Genome-wide identification of the CLAVATA3/EMBRYO SURROUNDING REGION (CLE) family in grape (Vitis vinifera L.)
- 274 Downloads
CLE genes play various biological roles in plant growth and development, as well as in responses to environmental stimuli.
In the present study, we identified nine CLE genes in the grape genome using an effective identification method. We analyzed the expression profiles of grape CLE genes in different tissues and under environmental different stimuli. VvCLE3 was expressed in shoot apical meristem (SAM) enriched regions, and VvCLE6 was expressed in shoot tissue without SAM. When grapes were infected with bois noir, VvCLE2 was up-regulated. Under ABA treatment, VvCLE3 was down-regulated. VvCLE6 was up-regulated under high temperature stress. We found that VvCLE6 and VvCLE1 were highly expressed in root tissue. In addition, we compared the characteristics of CLEs from grape and other plant species. The CLE family in Sphagnum fallax underwent positive selection, while the CLE family in grape underwent purifying selection. The frequency of optimal codons and codon adaptation index of rice and grape CLE family members were positively correlated with GC content at the third site of synonymous codons, indicating that the dominant evolutionary pressure acting on rice and grape CLE genes was mutation pressure. We also found that closely related species had higher levels of similarity in relative synonymous codon usage in CLE genes. The rice CLE family was biased toward C and G nucleotides at third codon positions. Gene duplication and loss events were also found in grape CLE genes.
These results demonstrate an effective identification method for CLE motifs and increase the understanding of grape CLEs. Future research on CLE genes may have applications for grape breeding and cultivation to better understand root and nodulation development.
KeywordsCLE gene Environmental stimuli CLE motif Codon usage bias
C-TERMINALLY ENCODED PEPTIDE
CLAVATA3/EMBRYO SURROUNDING REGION
ETHYLENE RESPONSE FACTOR
LEUCINE-RICH REPEAT RECEPTOR-LIKE KINASE
PHLOEM INTERCALATED WITH XYLEM
Root apical meristem
Shoot apical meristem
LRR-RLK TDIF RECEPTOR
Small signaling peptides (SSPs) are secreted chains of amino acids that typically have five to 50 residues, and are often encoded within a longer protein sequence of about 100 to 200 amino acids called a preproprotein. SSPs can act like hormones, and their cognate receptors can transmit local and systemic signals . SSP signals relay information that coordinate cell proliferation and differentiation during plant development, and SSPs often bind to corresponding families of receptors and are encoded by gene families . SSPs are known to include the CLAVATA3/EMBRYO SURROUNDING REGION (CLE) and the C-TERMINALLY ENCODED PEPTIDE (CEP) families . The CLE peptide family is a well-studied peptide family in plants and has been found in many plant species and some parasitic nematodes [3, 4, 5, 6, 7, 8, 9, 10, 11]. Thirty-two CLE genes been reported from Arabidopsis thaliana , while 15 and 50 CLE genes have been identified in tomato  and Populus trichocarpa , respectively.
CLEs contain a conserved 14-amino acid (aa) consensus sequence (conserved sequence motif KRXVPXGPNPLHNR), which is called the CLE motif or CLE domain . However, some reports indicate that the conserved CLE motif is 12–13 aa in length [10, 12, 14]. Several studies have shown that the conserved CLE motif is the functional region of CLE peptides, and the mature CLE peptide is derived directly from the CLE motif . The mature CLE peptides contain 12–13 amino acids that are proteolytically released from the precursors by a serine protease [15, 16]. Based on previous research that identified 12 amino acids as the conserved motif, the typical CLE motif has histidine (H1) or arginine (R1) at the first amino acid position. The research that indicated that the conserved motif is 13–14 amino acids long identified that the typical CLE motif has histidine (H2) or arginine (R2) at the second amino acid position . Whether the 12, 13, or 14 amino acids are conserved, most typical CLE conserved motifs have three prolines [6, 17]. The 12-aa motifs contain P4, P7, and P9, while 13–14 aa motifs often contain P5, P8, and P10. However, some CLEs lack one or two Ps [6, 17]. Goad et al. concluded that proteins without a typical CLE domain were not CLEs, positing that the Chlamydomonas reinhardtii green algal genome did not contain CLE genes because it lacked the typical motif . The putative candidate CLEs in C. reinhardtii contained a potential conserved CLE motif, ALVPSGPERRHH, but it lacks the typical histidine (H1) or arginine (R1) at the first amino acid position . Sawa et al. previously reported the chemical structure of two Arabidopsis CLE peptides in vivo, Arabidopsis Tracheary elements (TEs) differentiation inhibitory factor (TDIF) and CLV3 [17, 19]. The mature Arabidopsis TDIF peptide is a dodecapeptide, and the sequence was highly homologous to part of the CLE domain. The CLV3 peptide had two prolines that were hydroxylated, similar to TDIF . Ohyama et al. reported that CLV3 is a 13-amino-acid arabinosylated glycopeptide .
