Therefore, only bulges were present, showing typical fluorescence in the cytoplasm after FDA staining of rhd2-1 (Fig. 1F) and der1-3 (Fig. 1G) roots. Analysis of root hair FDA fluorescence at three time points-0 min (Fig. 1, H and K), 5 min (Fig. 1, I and L), and 10 min (Fig. 1, J and M) after the beginning of imaging-revealed the equal distribution of the fluorescence signal in apical and subapical parts of growing root hairs with negligible temporal variations (Fig. 1N). This favors FDA as a positive control for dye-loading, showing that any difference in signal intensity among time points or between mutants observed in further experiments is not related to prolonged or unequal dye accessibility or internalization during imaging. Root hair tip growth of untreated plants observed in time-lapsed imaging mode using 100× lens in the spinning disk microscope was not disturbed (Supplemental Movie S1). FDA staining partially reduced the root hair tip growth rate as compared to unstained Col-0 root hairs, but values around 1 µm·min⁻¹ recorded after FDA staining (Fig. 1O) were within a typical range of Arabidopsis root hair growth rate.

Figure 1.

Monitoring of root hairs, their viability, and tip growth during imaging by staining with FDA. A–C) Bulges and growing root hairs in root hair formation zone of A) Col-0, B)rhd2-1, and C)der1-3 roots. D–G) FDA staining of growing root hairs of D) Col-0, E) C24, and bulges of F)rhd2-1 and G)der1-3 mutants. H–M) Growing root hair of Col-0 after FDA staining (H–J), and pseudocolor-coded fluorescence intensity visualization (K–M). Root hair was imaged at time points of 0 min (H, K), 5 min (I, L) and 10 min (J, M) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where dark blue represents minimal intensity and red represents maximum intensity (inset in K). N) Mean fluorescence intensity of FDA measured in root hair tips (measured area is schematically illustrated in Supplemental Fig. S1B) at imaging times of 0, 5, and 10 min. O) Averaged root hair tip growth rate of control and FDA-treated root hairs of Col-0 plants measured during imaging within the time period of 10 min. N = 9–10 (N, O). Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). Scale bar = 50 µm (A to C), 10 µm (D to M).

We analyzed fluorescence signal distribution in apical parts of developing bulges and growing root hairs after ROS staining with CM-H₂DCFDA, CellROX Deep Red, and Amplex Red probes by qualitative and semiquantitative analyses. We characterized a fluorescence intensity distribution measured along a line profile oriented longitudinally at the apex of bulges and root hairs that was 10 µm long and reached the apical cell wall (Supplemental Fig. S1A), in a 10 µm segment encompassing a clear zone and subapical part of bulges and root hairs (Supplemental Fig. S1B), in the area covering only the cell wall within the clear zone and subapical part of root hairs (Supplemental Fig. S1C), and in the zone of small organelles distribution in the area located in the subapical region of root hairs (Supplemental Fig. S1D). Within the 10-min-long time-lapsed imaging with images captured every 30 s, individual frames from three imaging time points (0, 5, and 10 min) are presented in each experiment.

Differences in ROS detection in bulges and root hairs of control plants and root hair mutants

Staining with CM-H₂DCFDA showed that all epidermal cells in the elongation and differentiation zones of Col-0 roots were stained, while prominent labeling appeared in bulges, short root hairs, and apices of longer growing root hairs (Supplemental Fig. S2, A and B). In developing bulges, staining with CM-H₂DCFDA revealed a relatively low signal, distributed evenly in the cytoplasm without any preference for particular subcellular compartments (Supplemental Fig. S2E to G and Movie S2). An overview of the root hair formation zone showed a prominent labeling pattern observed in apices of growing root hairs (Supplemental Fig. S2A). Therefore, after documentation of signal intensity and distribution in bulges, we analyzed the subcellular distribution of fluorescence signal in growing root hairs. Staining with CM-H₂DCFDA led to the accumulation of specific signal in the cytoplasm, which was particularly prominent in root hair apical and subapical zones and decreased in the apical clear zone (Fig. 2, A to C; Supplemental Fig. S2H to J). Analysis at 0, 5, and 10 min from the beginning of imaging showed the highest signal still located in the apical and subapical parts of growing root hairs (Fig. 2, A to C; Supplemental Fig. S2H to J and Movie S3).

Figure 2.

Distribution of ROS in root hairs of Col-0 and bulges of rhd2-1 and der1-3 mutants after staining with CM-H₂DCFDA. A–I) Growing root hair of Col-0 (A–C), developing bulge of rhd2-1 mutant (D–F) and der1-3 mutant (G–I). Root hairs and bulges were imaged at time points of 0 min (A, D, G), 5 min (B, E, H), and 10 min (C, F, I) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where black represents minimal intensity and white represents maximum intensity (insets in A, D, G). J) Mean fluorescence intensity distribution of CM-H₂DCFDA measured along a 10-µm line oriented longitudinally at the central part of Col-0 growing root hairs, reaching the apical cell wall (schematically illustrated in Supplemental Fig. S1A). Error bars represent standard deviation (SD). K) Semiquantitative mean CM-H₂DCFDA fluorescence intensity analysis (measured area is schematically illustrated in Supplemental Fig. S1B) in bulges of rhd2-1 and der1-3 mutants, and respective Col-0 and C24 wild types. N = 10–21 (J, K). Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). Scale bar = 10 µm (A to I).

To study the relationship between subcellular sources of ROS generation and the mechanism of root hair tip growth, we analyzed two root hair mutants with phenotypes of short root hairs (bulges) that cannot elongate by tip growth. First, the rhd2-1 mutant bears a loss-of-function mutation in the AtRBOHC/RHD2 locus, resulting in missing tip-focused ROS and Ca²⁺ gradients (Schiefelbein and Somerville 1990; Foreman et al. 2003). Second, the der1-3 mutant bears a single-point mutation in the ACTIN2 gene and shows impairment in root hair elongation after bulge establishment (Ringli et al. 2002). Fluorescence ROS staining with CM-H₂DCFDA revealed a very low signal in the root hair formation zone of rhd2-1 mutant (Supplemental Fig. S2C), while it was prominent in this zone of der1-3 mutant (Supplemental Fig. S2D). Live-cell imaging of ROS in bulges of rhd2-1 mutant after staining with CM-H₂DCFDA confirmed a very low signal in the cytoplasm that remained unchanged at 0, 5, and 10 min (Fig. 2, D to F; Supplemental Fig. S2K to M and Movie S4). The fluorescence signal after staining with CM-H₂DCFDA was considerably stronger in bulges of der1-3 mutant (Fig. 2, G to I; Supplemental Fig. S2N to P and Movie S5). Analysis of mean fluorescence intensity distribution of this ROS marker along a profile (Supplemental Fig. S1A) in growing Col-0 root hairs showed equal distribution of the fluorescence signal in apical and subapical zones, with decreasing tendency in the clear zone (Fig. 2J). Accordingly, mean fluorescence intensity measurement (Supplemental Fig. S1B) in bulges of rhd2-1 and der1-3 mutants and their respective Col-0 and C24 wild types revealed a significantly reduced values in rhd2-1, while it was higher and similar to each other in bulges of Col-0, C24, and der1-3 (Fig. 2K). Fluorescence intensity measurement along the profile in apical parts of bulges and root hairs after ROS staining with CM-H₂DCFDA in Col-0 plants was acquired during their growth (Supplemental Fig. S3A and Movies S6 and S7). Fluorescence signal distributions after line profile measurements in growing bulges (Supplemental Fig. S3B), short root hairs (10 to 200 µm; Supplemental Fig. S3C), and long root hairs (>200 µm; Supplemental Fig. S3D) did not show differences among time points. However, profile fluorescence intensities increased during root hair development from bulges (Supplemental Fig. S2E to G) to long root hairs (Supplemental Figs. S2H to J and S3E). This analysis also revealed that CM-H₂DCFDA-specific signal was distributed equally in the subapical region, decreasing considerably in the apical clear zone. Semiquantitative mean CM-H₂DCFDA fluorescence intensity analysis in growing root hairs of Col-0 and C24 wild types did not show considerable differences (Supplemental Fig. S3F).

