Bipolar Plasma Membrane Distribution of Phosphoinositides and Their Requirement for Auxin-Mediated Cell Polarity and Patterning in Arabidopsis

Abstract

Cell polarity manifested by asymmetric distribution of cargoes, such as receptors and transporters, within the plasma membrane (PM) is crucial for essential functions in multicellular organisms. In plants, cell polarity (re)establishment is intimately linked to patterning processes. Despite the importance of cell polarity, its underlying mechanisms are still largely unknown, including the definition and distinctiveness of the polar domains within the PM. Here, we show in Arabidopsis thaliana that the signaling membrane components, the phosphoinositides phosphatidylinositol 4-phosphate (PtdIns4P) and phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P₂] as well as PtdIns4P 5-kinases mediating their interconversion, are specifically enriched at apical and basal polar plasma membrane domains. The PtdIns4P 5-kinases PIP5K1 and PIP5K2 are redundantly required for polar localization of specifically apical and basal cargoes, such as PIN-FORMED transporters for the plant hormone auxin. As a consequence of the polarity defects, instructive auxin gradients as well as embryonic and postembryonic patterning are severely compromised. Furthermore, auxin itself regulates PIP5K transcription and PtdIns4P and PtdIns(4,5)P₂ levels, in particular their association with polar PM domains. Our results provide insight into the polar domain-delineating mechanisms in plant cells that depend on apical and basal distribution of membrane lipids and are essential for embryonic and postembryonic patterning.

Figure 1.

PIs Are Polarly Localized in Root Cells. Root epidermal cells of 4-d-old seedlings harboring PI biosensors were imaged using confocal microscopy. Images are displayed using a color gradient to indicate low (black/dark blue) to high (yellow) fluorescence intensity. (A) and (B) PtdIns4P biosensor YFP-PH_FAPP1, driven by the cauliflower mosaic virus promoter 35S (p35S) (A) and the UBQ10 promoter (pUBQ10) (B). (C) and (D) PtdIns(4,5)P₂ biosensor YFP-PH_PLCδ1, driven by the p35S (C) and pUBQ10 (D) promoters. Arrowheads highlight the polar localization of the markers. Bars in the left and middle images in each panel = 20 μm; bars in the right images in each panel = 10 μm.

Figure 2.

PIP5K1 and PIP5K2 Expression and Subcellular Localization. (A) and (B) PIP5K1 (A) and PIP5K2 (B) expression under the control of their respective native promoters imaged in 4-d-old root tip seedlings using confocal microscopy. YFP-PIP5K1 is predominantly expressed in procambial cells and YFP-PIP5K2 in epidermal cells. Both PIP5Ks are localized to apical–basal PM (arrowheads), apart from a strong nuclear localization (asterisks). (C) Live imaging of 4-d-old seedlings using confocal microscopy of YFP-PIP5K1 in the apical root meristem expressed under the control of the weak constitutive promoter of Arabidopsis DNAJc17 (AT5G23590). YFP-PIP5K1 is asymmetrically localized to the apical and basal cell sides of procambial (proc; middle panel) and epidermal (epi; right panel) cells. (D) and (E) Asymmetrical localization of YFP-PIP5K1 in heart embryo cells (D) and at different developmental stages of lateral root primordia in 10-d-old seedlings (E). Lateral root primordia stages indicated in the top right corners are as described by Malamy and Benfey (1997). Arrowheads highlight polar localization, and asterisks indicate nuclear localization. Bars = 20 μm.

Figure 3.

PI and PIP5K Polarity Indexes. Apical/basal-to-lateral fluorescence ratio (polarity index) of PI biosensors expressed under the control of the 35S promoter and PIP5K1 proteins in root epidermis (epi) and procambial (proc) cells expressed under the control of the DNAJc17 promoter. As an apolar marker, we measured the aquaporin PLASMA MEMBRANE INTRINSIC PROTEIN2 (aqPIP2)-GFP in root epidermis cells, and we used PIN1-GFP and PIN2-GFP as polar markers in stele and epidermal root cells, respectively. The dotted line indicates a 1:1 ratio corresponding to a theoretical symmetrically localized PM protein. Root cells of seedlings 4 d after germination were quantified. Bars show means ± sd. A two-tailed Student’s t test compared data with aqPIP2-GFP values. *P < 0.05, **P < 0.005. n ˃ 20 cells corresponding to roots imaged under comparable conditions.

Figure 4.