CLEs, as short-range signaling molecules, are a class of plant peptides that control stem cell fate . CLE genes have also been shown to participate in various biological processes, such as plant growth, development, and responses to environmental stimuli . CLE family genes can be divided into two types: A- and B-types. Arabidopsis CLE41, CLE44, and CLE46 are B-type genes, while A-type genes include CLV3. A-type CLE peptides promote cell differentiation in root and shoot apical meristem . The B-type CLE peptide Arabidopsis CLE41/TDIF/CLE44 promotes proliferation of vascular cells, while delaying differentiation into phloem and xylem cell lineages and regulating vascular stem cells [21, 22]. Similarly, Arabidopsis CLE41 inhibits the differentiation of tracheary elements in Zinnia elegans . Arabidopsis CLE45 inhibits protophloem development and regulates pollen–pistil interactions under high temperature conditions [23, 24, 25]. The A-type CLE peptides of Arabidopsis (CLV3, CLE19, and CLE40) trigger consumption of the root meristem . Overexpression of CLE6 results in a short-root phenotype through inhibition of the root apical meristem (RAM) . Arabidopsis CLV3 was also shown to play roles in regulating development of stem cell niches of SAMs .
The mechanism by which CLE peptides regulate plant development remians unclear. CLEs primarily interact with LEUCINE-RICH REPEAT RECEPTOR-LIKE KINASEs (LRR-RLKs) [10, 27, 28] . In addition, CLEs regulate WUSCHEL-related homeobox proteins (WOXs) and that the CLE-RLK-WOX signaling system can regulate and maintain the meristem . For example, Arabidopsis CLV3 signaling through these receptor complexes represses the transcription of WUS in the organizing center of the SAM, restricting stem cell divisions . Arabidopsis CLE8 positively regulates Arabidopsis WOX8 either in the endosperm or in suspensor cells [12, 30]. Arabidopsis CLE40 may control WOX5 expression in the RAM through interactions with the RLK ARABIDOPSIS CRINKLY4 .
Plant CLE peptides are also related to hormones. For example, Arabidopsis CLE41/TDIF interacts with brassinosteroids to determine xylem vessel formation by regulating GSK3s activity . Arabidopsis CLE6 peptide can counter gibberellic acid (GA) deficiency to promote shoot growth . TDIF/CLE41/CLE44 can interact with the LRR-RLK TDIF RECEPTOR (TDR)/PHLOEM INTERCALATED WITH XYLEM (PXY) expressed in the procambium and cambium [28, 33, 34]. TDR/PXY peptides are required for the auxin-dependent stimulation of cambium activity . ETHYLENE RESPONSE FACTORs (ERFs) are required for normal vascular cell divisions, and in the absence of TDR/PXY and WOX4 genes, expression of several ERFs are induced, suggesting that there is an interaction between TDIF/CLE41/CLE44-TDR/PXY-WOX4 signaling and ethylene signaling . In addition, several CLE peptides are controlled by N-, P-, and S-responsive pathways in various ways .
However, CLE family members from many other species have not yet been identified and thus the functions of CLE family members from many species remain unknown. In 2013, the CLV3-like genes VviCLE6, VviCLE25, VviCLE1, and TDIF were identified from the grapevine by Tominaga-Wada et al. . Later, Goad et al. identified the grape CLE genes GSVIVT01024996001, GSVIVT01028623001, GSVIVT01020127001, and GSVIVT01016568001 . However, many grape CLE genes remain to be identified, and the roles of grape CLE genes in most tissues and under various abiotic stresses remain unclear. In this study, we conducted a comprehensive analysis of grape CLE genes using the newly released grape genome (Version 2.1), and we investigated their expression profiles in different tissues and under different abiotic stresses. We identified nine putative CLE genes from the grape genome using a bioinformatics approach. The gene expression profiles in different tissues, codon usage bias and cis-regulatory elements of grape CLE genes were also analyzed.