Staining with CellROX Deep Red led to the detection of fluorescence signal in root epidermal cells of Col-0 plants, and the signal increased in bulges and apices of growing root hairs (Supplemental Fig. S4, A and B). Comparable fluorescent staining with this ROS marker was observed also in the root hair formation zone of rhd2-1 (Supplemental Fig. S4C) and der1-3 (Supplemental Fig. S4D) mutants. After staining with CellROX Deep Red, the resulting signal was localized in distinct intracellular oval compartments in developing bulges. The size, number, and cellular distribution of ROS-positive compartments were not changed during bulge development, and this staining pattern was stable over the three imaging time points (Supplemental Fig. S4E to G and Movie S8). In growing root hairs, these compartments were distributed in the cytoplasm within the subapical zone and around the vacuole but were absent in the clear zone (Fig. 3, A to C; Supplemental Fig. S4H to J). The most prominent accumulation of such compartments was visible in the root hair subapical zone (Fig. 3, A to C; Supplemental Fig. S4H to J and Movie S9). Staining of ROS with CellROX Deep Red in root hair mutants showed a similar pattern as observed in bulges of wild type. Thus, ROS localization in compartments moving in the cytoplasm was observed in bulges of both rhd2-1 (Fig. 3, D to F; Supplemental Fig. S4K to M and Movie S10) and der1-3 (Fig. 3, G to I; Supplemental Fig. S4N to P and Movie S11) mutants.

Figure 3.

Distribution of ROS in root hairs of Col-0 and bulges of rhd2-1 and der1-3 mutants after staining with CellROX Deep Red. A–I) Growing root hair of Col-0 (A–C), developing bulge of rhd2-1 mutant (D–F) and der1-3 mutant (G–I). Root hairs and bulges were imaged at time points of 0 min (A, D, G), 5 min (B, E, H), and 10 min (C, F, I) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where black represents minimal intensity and white represents maximum intensity (insets in A, D, G). J) Mean fluorescence intensity distribution of CellROX Deep Red measured along a 10-µm line oriented longitudinally at the central part of Col-0 growing root hairs, reaching the apical cell wall (schematically illustrated in Supplemental Fig. S1A). Error bars represent standard deviation (SD). K) Semiquantitative mean CellROX Deep Red fluorescence intensity analysis (measured area is schematically illustrated in Supplemental Fig. S1B) in bulges of rhd2-1 and der1-3 mutants, and respective Col-0 and C24 wild types. N = 10 to 17 (J, K). Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). Scale bar = 10 µm (A to I).

Mean fluorescence intensity distribution in CellROX Deep Red-stained growing root hairs of Col-0 plants showed fluctuation of the fluorescence signal intensity, which was caused by the distribution pattern of ROS-positive subcellular compartments that were actively moving. Fluorescence was visible mainly in the subapical zones, but not in the clear zones of root hairs (Fig. 3J; Supplemental Fig. S5A). A similar pattern of distribution and active movement of CellROX Deep Red-stained subcellular compartments was recorded in bulges, and the pattern was stable among the three time points (Supplemental Fig. S5B and Movie S12). The mean fluorescence intensities were greater in growing root hairs than in bulges. Due to the active movement of ROS-positive compartments, the mean fluorescence intensities fluctuated among time points (Supplemental Fig. S5C to E and Movie S13). Similar to the staining pattern by CM-H₂DCFDA, the signal intensity after CellROX Deep Red staining decreased in the apical part of root hairs and was considerably low in their clear zones (Fig. 3J; Supplemental Fig. S5C to E). Mean fluorescence intensity measured in the whole apical part of bulges in rhd2-1 and der1-3 mutants, as well as in their respective Col-0 and C24 wild types, revealed no differences (Fig. 3K). The mean fluorescence intensity of CellROX Deep Red staining in whole apical and subapical parts of growing root hairs was higher in C24 than in Col-0 plants (Supplemental Fig. S5F).

Staining of Col-0 roots with Amplex Red revealed that root epidermal cells were negative, unlike CM-H₂DCFDA and CellROX Deep Red probes. However, bulges, short root hairs, and apices of longer and still growing root hairs were specifically labeled by Amplex Red (Supplemental Fig. S6, A and B). The fluorescence signal of ROS stained with this probe was very low in the root hair formation zone of rhd2-1 mutant (Supplemental Fig. S6C). On the other hand, it was strong and comparable to Col-0 in the root hair formation zone of the der1-3 mutant (Supplemental Fig. S6D). Imaging of the root hair formation zone in Col-0 at higher resolution revealed that developing bulges were outlined by Amplex Red-specific signal at the surface (Supplemental Fig. S6E to G and Movie S14). In growing root hairs, this ROS marker also outlined surface of root hairs, together with subcellular compartments showing weak fluorescence and distributed in the cytoplasm within the subapical part (Fig. 4, A to C; Supplemental Fig. S6H to J and Movie S15). Staining of root hair bulges with Amplex Red provided almost negative results in rhd2-1 mutant (Fig. 4, D to F; Supplemental Fig. S6K to M and Movie S16), while prominent fluorescent signal outlined the surface of bulges in der1-3 mutant (Fig. 4, G to I; Supplemental Fig. S6N to P and Movie S17).

Figure 4.

Distribution of ROS in root hairs of Col-0 and bulges of rhd2-1 and der1-3 mutants after staining with Amplex Red. A–I) Growing root hair of Col-0 (A–C), developing bulge of rhd2-1 mutant (D–F), and der1-3 mutant (G–I). Root hairs and bulges were imaged at time points of 0 min (A, D, G), 5 min (B, E, H), and 10 min (C, F, I) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where black represents minimal intensity and white represents maximum intensity (insets in A, D, G). J) Mean fluorescence intensity distribution of Amplex Red measured along a 10-µm line oriented longitudinally at the central part of Col-0 growing root hairs, reaching the apical cell wall (schematically illustrated in Supplemental Fig. S1A). Error bars represent standard deviation (SD). K) Semiquantitative mean Amplex Red fluorescence intensity analysis (measured area is schematically illustrated in Supplemental Fig. S1C) in bulges of rhd2-1 and der1-3 mutants, and respective Col-0 and C24 wild types. N = 10 (J, K). Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). Scale bar = 10 µm (A to I).

Mean fluorescence intensity distribution within the apex of growing Col-0 root hairs after staining with Amplex Red, analyzed at the distance of 10 µm longitudinally from the apical cell wall (Supplemental Fig. S7A), revealed a prominent signal located at the root hair periphery in the tip, indicating the cell wall (Fig. 4J). A similar pattern of fluorescence signal distribution at the surface was recorded in bulges (Supplemental Fig. S7B and Movie S18) and growing (both short and long) root hairs (Supplemental Fig. S7, C and D and Movie S19) with no differences among recorded time points at 0, 5, and 10 min. The signal intensity, however, increased from bulges to growing root hairs (Supplemental Fig. S7E). Mean fluorescence intensity measured in the cell wall of apical bulge parts (the area of measurement schematically illustrated in Supplemental Fig. S1C) showed a very low level in rhd2-1 mutant (Fig. 4K). It was significantly higher in the der1-3 mutant, which was along with C24 also higher in comparison to Col-0 (Fig. 4K). This difference was corroborated also by the comparison of Col-0 and C24 alone, showing higher mean area fluorescence intensity in C24 (Supplemental Fig. S7F).

