NTL8 Regulates Trichome Formation in Arabidopsis by Directly Activating R3 MYB Genes TRY and TCL1

Abstract

The NAM, ATAF1/2, and CUC (NAC) are plant-specific transcription factors that regulate multiple aspects of plant growth and development and plant response to environmental stimuli. We report here the identification of NTM1-LIKE8 (NTL8), a membrane-associated NAC transcription factor, as a novel regulator of trichome formation in Arabidopsis (Arabidopsis thaliana). From an activation-tagged Arabidopsis population, we identified a dominant, gain-of-function mutant with glabrous inflorescence stem. By using plasmid rescue and RT-PCR analyses, we found that NTL8 was tagged; thus, the mutant was named ntl8-1 Dominant (ntl8-1D). Recapitulation experiment further confirmed that the phenotype observed in the ntl8-1D mutant was caused by elevated expression of NTL8 Quantitative RT-PCR results showed that the expression level of the single-repeat R3 MYB genes TRIPTYCHON (TRY) and TRICHOMELESS1 (TCL1) was elevated in the ntl8-1D mutant. Genetic analyses demonstrated that NTL8 acts upstream of TRY and TCL1 in the regulation of trichome formation. When recruited to the promoter region of the reporter gene Gal4:GUS by a fused GAL4 DNA-binding domain, NTL8 activated the expression of the reporter gene. Chromatin immunoprecipitation results indicated that TRY and TCL1 are direct targets of NTL8. However, NTL8 did not interact with SQUAMOSA PROMOTER BINDING PROTEIN LIKE9, another transcription factor that regulates the expression of TRY and TCL1, in yeast and plant cells. Taken together, our results suggest that NTL8 negatively regulates trichome formation in Arabidopsis by directly activating the expression of TRY and TCL1.

Figure 1.

ntl8-1D/gpa1-2 is a gain-of-function, dominant mutant with glabrous inflorescence stems. A, Trichomes on the main inflorescence stems of the Ws wild-type plant (left) and the gpa1-2 mutant (middle), and glabrous stem of the ntl8-1D/gpa1-2 dominant mutant (right). Photographs were taken from the first internodes of 1-month-old soil-grown plants. B, Trichomes on the first two rosette leaves of the Ws wild-type plant (top), the gpa1-2 mutant (middle), and the ntl8-1D/gpa1-2 dominant mutant (bottom). Photographs were taken from 10-d-old soil-grown seedlings. C, Trichome density on the first two rosette leaves of the Ws wild-type plant, the gpa1-2 mutant and the ntl8-1D/gpa1-2 dominant mutant. Data represent the mean ± sd of 18 plants.

Figure 2.

Identification of the T-DNA insertion site in the ntl8-1D/gpa1-2 mutant and recapitulation of the mutant phenotypes. A, Diagram illustrating the activation-tagged T-DNA insertion site in the ntl8-1D/gpa1-2 mutant. Arrowheads indicate the orientation of the 4X35S enhancer repeats in the T-DNA situated 101 bp upstream of the start codon of NTL8, and 1898 bp downstream of the stop codon of At2g27310. B, Expression level of NTL8 and At2g27310 in the Ws wild-type, the gpa1-2 mutant, and the ntl8-1D/gpa1-2 mutant seedlings. Total RNA was isolated from 10-d-old seedlings grown on 0.5× MS plates, and RT-PCR was used to examine the expression of NTL8 and At2g27310. ACTIN2 (ACT2) was used as a control. C, Trichomes on the main inflorescence stems of the Col wild-type (left), the 35S:NTL8 (middle), and 35S:NTL8ΔC transgenic plants (right). Photographs were taken from the first internodes of 1-month-old soil-grown plants. D, Trichomes on the first two rosette leaves of the Col wild-type (left), the 35S:NTL8 (middle), and 35S:NTL8ΔC transgenic plants (right). Photographs were taken from 10-d-old soil-grown seedlings. E, Trichome density on the first two rosette leaves of the Col wild-type plant, the 35S:HA-NTL8 and 35S:HANTL8ΔC transgenic plants. Data represent the mean ± sd of 18 plants.

Figure 3.

Phenotypes of the loss-of-function mutants of NTL8. A, Diagram showing the T-DNA insertion sites in ntl8 and ntl5 single mutants. The T-DNA is inserted in the third and the second exon of NTL8, respectively, for the ntl8-1 and ntl8-2 mutants, and immediately after and before the second exon of NTL5, respectively, for the ntl5-1 and ntl5-2 mutants. B, Trichomes on the main inflorescence stems of the Col wild-type plant, the ntl8-1D dominant mutant, the ntl8 single mutants, the ntl5 single mutants, and the ntl5 ntl8 double mutant. Photographs were taken from the first internodes of 1-month-old soil-grown plants. C, Percentage of branched trichomes on the inflorescence stems of the Col wild-type plant, the ntl8-1D dominant mutant, the ntl8 single mutants, the ntl5 single mutants, and the ntl5 ntl8 double mutant. Data represent the mean ± sd of 11 or 12 plants. D, Trichomes on first two rosette leaves of the Col wild-type plant, the ntl8-1D dominant mutant, the ntl8 single mutants, the ntl5 single mutants, and the ntl5 ntl8 double mutant. Photographs were taken from 10-d-old soil-grown seedlings. E, Phylogenetic analysis of NTL8 and its closely related proteins. The entire amino acid sequences of the proteins were obtained from Phytozome (https://phytozome.jgi.doe.gov/pz/portal.html). “OneClick” mode with default settings on Phylogeny (www.phylogeny.fr) was used to generate the phylogenetic tree. Branch support values are indicated above the branches.

