Overexpression of Loose Plant Architecture 1 increases planting density and resistance to sheath blight disease via activation of PIN-FORMED 1a in rice.

Increasing yield and resistance to pathogens are important objectives in plant breeding. However, difficulties in breeding are encountered due to the antagonistic relationship between crop yield production and immunity pathways (Ning et al., 2017). Planting density and plant architecture are key factors determining crop yields in a given area. The plant traits of tiller and lamina joint angles have long attracted the attention of breeders due to their significant contributions to plant architecture by enhancing photosynthetic efficiency and facilitating enhanced planting density (Sakamoto et al., 2006; Wang and Li, 2008). Earlier reports have shown that Loose Plant Architecture 1 (LPA1), used for encoding an indeterminate domain (IDD) protein, negatively controls tiller and lamina joint angle in an expression level‐dependent manner (Liu et al., 2016; Wu et al., 2013). Overexpression of LPA1 significantly decreased tiller angle and resulted in the development of erect leaves and the generation of severe lines at an angle of ~1/5 that of the tiller angle, relative to the wild‐type (WT) plants. Also, LAP1 mRNA was highly accumulated in overexpressor lines and the levels were negatively associated with tiller angle (Figure 1a–c). Further inspection demonstrated that severe lines resulted in decreased tiller number and thousand grain weight, whereas the lines with moderate expressions sustained similar tiller numbers and thousand grain weight relative to the WT (Figure 1d,e), implying that moderate expression of LPA1 increases planting density without impacting tiller number and seed weight.

Figure 1

triggers to regulate tiller angle and resistance to sheath blight disease (SBD). (a) 2‐month‐old wild‐type (WT) and overexpressors (OX; 2, 5, 6, 7 and 8) were aligned according to the degree of tiller angles. (b) expression levels in WT and LPA1 overexpressors were analysed by northern blot analysis. EtBr staining of rRNA was used as a loading control. Tiller angles (c) and number (d) from the lines shown in (a) are shown. Data indicate average ± standard deviation (SD) (n > 10). (e) Thousand grain weight from the WT and lines were measured. Data indicate average ±SD (n = 6). Leaves (f) and sheath (g) from the WT and lines (OX5 and OX6) were inoculated with Rhizoctonia solani AG1‐1A and were photographed after infection. Six leaves from each line were examined. Each experiment was performed in triplicate. (h) The lesion areas on the leaf or sheath surfaces of R. solani AG1‐1A‐infected tissues were examined. Data indicate average ±standard error (SE) (n > 10). (i) and (j) expression levels in the WT and lines (5 and 6) after 0, 24, 48 and 72 hours of R. solani AG1‐1A inoculation using Quantitative Reverse Transcription Polymerase Chain Reaction (qRT‐PCR). The experiments were performed in triplicate. (k) The expression levels of were monitored in the WT and lines (2, 5, 6, 7 and 8) using qRT‐PCR. The experiments were performed in triplicate. (l) Schematic diagram indicating location of the putative IDD‐binding motif (red circle) within 1.5 kb of promoter and probes (P) used for chromatin immunoprecipitation (ChIP) assays. Relative ratios of immunoprecipitated DNA to input DNA were determined by qPCR. Input DNA was used to normalize the data. −Ab or +Ab: green fluorescent protein (GFP) antibody. Error bars represent ±SE (n = 3). (m) An electrophoretic mobility‐shift assay (EMSA) was conducted to evaluate affinities to P2 and mutated probe mP2. The probe was labelled with biotin and the band shifting was detected via western blot analysis using anti‐glutathione‐S‐transferase (GST) antibody. (n) A transient expression assay was conducted by co‐transfection with p35S: and each of the vectors expressing the beta‐glucuronidase gene (GUS) under the control of native () and IDD‐binding motif‐mutated (mp) promoters in protoplast cells. The luciferase gene driven by the 35S promoter was used as an internal control to normalize GUS expression. Error bars represent ± SE (n = 6). (o) PIN1a expression level in WT, lines (2 and 4) and lines (2 and 3) was examined using qRT‐PCR. The experiments were performed in triplicate. Leaves (p) and sheath (q) from the WT, lines (2 and 4) and lines (2 and 3) were inoculated with R. solani AG1‐1A and were photographed after infection. The leaves and sheath from each line were examined, and the experiments were performed in triplicate. (r) The lesion areas on the leaf surfaces were examined for R. solani AG1‐1A‐infected leaves and sheath. Data indicate average ± SE (n > 10). (s) and (t) expression levels in the WT, lines (2 and 4) and lines (2 and 3) after 0 and 48 hours of R. solani AG1‐1A inoculation using qRT‐PCR. The experiments were performed in triplicate. (u) and (Ri2) double‐mutant leaves and sheath, respectively, were inoculated with R. solani AG1‐1A (left and middle) and 2‐month‐old and plants were photographed (right). (v) The lesion area on the leaf and sheath surface of WT,, and plants was measured for R. solani AG1‐1A‐infected tissues. Data indicate average ± SE (n > 10). (w) Tiller angle of genetic combination between and were analysed. More than 10 plants from segregated WT,, and plants were used for measurement. Data indicate averages ±SE. (x) Leaves from 2‐month‐old WT plants with or without 100 nM IAA treatment for 3 days, were inoculated with R. solani AG1‐1A. (y) The lesion areas on the leaf surfaces of R. solani AG1‐1A‐infected leaves shown in (p) were examined. Data indicate averages ±SE (n > 10). (z) IAA content from the leaves of 2‐month‐old WT and lines (OX5 and OX6) were measured. Vertical bars indicate average values ±SE (n = 3). One‐way analysis of variance (ANOVA) followed by Bonferroni's multiple comparison tests were performed to assess significant differences between more than two groups. Different letters above the bars denote statistically significant differences (P < 0.05).