Subcellular identity of ROS-positive compartments in root hairs labeled with CellROX Deep Red

In developing bulges and growing root hairs, staining with CM-H₂DCFDA revealed that the fluorescence signal distributed evenly in cytoplasm without any preference to particular subcellular compartments (Fig. 2, A to C; Supplemental Fig. S2H to J and Movie S3). However, CellROX Deep Red-stained motile compartments distributed in the cytoplasm of root hairs with the exception of the clear zone (Supplemental Fig. S4H to J and Movie S9). This prompted us to define their nature by using colocalization analysis with known subcellular molecular markers. We have used transgenic lines carrying (i) GFP-RHD2, a GFP-tagged AtRBOHC/RHD2 (Takeda et al. 2008; Kuběnová et al. 2022), (ii) GFP-RabA1d marker for early endosomal/TGN compartments (Ovečka et al. 2010; Berson et al. 2014; von Wangenheim et al. 2016), (iii) RabF2a-YFP marker of late endosomes (von Wangenheim et al. 2016), and (iv) GFP-tagged mitochondria-targeting sequence of the N-terminus of the F1-ATPase g-subunit (Niwa et al. 1999). GFP-RHD2 is located in vesicular compartments belonging to early endosomes/TGN and to apical plasma membrane in growing root hairs (Fig. 5A). The detailed view did not reveal a colocalization between ROS-positive compartments and GFP-RHD2 (Fig. 5B), which was confirmed by a fluorescence intensity profile measurement through the compartments (Fig. 5C). GFP-RabA1d was located in early endosomal/TGN compartments accumulated in the apical and subapical regions of root hairs (Fig. 5D), and detailed analysis showed no colocalization with ROS-positive compartments (Fig. 5, E and F). RabF2a-YFP decorated larger late endosomes located mainly away from the apical and subapical regions of root hairs (Fig. 5G). Higher magnification (Fig. 5H) and a semiquantitative profile measurement of fluorescence signal intensities (Fig. 5I) revealed no colocalization with ROS-positive compartments. On the other hand, mitochondria identified by a mitochondrial GFP marker showed the same size, shape, and distribution pattern as ROS-positive compartments after CellROX Deep Red staining (Fig. 5J). A high degree of their colocalization was revealed by both high magnification visualization (Fig. 5K) and fluorescence intensity profile measurements (Fig. 5L).

Figure 5.

Colocalization analysis of subcellular molecular markers in root hairs with compartments accumulating ROS after staining with CellROX Deep Red. A–C) Colocalization with transgenic line for visualization of vesicular compartments containing GFP-RHD2, a GFP-tagged AtRBOHC/RHD2. D–F) Colocalization with transgenic line carrying early endosomal marker GFP-RabA1d. G–I) Colocalization with transgenic line carrying late endosomal marker RabF2a-YFP. J–L) Colocalization with transgenic line carrying GFP-tagged mitochondria. Overview of root hairs carrying molecular markers (in green) and ROS-accumulating compartments (in magenta) after CellROX Deep Red staining (A, D, G, J). Detailed images of compartments containing molecular markers and ROS (B, E, H, K) from the white boxes shown in (A, D, G, J). Fluorescence intensity profiles (C, F, I, L) of fluorescence signals of molecular markers (green lines) and ROS-containing compartments (magenta lines) along the interrupted white lines shown in (B, E, H, K). Arrows in (B, C, E, F, H, I, K, L) indicate the position of the compartments. RabF2a-YFP is artificially displayed in green. Scale bar = 5 µm (A, D, G, J), 1 µm (B, E, H, K).

High-resolution time-lapse video-recordings of growing root hairs in Arabidopsis plants expressing a mitochondrial GFP marker and stained with CellROX Deep Red at a given frequency of image acquisition every 30 s provided a dynamic pattern of colocalization (Fig. 6, A to C). We used a single particle dynamic colocalization analysis to track events when the signal of ROS accumulation after staining with CellROX Deep Red colocalized with mitochondria. However, some GFP-labelled mitochondria appeared temporarily without CellROX Deep Red-based ROS staining. This occurred mainly in the clear zone and just behind the clear zone, but not deeper in the subapical and shank regions of root hairs (Fig. 6, A to C). In general, the clear zone of growing root hairs was free of mitochondria, and only sparsely, they entered this region. Such GFP-labeled but initially ROS-free mitochondria were small, round, and highly motile. Therefore, they temporarily showed no CellROX Deep Red signal at the beginning, which was replenished after a short time (Fig. 6, A and D). Also, events resembling small round-shaped mitochondria separation from larger elongated ones already containing ROS stained with CellROX Deep Red (“fission”) were observed. In such cases, small newly separated mitochondria were not stained with CellROX Deep Red; however, they acquired it shortly after appearance (Fig. 6, B and E). Generally, ROS-free appearance was typical for small mitochondria showing dynamic movement that was visually different from larger and less mobile mitochondria already containing ROS stained with CellROX Deep Red (Fig. 6, C and F).

Figure 6.

Single particle dynamic colocalization analysis of GFP-labeled mitochondria with ROS-accumulating compartments after staining with CellROX Deep Red in root hairs. A to C) Three different time points from time-lapsed video-recordings of growing root hairs containing GFP-tagged mitochondria (in green) and ROS-accumulating compartments (in magenta) after CellROX Deep Red staining. D) Video-sequence captured from dashed box in (A) showing mitochondria (green arrows) and ROS-accumulating compartments (magenta arrows) colocalized (double arrows in merge), while small individual mitochondria in the clear zone are still free of ROS (green arrows only in merge). E) Video-sequence captured from dashed box in (B) showing small mitochondrion negative for ROS staining (green arrow) independent from larger mitochondria containing ROS (double arrows in merge). F) Video-sequence captured from dashed box in (C) showing large round and immobile mitochondria containing ROS (double arrows in merge) and small mobile individual mitochondria free of ROS (green arrows only in merge). Time sequences (in seconds) are shown in mito-GFP channels (D to F). Scale bar = 5 µm (A to C), 1 µm (D to F).

Pattern of ROS labeling with Amplex Red in root hairs

Staining pattern and fluorescence intensity distribution within the apex of growing root hairs in Col-0 after staining with Amplex Red revealed a prominent signal at the root hair periphery, indicating the cell wall (apoplastic) localization (Fig. 4, A to C and J). We decided to proof this localization pattern using a transgenic line carrying GFP-RHD2, a GFP-tagged AtRBOHC/RHD2, which is localized in early endosomes/TGN and apical plasma membrane in root hairs (Kuběnová et al. 2022). Staining of the GFP-RHD2 line for ROS with Amplex Red and treatment with 300 mmol·L^–1 D-mannitol showed protoplast detachment from the cell wall in root hairs by plasmolysis. The surface of the detached protoplast was visualized by GFP-RHD2 located in the plasma membrane. Amplex Red-positive surface signal appeared at the remaining apoplast upon protoplast detachment caused by plasmolysis (Fig. 7, A to C), thus supporting the cell wall localization. To support the cell wall-specific detection of ROS by Amplex Red, growing root hairs were double-labeled with cell wall-specific dye Calcofluor White (staining preferentially cellulose) and Amplex Red (Fig. 7, D to F). Qualitative colocalization (Fig. 7, G and I to K) and semiquantitative profile measurement (Fig. 7, H, L, and M) revealed overlapping colocalization patterns between Amplex Red and Calcofluor White. This was further validated by Calcofluor White and Amplex Red double labeling approach combined with root hair plasmolysis by 300 mmol·L^–1 D-mannitol (Fig. 7, N to Q).

Figure 7.

Determination of subcellular ROS accumulating in root hairs after staining with Amplex Red. A–C) Root hair of transgenic line carrying GFP-RHD2 plasmolyzed by 300 mmol·L^–1 D-mannitol (A) that was before stained with Amplex Red (B), shown in merged image (C). D–F) Growing root hair stained with Calcofluor White (D) that was after stained with Amplex Red (E), shown in merged image (F). G, H) Colocalization of Calcofluor White and Amplex Red signal in sequentially stained root hair (G) and fluorescence intensity profile (H) of Calcofluor White (turquoise line) and Amplex Red (magenta line) fluorescence signals along the interrupted white lines across the root hair shown in (G). I–M) Colocalization of Calcofluor White (I) and Amplex Red (J) signal in sequentially stained root hair (K) in the cell wall from the subapical region to the tip and fluorescence intensity profile of Calcofluor White (L) and Amplex Red (M) fluorescence signals along the interrupted white lines in (K). (N–Q) Root hair N) stained with Calcofluor White O) and plasmolyzed by 300 mmol·L^–1 D-mannitol solution containing Amplex Red P), shown in merged image Q). Scale bar = 10 µm.

