Edited by: Paloma Moncaleán, Neiker-Tecnalia, Spain
Reviewed by: Jan Max Bonga, Natural Resources Canada, Canadian Forest Service, Canada; Jens F. Sundström, Swedish University of Agricultural Sciences, Sweden
†These authors have contributed equally to this work
This article was submitted to Plant Development and EvoDevo, a section of the journal Frontiers in Plant Science
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Somatic embryogenesis techniques have been developed for most coniferous species, but only using very juvenile material. To extend the techniques’ scope, better integrated understanding of the key biological, physiological and molecular characteristics of embryogenic state is required. Therefore, embryonal masses (EMs) and non-embryogenic calli (NECs) have been compared during proliferation at multiple levels. EMs and NECs originating from a single somatic embryo (isogenic lines) of each of three unrelated genotypes were used in the analyses, which included comparison of the lines’ anatomy by transmission light microscopy, transcriptomes by RNAseq Illumina sequencing, proteomes by free-gel analysis, contents of endogenous phytohormones (indole-3-acetic acid, cytokinins and ABA) by LC-MS analysis, and soluble sugar contents by HPLC. EMs were characterized by upregulation (relative to levels in NECs) of transcripts, proteins, transcription factors and active cytokinins associated with cell differentiation accompanied by histological, carbohydrate content and genetic markers of cell division. In contrast, NECs were characterized by upregulation (relative to levels in EMs) of transcripts, proteins and products associated with responses to stimuli (ABA, degradation forms of cytokinins, phenols), oxidative stress (reactive oxygen species) and carbohydrate storage (starch). Sub-Network Enrichment Analyses that highlighted functions and interactions of transcripts and proteins that significantly differed between EMs and NECs corroborated these findings. The study shows the utility of a novel approach involving integrated multi-scale transcriptomic, proteomic, biochemical, histological and anatomical analyses to obtain insights into molecular events associated with embryogenesis and more specifically to the embryogenic state of cell in Douglas-fir.
Douglas-fir [
Embryogenic cultures of conifers consist of EMs composed of early differentiated cells forming immature somatic embryo (SEs) that proliferate via cleavage polyembryony (
A characteristic cytological feature of somatic embryogenesis in Douglas-fir is interspersion of proliferating EMs with non-embryogenic cell clusters (
Conifers are considered highly recalcitrant to somatic embryogenesis from explants (e.g., shoot apices or needles) of selected trees in their adult vegetative and reproductive phases (
Thus, detailed characterization of embryogenic state, especially at molecular level, is required to complement macromorphological and cytological observations of proliferating structures (EMs, NECs, polyembryogenic centers, meristemoids and nodules) generated following somatic embryogenesis induction in conifers (
Several previous studies have addressed the ability of cotyledonary SEs to undergo secondary somatic embryogenesis in various conifers, such as
The objective in this study was to gain insights into molecular changes leading to the differentiation of immature SEs (EMs, embryogenic lines) rather than only calli (NECs). More knowledge of gene expression profiles associated with embryogenic state may help early detection of EMs and develop ways to overcome conifers’ recalcitrance to somatic embryogenesis and improve initiation rates of embryogenic lines. For this, we have compared isogenic EMs and NECs originating from a single SE, strictly initiated and then proliferated in the same media and under the same environmental conditions. We have applied a novel approach, involving integrated multi-scale analyses, combining genome-wide transcriptomic (RNAseq) and proteomic (free-gel/free-label) profiling followed by Sub-Network Enrichment Analyses (SNEA) to elucidate differences in functions and interactions of significant factors (transcripts and proteins) between EMs and NECs. Molecular data were complemented with morphological and histo-cytological, biological and biochemical analyses. We have also quantified endogenous phytohormones-auxin, ABA and CKs-with known importance in the control of cell proliferation and differentiation during somatic embryogenesis in conifers (
Parental
Origins and sampling scheme of the Douglas-fir plant material (EM and NEC) used for multi-scale characterization of the embryogenic state in three unrelated genotypes (SD4, TD15, and TD17). Cotyledonary SEs regenerated from primary SD4, TD15, and TD17 embryogenic lines were used as explants for induction of secondary somatic embryogenesis. Secondary EMs (SD4-8 EM, TD15-1 EM, TD17-1 EM) were obtained from the root cap region of SEs whereas NEC lines (SD4-8 NEC, TD15-1 NEC, TD17-1 NEC) were simultaneously obtained from the hypocotyl region of the same SEs during the same initiation experiment. All the EM and NEC material required (overall 1.7 g) for multi-scale comparison (soluble carbohydrate, starch, total protein, phytohormone, transcriptomic and proteomic), proliferated for 2 weeks and then samples were taken from the same Petri dishes, each treated as one biological replicate (3–7 replicates per type of analysis). Coty SE, cotyledonary somatic embryo; EM, embryonal mass; NEC, non-embryogenic callus.
