A CYP78As-small grain4-coat protein complex Ⅱ pathway promotes grain size in rice

Abstract

CYP78A, a cytochrome P450 subfamily that includes rice (Oryza sativa L.) BIG GRAIN2 (BG2, CYP78A13) and Arabidopsis thaliana KLUH (KLU, CYP78A5), generate an unknown mobile growth signal (referred to as a CYP78A-derived signal) that increases grain (seed) size. However, the mechanism by which the CYP78A pathway increases grain size remains elusive. Here, we characterized a rice small grain mutant, small grain4 (smg4), with smaller grains than its wild type due to restricted cell expansion and cell proliferation in spikelet hulls. SMG4 encodes a multidrug and toxic compound extrusion (MATE) transporter. Loss of function of SMG4 causes smaller grains while overexpressing SMG4 results in larger grains. SMG4 is mainly localized to endoplasmic reticulum (ER) exit sites (ERESs) and partially localized to the ER and Golgi. Biochemically, SMG4 interacts with coat protein complex Ⅱ (COPⅡ) components (Sar1, Sec23, and Sec24) and CYP78As (BG2, GRAIN LENGTH 3.2 [GL3.2], and BG2-LIKE 1 [BG2L1]). Genetically, SMG4 acts, at least in part, in a common pathway with Sar1 and CYP78As to regulate grain size. In summary, our findings reveal a CYP78As-SMG4-COPⅡ regulatory pathway for grain size in rice, thus providing new insights into the molecular and genetic regulatory mechanism of grain size.

Figure 1.

Phenotypes of the smg4 mutant. A) Plant architecture of wild-type (WT) and smg4 mutant plants at the mature stage. Scale bar, 10 cm. B) Mature panicles of WT and smg4. Scale bar, 5 cm. C) Mature rice grains of WT and smg4. Scale bar, 1 cm. D) Brown rice grains of WT and smg4. Scale bar, 1 cm. E) to (J) plant height (n = 20) (E), panicle length (n = 15) (F), grain length (n = 30) (G), grain width (n = 30) (H), grain thickness (n = 30) (I), and thousand-grain weight (n = 3) (J) in WT and smg4. Values are means ± SD. Student's t-test was used to generate the P-values, **P < 0.01.

Figure 2.

SMG4 promotes cell expansion and cell proliferation. A) Spikelet hulls of WT and smg4 just before a thesis. The boxes represent the observation sites of (B) and (C). The dotted lines indicate the sites of cross sections in (D) and (E). The double-headed arrows indicate the orientation for the cell number statistics in (H). Scale bars, 3 mm. B, C) Microscopy observation of WT (B) and smg4(C) inner epidermal cells. Scale bars, 100 µm. D, E) Cross sections of WT (D) and smg4(E) spikelet hulls. Scale bars, 500 µm. F, G) The 5× enlargement of the regions outlined by the box in (D) and (E), respectively. The arrows indicate the outer cell layers that were compared in (J) to (L). Scale bars, 100 µm. H, I) Longitudinal cell number (n = 25) (H) and cell length (n = 30) (I) of inner epidermal cells in WT and smg4. J–L) Total cell length (n ≥ 6) (J) cell number (n ≥ 6) (K) and cell length of each cell (n ≥ 6) (L) in the outer parenchyma cell layer in WT and smg4. M, N) Relative expression levels of cell expansion-related genes (M) and cell cycle-related genes (N) in spikelet hulls BH of WT and smg4 (n = 3). The UBIQUITIN gene was used as an internal control. Values are means ± SD. Student's t-test was used to calculate the P-values, **P < 0.01.

Figure 3.