In addition to apoplastic signal, we also observed weak labeling of intracellular compartments by Amplex Red in growing root hairs (Fig. 7, B and G). This was supported also by a pseudocolor-coded visualization of signal distribution (Fig. 4, A to C). Subsequently, we have used the colocalization analysis to characterize the nature of intracellular ROS-positive compartments in root hairs after Amplex Red staining. GFP-RHD2 in moving vesicular compartments (Fig. 8A) did not colocalize with ROS-positive compartments (Fig. 8, B and C). Early endosomal/TGN compartments marked by GFP-RabA1d (Fig. 8D) did not show colocalization with ROS-positive compartments as well (Fig. 8, E and F). No colocalization between late endosomes marked by RabF2a-YFP (Fig. 8G) and ROS-positive compartments (Fig. 8H) was detected on merged images (Fig. 8I). However, a GFP marker identifying mitochondria (Fig. 8J) showed an overlapping distribution pattern with ROS-positive compartments after Amplex Red staining (Fig. 8K), which was proved by fluorescence intensity profile measurements (Fig. 8L). Colocalization analysis thus revealed that distinct motile compartments weakly stained by Amplex Red in developing bulges and growing root hairs are indeed mitochondria.

Figure 8.

Colocalization analysis of subcellular molecular markers in root hairs with compartments accumulating ROS after staining with Amplex Red. A–C) Colocalization with transgenic line for visualization of vesicular compartments containing GFP-RHD2, a GFP-tagged AtRBOHC/RHD2. D–F) Colocalization with transgenic line carrying early endosomal marker GFP-RabA1d. G–I) Colocalization with transgenic line carrying late endosomal marker RabF2a-YFP. J–L) Colocalization with transgenic line carrying GFP-tagged mitochondria. Overview of root hairs carrying molecular markers (in green) and ROS-accumulating compartments (in magenta) after Amplex Red staining (A, D, G, J). Detailed images of compartments containing molecular markers and ROS (B, E, H, K) from the white boxes shown in (A, D, G, J). Fluorescence intensity profiles (C, F, I, L) of fluorescence signals of molecular markers (green lines) and ROS-containing compartments (magenta lines) along the interrupted white lines shown in (B, E, H, K). Arrows in (B, C, E, F, H, I, K, L) indicate the position of the compartments. RabF2a-YFP is artificially displayed in green. Scale bar = 5 µm (A, D, G, J), 0.5 µm (B, E, H, K).

Spatial changes of ROS production and compartmentalization induced by external modulators

To further validate the specificity of probes in growing root hairs, we have used external modulators of ROS production and compartmentalization. Therefore, we tested the effect of ethylene, which is known to cause a ROS burst in root hairs. Considering physiological relevance of the experiment and the efficient conversion of ACC to ethylene necessary for ROS production in trichoblasts, only low doses of ACC precursor have been used (according to Martin et al. 2022). Arabidopsis seedlings growing on media containing 0.7 µmol·L⁻¹ ACC for 24 h displayed a typical phenotype with shorter roots and a high density of root hairs (Supplemental Fig. S8). Typical amount of ROS accumulation was observed in the apical and subapical cytoplasm of growing root hairs after staining of Col-0 plants with CM-H₂DCFDA (Fig. 9, A to C), which was significantly increased in root hairs pretreated with ACC. Evaluation of images acquired at three different time points (0, 5, and 10 min from the beginning of imaging) revealed enhanced fluorescence, which was distributed in whole apical and subapical parts of growing root hairs (Fig. 9, D to F). Measurement in root hair tips (Supplemental Fig. S1B) confirmed considerable fluorescence intensity increases in ACC-treated root hairs (Fig. 9G). Averaged root hair tip growth rate compared between control and ACC-treated Col-0 plants within 10 min showed reduction caused by ACC treatment (Fig. 9H). This is supported by kymographs showing the velocity of the tip growth in representative individual root hairs and differences between control nonstained root hair (Fig. 9I), CM-H₂DCFDA-stained root hair (Fig. 9J), and ACC-pretreated and CM-H₂DCFDA-stained root hair (Fig. 9K).

Figure 9.

Accumulation of ROS in root hairs of Col-0 stained with CM-H₂DCFDA after pretreatment with the ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC). A–F) Growing control root hair of Col-0 (A–C), and root hair pretreated with ACC (D–F) stained with CM-H₂DCFDA for ROS localization and quantification. Root hairs were imaged at time points of 0 min (A, D), 5 min (B, E), and 10 min (C, F) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where black represents minimal intensity and white represents maximum intensity (insets in A, D). G) Semiquantitative measurement of mean CM-H₂DCFDA fluorescence intensity accumulation in tips of root hairs (measured area is schematically illustrated in Supplemental Fig. S1B) in control and ACC-treated Col-0 root hairs. H) Averaged root hair tip growth rate of control and ACC-treated root hairs of Col-0 plants measured during imaging within the time period of 10 min. N = 9–10. Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). I–K) Kymographs presenting a velocity of root hair tip growth of control nonstained root hair (I), CM-H₂DCFDA-stained root hair (J), and ACC-treated and CM-H₂DCFDA-stained root hair (K). Scale bar = 10 µm (A to F).

Mitochondria are very sensitive to any changes in their membrane potential, representing a driving force for the uptake of different cations. Valinomycin is a potent ionophore causing inner mitochondrial membrane depolarization and malfunctioning (Kolisek et al. 2003). We pretreated Col-0 plants with valinomycin and then analyzed CellROX Deep Red-based staining and distribution of ROS in growing root hairs. Control root hairs of Col-0 stained with CellROX Deep Red showed ROS-positive mitochondria distributed in cytoplasm mainly in the root hair subapical zone (Fig. 10, A to C; Supplemental Fig. S9A to C). However, valinomycin pretreatment caused the removal of mitochondria-resident ROS fluorescence signal (Fig. 10, D to F; Supplemental Fig. S9D to F). When we measured a relative distribution of pixels in the subapical part of root hairs and displayed them in a graph according to their fluorescence intensity, we found a shallow peak of pixels with low fluorescence intensity and many counts of pixels with high fluorescence intensity in root hairs stained with CellROX Deep Red alone (Fig. 10G). Pixels with high fluorescence intensity in the graph (Fig. 10G) from root hairs stained with CellROX Deep Red alone are represented by white color in pseudocolor-coded images (Fig. 10, A to C). A similar analysis from valinomycin pretreated root hairs revealed strong shifting of the curve to pixels with low fluorescence intensity, while high fluorescence intensity pixels were missing (Fig. 10G). This is indicated by the absence of white color in pseudocolor-coded images of valinomycin pretreated root hairs (Fig. 10, D to F). Fluorescence pixel intensity was measured in the area located in the subapical region of root hairs (Supplemental Fig. S1D). Accordingly, measurement of mean fluorescence intensity accumulation in the subapical part of root hairs confirmed a considerable reduction in valinomycin-pretreated root hairs (Fig. 10H). Measurement during the recording time of 10 min revealed that the averaged root hair tip growth rate of valinomycin-treated and CellROX Deep Red-stained root hairs was significantly reduced in comparison to root hairs labeled by CellROX Deep Red without valinomycin (Fig. 10I). This was supported by kymographs showing the velocity of the tip growth in representative individual root hairs and differences between control nonstained root hair (Fig. 10J), CellROX Deep Red-stained root hair (Fig. 10K), and valinomycin-treated and CellROX Deep Red-stained root hair (Fig. 10L). The representative examples of root hair tip growth rate analysis by kymographs are provided for the control nonstained root hair (Supplemental Fig. S10A) and root hairs stained by CM-H₂DCFDA (Supplemental Fig. S10B), CellROX Deep Red (Supplemental Fig. S10C), Amplex Red (Supplemental Fig. S10D), or FDA (Supplemental Fig. S10E), along with averaged root hair tip growth rates (Supplemental Fig. S10F).

Figure 10.