Both the EMs and NECs were sub-cultured every 2 weeks on Glitz proliferation medium, supplemented with 4.5 μM 2,4-D, 2.2 μM BA and 0.087 M maltose solidified with 4 g L-1 of gellan gum (
EMs and NECs samples of all three investigated genotypes were collected after 10 days of multiplication for morphological and histo-cytological characterizations. Their morphology was documented using a SMZ 1500 stereomicroscope (Nikon, Tokyo, Japan). Fresh material was stained with either 0.4% (w/v) Trypan Blue (Sigma-Aldrich) as described by
Proliferating plant material (EMs and NECs) collected from filter papers was weighed, dispersed in liquid Glitz medium with no plant growth regulator and distributed on a filter paper disk placed on the surface of Glitz maturation medium supplemented with 0.2 M sucrose, 60 μM
Samples of proliferating EMs and NECs (about 100 mg f.w.) collected from filter papers were immediately weighed to estimate their f.w. Their d.w. was determined after oven-drying at 70°C for 24 h (
Extracts of 3–4 biological replicates of each sample were used to analyze contents of the following phytohormones: ABA, ABA-GE, free auxin IAA and an array of CKs. The CK analysis included both isoprenoid forms—
Accurately weighed EMs and NECs samples (around 0.3 g f.w.) were placed in 2 mL Eppendorf tubes and mixed with 1.6 mL of modified Bieleski solution (methanol:formic acid:water, 15:1:4), supplemented with deuterated standards (Olchemim, Olomouc, Czechia) and ground using a MM 400 mixer mill (Retsch, Haan, Germany). The samples were left to extract overnight in a refrigerator at 4°C, then centrifuged and the pellets were re-suspended and re-extracted in the same volume (1.6 ml) of the same solution for 10 min in an DT 100 H ultrasonic bath (Bandelin electronic, Bandelin, Germany) at room temperature. After further centrifugation both supernatants were combined and loaded on a Strata C18-T SPE column (Phenomenex, Torrance, CA, United States) to remove non-polar compounds. The eluate was partly evaporated in an RVC Alpha rotary vacuum concentrator (Martin Christ, Osterode am Hartz, Germany) to approximately half the original volume and acidified by adding 1 mL of 1 M formic acid. The concentrated samples were loaded on an Oasis MCX SPE column (Waters, Milford, MA, United States) and further proceeded according to
Each dried extract was dissolved in 100 μL of 10% (v/v) acetonitrile, filtered through a nylon Micro-Spin 0.2 μm centrifugal filter (Grace, Columbia, MD, United States) and placed in a cooled sample stack. A portion of the extract (5 μL) was analyzed by a LC-MS system consisting of a Rheos 2200 HPLC pump (Flux Instruments, Basel, Switzerland) and HTS-Pal auto-sampler with cooled sample stack (CTC Analytics, Zwingen, Switzerland) coupled to a TSQ Quantum Ultra AM triple-quad high resolution mass spectrometer (Thermo Electron, San Jose, CA, United States) equipped with an electrospray interface. The HPLC column was tempered in a Delta Chrom CTC 100 Column oven (Watrex, Praha, Czechia).
The mass spectrometer was operated in multiple SRM (single reaction monitoring) mode (positive for CK analysis, negative for IAA, ABA and ABA-GE analysis) with acquisition of 2–4 transitions per compound. The most intense ion was used to quantify the analyte, the others to confirm its identity. The analytes were quantified using multilevel calibration curves with stable isotope labeled compounds used as internal standards. Each sample was analyzed twice.
