Fig 1.
Phylogenetic analysis of LRE/LLG in the Fabaceae family and modeling of the LLG1 and LLG2 proteins.
(A) Alignment of the amino acid sequence of LRE/LLG proteins from 7 members of Fabaceae family. The 8 conserved cysteines are indicated with purple triangles, the conserved Asn-Asp dipeptide (ND) is indicated with blue tringles, the 13 amino acids involved in binding with RALF23 are indicated with navy blue asterisks, the conserved motif KEGKEGLE/D for binding with RALF23 is indicated with a green line, the omega site for PvLLG1 and PVLLG2 is indicated with a dark red square, and the hydrophobic tail is indicated with the blue box. (B) Maximum Likelihood tree constructed using IQTree and iTOL software with nodes showing percent of 1000 boot strap replicates. (C) Surface models for LLG1 and LLG2 from P. vulgaris and A. thaliana with the exposed amino acids involved in RALF23 binding indicated in red, the exposed amino acids involved in complementary binding with RALF23 indicated in gray, and the exposed amino acids in common between the two binding motifs binding with RALF23 in purple.
Fig 2.
PvLLG1 and PvLLG2 expression in P. vulgaris.
(A) Heatmap of expression based on data at the Phytozome v.13.0. (B) Heatmap of expression reported by O’Rourke et al. (2014) and Nanjareddy et al. (2017). (C) mRNA levels in P. vulgaris tissues determined by RT-qPCR. (D) PvLLG1 and PvLLG2 transcript accumulation under R. giardini inoculation. Quantification by RT-qPCR is expressed as relative expression (2−ΔCt) calculated with normalization to the P. vulgaris housekeeping gene, PvEf1α. For each sample, three biological replicates were analyzed with three technical replicates each. Different letters indicate significant differences among samples according to the Two -Way ANOVA analysis at p <0.0001.
Fig 3.
PvLLG1 transcript accumulation during R. tropici interaction.
(A) PvLLG1, (B) PvLLG2, (C) PvEnod40, and (D) PvLgHb2 transcript accumulation in P. vulgaris roots inoculated (green, purple, and orange bars) and uninoculated (brown, violet and beige bars), at 3, 5-, 7-, 14-, 18-, 21-, 25- and 30- days post-germination. Quantification by RT-qPCR is normalized to PvEf1α. Bars represent ± SEM (Standard Error of Mean) of 2 biological replicates with 3 technical replicates each. Different letters indicate significant differences among samples according to the Two -Way ANOVA analysis at p <0.0001.
Fig 4.
Analysis of PvLLG1 promoter and expression in P. vulgaris roots and root hairs.
(A) PvLLG1 cis-regulatory elements in a promotor region of 2200 bp with ATATT (ROOTMOTIFTAPOX1) indicated with light blue boxes, CTCTT (OSE2ROOTNOULE) purple box, KCACGW (RHERPATEXPA7) violet blue boxes, GNATATNC (P1BS) magenta boxes, TGAC (WRKY71OS) pink boxes, ACGT (CATATGGMSAUR) green boxes, ACTTTA (NTBBF1ARROLB) ocher boxes, MACGYGB (ABRERATCAL) beige boxes. (B-N) pPvLLG1:GFP-GUS expression in non-inoculated P. vulgaris roots and nodules. (B) GUS in main and lateral roots, (C, D and E) GUS and (F, G, and H) GFP at different stages of a lateral root development, (I and K) GUS and (L) GFP at the root tip in the lateral root cap, (J) GUS and (M) GFP in root hair cells, (N) GUS in non-inoculated roots. (O-R) pPvLLG1:GFP-GUS expression in R. tropici-inoculated P. vulgaris roots. (O) GUS in inoculated roots at 3 dpi, (P) GUS in root nodules at 21 dpi, (Q) GUS in root hairs with infection threads at 4 dpi, and (R) cross sections of nodules at 5, 7, 14 and 21 dpi.
Fig 5.
Subcelular localization of PvLLG1 under control and rhizobia inoculated roots.
A-C) Subcelular localization of LLG1-Neon under its own promoter (pLLG1::LLG1-Neon). Note the plasma membrane localization in root hairs and cortical cells (see arrows). D) Subcellular localization of cytoplasmic GFP used as a control (see arrow). E-G) Subcellular localization during the infection thread formation (arrow). H-J) Subcellular localization during early cortical cell divisions induced by rhizobia inoculation (arrow). K-M) Subcellular localization in well-developed nodules (see arrow). N-P) Subcellular localization of cytoplasmic GFP in nodules used as a control.
Fig 6.
Effects of silencing PvLLG1 on P. vulgaris in non-inoculated or R. tropici-inoculated roots.