Redistribution of ROS stained with CellROX Deep Red in root hairs of Col-0 after pretreatment with the mitochondrial ionophore valinomycin. A–F) Growing control root hair of Col-0 stained with CellROX Deep Red (A–C), and root hair stained with CellROX Deep Red that was pretreated with valinomycin (D–F). Root hairs were imaged at time points of 0 min (A, D), 5 min (B, E), and 10 min (C, F) of growth. Fluorescence intensity distribution is visualized in a pseudocolor-coded scale, where black represents minimal intensity and white represents maximum intensity (insets in A, D). G) Relative distribution of pixels in apical part of root hairs according to their fluorescence intensity. Normalized pixel number was evaluated using 50 A.U. intervals distribution for control and valinomycin-treated root hairs. N = 4–6. Error bars represent standard deviation (SD). H) Semiquantitative measurement of mean CellROX Deep Red fluorescence intensity accumulation in subapical part of root hairs (measured area is schematically illustrated in Supplemental Fig. S1D) in control and valinomycin-treated Col-0 root hairs. I) Averaged root hair tip growth rate of control and valinomycin-treated root hairs of Col-0 plants measured during imaging within the time period of 10 min. N = 10 (H, I). Box plots display the first and third quartiles, split by the median; the crosses indicate the mean values; whiskers extend to include the max/min values. Lowercase letters indicate statistical significance between lines according to one-way ANOVA with Fisher's LSD tests (P < 0.05). J–L) Kymographs presenting a velocity of root hair tip growth of control nonstained root hair (J), CellROX Deep Red-stained root hair (K), and valinomycin-treated and CellROX Deep Red-stained root hair (L). Scale bar = 10 µm (A to F).

Altogether, experiments with external ROS modulators ACC and valinomycin confirmed the sensitivity of used fluorescent probes to specific changes in ROS production (induced by ACC) or subcellular compartmentalization (induced by valinomycin). The later experiments also indicate that the physiological status of mitochondria might be important for ROS homeostasis during root hair tip growth.

Discussion

Production, accumulation, and localization of ROS in plant cells can be documented by different methods. Popular are small-molecule fluorescent probes that can assess ROS produced within cells or even those released from cells. Because ROS represent a range of chemical oxygen species with different properties, investigating ROS in biological systems should selectively concern a particular ROS of interest. It is obvious that most of the ROS probes do not capture quantitatively intracellular ROS formed under investigation. In addition, the specific reaction of the probe with the ROS generated in cells may interfere with the cellular redox state and, indeed, affect experimental results (Murphy et al. 2022). Small-molecule fluorescent probes are thus frequently used for ROS documentation. However, some limitations, such as selectivity, ability to quantify, linearity in response, or some possible artifacts, should be considered (Kalyanaraman et al. 2012). Fluorescent probe 2′,7′-dichlorodihydrofluorescein (H₂DCF), its diacetate form (CM-H₂DCFDA), rapidly enters living cells and is used as an indicator of ROS in living cells. After its passive diffusion into the cytoplasm, its acetate groups are cleaved by esterases and chloromethyl groups react with glutathione and other thiols. It is subsequently oxidized to a highly fluorescent product 2′,7′-dichlorofluorescein (DCF), which is entrapped inside the cell. This allows long-term observation (Eruslanov and Kusmartsev 2010). It is often used as a probe for the indirect detection of H₂O₂ and oxidative stress (Kalyanaraman et al. 2012; Winterbourn 2014). However, oxidation to the fluorescent 2′,7′-dichlorofluorescein could be catalyzed by several ROS. Thus, this probe is not always considered specific for any particular ROS (Murphy et al. 2022). It is very important to assess physiologically relevant conditions for cells before ROS are detected, visualized, and measured. In our system, undisturbed tip growth of root hairs, reflected by the speed of growth and root hair morphology, secured biologically relevant conditions in our experiments (Supplemental Movie S1). In addition, we stained growing root hairs with a cell viability marker FDA, which served as a positive control for monitoring root hair viability and tip growth. Thus, viability and fitness of treated plants in the microscope were sustainably controlled. A slight increase in ROS levels during imaging of growing root hairs could be induced by the energy of lasers used in long-term observation (Khan et al. 2015). Light is absorbed by organic molecules such as flavins, and they degrade in reaction with oxygen (Gorgidze et al. 1998; Hockberger et al. 1999; Rösner et al. 2016). This produces ROS, mainly O₂^•− and hydroxyl radicals (Icha et al. 2017), which cause the formation of DCF (Eruslanov and Kusmartsev 2010) and the subsequent increase in fluorescence intensity. Moreover, an increasing amount of ROS is typical during root hair elongation. The generation of ROS is important for the import of Ca²⁺ ions, which are required for apical expansion of root hair (Foreman et al. 2003; Carol et al. 2005). Through the control of root hair morphology, undisturbed tip growth, as well as fluorescent dyes loading, any difference in signal intensity among different dyes, genotypes, and recording time-points can be qualitatively and semiquantitatively analyzed, with minimizing the side effects such as prolonged, delayed, or unequal dye internalization during imaging.

The level of H₂O₂ in the apoplast can be determined by oxidizing peroxidase substrates, such as Amplex Red. In addition to H₂O₂ produced in the apoplast, this probe can be sensitive to H₂O₂ released from cell, since H₂O₂ can cross plasma membrane directly or via aquaporins (Bienert and Chaumont 2014). Importantly, the measurement could reveal a real balance between H₂O₂ being produced or removed by intracellular or apoplastic enzymes, and it may reflect the rate of diffusion not only inside but also outside of the cell. Therefore, Amplex Red with horseradish peroxidase can be used for the detection of H₂O₂ release from cells if other reducing agents or peroxidase substrates are absent (Murphy et al. 2022).

The process of ROS production in cells, in relation to their subcellular source, reflects either a passive production, where ROS are generated as byproducts of metabolic pathways during photosynthesis and respiration, or active production by oxidases for signaling, such as RBOHs. The cellular sources of ROS production for signaling can thus be defined as extrinsic (in apoplast and/or cell wall), cytoplasmic and nuclear, or organellar (in chloroplasts, mitochondria, or peroxisomes) (Yao et al. 2002; Datt et al. 2003; Mittler et al. 2022). The NADPH oxidases are plasma membrane proteins accepting electrons from NADPH. This happens at the cytosolic side of the plasma membrane with the formation of molecular oxygen at the outer side, leading to the production of apoplastic O₂^•− (Mittler 2017; Smirnoff and Arnaud 2019). Fluorescent probe 2′,7′-dichlorodihydrofluorescein-diacetate (H₂DCFDA) that is particularly sensitive to H₂O₂ was used for the estimation of time-course production and intracellular localization of ROS in Arabidopsis root tips exposed to salt stress. The non-fluorescent H₂DCFDA becomes entrapped inside the cells by the surrounding plasma membrane once hydrolyzed by the intracellular esterases to H₂DCF. After reaction with H₂O₂, fluorescent dichlorofluorescein (DCF) is produced, which was utilized for ROS localization in endosomes (Leshem et al. 2007). Essential processes, such as vesicle trafficking and ROS formation, were compromised by LY294002, a PI3K-specific inhibitor. ROS generation inside endosomes and endocytosis were indispensable for tip growth. Preincubation of seedlings with LY294002 considerably reduced ROS levels in the root hairs, as estimated using the dye 2′,7′-dichlorodihydrofluorescein-diacetate (Lee et al. 2008). On the other hand, the use of the CellROX Deep Red revealed that the signal was excluded from the apex of growing pollen tube. Instead, it was distributed throughout the body of the pollen tube. This analysis suggested that the tip-localized ROS signal does not represent O₂^•− or hydroxyl radicals, but H₂O₂ in growing pollen tubes. Such results indicate that ROS generated from mitochondria are not major contributors to specific spatial distribution of H₂O₂ in the apex of growing pollen tubes (Do et al. 2019). Interestingly, both rhd2-1 and der1-3 mutants that differed in ROS accumulation in root hair bulges after CM-H₂DCFDA and Amplex Red staining did not show any difference in the pattern of ROS distribution after CellROX Deep Red staining (Fig. 3).