Cytokinins were analyzed using an HPLC system with a Synergi 4 μm Hydro-RP 80 Å, 250 × 2.1 mm column (Phenomenex, Torrance, CA, United States) and a mobile phase consisting of a 30-min ternary gradient of water, acetonitrile and 0.01% of acetic acid (flow rate, 200 μL min-1). The proportion of acetonitrile was linearly increased from 8 to 50% and the acetic acid solution was maintained at 35% throughout each run, after which the column was washed with 90% acetonitrile.
A Kinetex 2.6 μ C18 100 Å, 50 × 2.1 mm HPLC column (Phenomenex, Torrance, CA, United States) was used to analyze IAA, ABA and ABA-GE, using a 13-min water, acetonitrile and 1% (v/v) acetic acid ternary gradient as the mobile phase (flow rate, 200 μL min-1). The proportion of acetonitrile was linearly increased from 5 to 90%, and the proportion of acetic acid solution was kept at 10% during each run, after which the column was washed with 90% acetonitrile.
Soluble proteins were extracted from five biological replicates of both EMs and NECs (200 mg f.w. of frozen material) with 1 mL of buffer containing 4 M urea, 0.1% v/v SDS, 0.1 M DTT, 80 mM Tris HCl (pH 6.8) and 10% (v/v) glycerol. Protein content was determined using the Bradford assay with bovine serum albumin as a standard. Results were expressed as soluble protein content in μg g-1 d.w.
Carbohydrates and starch were extracted following
RNA were extracted from seven biological replicates of both EMs and NECs (100 mg f.w. of frozen material). Samples were ground for 5 min with a mortar and pestle to a fine powder. Total RNA was extracted following
Five samples (1 of SD4-8 EM and TD15-1 EM, and 3 of TD17-1 EM) were removed after RNA extraction quality control made by the outsourced company (GATC Biotech, Germany) and before sequencing, resulting in a total of 37 samples. Libraries were constructed and sequenced by GATC after polyA selection. Strand specific cDNA libraries were synthesized and all samples were paired-end sequenced using an Illumina HiSeq 2500 system and the manufacturer’s recommended procedures. Raw sequencing reads (between 40,000,000 and 50,000,000 reads per library) are available through SRA BioProject accession N°
The short-reads were mapped using reference
Four biological replicates per sample were subjected to proteomic and nLC-MS/MS analyses following
Significantly differentially expressed proteins were functionally classified using Gene Ontology Consortium (GO) codes
Changes in protein expression were calculated in comparison with corresponding control based on the cumulative intensity in each peptide. All sequences have been mapped with GO terms against
SNEA of selected genes was performed using Plant Pathway Studio® version 12 (Elsevier B.V.), with a threshold
R version 3.3.2 (
Transcriptome mapping results were analyzed using DESeq2 library (
For proteomic analysis, differential expression of proteins in the tissue types was analyzed by ANOVA using the Limma R-package
EMs and NECs cultures derived from all three Douglas-fir genotypes (SD4-8, TD15-1, and TD17-1) fundamentally differed in color, morphology, cell arrangement and levels of secondary metabolites. EMs were usually whitish, yellowish (
Histological characterization of isogenic embryonal mass (EM,
Non-embryogenic calli were usually brownish-yellow, light or dark brown, indicating that they had higher levels of phenolic compounds than EMs, as confirmed by histochemical staining (
Multifactorial analysis of the water contents of EMs and NECs cultures indicated a significant effect of tissue type (
Water content
In maturation experiments, NECs cultures never gave rise to SEs, while all EMs yielded cotyledonary SEs. Mean SE production significantly (
Soluble sugar levels varied between genotypes and tissue types (
Soluble sugar, starch and soluble protein contents of isogenic embryonal mass (EM) and non-embryogenic callus (NEC) of Douglas fir after 2 weeks of multiplication: means obtained from analyses of three genotypes (SD4-8, TD15-1, and TD17-1),
Compounds | EM | NEC |
---|---|---|
Fructose (Fru) | 17.14 ± 4.27 ab | 10.78 ± 12.80 ab |
Galactose | 2.68 ± 0.77 a | 0.00 a |
Glucose (Glc) | 208.40 ± 32.74 d | 90.79 ± 76.97 c |
Sucrose (Suc) | 41.23 ± 19.53 b | 26.87 ± 12.93 ab |
Maltose | 6.09 ± 0.76 a | 8.95 ± 12.53 a |
Melibiose | 3.46 ± 0.82 a | 0.00 a |
Myo-inositol | 3.30 ± 0.82 a | 4.03 ± 0.66 a |
Raffinose | 11.58 ± 5.12 ab | 5.33 ± 3.92 a |
Carbohydrate total | 869.47 ± 70.00 e | 440.25 ± 43.32 f |
[(Glc + Fru)/Suc] | 7.50 ± 1.49 * | 3.14 ± 0.59 ** |
Starch (mg g-1 d.w.) | 9.03 ± 0 0.89 α | 19.01 ± 1.37 β |
Soluble proteins (mg g-1 d.w.) | 94.19 ± 7 0.54 A | 92.27 ± 27.85 A |
Considerable differences were detected in concentrations of measured phytohormones (auxins, ABA and CKs) between EMs and NECs of all three genotypes (SD4-8, TD15-1, and TD17-1).