Identification of grape CLEs
BLASTP analyses were performed using the previously reported Arabidopsis thaliana CLE proteins and a conserved CLE motif (KRXVPXGPNPLHNR) as queries against proteins in the newest grape genome database (http://genomes.cribi.unipd.it/grape/). Previous reports have shown that CLE precursors are less than 300 aa in length , so candidates exceeding 300 aa in length were removed.
Grape CLEs that were identified
Our finding included all four CLE genes reported by Goad et al. . Here, GSVIVT01016568001 was named VvCLE8, GSVIVT01024996001 was named VvCLE5, GSVIVT01028623001 was named VvCLE6, and GSVIVT01020127001 was named VvCLE3. Tominaga-Wada et al. identified grape CLE genes VvCLE6, VvCLE25, VvCLE1, and TDIF. Here, VvCLE25 was named VvCLE3, and VvCLE1 was named VvCLE1 . As “VviCLE6” identified by Tominaga-Wada et al. exceeded 300 aa in length, it was filtered out. TDIF did not contain a typical CLE motif, and it was also excluded.
Analysis of gene expression and cis-regulatory elements
CLEs play roles as short-range signals to control the fate of stem cells . Some WOXs can be regulated by CLEs, and the CLE-RLK-WOX signaling system can regulate and maintain the meristem . Accordingly, we performed qRT-PCR to analyze the expression profiles of grape CLE genes in SAM-enriched regions, as well as shoot tissue without SAMs. VvCLE3 was expressed in the SAM-enriched regions and not expressed in the shoot tissue without SAMs. VvCLE9, VvCLE2, VvCLE8, VvCLE1, and VvCLE6 were not expressed in SAM-enriched regions, but were expressed in the shoot tissue without SAMs. Other CLEs were not expressed in both SAM-enriched regions and shoot tissues without SAMs (Additional file 3: Figure S1).
In addition, we analyzed the cis-regulatory elements of grape CLE promoters. Promoters of some A-type CLE genes, such as VvCLE1, VvCLE2, and VvCLE3, contained CAT-box elements, which are cis-acting regulatory elements involved in meristem expression . Among the B-type CLEs, only the promoter of VvCLE6 contained dOCT, a cis-acting regulatory element involved in meristem-specific activation . Most grape CLE promoters contained the GCN4 motif or Skn-1 motif, which are cis-regulatory elements involved in endosperm expression [42, 43]. The promoters of some A-type grape CLE genes, such as VvCLE1, VvCLE2, and VvCLE9, contained ABRE, which is involved in responses to ABA . However, the VvCLE3 promoter, which has been shown to respond to ABA treatment in grape berry skin, did not contain ABRE. All B-type grape CLE promoters did not contain ABRE. Some grape CLE promoters contained GA-related cis-regulatory elements, such as the GARE motif . VvCLE1, VvCLE9, and VvCLE6 contained an as-2-box element, which is involved in shoot-specific expression . Most grape CLE promoters contained HSE sequences, and some contained MeJA-responsiveness and salicylic acid responsiveness elements. VvCLE1 and VvCLE6 contained LTR elements, which are involved in cold stress responses . VvCLE9, VvCLE8, and VvCLE6 contained ERE elements, which are ethylene responsive . VvCLE9, VvCLE2, and VvCLE8 contained O2 sites, which are cis-acting regulatory elements involved in zein metabolism regulation  (Additional file 2: Table S2).
CLE comparison between grape and other species
Selection pressure analysis of CLE families in different species
Ka/Ks value of CLE family
Positively selected site number
Codon preference index of CLE families in different species
Gene duplication and loss event analysis
Compared with A. thaliana, tomato, and Populus trichocarpa, grape had fewer CLE genes [3, 12, 13]. The previous study found that CLE genes in grape, S. fallax (nine CLE genes) and P. patens (nine CLE genes) were outnumbered relative to those in most species .
We used the Notung software package to identify gene duplication and loss events. By comparing grape and P. patens, Notung identified the duplication and loss events in both grape and P. patens (Additional file 6: Figure S4a). By comparing grape and S. fallax, Notung identified the duplication events in both grape and Sphagnum fallax but only identified loss events in S. fallax (Additional file 6: Figure S4b). By comparing grape and A. thaliana, Notung only identified loss events in grape (Additional file 6: Figure S4c).