Root hair tip growth is regulated to secure the correct shape and size of bulges and elongating root hairs. Factors such as Ca²⁺, pH, and ROS play an important role in this process (Mangano et al. 2016). In growing root hairs, ROS are produced by RHD2, a plasma membrane-resident NADPH oxidase. ROS regulate polar root hair expansion by activating calcium channels allowing Ca²⁺ uptake. Arabidopsis rhd2-1 mutant is defective in Ca²⁺ uptake and have impaired cell expansion, causing a phenotype of very short root hairs (Foreman et al. 2003). Our results show that the rhd2-1 mutant possesses a considerably lower amount of ROS in the bulge as analyzed by a CM-H₂DCFDA fluorescence. These results are not surprising, as a mutation of the RHD2 gene targets the NADPH oxidase AtRBOHC/RHD2, responsible for the production of ROS in root hairs (Foreman et al. 2003). It proves previous results that intracellular ROS levels in short root hairs of the rhd2 mutant, detected using CM-H₂DCFDA, were below the detection limit at pH 5 (Lee et al. 2008). Activity of AtRBOHC/RHD2 leads to the generation of apoplastic O₂^•−, from which H₂O₂ is subsequently produced by a reaction catalyzed by SOD (Mangano et al. 2016). Our data show a dramatic reduction in the formation of apoplastic ROS, particularly H₂O₂, which has been clearly documented by very low level of Amplex Red staining. In turn, a reduction in the formation of apoplastic ROS can also result in reduced levels of H₂O₂ in the cytoplasm, which was in our results reflected by a lowering in fluorescence intensity after CM-H₂DCFDA staining. The actin cytoskeleton participates in the polar elongation of cells and is indispensable also for determining the tip growth in root hairs (Zheng et al. 2009; Wang et al. 2010; Pei et al. 2012). Together with actin, indispensable is the activity of actin-binding proteins. Actin-depolymerization factor (ADF) and profilin are accumulated in the bulge and in the growing tips of root hairs, where they are responsible for the dynamic remodeling of the filamentous actin network (Staiger et al. 1997; Braun et al. 1999). A high degree of actin dynamics in the bulge selectively concentrates endomembranes and leads to the direction of vesicular transport preferentially to this subcellular domain (Baluška et al. 2000). Mutagenesis of the ACTIN2 gene led to the isolation of der1-3 mutant line, which shows a phenotype of very short root hairs not able to elongate after a bulge formation (Ringli et al. 2002; Vaškebová et al. 2018). Since der1-3 mutant is more resistant to oxidative stress (Kuběnová et al. 2021), it justifies our interest to determine ROS accumulation and distribution in bulges of this mutant. Determination of ROS related to bulge stage during the root hair formation process revealed, however, that ROS levels selectively detected by CM-H₂DCFDA and Amplex Red staining were not altered in der1-3 mutant. Apart from striking differences in cytoplasmic and apoplastic ROS levels between wild-type and rhd2-1, clear differences were found also between rhd2-1 and der1-3, suggesting principally different spatiotemporal distribution, production, delivery, and utilization of ROS in these mutants.

The results show that the employed dyes have high sensitivity for ROS localization in different subcellular compartments and, therefore, are suitable for high-resolution imaging of ROS specific distribution in living cells. Morphological and molecular identification of ROS-positive compartments in root hairs after the use of different dyes is thus important aspect of the study. Our colocalization analysis identified motile compartments labeled with CellROX Deep Red as mitochondria. In addition, it provided biological insight into possible participation of mitochondria in the maintenance of subapical region of root hair. In particular, the clear zone of growing root hairs is usually free of larger organelles including mitochondria that only sparsely invade this region (Zheng et al. 2009); thus, their appearance and behavior in the clear zone is rather enigmatic. Here, we document presence of mitochondria in the clear zone of growing root hairs. Although these were rather unique events, mitochondria appeared there occasionally and were temporarily not stainable for ROS with CellROX Deep Red. Importantly, such mitochondria were small, round, and highly motile. They were negative to CellROX Deep Red staining at the beginning; however, this labeling was replenished in them after short time. In contrast, larger and less mobile mitochondria were intensely stained for ROS with CellROX Deep Red. Appearance of small ROS-free mitochondria was not observed in root hair basal parts. Hypothetically, they may represent premature mitochondria created by fission from mother ones (Westrate et al. 2014; Rose 2021), while ROS producing capacity is activated in them only after some time. Experimental proof of this hypothesis and clarification of its functional significance will require next detailed study.

Based on high sensitivity of employed dyes, external modulation of ROS production or compartmentalization should be microscopically detectable and directly analyzed in growing root hairs. Plant gas hormone ethylene affects root growth and causes elevated production of ROS in trichoblasts (Martin et al. 2022). An ethylene precursor, ACC, used in a high concentration may directly alter several developmental processes without its own conversion to ethylene (Li et al. 2022). Therefore, we applied recommended low doses of ACC (Martin et al. 2022) to keep physiological relevance in our experiments. Qualitative and quantitative analysis of mean fluorescence intensity accumulation in tips of growing root hairs after staining with CM-H₂DCFDA revealed a significant increase in ACC-treated root hairs. This experiment confirmed the sensitivity of the dye CM-H₂DCFDA to the external modulation of ROS production in cytoplasm.

Rapid influx of different cations (e.g. Mg²⁺) to mitochondria is driven by the mitochondrial membrane potential. Upon addition of ionophore valinomycin, the depolarization of mitochondrial membranes leads to drastic reduction of Mg²⁺ influx to mitochondria (Kolisek et al. 2003). We used this approach to depolarize the mitochondrial membrane potential using valinomycin in root hairs. Roots of Col-0 plants were pretreated with valinomycin and stained with CellROX Deep Red. Compared to nontreated Col-0 plants, where root hairs contained mitochondria stained with CellROX Deep Red, such specific staining pattern disappeared after pretreatment with valinomycin. This result indicates that depolarization of mitochondrial membranes potential by valinomycin likely prevents CellROX Deep Red oxidation inside mitochondria, which might be impaired in ROS production. This may suggest that in control root hairs, CellROX Deep Red internalization to fully functional mitochondria is needed for its ROS-dependent oxidation.

We have used a semiquantitative measurement approach, which is compatible with high-resolution live-cell imaging microscopy. A quantitative approach would also be feasible, considering the biochemical principle of the reactions required for fluorescence detection. However, the reaction of the probe with the ROS generated in cells may interfere with the cellular redox state. As a consequence, the ROS equilibrium can be changed by reducing the number of molecules to be detected due to their use for a probe oxidation (Kalyanaraman et al. 2012). For example, the Amplex Red is an oxidizing substrate for peroxidases in the apoplast and can be used with horseradish peroxidase for detection of H₂O₂ release from cells, but only if other reducing agents or peroxidase substrates are absent (Murphy et al. 2022). Therefore, the influence of employed dyes on ROS homeostasis in root hairs may affect directly or indirectly root hair tip growth rate. Indeed, our data show that averaged root hair tip growth rate was highest in the untreated root hairs, while it was decreased after FDA, ACC, CM-H₂DCFDA, and Amplex Red application. The growth reduction was considerable in CellROX Deep Red-stained root hairs, and the highest one was recorded in valinomycin-treated and CellROX Deep Red-stained root hairs (Figs. 9 and 10; Supplemental Fig. S10). Therefore, the possibility of partially affecting root hair physiology by the employed dyes through their scavenging potential to ROS should be taken into account.

In conclusion, our high-resolution live-cell imaging provides a dynamic mapping of ROS in cytoplasm, mitochondria, and apoplast, highlighting subcellular location of the ROS production and dynamics in tip-growing root hairs (Fig. 11). The data show a sensitivity of CM-H₂DCFDA to ROS in cytoplasm, CellROX Deep Red to ROS in mitochondria, and root hair-specific sensitivity of Amplex Red to ROS in apoplast and mitochondria. The high sensitivity of CellROX Deep Red combined with high spatiotemporal resolution imaging method revealed small, highly motile premature mitochondria in the growing tip of root hairs that are free of ROS at the early stages of their formation. We also provide characteristic spatial changes in ROS production and compartmentalization that were induced by external ROS modulators, by inducer, ethylene precursor ACC, or membrane depolarization agent, ionophore valinomycin. Thus, this study also shows physiological characterization of ROS probes employed herein. Moreover, subcellular localization and quantification of ROS production in rhd2-1 and der1-3 root hair mutants, differences between them, and their distinct performance from corresponding wild types shed light on the relationship between different ROS sources in the root hair tip growth mechanism requiring both RBOHC and actin cytoskeleton. Overall, this study establishes fluorescent ROS probes, CM-H₂DCFDA, CellROX Deep Red, and Amplex Red, for functional characterization of spatiotemporal mode of ROS production, delivery, and utilization in Arabidopsis wild-type bulges and root hairs, but also in diverse root hair mutants.

Figure 11.