Free IAA was the only auxin detected, at levels ranging from 241.81 to 301.45 pmol g-1 d.w. in EMs (in TD15-1 EM and SD4-8 EM, respectively) and 110.15–266.58 pmol g-1 d.w. in NECs (in TD15-1 NEC and SD4-8 NEC, respectively). IAA levels were 1.1- to 2.2-fold higher in EMs than in NECs of all genotypes with statistically significant differences in two out of three lines (TD15-1 and TD17-1) (
Concentrations of the auxin indole-3-acetic acid (IAA,
In contrast, contents of the “stress hormone” ABA and its conjugated storage form ABA-GE varied from tens to ten thousands of picomols per gram d.w. and were respectively 3.8- to 22.7-fold and 41.7- to 112.4-fold higher in NECs than EMs. In all NECs, ABA-GE was more abundant than ABA (1.9-, 3.3-, and 9.2-fold in TD15-1 NEC, TD17-1 NEC, and SD4-8 NEC, respectively). ABA-GE levels exceeded ABA levels in the SD4-8 EMs (1.9-fold), but ABA was ca. 4-fold more abundant than ABA-GE in the other two embryogenic lines (
Wide spectra of endogenous CKs were detected in both EMs and NECs of all three genotypes, including bioactive (free bases), transport (ribosides) and immediate CK biosynthetic precursor (nucleotides) forms, irreversibly deactivated derivatives (
The concentrations of isoprenoid cytokinins (CKs;
The most abundant isoprenoid CKs were
Only two forms of
Proportions of bioactive and transport CKs (free bases and ribosides) in the total CK pool were lower in NECs than in EMs of all three genotypes (
Overall contents of aromatic CKs were considerably higher in SD4-8 NEC and TD15-1 NEC than in corresponding EMs, but slightly higher in TD17-1 EM than in TD17-1 NEC (
Moderate concentrations (picomols to tens of picomols per gram d.w.) of BA derivatives hydroxylated on the sidechain phenyl ring in
To summarize, EMs of all three genotypes were characterized by higher content of auxin IAA (although statistically significant differences only in two out of three lines) whereas the levels of ABA and its conjugated storage form ABA-GE in NECs exceeded those in EMs. Concentrations of bioactive, transport and prevailing biosynthetic precursor forms of CKs were higher in EMs compared to NECs while the inactive and/or weakly active CK forms were detected mainly in NECs.
RNAseq analysis with 4–7 biological repetitions revealed very similar differences between EMs and NECs of all three genotypes. Our Illumina short-reads mapped to 53,185 (97%) of the 54,830 sequences compiled in the Douglas-fir reference transcriptome (
Of the mapped transcripts, 8,955 were differentially expressed in EMs and NECs (4,149 more strongly expressed in EMs than in NECs, and 4,806 more strongly in NECs). Of these, 8,152 (91%) were assigned at least one functional annotation by Blast2GO. Complete GO annotation, in terms of all three ontologies (Molecular Function, MF; Biological Process, BP; and Cellular Component, CC), was obtained for 2,029 transcripts (23%) of the transcripts.