(A) Transgenic hairy roots showing expression of red fluorescent protein from nucleotide-scrambled sequence control (SAC) and RNAi:PvLLG1 constructs. (B) PvLLG1 transcript accumulation in P. vulgaris roots expressing SAC or RNAi:PvLLG1 constructs under non-inoculated and R. tropici-inoculated condition. (C) Root hair length in R. tropici-inoculated roots with SAC or RNAi:PvLLG1 constructs. (D) Nodule number at 7, 14, and 21 dpi in R. tropici-inoculated roots with SAC or RNAi:PvLLG1 constructs. (E) Nodule diameter at 7, 14 and 21 dpi in R. tropici-inoculated roots with SAC or RNAi:PvLLG1 constructs. (F) Number of nodules per composite plant with different ranges of diameters at 7, 14 and 21 dpi in R. tropici-inoculated roots with SAC or RNAi:PvLLG1 constructs. (G) Images of root nodules from hairy roots expressing the SAC or RNAi:PvLLG1 constructs at 7, 14 and 21 dpi in R. tropici-inoculated roots. Replications for PvLLG1 transcript accumulation were 15, root hair length were 150, nodule number were 5, and nodule diameter were 150. Different letters or asterisk indicate significant differences among samples according to the Two -Way ANOVA analysis (*p = 0.05, ** p = 0.01, *** p = 0.001) or a pair-wise comparison with a Student´s T- test (p <0.0001), respectively.
Fig 7.
Effect of overexpression of PvLLG1 in P. vulgaris for non-inoculated or R. tropici-inoculated roots.
(A) Transgenic hairy roots show the expression of green fluorescent protein from the 35S::GFP (control) and 35S:PvLLG1-GFP (overexpression) constructs. (B) PvLLG1 transcript accumulation in non-inoculated and R. tropici -inoculated P. vulgaris roots expressing 35S::GFP or 35S:PvLLG1-GFP. (C) Root hair length in R. tropici-inoculated roots with 35S::GFP and 35S:PvLLG1-GFP constructs. (D) Nodule number at 7, 14, and 21 dpi in R. tropici-inoculated roots with 35S::GFP and 35S:PvLLG1-GFP constructs. (E) Nodule diameter at 7, 14 and 21 dpi in R. tropici-inoculated roots with 35S::GFP or 35S:PvLLG1-GFP constructs. (F) Number of nodules per composite plant with different ranges of diameters at 7, 14 and 21 dpi in R. tropici-inoculated roots with 35S::GFP and 35S:PvLLG1-GFP constructs. (G) Images of representative root nodules from hairy roots at 7, 14 and 21 dpi in R. tropici-inoculated roots with 35S::GFP and 35S:PvLLG1-GFP constructs. Replications for PvLLG1 transcript accumulation were 15, root hair length were 150, nodule number were 5, and nodule diameter were 150. Different letters or asterisk indicate significant differences among samples according to the Two -Way ANOVA analysis (*p = 0.05, ** p = 0.01, *** p = 0.001) or a pair-wise comparison with a Student´s T- test (p <0.0001), respectively.
Fig 8.
Rhizobium tropici colonization and nitrogen fixation in PvLLG1 silenced and overexpressing P. vulgaris nodules.
(A) Cross sections of hairy root nodules colonized by R. tropici-GUS showing bacterial colonization at 7, 14 and 21 dpi in PvLLG1 silenced plants (RNAi:PvLLG1 construct) compared to control (SAC). (B) Infection thread formation at 7 dpi in PvLLG1 silenced plants compared to control. (C) Nitrogen fixation (acetylene reduction) at 21 dpi in PvLLG1 silenced plants compared to control. (D) Cross section of nodules from hairy root colonized by R. tropici-GFP showing bacterial colonization at 7, 14 and 21 dpi in PvLLG1 overexpressing plants (35S:PvLLG1-GFP construct) compared to control (35S::GFP construct). (E) Infection thread formation at 7 dpi in PvLLG1 overexpressing plants compared to control. (F) Nitrogen fixation (acetylene reduction) at 21 dpi in PvLLG1 overexpressing plants compared to control. Asterisks indicate significant differences among samples according to the Two -Way ANOVA analysis (*p = 0.05, ** p = 0.01, *** p = 0.001).
Fig 9.
ROS distribution in PvLLG1 silenced and overexpressing P. vulgaris roots.
(A) Control (SAC) and PvLLG1 silenced (RNAi:PvLLG1 construct) roots with NBT staining to indicate superoxide distribution (upper panel), and H2CDFDA fluorescent probe to indicate ROS distribution in roots (middle panel), and epidermal zone where root hairs emerge (bottom panel). (B) Control (35S::GFP) and PVLLG1 overexpressing (35S:PvLLG1-GFP construct) roots with superoxide distribution (upper panel), ROS distribution (middle panel) and ROS distribution in emerging root hair cells (bottom panel).