To determine whether these lines not only increase planting density but also pathogen resistance, Rhizoctonia solani AG1‐1A, which is the cause of sheath blight disease (SBD), one of the major rice diseases, was inoculated to the leaves of the WT and LPA1 overexpressors (OX5 and OX6, the tiller number and thousand grain weight of which were not impacted). SBD imperils rice throughout its growth cycle, from seedling to heading, and causes lesions on leaves, sheaths, and panicles that can decrease rice yield by 8%–50%, depending on disease severity (Savary et al., 1995, 2000). However, resistant cultivars and gene sources against SBD are currently lacking. Interestingly, LPA1 overexpressors are less vulnerable to R. solani AG‐1 than WT plants (Figure 1f,g). 46% of the leaf area was covered with lesions in the WT, 30% in LPA1 OX5, and 29% in LPA1 OX6. While 54% of the sheath area was covered with lesions in the WT, 37.8% in LPA1 OX5 and 38.6% in LPA1 OX6 (Figure 1h), implying that LPA1 overexpression enhanced plant resistance to SBD. Further examination indicated that expression of PBZ1 and PR1b, two pathogen resistant genes were more highly induced in LPA1 OXs than in WT after inoculation of R. solani AG1‐1 (Figure 1i,j). Our earlier work demonstrated that hormonal signals play key roles in rice resistance to SBD (Yuan et al., 2018). In addition, LPA1 positively controls the expressions of the auxin efflux carrier gene PIN‐FORMED 1a (PIN1a) in the lamina joint (Liu et al., 2016). Phenotypically, PIN1a knock‐down plants exhibited increased tiller angle whereas PIN1a overexpression lines slightly decreased tiller angle relative to that of the WT (Xu et al., 2005), similar to differences in comparisons between LPA1 mutants and overexpressors for plant shape (Wu et al., 2013). Quantitative Reverse Transcription Polymerase Chain Reaction (qRT‐PCR) analysis identified that LPA1 overexpression up‐regulated PIN1a expressions in leaves (Figure 1k). As PIN1a was positively regulated by LPA1 and IDD proteins are known to function as a transcription factor (Kozaki et al., 2004), the PIN1a promoter sequences were examined to identify the presence of putative IDD‐binding motif. A single IDD‐binding motif was located within 1.5 kb of the PIN1a promoter (Figure 1l). To determine the binding affinity of LPA1 to the IDD‐binding motif, a chromatin immunoprecipitation (ChIP) assay was conducted using 35S: green fluorescent protein (GFP) and 35S:LPA1:GFP transgenic plant calli. Without addition of GFP, the antibody was used as the control for the GFP antibody to immunoprecipitate DNA. Data showed that LPA1 bound to the P2, but not to the P1 region (Figure 1l). We further conducted an electrophoretic mobility shift assay (EMSA) to verify binding affinity of P2 to LPA1. We found that LPA1 bound to biotin‐labelled P2; however, it failed to bind to the mutated probe mP2 that were detected by western blot analysis using GST antibody (Figure 1m). To verify whether these cis‐elements were responsible for the transcriptional initiation of PIN1a promoter by LPA1, we conducted transient expression assays using the protoplast system. Protoplast cells were co‐transformed with the 35S:LPA1 plasmid and a vector expressing the beta‐glucuronidase gene (GUS) under the control of pPIN1a or mpPIN1a. In the mutated promoter (mpPIN1a), IDD‐binding motif sequences TTTGTCG were substituted by the sequence AAAAAAA. Using 35S:luciferase (LUC) as an internal control to normalize the transformation efficiency in each assay, protoplasts expressing LPA1 had approximately twice the levels of activated pPIN1a; however, LPA1 was unable to activate mpPIN1a (Figure 1m). These results show that LPA1 directly triggers PIN1a via promoter binding.