Goad et al.  found four CLE genes in grape. Their study identified fewer CLE in various species than previously expected. For example, one previous study found 32 Arabidopsis thaliana CLE genes, while Goad et al. found 31. Another previous study found 44 soybean CLE genes, while Goad et al. found 43 [12, 18, 54]. Our method was able to find more reliable grape CLE genes at a genome-wide scale in grape.
In grape, fewer CLE genes were found compared with other species, such as A. thaliana, P. trichocarpa, and tomato [3, 12, 13]. AtCLV3 was shown to play roles in regulating development of stem cell niches of SAMs . Over-expression of AtCLE6 in a GA-deficient mutant partially rescued the mutant phenotype, suggesting that CLE6 can compensate for GA deficiency to promote shoot growth in Arabidopsis  and may play a role in the GA pathway. AtCLE6 is also associated with procambium proliferation . We found grape CLEs similar to AtCLE6 that may have implications for grape cultivation, such as elongating grape spikes and grape flowers, as well as fruit thinning. However, we did not find homologs of AtCLV3 and AtCLE6 in grape. We analyzed the expression profile of grape CLE genes and found that grape CLE genes likely play a role in shoot or SAM development. We found that grape VvCLE3 was expressed in most tissues, including SAM tissue. In our study, other grape CLE genes were not expressed in SAM tissue. We also found that VvCLE3 was down-regulated under exogenous ABA treatment. Previous studies have shown that some CLEs interact with hormones or are related to hormones. For example, AtCLE41/TDIF can interact with brassinosteroids (BR) , and AtCLE6 peptide can counter GA deficiency to promote shoot growth . TDR/PXY are required for the auxin-dependent stimulation of cambial activity . ERFs are required for normal vascular cell divisions, and in the absence of TDR/PXY and WOX4 genes, expression of several ERFs are induced, suggesting an interaction between TDIF/CLE41/CLE44-TDR/PXY-WOX4 signaling and ethylene signaling .
A previous study showed that the CLE45 peptide can mediate environmental signals . CLE45 is preferentially expressed in the stigma at normal temperatures, whereas its expression domain expands into the transmitting tract at elevated temperatures, suggesting that CLE45 has temperature-dependent functions . With the intensification of the greenhouse effect and global warming, plant responses to heat stress are particularly important. We found that only VvCLE6 responded to heat stress, so we speculate that VvCLE6 was the only CLE gene involved in heat stress responses in grape. Under bois noir or GLRaV-3 virus infections, most grape CLE genes did not exhibit significant changes compared with normal conditions.
We found that the A-type grape CLE genes VvCLE1, VvCLE2, and VvCLE3 contained a CAT-box, which is a cis-acting regulatory element involved in meristem expression. However, VvCLE1 and VvCLE2 were not expressed in grape SAM, though they may be expressed in other meristem tissues. In Arabidopsis, CLE43 peptides suppress xylem differentiation [18, 21]. CLE17 is expressed in the RAM, the lateral root cap cells, and the epidermis, and CLE17 is involved in root development .
In B-type CLE genes, only VvCLE6 contained dOCT, which is a cis-acting regulatory element related to meristem-specific activation. In Arabidopsis, CLE44 or CLE41 are involved in the regulation of vascular stem cells [22, 28, 34, 55]. VvCLE1, VvCLE2, and VvCLE9 contained the ABRE element, while VvCLE3 did not. However, only VvCLE3 had altered expression under exogenous ABA treatment. B-type CLE genes did not contain ABRE. VvCLE6 and some other grape CLE genes contained HSE. In addition, many grape CLE promoters contained some hormone responsiveness elements, indicating that they may be involved in hormone signal pathways, such as those of GA, jasmonic acid, salicylic acid, and ethylene (Additional file 2: Table S2).