Most GSEA results were obtained from Molecular Function (MF) ontology annotations. In some cases, BP ontology provided support at another level of complexity, and in a single case, Cellular Component (CC) ontology supported MF ontology results. Briefly, early EM events seem to include reorganization of sugar metabolism at transcriptome level and deep reprogramming of the ribosomic protein production and post-maturation system. In contrast, key features of NEC formation seem to include upregulation of genes encoding diverse trans-membrane transporters for water, ions, metals, sugars, amino acids and lipids, numerous homeobox leucine zipper and bHLH transcription factors as well as genes involved in phenolic secondary metabolism.
The most strongly upregulated sets of transcripts in EMs (relative to levels in NECs), were 45, 29 and 96 transcripts annoted with MF GO terms GO:0003735 (structural constituent of ribosomes,
Upregulated transcripts in EM (relative to NEC) annotated to MF GO:0003735, “structural constituent of ribosome.” Expression values are normalized mapping scores from DESeq2 analysis: means for NEC and EM obtained from analyses of 16 and 21 samples, respectively. Transcript labels are from annotations obtained by Blast2GO. The dashed red line indicates non-differential expression between EM and NEC. Differential expression
The most strongly upregulated transcripts of enriched group MF GO:0003735 were PSME_00003749-RA (“seed maturation PM30”; 580-fold differentially expressed, mapping score 1,446 in EMs) (
The most differentially (65-fold) and highly expressed (mapping score 350) transcript in the MF GO:0016757 group (
Enrichment of 19 GO terms was detected in NECs (
MF ontology terms enriched in non-embryogenic callus (NEC), numbers of transcripts annotated to the terms detected in NEC and isogenic embryonal mass (EM), GOid and GO code for each term obtained by BLAST2GO and
GO id | EM | NEC | GO terms | |
---|---|---|---|---|
GO:0005506 | 26 | 131 | 3 10-19 | |
GO:0016705 | 13 | 117 | 5 10-18 | |
GO:0016887 | 50 | 109 | 4 10-5 | |
GO:0043531 | 16 | 94 | 8 10-32 | |
GO:0050662 | 22 | 86 | 3 10-12 | |
GO:0016616 | 24 | 79 | 1 10-9 | |
GO:0003854 | 16 | 66 | 2 10-22 | |
GO:0046983 | 19 | 51 | 1 10-3 | |
GO:0005215 | 13 | 49 | 5 10-5 | |
GO:0043565 | 17 | 41 | 9 10-5 | |
GO:0016773 | 19 | 41 | 2 10-53 | |
GO:0008289 | 9 | 33 | 1 10-8 | |
GO:0022857 | 15 | 30 | 3 10-3 | |
GO:0008236 | 4 | 28 | 9 10-16 | |
GO:0016829 | 12 | 27 | 2 10-3 | |
GO:0019001 | 3 | 21 | 1 10-5 | |
GO:0016614 | 2 | 18 | 1 10-9 | |
GO:0008233 | 6 | 17 | 6 10-8 | |
GO:0016798 | 2 | 13 | 9 10-6 | |
Upregulated transcripts in NEC (relative to EM) annotated to MF GO:0003854, “3-beta-hydroxy-delta5-steroid dehydrogenase activity.” Expression values are normalized mapping scores from DESeq2 analysis: means for NEC and EM obtained from analyses of 16 and 21 samples, respectively. Transcript labels are from annotations obtained by Blast2GO. The dashed red line indicates non-differential expression between EM and NEC. Differential expression p-values are color-coded, and sizes of dots are proportional to mapping scores.
Four upregulated MF GO terms group transcripts associated with membrane transport activity (GO:0005215,
Upregulated transcripts in NEC (relative to EM) annotated to MF GO: 0005215, “transporter activity.” Expression values are normalized mapping scores from DESeq2 analysis: means for NEC and EM obtained from analyses of 16 and 21 samples, respectively. Transcript labels are from annotations obtained by Blast2GO. The dashed red line indicates non-differential expression between EM and NEC. Differential expression
Thirty upregulated transcripts are annotated to MF GO:0022857, “transmembrane transporter activity” (
Upregulated GO terms in NECs included two classes of transcription factors: MF GO:0008289 and MF GO:0046963 (
Further differences between EMs and NECs of the three genotypes were explored by nLC-MS/MS-based quantitative proteome characterization. A global profiling of quantitative proteome was obtained for these tissues. In total, 3,028 proteins were identified (
Venn diagram showing overlaps of the identified proteins (3028) in isogenic embryonal mass (EM) and non-embryogenic callus (NEC) of the three genotypes SD4-8, TD15-1 and TD17-1 of Douglas-fir. In total 413 proteins were significantly differentially expressed between EMs and NECs of all three genotypes: 236 more strongly expressed in EMs than in NECs and 177 more strongly expressed in NECs. Interpretation of results of the proteomic study was based on expression patterns of these 413 proteins.
