Auxin regulates distal stem cell differentiation in Arabidopsis roots

Abstract

The stem cell niche in the root meristem is critical for the development of the plant root system. The plant hormone auxin acts as a versatile trigger in many developmental processes, including the regulation of root growth, but its role in the control of the stem cell activity remains largely unclear. Here we show that local auxin levels, determined by biosynthesis and intercellular transport, mediate maintenance or differentiation of distal stem cells in the Arabidopsis thaliana roots. Genetic analysis shows that auxin acts upstream of the major regulators of the stem cell activity, the homeodomain transcription factor WOX5, and the AP-2 transcription factor PLETHORA. Auxin signaling for differentiation of distal stem cells requires the transcriptional repressor IAA17/AXR3 as well as the ARF10 and ARF16 auxin response factors. ARF10 and ARF16 activities repress the WOX5 transcription and restrict it to the quiescent center, where WOX5, in turn, is needed for the activity of PLETHORA. Our investigations reveal that long-distance auxin signals act upstream of the short-range network of transcriptional factors to mediate the differentiation of distal stem cells in roots.

Fig. 1.

Local auxin levels controlling DSC differentiation. (A–G) Differentiation status of columella cells in 5-d-old seedlings. Nondifferentiated DSC (yellow arrowheads) below the QC (red arrowheads) are characterized by the absence of starch, whereas Lugol's solution–stained starch is visible in differentiated columella cells. WT roots show typically one tier of DSC (A), but increasing auxin levels by growing seedlings on 1 μM NAA (B) (96%, n = 288) or 1 μM NPA (80%, n = 286) (C) leads to DSC differentiation. In contrast, mutants defective in auxin biosynthesis yuc1-/yuc4,10,11/± (D) (53%, n = 208) and wei8-1-/wel1-1±/wel2-2± (E) (22%, n = 246) or auxin transport pin3 (F) (33.6%, n = 286) and pin4 (G) (43%, n = 323) show defects in columella differentiation as manifested by multiple tiers of DSC. (H and I) DSC marker J2341 in pin3-4 roots (I) confirms multiple tiers of DSC as compared with the single tier in WT roots (H). (J) Gradual differentiation of DSC revealed by the frequency of roots without DSC (gray) and with QC divisions (white) after different times of treatment with 5 μM NAA. At least 100 seedlings were examined for each time point for each biological repeat. Error bars depict SD from three biological repeats (Student's t test, *P < 0.01). (K) Quantitative evaluation of DSC differentiation confirms that elevated auxin levels promote DSC differentiation (gray) and conversely that defective auxin biosynthesis or transport inhibit DSC differentiation (white). At least 100 seedlings were examined for each genotypes or drug treatments for each repeat. Error bars depict SDs from three biological repeats (Student's t test, *P < 0.01).

Fig. 2.

Regulation of stem cell activity by auxin upstream of the WOX5 homeodomain factor. (A–F) Differentiation status of DSC (yellow arrowhead) below the QC (red arrowhead) in 5-d-old seedlings as inferred from the Lugol's solution staining of differentiated columella cells. The auxin transport mutant pin4 shows defects in columella differentiation displaying typically two tiers of DSC (42.8%, n = 323) (A); wox5 (B), wox5 pin4 (C), and wox5 pin3 (D) promote columella differentiation as shown by the absence of DSC. The 35S::WOX5-GR line shows an inhibited DSC differentiation in the presence of 1 μM DEX, as revealed by multiple tiers of DSC (E) that cannot be rescued by 1 μM NAA (F). (G and H) WOX5::ERGFP signal detected in endogenous expression domain in the QC of the primary root (G), whereas this signal was strongly decreased in the presence of 1 μM NAA. Residual very weak WOX5::ERGFP activity was detected in the broader domain encompassing endodermis and cortex initials (H). (I) Transcription of WOX5 repressed by auxin as detected by quantitative PCR analysis of RNA from primary distal root tips treated with 5 μM NAA for the indicated time (h) or grown at 1 μM NAA. Tubulin is used as a constitutively expressed control. Error bars mark SDs from three biological repeats (Student's t test, *P < 0.05; **P < 0.05; ***P < 0.01).

Fig. 3.

Regulation of stem cell activity by auxin and WOX5 through PLT1. (A–E) Differentiation status of DSC (yellow arrowhead) below QC (red arrowhead) in 5-d-old seedlings as inferred from Lugol's solution staining of differentiated columella cells. Blue arrows mark ectopic cell division in QC or columella cells. Auxin transport mutant pin3 is defective in columella differentiation, displaying typically two tiers of DSC (33.6%, n = 286) (A). Both plt1 (47%, n = 66) (B) and plt1 pin3 (45.1%, n = 76) (C) promoted columella differentiation, as indicated by absence of DSC. Both plt1 plt2 (100%, n = 87) (D) and pin3 plt1 plt2 (100%, n = 101) (E) show strong columella differentiation as indicated by absence of DSC. (F) Both plt1 plt2 and pin3 plt1 plt2 have very short roots compared with Col-0 and pin3. (G and H) Strongly repressed PLT1::ERCFP expression in tin allele of wox5 (H) as compared with WT control (G). (I) Induced PLT1 transcription after overexpression of WOX5 as detected by quantitative PCR analysis. Tubulin is relative control. RNA was isolated from roots of 5-d-old 35S::WOX5-GR seedlings without and with 1 μM DEX induction. Error bars mark SDs from three biological repeats (Student's t test, *P < 0.01).

Fig. 4.

Auxin-promoted root DSC differentiation mediated by ARF proteins. (A–F) Differentiation status of DSC (yellow arrowheads) below QC (red arrowheads) in 4-d-old seedlings as inferred from Lugol's solution staining of differentiated columella cells. WT roots show typically one tier of DSC (A), but increasing auxin levels by growing seedlings on 1 μM NAA (B) leads to DSC differentiation. Pro_35S:MIR160 line shows highly inhibited DSC differentiation as indicated by multiple tiers of DSC (D) that cannot be fully rescued by 1 μM NAA (E). The wox5 Pro_35S:MIR160 double mutant enhanced columella differentiation as shown by the absence of DSC (F), similar to wox5 (C). (G–J) WOX5::ERGFP signals were detected in endogenous expression domain of QC in primary root (G), whereas these signals were strongly inhibited in the presence of 1 μM NAA (H). WOX5::ERGFP signals in Pro_35S:MIR160 are comparable to those in WT controls, whereas very weak ectopic WOX5::ERGFP signals could be observed in columella cells (Inset) (I). Repression of WOX5::ERGFP expression in the QC by auxin was less obvious in Pro_35S:MIR160 (Inset) (J).