Since PIN1a is a target of LPA1, the role of PIN1a in resistance to SBD was examined. PIN1a RNAi lines and overexpression plants were used to evaluate the response of PIN1a to R. solani AG1‐1A. qRT‐PCR results showed that PIN1a level was obviously lower in PIN1a RNAi lines (Ri2 and Ri4) and higher in PIN1a overexpression lines (OX2 and OX3) compared to wild‐type one (Figure 1o). Furthermore, the results demonstrated that PIN1a RNAi lines (Ri2 and Ri4) were more vulnerable, whereas overexpression lines (OX2 and OX3) were less vulnerable to R. solani AG1‐1A (Figure 1p,q). 47% of the leaf area was covered with lesions in the WT, 58% in PIN1a Ri2, 59% in PIN1a Ri4, 30% in PIN1a OX2, and 29% in PIN1a OX3. While 58% of the sheath area was covered with lesions in the WT, 64% in PIN1a Ri2, 65% in PIN1a Ri4, 40% in PIN1a OX2, and 41% in PIN1a OX3 (Figure 1r), implying that PIN1a positively controls rice resistance to SBD, similar to the degree of LPA1 regulation on SBD resistance. In addition, PBZ1 and PR1b expression levels were less induced in PIN1a RNAi lines while more highly induced in PIN1a OXs than in WT after inoculation of R. solani AG1‐1 (Figure 1s,t). Next, we investigated whether LPA1 controls planting density and resistance to SBD via initiation of PIN1a by genetic combination between LPA1 OX6 and PIN1a Ri2. Analysis of the inoculation of R. solani with AG1‐1A showed that LPA1 OX6 is less vulnerable, whereas PIN1a Ri2 is more vulnerable to SBD. Furthermore, PIN1a Ri2 enhanced LAP1 OX6 vulnerability to SBD, and PIN1a Ri2 and LPA1 OX6/PIN1a Ri2 exhibited a similar degree of vulnerability response to R. solani AG1‐1A. 47% of the leaf area was covered with lesions in the WT, 30% in LPA1 OX6, 58% in PIN1a Ri2, and 56% in LPA1 OX6/PIN1a Ri2. While 48% of the leaf area was covered with lesions in the WT, 40% in LPA1 OX6, 59% in PIN1a Ri2, and 57% in LPA1 OX6/PIN1a R i2 (Figure 1u,v). In parallel, tiller angle was compared between the WT, LPA1 OX6, PIN1a Ri2, and LPA1 OX6/PIN1a Ri2 from the same siblings. The LPA1 OX6 decreased tiller angle whereas PIN1a Ri2 enlarged tiller angle relative to that of the WT. LPA1 OX6/PIN1a Ri2 exhibited enhanced tiller angle relative to LPA1 OX6 and the WT; however, the degree of increase was less than that resulting from PIN1a Ri2 (Figure 1u,w). These data suggest that LPA1 positively controls resistance to SBD via initiation of PIN1a. In addition, the regulation of the tiller angle by LPA1 may partially regulate resistance to SBD via PIN1a‐dependent auxin transport. We further examined the effect of auxin on resistance to SBD. 3‐Indole acetic acid (IAA), a natural form of auxin, was exogenously applied to plants, after which R. solani AG1‐1A was inoculated. The data showed that IAA treatment enhanced rice resistance to SBD (Figure 1x,y). Next, the endogenous IAA levels of the WT, LPA1 OX5 and LPA1 OX6 plants were measured. The data demonstrated that LPA1 overexpressors contain higher levels of IAA than that of WT plant leaves (Figure 1z), implying that LPA1 might activate PIN1a to accumulate more IAA.

Overall, our analyses identified that LPA1 overexpression enhanced planting density by decreasing tiller and lamina joint angles. However, strong lines decreased tiller number and seed weight. The LPA1 overexpression lines with moderate expressions not only sustained normal tiller angle, but also increased resistance to SBD, a major disease affecting rice cultivation. The molecular and biochemical data showed that LPA1 triggers PIN1a via promoter binding. Interestingly, PIN1a controls tiller angle and resistance to SBD, and genetic combination between LPA1 overexpressor and PIN1a knock‐down mutants revealed that mediation of planting density and resistance to SBD through overexpression of LPA1 requires PIN1a. Exogenous auxin treatment enhanced rice resistance to SBD and resulted in LPA1 overexpressors accumulating higher IAA than that of the WT. Rhizoctonia solani AG1‐1A‐mediated induction of Pathogen resistant genes PBZ1 and PR1b levels were higher in LPA1 OX and PIN1a OX while lower in PIN1a RNAi lines than in wild‐type one, implying that LPA1 might control auxin transport via initiation of PIN1a to increase planting density and activate plant defense gene expressions.