We constructed a 2-D clustering diagram of all CLEs using the grape, P. patens, S. fallax, A. thaliana, and rice CLEs as well as two soybean CLEs. These CLEs were divided into three groups: A1, A2, and B. The phylogenetic analysis showed a similar pattern for the A2 and B groups. A2 was divided into two clusters, the ancestral and modern clusters. In the ancestral cluster, there were only non-angiosperms P. patens and S. fallax CLE genes, matching findings by Goad et al. , who found that CLE genes from non-angiosperms P. patens and S. fallax could be divided into an independent cluster . Dicotyledon or monocotyledon CLEs could not be divided into an independent cluster, also consistent with the findings of Goad et al. . The A2 group contained GmNIC1 and GmNIC2, which are regulated by nodulation [18, 40, 41]. Goad et al. concluded that the group with GmNIC1 and GmNIC2 could be involved in nodulation or mycorrhizal development, as mycorrhizae and nodulation share overlapping gene regulatory networks . However, the A2 group did not contain any grape, rice, or moss CLE genes. A previous study indicated that some CLE genes were related to nodulation and could modulate nodulation . Nematodes may have also secreted nematode CLEs, and nematode CLEs function like endogenous plant CLE peptides. Once nematode CLEs are delivered into plant cells, they can function similarly to endogenous plant CLEs to redirect plant CLE signaling pathways to establish a successful parasitic association with host plants . Grapes are also invaded by rhizobia, nematodes, and Phylloxera. While group A2 did not contain any grape CLE genes, we found that some grape CLE genes, such as VvCLE1 and VvCLE6, were expressed at a higher level in grape root tissue. Furthermore, these CLE genes should be studied to better understand the relationship of grape plants with rhizobia, nematodes, and Phylloxera. Groups B and A2 did not contain any moss CLE genes. Because moss plants do not have true root or vascular tissues, moss CLEs cannot be exclusively involved in functions related to root tissue, vascular tissue, or nodulation. This may explain why groups B and A2 did not contain any moss CLE genes (Fig. 4).
We found P. patens, A. thaliana, and rice CLEs that contained a 12-aa-long motif. The most conserved part of the motif was “LLLL.” The function of the motif is unclear (Additional file 4: Figure S2). A previous study has shown some CLE genes contain two CLE motifs . In rice, we found the LOC_Os05g29870 contained six repeated CLE motifs. Another peculiarity of the rice CLE family was its strong codon bias compared with the other four species. Its average Fop and CAI were highest and its GC3s and GC content were highest as well. In contrast, the S. fallax CLE family was the only family that underwent positive selection. If codon usage bias is correlated with GC content, we could deduce whether codon usage bias was affected by mutation pressure during its evolutionary history [59, 60, 61]. The Fop and CAI values of rice and grape CLE family members were positively correlated with GC and GC3s content, suggesting that rice and grape CLE genes mainly evolved by mutation pressure. The Fop and CAI values of P. patens and A. thaliana CLE family members were not positively correlated with GC and GC3s content, indicting their CLE genes evolved by other pressure, such as natural pressure [59, 60].
We identified some gene duplication events in CLE genes. The analysis conducted with Notung identified gene duplication events in grape, P. patens, S. fallax, and A. thaliana therefore gene duplication may have contributed to the expansion of the CLE family. Both rice and Arabidopsis thaliana have undergone one γ events (ancient whole-genome replication events) and at least two whole-genome duplication (WGD) events, and grapes appear to have only undergone one γ event . P. patens appear to have experienced at least two WGD events , and S. fallax has undergone at least one WGD event . More gene duplication and WGD events have been identified in A. thaliana. This result indicates that WGD may have played an important role in gene duplication events in the CLE family. The analysis with the Notung software indicated that gene loss events have occurred in grape and moss. Gene loss events or lower gene duplication rates may have led to grape and moss containing fewer CLE gene family members than Arabidopsis thaliana.
The present study provides an effective method for identifying CLE motifs and increases the understanding of grape CLEs. Moreover, our systematic analysis provided comprehensive information for further research investigating the functions of grape CLEs. Future research on CLE genes may have applications for grape breeding and cultivation to better understand root and nodulation development.
Data collection and identification of grape CLE genes
Grape protein data was obtained from the grape genome database (http://genomes.cribi.unipd.it/grape/). CLEs that were previously identified in Arabidopsis thaliana and a CLE conserved motif (KRXVPXGPNPLHNR) were used as queries to perform BLASTP analysis (E < 10− 20). Candidate proteins without a conserved C-terminal CLE motif and with candidate protein lengths exceeding 300 aa were removed.
Multiple sequence alignment, phylogenetic, and two-dimensional clustering analysis
Protein multiple sequence alignment was performed using MAFFT v7 (https://mafft.cbrc.jp/alignment/server/), and Neighbor-joining (NJ) trees were constructed using MEGA 6.0 based on full-length protein sequences . To support the inferred relationships, 1000 bootstrap samples were generated. CLANS was used to construct the two-dimensional clustering diagram .