pip5k1^−/− pip5k2^−/− Mutants Show Multiple Developmental Abnormalities and Ectopic Auxin Response. Siliques of 6-week-old plants were used to isolate ovules using a binocular stereomicroscope and mounted either directly in a drop of chloral hydrate clearing solution ([A] and [B]) or an 8% Suc solution ([C] and [D]). The images were obtained using a DIC-equipped microscope ([A] and [B]) or a confocal microscope ([C] and [D]). (A) and (B) Two-celled (left), dermatogen (middle), and torpedo (right) stages of embryo development in wild-type (A) and pip5k1^−/− pip5k2^+/− (B) plants. An example of an arrested embryo is shown in the right panel in (B). Arrowheads indicate aberrant cell divisions. Bars = 20 μm. (C) and (D) Auxin-response maxima in embryo visualized by DR5rev:GFP in wild-type (C) and pip5k1^−/− pip5k^−/− (D) embryos. Arrowheads highlight the normal DR5 maxima in incipient cotyledons and the future root pole, and arrows indicate the ectopic DR5 response in pip5k1^−/− pip5k2^−/− embryo suspensor cells. The asterisks mark the position of the hypophysis. Bars = 20 μm. (E) to (G) Seven-day-old seedlings were used to visualize the vascular tissues in cotyledons (E) and the root apical meristem size (F) and auxin response (G). Seedlings were cleared in 100% ethanol and mounted directly in a drop of chloral hydrate solution (E) or mounted directly in the clearing solution after 1 min of Lugol staining (F) or overnight GUS staining (G). (E) and (F) pip5k1^−/− pip5k2^−/− mutants show open vascular loops and vascular islands ([E], arrowhead) in cotyledons. The root apical meristem of the double mutant is smaller than that of the wild type (F). Bars = 50 μm. (G) Auxin-response maxima visualized by DR5:GUS in wild-type and pip5k1^−/− pip5k2^−/− root tips. Bars = 50 μm.

Figure 5.

pip5k1^−/− pip5k2^−/− Mutants Have PIN Polarity Defects. (A) and (B) Immunolocalization of PIN1 in wild-type and pip5k1^−/− pip5k2^−/− mutant embryos. In contrast with the mainly apical polarity of PIN1 in epidermal cells and the basal polarity in procambial cells of wild-type embryos ([A], top panels), the pip5k1^−/− pip5k2^−/− embryos displayed more diffuse PIN1 polarizations ([A], bottom panels) and ectopic PIN1 expression in the upper suspensor cells ([B], arrowhead in right panel). Asterisks indicate the hypophysis, and arrowheads highlight PIN polarity. Bars = 20 μm. (C) PIN immunolocalization in wild-type and pip5k1^−/− pip5k2^−/− 7-d-old seedling root tips. In the wild type, PIN1 in endodermis (en) and PIN2 in cortex (co) and epidermis (epi) cells show clear basal and apical localizations, respectively. By contrast, in pip5k1^−/− pip5k2^−/− root tips, PIN2 accumulates abnormally at the inner and outer lateral sides of cortex cells (arrowheads). Bars = 50 μm. (D) Immunolocalization of PIN1 in wild-type (left) and pip5k1^−/− pip5k2^−/− (right) primary leaves of 9-d-old seedlings. In the wild type, PIN1 is clearly polarized in the developing vascular tissue pointing toward the leaf base. Newly forming, not yet connected, provascular cells show PIN1 polarization toward older vascular strands (inset). In the pip5k1^−/− pip5k2^−/− mutant, PIN1 localization toward the older vascular strands is lost in young, developing vascular tissues (inset). Arrowheads indicate PIN polarity, and asterisks highlight the existing/older vascular strands. Bars = 50 μm.

Figure 6.

Auxin Modulates PIP5K Expression and Transcript Accumulation. (A) and (B) Four-day-old seedlings expressing the transcriptional reporters pPIP5K1:nlsGUS-GFP (A) and pPIP5K2:nlsGUS-GFP (B) were treated in liquid Arabidopsis medium containing 10 μM IAA for the times indicated in the bottom left corners and imaged immediately using confocal microscopy. Both transcriptional reporters were induced within 1 h of auxin treatment and maintained their induction over the period of 4 h studied. The pPIP5K2:nlsGUS-GFP marker was induced in the epidermal cell file (arrowheads in [B]) and all over the differentiation zone after 2 h of auxin treatment (asterisks in [B]). The fluorescence intensity is indicated as a gradient of color from low (black/dark blue) to high (yellow) fluorescence. (C) and (D) Time-course experiment using 10 μM IAA for the times indicated in (C) and for 4 h in (D). RNA extracted from a 1-mm distal section of the root tips of 4-d-old seedlings was used. Means ± se are presented for three independent experiments. PIP5K1 transcript accumulated after 0.5 h of IAA treatment, and these levels were maintained over the time course of the experiment; this pattern of transcript accumulation is similar to that observed for the early auxin-inducible genes BDL and IAA3 (C). The PIP5K1 transcript accumulation after auxin treatment was abolished in arf7 arf19 double mutants (D).

Figure 7.

Auxin Modulates PI Levels at the PM. (A) Total PI amounts measured in 5-d-old seedlings of the wild type and pip5k mutants before and after treatment with 10 μM NAA for 30 min. Auxin reduced the PtdIns(4,5)P₂ levels, indicated here by a 28% reduction in the PtdIns4P:PtdIns(4,5)P₂ ratio in the wild type (see Supplemental Figure 10 for specific data). The single and double pip5k mutants failed to respond to auxin by increasing the amounts of PIs. A two-tailed Student’s t test was used; *P < 0.05. (B) Quantification of the effect of auxin on PtdIns(4,5)P₂ and PtdIns4P at the PM as visualized by the PI markers p35S:YFP-PH_PLCδ1 and p35S:YFP-PH_FAPP1, respectively. A two-tailed Student’s t test was used; *P < 0.05, **P < 0.01. See Methods for details.