Model of subcellular ROS localization in root hairs with CM-H₂DCFDA, CellROX Deep Red and Amplex Red. CM-H₂DCFDA represents a sensitive probe for ROS in the cytoplasm. CellROX Deep Red localizes specifically ROS in mitochondria. Amplex Red serves as a highly sensitive marker for root hair-specific apoplastic ROS. In addition, root hairs are able to internalize Amplex Red, which sensitively detects ROS in mitochondria. H₂O₂, hydrogen peroxide; O₂^•−, superoxide; PIP, plasma membrane intrinsic protein aquaporin; RHD2, ROOT HAIR-DEFECTIVE 2; ROS, reactive oxygen species; SOD, superoxide dismutase.

Materials and methods

Plant material and cultivation in vitro

Experiments were performed with Arabidopsis (Arabidopsis thaliana) ecotypes Col-0 (corresponding wild type for rhd2-1) and C24 (corresponding wild type for der1-3); rhd2-1 (Foreman et al. 2003) and der1-3 mutants (Ringli et al. 2002); and stably transformed Arabidopsis lines carrying the construct GFP-RHD2 for NADPH oxidase type C (Takeda et al. 2008), GFP-RabA1d for early endosomes/TGN (Ovečka et al. 2010; Berson et al. 2014; von Wangenheim et al. 2016), RabF2a-YFP for late endosomes (von Wangenheim et al. 2016), and the GFP-tagged mitochondria-targeting sequence of the N-terminus of the F1-ATPase g-subunit (Niwa et al. 1999). After surface-sterilization, the seeds of control, mutant and transgenic lines were planted on half-strength Murashige and Skoog (MS) medium (Murashige and Skoog 1962) without vitamins, solidified with 0.6% (w/v) gellan gum (Alfa Aesar, Thermo Fisher Scientific, Heysham, UK), the pH was adjusted to 5.7. To synchronize seed germination, Petri dishes were stratified for 3 days at 4 °C and then cultivated vertically in an environmental chamber at 21 °C, 70% humidity, and 16-h/8-h light/dark cycle. The illumination intensity was 130 µmol·m⁻²s⁻¹.

Application of fluorescent probes for ROS detection

CM-H₂DCFDA, CellROX Deep Red, and Amplex Red reagents (Thermo Fisher Scientific, Waltham, MA, USA) were used as vital probes for ROS detection in growing root hairs. At least in five independent plants per treatment, CM-H₂DCFDA (5 µmol·L^–1), CellROX Deep Red (10 µmol·L^–1), and Amplex Red (1 µmol·L^–1) in modified liquid MS medium (Ovečka et al. 2014) were individually applied by perfusion directly to plantlets in the microscopic chamber. In total, 60 µL of the modified liquid MS medium with CM-H₂DCFDA, CellROX Deep Red, or Amplex Red was applied in six separate steps of 10 µL each. Plants were then observed using a spinning disk microscope. The root hairs were scanned at different times after perfusion of the probes. Several hairs were usually scanned consecutively, with at least 10-min pauses between scans to avoid induction of fluorescence caused by overloaded laser exposure.

Plasmolysis of root hairs

Transgenic plants carrying GFP-RHD2, 2- to 3-day-old were transferred to microscopic chambers containing liquid MS medium modified according to Ovečka et al. (2005, 2014). After manipulation with plants during sample preparation, the subsequent stabilization period for 24 h allowed undisturbed growth of the root and the formation of new root hairs. The plasmolysis was induced with 300 mmol·L^–1 D-mannitol (Sigma-Aldrich, St Louis, MO, USA) by perfusion method. Amplex Red Reagent (1 µmol·L^–1) was added to the mannitol solution. Total volume of solution was 100 µL divided into 10 steps of 10-µL applications. Plasmolyzed root hairs were immediately observed using Airyscan microscope.

Epifluorescence microscopy

Col-0 plantlets (prepared in the microscopic chambers) were observed using an epifluorescence microscope Axio Imager M2 equipped with DIC optics and epifluorescence metal halide source illuminator HXP 120 V (Zeiss, Oberkochen, Germany) and analyzed using Zeiss ZEN 2012 Blue software (Zeiss, Germany). Root tips were labeled with fluorescent probes for ROS detection (CM-H₂DCFDA, CellROX Deep Red, and Amplex Red). Imaging was performed with N-Achroplan 5×/0.15 NA dry objective and documented with a Zeiss AxioCam ICm1 camera. A filter set providing a wavelength of 450–490 nm for the excitation and 515–565 nm for the emission at 200-ms exposure time was used to visualize the CM-H₂DCFDA signal. For the CellROX Deep Red and Amplex Red, the filter set provided an excitation wavelength of 533–558 nm and an emission wavelength of 570–640 nm. The exposure time was set up to 1 s for both probes. The image scaling for all signals was 0.93 × 0.93 µm in x × y dimensions and 15.12 µm in the z dimension. Final images were processed by orthogonal projection with weighted average function.

Spinning disk microscopy

Samples were prepared according to Ovečka et al. (2005, 2014), by transferring 2- to 3-d-old plants to microscopic chambers containing modified liquid MS medium. As the root hairs are very sensitive to stress during the manipulation, overnight cultivation for plant stabilization and undisturbed root growth and the formation of new root hairs was necessary. For quantitative and colocalization analyses, a Cell Observer SD Axio Observer Z1 spinning disk (SD) microscope (Carl Zeiss, Germany) was used, equipped with alpha Plan-Apochromat 100×/1.57 NA DIC Korr oil immersion objective (Carl Zeiss, Germany). The samples were imaged using an excitation laser line of 405 nm and emission filter BP450/50 for Calcofluor White signal detection, excitation laser line of 488 nm and emission filter BP525/50 for CM-H₂DCFDA, FDA and GFP signal detection, excitation laser line of 514 nm, and emission filter BP535/30 for YFP signal detection, excitation laser line of 639 nm and emission filter BP690/50 for CellROX Deep Red signal detection, and excitation laser line of 561 nm and emission filter BP629/62 for the Amplex Red signal detection. The excitation laser power level for all lasers used was set up to 50%, and the image scaling for all channels was 0.133 × 0.133 µm in x × y dimensions, and with the z dimension 0.50 µm for the Calcofluor White signal, 0.63 µm for the GFP, FDA, and CM-H₂DCFDA signal, 0.66 µm for the YFP channel, 0.75 µm for the CellROX Deep Red and Amplex Red signal. Images were acquired sequentially or simultaneously with two Evolve 512 EMCCD cameras (Photometrics) with an exposure time of 700 ms. The samples were scanned with 700 ms of exposure time for GFP, YFP and CM-H₂DCFDA signal, with 50 ms of exposure time for CellROX Deep Red and Calcofluor White signal, and with 500 ms of exposure time for Amplex Red signal (for quantitative analysis) or using a camera-streaming mode (for colocalization studies).

Airyscan confocal laser scanning microscopy

Root hairs for ROS staining after plasmolysis were observed using a confocal laser scanning microscope LSM880 equipped with Airyscan (Carl Zeiss, Germany). Image acquisition was performed with a 20×/0.8 NA dry Plan-Apochromat objective (Carl Zeiss, Germany). The samples were imaged with an excitation laser line of 488 nm and BP420-480 + BP495-550 emission filters for GFP detection and an excitation laser line of 561 nm and BP420-480 + BP600-650 emission filters for Amplex Red detection. The laser power did not exceed 2% for GFP and 1% for Amplex Red of the available laser intensity range. The samples were scanned with 700 ms of exposure time for both signals using a 32 GaAsP detector. Pixel dwell time was set up to 1.6 µs, and with default settings of the gain level the image scaling was set up to 0.07 × 0.07 µm (x × y).

ROS induction

For ROS induction in root hairs, ACC acid, an ethylene precursor, was used according to Martin et al. (2022). Two-day-old Col-0 seedlings, previously cultured on the control medium, were carefully transferred to the medium containing 0.7 µmol·L⁻¹ ACC. After transfer to medium with ACC, plantlets were cultured in vitro overnight, vertically in a phytotron at 21 °C, 70% humidity, and a photoperiod of 16/8 h light/dark (illumination intensity, 130 µmol·m⁻²·s⁻¹). The next day, samples were prepared as described above by transferring to microscopic chambers containing modified liquid MS medium (Ovečka et al. 2014), containing 0.7 µmol·L⁻¹ ACC. The prepared chambers were cultured overnight in a phytotron at 21 °C, 70% humidity, and a photoperiod of 16/8 h light/dark (illumination intensity 130 µmol·m⁻²s⁻¹). Next day, the CM-H₂DCFDA dye (5 µmol·L⁻¹) was applied to prepared chambers by perfusion and examined at the microscope.