Sub-network enrichment analysis of the significant proteins, based on the
Results of Sub-Network Enrichment Analysis (SNEA) connecting significantly differentially expressed (
Significantly differentially expressed proteins related to the biological processes identified in the SNEA (see
Biological process | Overlap | % overlap | Overlapping Entities | Exp. | |
---|---|---|---|---|---|
Plant growth | 18 | 1 | AGO10; CR88; DWF1; ABH1; EBS1; ELF5A-1; AOX1A; DET3; APX1; TH9; DL1; CYT1; APY2; RGP1; GSL8; HMGB1; FER; VDAC1 | 2.7 10-02 | EM |
Cell death | 13 | 1 | LIN2; mMDH1; AOS; HCHIB; SGT1B; HOT5; AER; GSTF10; FDH; TED4; PR4; PLDALPHA1; CPN60A | 6.6 10-03 | NEC |
Cell division | 13 | 1 | AGO10; MIRO1; SMT2; DWF1; ELF5A-1; ACT7; IMPA-1; COB; APY2; AVP1; RGP1; HDA3; NRPB3 | 5.8 10-03 | EM |
Defense response | 13 | 1 | LIN2; GLU1; TPS5; AOS; HCHIB; SGT1B; LOX5; HOT5; AER; FDH; HB1; PR4; OPR3 | 1.3 10-03 | NEC |
Flower development | 13 | 1 | P5CS1; PRMT4A; AIM1; DWF1; APX1; PP5.2; CYT1; RGP1; ADG1; PRP39; HDA3; PKT3; CEP1 | 4.9 10-02 | EM |
Photosynthesis | 13 | 1 | PGM; PGK1; LIN2; mMDH1; GLU1; AOS; APX1; PGK; PAS2; PRX72; NIR1; RPI2; OPR3 | 1.3 10-03 | NEC |
Cell expansion | 12 | 3 | CSI1; SMT2; THE1; ELF5A-1; DET3; DL1; ACT7; COB; CYT1; VHA-A; APY2; FER | 5.1 10-05 | EM |
Detoxification (process) | 12 | 4 | GLU1; AOS; G6PD6; HOT5; SFGH; AER; NIR1; FDH; ALDH2B4; NIT4; ADH1; AOR | 2.0 10-07 | NEC |
Developmental process | 9 | 1 | DL1; CYT1; AGO10; AVP1; ABH1; HDA3; APX1; FER; CEP1 | 1.6 10-02 | EM |
Plant defense | 9 | 1 | HOT5; LIN2; ATCAD4; CM1; AOS; HCHIB; PR4; ADH1; SGT1B | 2.2 10-02 | NEC |
Root development | 9 | 1 | HOT5; PGM; GAPCP-2; TRN2; AOS; HB1; TED4; SGT1B; LOX5 | 3.1 10-02 | NEC |
Embryonal development | 9 | 2 | DL1; COB; CYT1; AGO10; CR88; MIRO1; TUF; WHY1; CLPC1 | 6.8 10-03 | EM |
mRNA splicing | 9 | 4 | P5CS1; PMH2; PRP39; WHY1; ABH1; HEN2; RHD4; CCR1; HSP70 | 4.6 10-05 | EM |
Cell elongation | 8 | 2 | ACT7; DWF1; CSI1; SMT2; THE1; ABH1; FER; DET3 | 1.1 10-02 | EM |
Disease resistance | 7 | 1 | HOT5; LIN2; FDH; HSP90.1; HB1; PR4; SGT1B | 2.2 10-02 | NEC |
Plant response | 7 | 1 | HOT5; AOS; HB1; ADH1; PLDALPHA1; TED4; OPR3 | 2.9 10-02 | NEC |
Ripening | 7 | 1 | HOT5; DFR; FDH; ASD1; ADH1; BAN; PAS2 | 9.7 10-03 | NEC |
Senescence | 7 | 1 | LIN2; NIR1; AOS; TED4; PLDALPHA1; OPR3; APX1 | 4.5 10-02 | NEC |
Pollen development | 7 | 2 | DL1; MIRO1; RGP1; GSL8; GLT1; CEP1; VDAC1 | 4.7 10-03 | EM |
Cellulose biosynthesis | 7 | 6 | DL1; COB; SMT2; CSI1; THE1; DET3; FER | 3.9 10-05 | EM |
Heat tolerance | 6 | 2 | HOT5; HSP90.1; TPS5; AOS; APX1; HSP101 | 3.1 10-03 | NEC |
Plant morphology | 5 | 2 | ACT7; VLN2; DWF1; ABH1; ADF8 | 1.8 10-02 | EM |
Pollen germination | 5 | 2 | MIRO1; MAT3; ACT11; APY2; VDAC1 | 1.6 10-02 | EM |
Pollen tube growth | 5 | 2 | VLN2; MAT3; MIRO1; ACT11; FER | 4.2 10-02 | EM |
