Fruit indehiscence caused by enhanced expression of NO TRANSMITTING TRACT in Arabidopsis thaliana

Abstract

In flowering plants, fruit dehiscence enables seed dispersal. Here we report that ntt-3D, an activation tagged allele of NO TRANSMITTING TRACT (NTT), caused a failure of fruit dehiscence in Arabidopsis. We identified ntt-3D, in which the 35S enhancer was inserted adjacent to AT3G-57670, from our activation tagged mutant library. ntt-3D mutants showed serrated leaves, short siliques, and indehiscence phenotypes. NTT-overexpressing plants largely phenocopied the ntt-3D plants. As the proximate cause of the indehiscence, ntt-3D plants exhibited a near absence of valve margin and lignified endocarp b layer in the carpel. In addition, the replum was enlarged in ntt-3D mutants. NTT expression reached a peak in flowers at stage 11 and gradually decreased thereafter and pNTT::GUS expression was mainly observed in the replum, indicating a potential role in fruit patterning. NTT:GFP localized in the nucleus and cytoplasm. FRUITFULL (FUL) expression was downregulated in ntt-3D mutants and ntt-3D suppressed upregulation of FUL in replumless mutants. These results indicate that NTT suppresses FUL, indicating a potential role in patterning of the silique. In seed crops, a reduction in pod dehiscence can increase yield by decreasing seed dispersal; therefore, our results may prove useful as a basis to improve crop yield.

Fig. 1

Phenotypes and characterization of the ntt-3D. mutants under long-day (LD) conditions. (A) Overall morphology of 25-day-old wild-type (WT) (left) and ntt-3D. plants (right). (B) Leaf morphology of wild-type plants (Top) and ntt-3D plants (Bottom). (C) Inflorescence morphology of wild-type (left) and ntt-3D plants (right). (D) Silique phenotypes of wild-type (left) and ntt-3D plants (right). (E) Indehiscent silique phenotype of ntt-3D plants (Bottom). (F) A T-DNA insertion site in ntt-3D (S6-003) plants. (G) Upregulation of NTT (AT3G57670) in ntt-3D whole seedlings grown for 10 days under LD conditions. UBQ10 was used as an internal control. (H-I) Rosettes (H), bolted plants (I), and siliques (J) of wild-type, ntt-3D, and 35S::NTT plants.

Fig. 2

Expression of NTT in wild-type plants. (A) Spatial expression pattern of NTT. UBQ10 was used as a control. RL, rosette leaves; CL, cauline leaves; S, stems; OF, old flowers at stage 15 (Smyth et al., 1990); YF, young flowers at stages 13 and 14; FB, floral buds; R, roots. Floral stages were previously defined (Smyth et al., 1990). (B) pNTT:: GUS expression in whole seedlings. Magnified views of shoot apex (Bottom right) and root apex (Bottom left) are shown. (C) Time-course expression of NTT during flower developmental stages revealed via RT-qPCR. Floral stages were previously defined (Smyth et al., 1990). (D) pNTT:: GUS expression in the carpel of flowers at stage 12 (Smyth et al., 1990) (Top) and at stage 14 (Bottom). (E) Subcellular localization of NTT::GFP protein in Arabidopsis epidermal cells.

Fig. 3

Defects in endocarp b layer, valve margin, and replum seen in the carpel of ntt-3D mutants. (A, B) Scanning electron microscopy (SEM) images of the carpel in a flower at stage 12 (Smyth et al., 1990) (A) and at stage 18 (B) of wild-type (WT) and ntt-3D plants. Dotted line: putative dehiscence zone. (C) Toluidine Blue O-stained sections of the silique at stage 17 (Smyth et al., 1990) of wild-type (left) and ntt-3D (right) plants. Note that the valve margin and the endocarp b layer was almost absent in ntt-3D plants (arrowhead). (D) Phloroglucinol-stained sections of the silique of a flower at stage 17 (Smyth et al., 1990) of wild-type, ntt-3D, and ntt-1 plants. Arrow: lignified cells stained with phloroglucinol. Note the near absence of lignified cells in ntt-3D fruits (arrowhead). enb, endocarp b layer; DZ, dehiscence zone; LL, lignified layer; R, replum; SL, separation layer; V, valve; VM, valve margin.

Fig. 4

Downregulation of FUL in ntt-3D mutants. (A) Expression levels of FUL, RPL, and BP in the carpel of flowers at stage 12 (Smyth et al., 1990) in ntt-3D mutants. (B) Expression levels of FUL in rpl-2 and rpl-2 ntt-3D double mutants. Each gene expression level in wild type plants was arbitrarily set to 1.0. Statistical analysis was done with an unpaired Student’s t-test. The results are considered statistically significant (p < 0.05).

Fig. 5

A proposed model of NTT (a replum factor) in regulating FUL (a valve factor) in carpel development. Arrows represent promotion effects, whereas T-bars indicate repression effects. A lignified layer (LL) is shown in grey. enb, endocarp b layer; SL, separation layer.