Analysis of selective pressure
The selective pressures on sequences were determined using Codeml in PAML (phylogenetic analysis maximum likelihood) version 4.7 software .
Gene duplication and loss event analysis
Notung 2.9  was used to analyze gene duplication and loss events.
Codon usage bias analysis
The frequency of optimal codons (FOP), GC content, GC content at the third site of synonymous codons (GC3s content), relative synonymous codon usage (RSCU) and codon adaptation index (CAI) were analyzed using the coding sequences of CLE genes from grape and other species with CodonW 1.4.2.
Plant materials, and RNA isolation
SAM enrichment regions and shoot tissue samples without SAMs were collected from grapes cv. ‘Cabernet Sauvignon.’ RNA was isolated using the CTAB method as described previously . DNase I was used to digest genomic DNA and to eliminate DNA contamination. Reverse transcription was also performed as previously described .
Gene expression analysis
Quantitative real time PCR for grape CLE genes was performed as previously described . The qRT-PCR primer sequences are provided in Additional file 1: Table S1. VvActin7 (VIT_204s0044g00580) was used as the reference sequence . The VvActin7 forward and reverse primers were 5′-CTTGCATCCCTCAGCACCTT-3′ and 5′-TCCTGTGGACAATGGATGGA-3′, respectively . Three biological replicates were included for each analysis, and each biological replicates were analyzed through three technical replicates. Gene relative expression levels were calculated using the ΔΔCT method.
The microarray expression profiles of bois noir–infected samples (GSE12842), GLRaV-3 virus–infected samples (GSE31660), high-temperature-treated samples (GSE31675), ABA-treated samples (GSE31664 and GSE31662), and 49 tissue samples (GSE36128) from grape plants were retrieved from the Plexdb (http://www.plexdb.org/) and GEO databases (https://www.ncbi.nlm.nih.gov/gds/). The RMA-normalized signal intensity values of grape CLE genes were used to represent the expression level. The average values of replicates were used to calculate the fold changes. T-test p-values < 0.05 were considered to be significantly different. The pheatmap R package was used to generate heatmaps.
Analysis of cis-acting regulatory elements in promoters
We identified cis-acting regulatory elements of grape CLE promoters using Plantcare software (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/), as previously described .
PW, YW and FR designed the study. PW wrote the manuscript and carried out most of the experiment and data analysis. All authors have read and approved the manuscript, and ensure that this is the case.
This work was supported by the Agricultural scientific and technological innovation project of Shandong Academy of Agricultural Sciences (CXGC2018E17; CXGC2016D01), Agricultural scientific and technological innovation project of Shandong Academy of Agricultural Sciences-cultivating project for National Natural Science Foundation of China in 2018 “Identification and function research of Vitis vinifera and Vitis amurensis cold stress response-related microRNAs”, Major Agricultural Application Technology Innovation Project of Shandong Province “Research and Application of Precision Control of Maturation and Product Innovation of Featured Brewing Grape”, Major Agricultural Application Technology Innovation Project of Shandong Province “Development of Landmark Wines and Integrated Application of Key Technologies in Shandong Province” and Fruit innovation team of modern agricultural industry technology system in Shandong Province-Jinan comprehensive test station (SDAIT-06-21). The funding agencies did not have a role in the design of the study and in the data generation, analysis, and interpretation of data and in writing the manuscript.
Ethics approval and consent to participate
Consent for publication
All authors declare that they consent for the publication.
The authors declare that they have no competing interests.
- 47.Wu C, Zheng C, Ji G, Jiang P. Synergistic effects of HSE and LTR elements from hsp70 gene promoter of Ulva prolifera (Ulvaphyceae, Chlorophyta) upon temperature induction. J Phycol. 2019; Epub ahead of print.Google Scholar
- 52.Yang Z, Nielsen R, Goldman N, Pedersen AM. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics. 2000;155:431–49.Google Scholar
- 57.Osipova MA, Mortier V, Demchenko KN, Tsyganov VE, Tikhonovich IA, Lutova LA, et al. Wuschel-related homeobox5 gene expression and interaction of CLE peptides with components of the systemic control add two pieces to the puzzle of autoregulation of nodulation. Plant Physiol. 2012;158:1329–41.CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.