Response to osmotic stress | 5 | 3 | HCHIB; PLDALPHA1; TED4; PGK; HSP101 | 2.0 10-03 | NEC |
Male sterility | 5 | 4 | E1 ALPHA; PGM; GAPCP-2; GAMMA CA2; OPR3 | 1.2 10-03 | NEC |
Photorespiration | 5 | 6 | HPR; mMDH1; NIR1; GLU1; PGK | 1.7 10-04 | NEC |
Lignification | 4 | 2 | PRX72; ATCAD4; PRX52; CAD1 | 2.6 10-02 | NEC |
Regulation of cell size | 4 | 2 | CYT1; ERMO2; THE1; FER | 3.7 10-02 | EM |
Systemic acquired resistance | 4 | 2 | HOT5; AOS; OPR3; SGT1B | 3.9 10-02 | NEC |
Response to wounding | 4 | 3 | FDH; AOS; PLDALPHA1; OPR3 | 6.7 10-03 | NEC |
Grain filling | 4 | 4 | PGM; GLU1; SUS3; F5O11.31 | 3,0 10-03 | NEC |
Tricarboxylic acid cycle | 4 | 4 | MAT3; AOX1A; FUM1; PGK | 9.2 10-03 | EM |
RNA processing | 4 | 7 | PRMT4A; PRP39; ABH1; CLPC1 | 1.0 10-03 | EM |
Anthesis | 3 | 2 | PGM; SUS3; PGK | 3.4 10-02 | NEC |
Cell redox homeostasis | 3 | 2 | FDH; AOS; GSTF9 | 3.0 10-02 | NEC |
Glycolysis | 3 | 2 | PGM; mMDH1; PGK | 3.4 10-02 | NEC |
Pigmentation | 3 | 2 | DFR; PGLCT; BAN | 4.0 10-02 | NEC |
Fatty acid oxidation | 3 | 3 | AIM1; ACX1; PKT3 | 3.9 10-02 | EM |
Meristem maintenance | 3 | 3 | DL1; AGO10; AIM1 | 4.2 10-02 | EM |
Tricarboxylic acid cycle | 3 | 3 | mMDH1; ATCS; PGK | 2.5 10-02 | NEC |
Cell wall biosynthesis | 3 | 4 | DL1; COB; RGP1 | 1.7 10-02 | EM |
Regulation of cell shape | 3 | 4 | ACT7; COB; DWF1 | 1.7 10-02 | EM |
Gluconeogenesis | 3 | 5 | PGM; mMDH1; PGK | 6.4 10-03 | NEC |
Response to brassinosteroid stimulus | 3 | 5 | DWF1; EBS1; FER | 1.3 10-02 | EM |
Post-embryonic development | 3 | 6 | COB; UBC13; AGO10 | 6.7 10-03 | EM |
mRNA processing | 3 | 12 | PAB2; ECT2; ABH1 | 1.1 10-03 | EM |
Anther dehiscence | 2 | 3 | GAMMA CA2; OPR3 | 4.8 10-02 | NEC |
Autotrophy | 2 | 3 | GLU1; PGK | 5.0 10-02 | NEC |
Nitrate assimilation | 2 | 4 | NIR1; GLU1 | 2.8 10-02 | NEC |
Photooxidation | 2 | 4 | HOT5; HB1 | 3.5 10-02 | NEC |
Plant pathogen interaction | 2 | 4 | SFGH; AOS | 4.1 10-02 | NEC |
Translation initiation | 2 | 4 | FDH; HSP101 | 3.6 10-02 | NEC |
Cell polarity | 2 | 5 | DL1; SMT2 | 4.2 10-02 | EM |
Saccharification | 2 | 5 | ATCAD4; CAD1 | 2.2 10-02 | NEC |
Spindle assembly | 2 | 5 | IMPA-1; RANBP1 | 3.6 10-02 | EM |
Microtubule depolymerization | 2 | 6 | TUA5; PLDALPHA1 | 1.7 10-02 | NEC |
RNA metabolism | 2 | 6 | ABH1; ELF5A-1 | 3.1 10-02 | EM |
Response to hypoxia | 2 | 8 | ADH1; PGK | 1.1 10-02 | NEC |
Specification of stamen identity | 2 | 11 | AGO10; HEN2 | 8.1 10-02 | EM |
Bacterial disease resistance | 2 | 20 | FDH; HCHIB | 1.5 10-03 | NEC |
Endomitotic cell cycle | 2 | 22 | SMT2; GSL8 | 2.0 10-03 | EM |
Cell recognition | 2 | 25 | THE1; FER | 1.5 10-03 | EM |
In order to compare the proteomic and transcriptomic results, mapping of significantly differentially expressed transcripts (
Mapping of significantly differentially expressed (