Map-based cloning of SMG4.A) Fine mapping of the SMG4 locus. The molecular markers and numbers of recombinants are indicated above and below the filled bars, respectively. Chr., chromosome; Recs, recombinants. B) Gene structure of SMG4 (LOC_Os03g62270) and the mutation in smg4. The boxes and lines indicate exons and introns, respectively. ATG and TAG represent the start and stop codons, respectively. C) The genotypes and phenotypes co-segregate. D) Grain morphologies of WT (SMG4), smg4, and the complementation transgenic lines (gSMG4^Com #1 and gSMG4^Com #2). Scale bar, 5 mm. E, F) Grain length (n = 20) (E) and grain width (n = 20) (F) of WT, smg4, and the complementation transgenic lines (gSMG4^Com #1 and gSMG4^Com #2). G) Grain morphologies of Kitaake (Kit), SMG4 knockout lines (smg4-2 and smg4-3), and overexpression lines (SMG4-OE#2 and SMG4-OE#3). Scale bar, 5 mm. H) Identification of SMG4 knockout lines generated by the CRISPR/Cas9 system. The sgRNA-targeted site and protospacer adjacent motif (PAM) are indicated in different colored fonts, respectively. The dashed line represents deleted nucleotides. The added base is highlighted. I–K) Grain length (n = 20) (I), grain width (n = 20) (J), and grain thickness (n = 20) (K) of Kit, SMG4 knockout and overexpression transgenic lines. Values are means ± SD. Student's t-test was used to calculate the P-values, **P < 0.01. ns, no significance.

Figure 4.

Expression of SMG4 and topology analysis of SMG4. A) Relative SMG4 expression levels in roots (R), stems (S), leaves (L), leaf sheaths (LS), young panicles (numbers indicate the length of young panicles, in cm), spikelet hulls (BH; numbers indicate the days after heading), and caryopses (numbers indicate the days after heading) of 9,311. The UBIQUITIN gene was used as an internal control. Values are means ± SD (n = 3). B, C) SMG4 is an integral membrane protein. Total protein extract from rice protoplasts was ultracentrifuged at 100,000 × g for 1 h to obtain the pellet (P100) and supernatant (S100) fraction, followed by immunoblot analysis with anti-GFP and specific antibodies for the cytosol marker anti-UGPase and the tonoplast marker anti-γTIP (tonoplast intrinsic protein) (B). The P100 fraction was resuspended in various buffers as indicated. These suspensions were ultracentrifuged to obtain pellet (P) and supernatant (S), followed by immunoblot analysis with anti-GFP and anti-γTIP antibodies (C). TX100, Triton X-100. D) Protease digestion assay. The microsomal pellets containing SMG4 tagged with GFP at either the C or N terminus were digested with or without proteinase K in the presence or absence of detergent (Triton X-100) and then analyzed by immunoblot with anti-GFP. E) Proposed topology of SMG4.

Figure 5.

Subcellular localization of SMG4 in the leaf epidermal cells of N. benthamiana and rice root tip cells. A to E) Confocal microscopy images showing that SMG4-GFP localizes as puncta in the cytosol and these punctate signals partially colocalize with the marker proteins targeted to the ER (mCherry-HDEL) (A), and Golgi (GmMan1-mCherry) (C), strongly colocalize with ERES (AtSar1b-mCherry) (B), but show a distinct localization from that of marker proteins targeted to the TGN (mCherry-SYP61) (D), and PVC (mCherry-AtVSR2) (E). PSC coefficients (r_s) between SMG4-GFP and each marker is shown in the right panel. Values are means ± SD (n ≥ 4 images). Scale bars, 10 µm. F–I) Immunoelectron microscopy localization of SMG4-GFP in root tip cells of rice. G and I) are the magnified images of the boxed areas in images (F) and (H), respectively. Gold particles are highlighted with arrows. G, Golgi. ER, endoplasmic reticulum. Scale bars, 200 nm.

Figure 6.

SMG4 interacts with COPⅡ components to regulate grain size. A) Confocal imaging of the roots from 5-d-old SMG4-GFP transgenic seedlings treated with or without 100 μM H89 for at least 12 h. Scale bars, 10 µm. B) Quantification of the number of SMG4-GFP puncta in (A) (n > 20 cells). C) In vivo Co-IP assay showing that SMG4 interacts with Sar1a, Sar1b, Sar1c, Sec23a, Sec23b, Sec23c, Sec24a, Sec24b, and Sec24c in rice protoplasts. The symbols “+” and “–” represent the presence and absence of the corresponding proteins. D) Grain morphologies of Kit, smg4-2, and Sar1 (Sar1a + Sar1b + Sar1c) RNAi lines in the Kit and smg4-2 backgrounds. Scale bar, 5 mm. E) Grain length (n = 30) and grain width (n = 30) of kit, smg4-2, and Sar1 (Sar1a + Sar1b + Sar1c) RNAi lines in the Kit and smg4-2 backgrounds. Values are means ± SD. Student's t-test was used to calculate the P-values, **P < 0.01.

Figure 7.

