Tomato POLLEN DEFICIENT 2 encodes a G-type lectin receptor kinase required for viable pollen grain formation

Abstract

Pollen development is a crucial biological process indispensable for seed set in flowering plants and for successful crop breeding. However, little is known about the molecular mechanisms regulating pollen development in crop species. This study reports a novel male-sterile tomato mutant, pollen deficient 2 (pod2), characterized by the production of non-viable pollen grains and resulting in the development of small parthenocarpic fruits. A combined strategy of mapping-by-sequencing and RNA interference-mediated gene silencing was used to prove that the pod2 phenotype is caused by the loss of Solanum lycopersicum G-type lectin receptor kinase II.9 (SlG-LecRK-II.9) activity. In situ hybridization of floral buds showed that POD2/SlG-LecRK-II.9 is specifically expressed in tapetal cells and microspores at the late tetrad stage. Accordingly, abnormalities in meiosis and tapetum programmed cell death in pod2 occurred during microsporogenesis, resulting in the formation of four dysfunctional microspores leading to an aberrant microgametogenesis process. RNA-seq analyses supported the existence of alterations at the final stage of microsporogenesis, since we found tomato deregulated genes whose counterparts in Arabidopsis are essential for the normal progression of male meiosis and cytokinesis. Collectively, our results revealed the essential role of POD2/SlG-LecRK-II.9 in regulating tomato pollen development.

Keywords: Male sterility; RNA interference; mapping-by-sequencing; microgametogenesis; microsporogenesis; parthenocarpic fruits; pollen formation; receptor kinases; tapetum development; tomato.

Fig. 1.

Phenotypic characterization of the pod2 mutant. (A, B) Side view (A) and cross section (B) of wild-type and pod2 fruits. (C–H) Comparison of yield- and fruit-related traits in wild-type and pod2 plants: fruit set (C), total fruit yield (D), fruit weight (E), fruit diameter (F), fruit length (G), and seeds per fruit (H). Data are presented as means ±SD. Asterisks indicate significant differences between pod2 and wild-type plants (*P≤0.001; Student’s t-test). (I, J) Hand-pollination of wild-type and pod2 stigmas with pod2 and wild-type pollen, respectively. (K, L) Pollen of wild-type (K) and pod2 (L) plants stained with TTC. Scale bars=1 cm (A, B), 100 µm (I, J), and 50 µm (K, L).

Fig. 2.

Histological characterization of anther development in wild-type and pod2 flowers. Cross sections of anthers at pre-meiotic (A, D), meiotic (B, E), tetrad (C, F), microspore (G, J), mitotic (H, K), and dehiscence (I, L) stages. Aw, anther wall; M, microspores; MMC, microspore mother cells; P, pollen grains; T, tapetum; Te, tetrads. Red arrows indicate tapetum tissue. Scale bars=50 μm (main images) and 25 μm (inset images).

Fig. 3.

Subcellular features of selected stages of pollen development in wild-type and pod2 anthers. Cross sections of anthers at the microspore (A, D, G–L), young pollen grain (B), and medium pollen grain (C, E, F) stages. Ap, aperture; Aw, anther wall; Ec, ectexine; En, endexine; EO, exine oncus; Ex, exine; GC, generative cell; In, intine; IO, intine oncus; La, lamella; M, microspores; N, nucleus; Nu, nucleolus; T, tapetum; V, vacuole, VC, vegetative cell cytoplasm; VN, vegetative nucleus; Asterisks, orbicules. Scale bars=5 µm (A, B, G–I), 2 µm (C), and 1 µm (D–F, J–L).

Fig. 4.

Molecular characterization of the pod2 mutant. (A) Average allele frequency distribution per chromosome of wild-type (WT; blue line) and pod2 (red line) pools. (B) Structure and position of the mutation that affects the POD2 gene in pod2 mutant plants. The coding sequence is depicted as a grey box. White boxes represent the 5ʹ and 3ʹ untranslated regions. The predicted translation start (ATG) and stop (TAG) codons are indicated. The red arrow indicates the position of the pod2 point mutation. (C) Relative expression of the POD2 gene in cv. Moneymaker plants at 10 different stages of plant development. FB0, floral buds of 0–2 mm; FB1, floral buds of 3–4 mm; FB2; floral buds of 4–5 mm; PA, flowers at pre-anthesis stage; A; flowers at anthesis stage; MG; fruits at mature green stage, BR; fruits at breaker stage; M, fruits at mature stage. (D) In situ hybridization analysis of POD2 expression in wild-type entire anthers from meiotic to mitotic developmental stages. (E) In situ hybridization analysis of POD2 expression in wild-type entire anthers at the late tetrad stage. Transverse sections were hybridized with antisense or sense digoxygenin-labeled probe of the POD2 gene. Aw, anther wall; Ca, callose; M, microspores; T, tapetum; Te, tetrads. Black arrows indicate microspores at the late tetrad stage. Scale bars=200 µm (D) and 50 µm (E).

Fig. 5.

POD2 gene silencing by RNAi. (A) Position of the sequence targeted by POD2:RNAi (indicated by the red box). The coding sequence is depicted as a grey box. White boxes represent the 5ʹ and 3ʹ untranslated regions. The predicted translation start (ATG) and stop (TAG) codons are indicated. (B) Quantitative RT–PCR analysis of POD2 expression in wild-type cv. Moneymaker (MM) plants and 11 independent POD2:RNAi lines. Asterisks indicate significant differences between MM and RNAi plants (*P≤0.001; Student’s t-test). (C, D) Pollen of wild-type (C) and POD2:RNAi (D) plants stained with TTC. (E) Cross section of wild-type and POD2:RNAi tomato fruits. Scale bars=50 µm (C, D) and 1 cm (E).

Fig. 6.

Differentially expressed genes in pod2 plants. (A) Venn diagrams showing the overlap between differentially expressed genes across three pollen developmental stages. (B) Heatmap of the expression levels of the 10 differentially expressed genes between the wild type (WT) and pod2 (Mut) in all developmental stages evaluated (Te, tetrad; Mc, microspore; YP, young pollen). The heatmap color scale ranges from blue (low expression) to red (high expression).

Fig. 7.

Significantly enriched GO terms in the Biological Process category for differentially expressed genes across developmental stages (Te, tetrad; Mc, microspore; YP, young pollen). The size of the circles indicates the number of genes in the given GO term. Enriched GO terms were determined by FDR adjusted P-value (adj.p) ≤0.05. FDR is represented as the –log (adj.p), where the darkest blue are the most significant GO terms. The full list of significantly enriched GO terms is available in Supplementary Dataset S2.