Inhibition of mitochondria

Valinomycin (Sigma-Aldrich, St Louis, MO, USA), applied directly to microscopic chambers by perfusion, at a final concentration of 5 µmol·L⁻¹ (Kolisek et al. 2003), was used to induce membrane potential depolarization in mitochondria. Subsequently, the prepared chambers with plants and applied valinomycin were incubated for 30 min and then the CellROX Deep Red dye (10 µmol·L⁻¹) was applied by perfusion and treated plants in chambers were examined at the microscope.

Cell wall staining

Calcofluor White stain binding to cellulose in plant cell walls, dissolved in a modified culture medium, was applied to the plants in microscopic chambers using the perfusion method in a final concentration of 100 µmol·L⁻¹. The chambers were then incubated in the dark for 90 min, according to Bidhendi et al. (2020). Then, root hair plasmolysis was induced using 300 mmol·L^–1 D-mannitol. Amplex Red (1 µmol·L^–1) was added directly to the solution with mannitol. For experiments without plasmolysis, after incubation with Calcofluor White (100 µmol·L⁻¹), only Amplex Red dye was applied to plants in microscopic chambers.

Data analysis and measurements

The signal intensity was analyzed using a SD microscope. The root hairs were scanned for 10 min at 30-s intervals and sorted into groups according to developmental stages. For Col-0 and C24 lines at the stages of bulge, short root hair (10 to 200 µm) and long root hair (>200 µm), while in der1-3 and rhd2-1 mutants at the bulge stage. Zen 3.3 blue edition software (Carl Zeiss, Germany) was used to obtain profiles for quantitative signal intensity distribution of different fluorescent probes in growing root hairs. Fluorescence intensity distribution was measured in single optical section, along the profile in a 10-µm segment from the center of the root hair to its tip (Supplemental Fig. S1A) or as the mean fluorescence intensity in area of the tip (Supplemental Fig. S1B). For the Amplex Red probe, an area including only the cell wall was measured (Supplemental Fig. S1C). Distribution of the CellROX Deep Red before and after valinomycin treatment was measured in area located in the subapical region of the root hair (Supplemental Fig. S1D). Measurements were always performed at the start (0 min), in the middle (5 min) and at the end (10 min) of imaging. The mean fluorescence intensity in area of the tip (Supplemental Fig. S1B) was always analyzed in the middle of the imaging time (5 min). In vacuolated bulges, the area was adjusted individually to not include vacuoles to the measurement.

Acquired data were processed in Microsoft Excel. Individually, the mean background intensities were subtracted from the intensities in the segment. Root hairs with abnormal intensities (exceeding double intensity over the average) were discarded from the statistics. All charts were done in Microsoft Excel software, and statistical analyses were obtained using STATISTICA 13.4 (StatSoft) software using analysis of variance (ANOVA) and subsequent Fisher's LSD tests (P < 0.05). The colocalization studies of fluorescence signals were analyzed using a SD microscope by simultaneous signal acquisition with two independent Evolve 512 EMCCD cameras. The colocalization was analyzed in transgenic lines GFP-RHD2, mito-GFP, GFP-RabA1d, and RabF2a-YFP in combination with the fluorescent probes for ROS detection described earlier. Profiles for colocalization were generated using Zen Blue 2014 software (Carl Zeiss, Germany) and graphically edited in Microsoft Excel.

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL (https://www.arabidopsis.org) data libraries under accession numbers: AT5G51060 (RHD2), AT4G18800 (RABA1D), and AT5G45130 (RABF2A).

Author contributions

L.K. and J.H. performed experiments, microscopic data acquisition and analysis, semiquantitative evaluation, and statistical analyses. P.D. analyzed data from fluorescent probe labeling. J.Š. and M.O. contributed to the experimental plan and data interpretation. M.O. wrote the manuscript with input from all coauthors. J.Š. provided infrastructure and secured funding.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Topology of the quantitative signal intensity measurement in bulges and growing root hairs.

Supplemental Figure S2. Distribution of ROS in Col-0 wild-type, rhd2-1, and der1-3 mutants after staining with CM-H₂DCFDA.

Supplemental Figure S3. Fluorescence intensity measurements in apical parts of bulges and root hairs after ROS staining with CM-H₂DCFDA.

Supplemental Figure S4. Distribution of ROS in Col-0 wild-type, rhd2-1, and der1-3 mutants after staining with CellROX Deep Red.

Supplemental Figure S5. Fluorescence intensity measurements in apical parts of bulges and root hairs after ROS staining with CellROX Deep Red.

Supplemental Figure S6. Distribution of ROS in Col-0 wild-type, rhd2-1, and der1-3 mutants after staining with Amplex Red.

Supplemental Figure S7. Fluorescence intensity measurements in apical parts of bulges and root hairs after ROS staining with Amplex Red.

Supplemental Figure S8. Phenotype of Col-0 wild-type seedlings treated with the ethylene precursor ACC.

Supplemental Figure S9. Distribution of ROS stained with CellROX Deep Red in root hairs of Col-0 after pretreatment with the mitochondrial ionophore valinomycin.

Supplemental Figure S10. Speed of the root hair tip growth analyzed by kymographs.

Supplemental Movie S1. Growing unstained root hairs.

Supplemental Movie S2. Developing bulge of Col-0 stained with CM-H₂DCFDA.

Supplemental Movie S3. Growing root hair of Col-0 stained with CM-H₂DCFDA.

Supplemental Movie S4. Developing bulge of rhd2-1 stained with CM-H₂DCFDA and profile measurement of fluorescence intensity.

Supplemental Movie S5. Developing bulge of der1-3 stained with CM-H₂DCFDA and profile measurement of fluorescence intensity.

Supplemental Movie S6. Developing bulge of Col-0 stained with CM-H₂DCFDA and profile measurement of fluorescence intensity.

Supplemental Movie S7. Growing root hair of Col-0 stained with CM-H₂DCFDA and profile measurement of fluorescence intensity.

Supplemental Movie S8. Developing bulge of Col-0 stained with CellROX Deep Red.

Supplemental Movie S9. Growing root hair of Col-0 stained with CellROX Deep Red.

Supplemental Movie S10. Developing bulge of rhd2-1 stained with CellROX Deep Red and profile measurement of fluorescence intensity.

Supplemental Movie S11. Developing bulge of der1-3 stained with CellROX Deep Red and profile measurement of fluorescence intensity.

Supplemental Movie S12. Developing bulge of Col-0 stained with CellROX Deep Red and profile measurement of fluorescence intensity.

Supplemental Movie S13. Growing root hair of Col-0 stained with CellROX Deep Red and profile measurement of fluorescence intensity.

Supplemental Movie S14. Developing bulge of Col-0 stained with Amplex Red.

Supplemental Movie S15. Growing root hair of Col-0 stained with Amplex Red.

Supplemental Movie S16. Developing bulge of rhd2-1 stained with Amplex Red and profile measurement of fluorescence intensity.

Supplemental Movie S17. Developing bulge of der1-3 stained with Amplex Red and profile measurement of fluorescence intensity.

Supplemental Movie S18. Developing bulge of Col-0 stained with Amplex Red and profile measurement of fluorescence intensity.

Supplemental Movie S19. Growing root hair of Col-0 stained with Amplex Red and profile measurement of fluorescence intensity.

Funding

This work was supported by the Czech Science Foundation GAČR (project Nr. 19-18675S) and by student project IGA_PrF_2022_014 from Palacky University Olomouc.

Data availability

Data that support the findings of this study are available from the corresponding author upon reasonable request.

Dive Curated Terms

The following phenotypic, genotypic, and functional terms are of significance to the work described in this paper:

• root AmiGo: PO:0009005

• NADPH CHEBI: CHEBI:16474

• valinomycin CHEBI: CHEBI:28545

• pollen AmiGo: PO:0025281

• RHD2 Gramene: AT5G51060

• RHD2 Araport: AT5G51060