SMG4 acts genetically with BG2 to regulate grain size. A) Yeast 2-hybrid assay showing the interaction between SMG4 and BG2. DDO, synthetic defined medium lacking Trp and Leu (SD/–Trp–Leu); QDO, SD medium lacking Trp, Leu, His, and Ade (SD/–Trp/–Leu–His/–Ade). B) Firefly LCI assay showing the interaction between SMG4 and BG2 in N. benthamiana leaf cells. CL, C terminus of LUC; NL, N terminus of LUC. The scale bar indicates the luminescence intensity in counts per second (CPS). C) BiFC assay showing that SMG4 interacts with BG2 in N. benthamiana leaf cells. Sec12b was used as a negative control. Scale bars, 10 µm. D) In vivo Co-IP assay showing that SMG4 interacts with BG2 in rice protoplasts. The symbols “+” and “–” represent the presence and absence of the corresponding proteins. E) Grain morphologies of Kit, smg4-2, and BG2 overexpression lines in the Kitaake and smg4-2 backgrounds. Scale bar, 5 mm. F) Relative BG2 transcript levels in spikelet hulls of Kit, smg4-2, and BG2 overexpression lines in the kit and smg4-2 backgrounds (n = 3). The UBIQUITIN gene was used as an internal control. Values are means ± SD. Student's t-test was used to calculate the P-values, **P < 0.01. (G–I) Grain length (n = 30) (G), grain width (n = 30) (H), and grain thickness (n = 30) (I) of kit, smg4-2, and BG2 overexpression lines in the Kit and smg4-2 backgrounds. Values are means ± SD. Different letters indicate significant differences ranked by pairwise multiple comparison followed by Tukey's test (P < 0.05).

Figure 8.

SMG4 acts in a common pathway with CYP78As to regulate grain size. A and B) Firefly LCI assays shows that SMG4 interacts with GL3.2 (A) and BG2L1 (B) in N. benthamiana leaves cells. CL, C terminus of LUC; NL, N terminus of LUC. The scale bar indicates the luminescence intensity in CPS. C) In vivo Co-IP assay showing that SMG4 interacts with GL3.2 and BG2L1 in rice protoplasts. The symbols “+” and “–” represent the presence and absence of the corresponding proteins. D) Grain morphologies of CYP78A (BG2 + GL3.2 + BG2L1) RNAi lines in the kit and SMG4-OE#3 backgrounds. Scale bar, 5 mm. E–G) Grain length (n = 30) (D), grain width (n = 30) (E), and grain thickness (n = 30) (F) of kit and CYP78A (BG2 + GL3.2 + BG2L1) RNAi lines in the Kit and SMG4-OE#3 backgrounds. Values are means ± SD. Different letters indicate significant differences ranked by pairwise multiple comparison followed by Tukey's test (P < 0.05).

Figure 9.

BIGE1A interacts with KLUH in Arabidopsis. A) Subcellular localization of the KLUH-GFP and BIGE1A-GFP fusion proteins in the leaf epidermal cells of N. benthamiana. mCherry-HDEL and AtSar1b-mCherry were used as the ER marker and ERES marker, respectively. Scale bars, 10 µm. B) Firefly LCI assay showing that BIGE1A interacts with KLUH in N. benthamiana leaf cells. CL, C terminus of LUC; NL, N terminus of LUC. The scale bar indicates the luminescence intensity in CPS. C) BiFC assay showing the interaction between BIGE1A and KLUH in N. benthamiana leaf cells. mCherry-HDEL was used as the ER marker. Scale bars, 10 µm. D) In vivo Co-IP assay showing that BIGE1A interacts with KLUH in rice protoplasts. The symbols “+” and “–” represent the presence and absence of the corresponding proteins. Arrows indicate the corresponding protein bands.

Figure 10.

A proposed working model for SMG4's role in regulating rice grain size. A) CYP78As may catalyze and generate a growth signal (CYP78A-derived signal) in the ER, and SMG4 interacts with CYP78As to receive the CYP78A-derived signal. Then, SMG4 interacts with COPⅡ components to transmit the CYP78A-derived signal from the ER to Golgi. B) In WT, CYP78A-derived signals are transported from the ER to Golgi normally to promote cell expansion and cell proliferation in spikelet hulls, thus leading to normal grains. In smg4, the transport of CYP78A-derived signal from the ER to Golgi is disrupted, thus restricting cell expansion and cell proliferation in spikelet hulls and finally leading to small grains.