Figure 4.

Expression of the known key trichome formation-regulating transcription factor genes in the wild type and gain- and loss-of- function mutants of NTL8. A, Expression of the known key trichome formation regulating transcription factor genes in the Ws wild-type, the gpa1-2 mutant, and the ntl8-1D/gpa1-2 mutant seedlings. Total RNA was isolated from 10-d-old seedlings grown on 0.5× MS plates and qRT-PCR was used to examine the expression of the key transcription factor genes involved in the regulation of trichome formation. ACT2 was used as reference gene, and expression of each gene in the Ws wild-type seedlings was set as 1. Data represent the mean ± sd of three replicates. B, Expression of TRY and TCL1 in the Col wild type, the 35S:HA-NTL8 transgenic plant, and the ntl8 loss-of-function mutants. Total RNA was isolated from 10-d-old seedlings grown on 0.5× MS plates, and qRT-PCR was used to examine the expression of TRY and TCL1. ACT2 was used as reference gene, and expression of each gene in the Col wild-type seedlings was set as 1. Data represent the mean ± sd of three replicates. C, Expression of NTL8, TRY, and TCL1 in the internodes of the main inflorescence stem of Col wild-type plants. Total RNA was isolated from different internodes of soil-grown Col wild-type plants, and qRT-PCR was used to examine the expression of NTL8, TRY, and TCL1. ACT2 was used as reference gene, and expression of each gene in first internode was set as 1. Data represent the mean ± sd of three replicates.

Figure 5.

NTL8 functions upstream of TRY and TCL1 in regulating trichome formation. Photographs of first two rosette leaves were taken from 10-d-old soil-grown seedlings. Photographs of flowers and inflorescence stems were taken from 5-week-old soil-grown plants of the Col wild type, try, tcl1, and ntl8-1D single mutants, try tcl1, try ntl8-1D, and tcl1 ntl8-1D double mutants, and try tcl1 ntl8-1D triple mutant plants.

Figure 6.

TRY and TCL1 are direct target genes of NTL8 transcriptional activator. A, NTL8 is a transcriptional activator. Effectors and the Gal4:GUS reporter plasmids were cotransfected into protoplasts isolated from 3- to 4-week-old Col wild-type plants. The protoplasts were incubated in darkness at room temperature for 20 to 22 h, and then GUS activity was measured. Data represent the mean ± sd of three replicates. B, ChIP assay. The 35S:HA-NTL8 transgenic plants were used for ChIP assay with rabbit anti-HA antibodies. Rabbit preimmune serum was used as mock control. Primers spanning putative NAC transcription factor binding sites in the promoters of TRY and TCL1 were used for qRT-PCR analysis. ACT2 was used as reference gene and DFR as a negative control. Data represent the mean ± sd of three replicates. C, Mutation of the NAC transcription factor binding sites affects the binding of NTL8 to the promoters of TRY and TCL1. Plasmids of NTL8 and the wild-type or mutated TRY or TCL1 promoter:GUS were cotransfected into protoplasts isolated from 3- to 4-week-old Col wild-type plants. The protoplasts were incubated in darkness at room temperature for 20 to 22 h, and then GUS activity was measured. Cotransfection of CAT (CHLORAMPHENICOL ACETYLTRANSFERASE) was used as a negative control. Data represent the mean ± sd of three replicates.

Figure 7.

NTL8 may function independent of SPL9 in regulating trichome formation. A, NTL8 does not interact with SPL9 in yeast cells. Bait and prey plasmids were cotransformed into NMY51 yeast cells and grown in SD-Trp-Leu plate or SD-Trp-Leu-His-Ade plate containing 15 mm 3-AT for 2 to 4 d before photographs were taken. Cotransformation of empty vector pDHB1 and pPR3-N was used as negative control, and cotransformation of pNubG-Fe56 and pOst1-Nub1 as positive control. B, NTL8 does not interact with SPL9 in Arabidopsis protoplasts. Effectors and the Gal4:GUS reporter plasmids were cotransfected into protoplasts isolated from 3- to 4-week-old Col wild-type plants. The protoplasts were incubated in darkness at room temperature for 20 to 22 h, and then GUS activity was measured. Cotransfection of CAT was used as a negative control. Data represent the mean ± sd of three replicates. C, Trichomes on the rosettes and main inflorescence stems of the SPL9p:rSPL9 (left), the ntl8-1D (middle) and SPL9p:rSPL9 ntl8-1D plants (right). Photographs of rosettes were taken from 3-week-old and stems from the first internodes of 1-month-old soil-grown plants.