Maturation yields of embryogenic lines of the three unrelated genotypes (SD4-8, TD15-1, TD17-1) varied from just 268 SEs g-1 f.w. (TD17-1) to 3942 SEs g-1 f.w. (SD4-8). Similar “genotype” effects are often observed in conifers (
The (Glc + Fru)/Suc ratio (
Phytohormones of auxin, ABA and CK classes were analyzed in EMs and NECs of the three Douglas-fir genotypes. Auxins and CKs were examined because they regulate cellular multiplication and differentiation, and changes in their levels correlate with multiplication phases of EMs and NECs. ABA is particularly involved in embryos’ differentiation and maturation (
Indole-3-acetic acid concentrations were higher in the EMs than in NECs although statistically significant differences were found only in two out of three lines. It suggests an important role of auxins in somatic embryogenesis in Douglas-fir. Auxins’ participation in maintenance of multiplying cells has been demonstrated in somatic embryogenesis of rubber tree (
Upregulation of 10 transcription factors of the
CKs play key roles in the initiation and further development of embryogenic cultures, as demonstrated by the requirement for aromatic derivatives in media during the induction of conifer somatic embryogenesis and subsequent proliferation (
Detected concentrations of isoprenoid CKs (largely
Among the differentially expressed genes, we identified an adenosine kinase (AK) upregulated in EMs, represented by a single upregulated (2-fold) transcript, PSME_00004950-RA, with a mapping score of 36556 ± 1772 (
To summarize, using advanced HPLC-MS methodology we analyzed differences in patterns and levels of about 20 CKs and CK derivatives in EMs and NECs of three Douglas-fir genotypes. To our knowledge, this provides the most comprehensive overview to date of profiles of endogenous CKs, including aromatic CKs, during proliferation of conifer SEs.
In addition to phytohormones involved in cell cycle regulation, levels of ABA and its (reversible) storage form ABA-GE were examined in EMs and NECs of the three tested genotypes. Levels of ABA and its derivatives are typically low in initial phases of embryonic development in conifers (
Transcripts upregulated in NECs included six (one, PSME_00028832-RA, with a 6-fold difference in expression and 9544 ± 2560 mapping score,
A set of 45 transcripts upregulated in EMs annotated to “structural constituent of ribosome” (MF GO:0003735,
Another set of 29 transcripts upregulated in EMs, annotated to GO:0016757, encodes CAZymes (
Transcripts upregulated in NECs included 48 involved in production of flavonoids (
Putative functions of transcripts associated with another GO term enriched in NECs, GO:0005215 (
Nearly all of 24 transcripts involved in production of monosaccharide sugar transporters annotated to the GO term MF GO:0022857 were significantly upregulated in NECs (
A set of upregulated ABC transporter transcripts associated with MF GO:0019001 (
Enrichment of MF GO:0008289 (
Among the proteins significantly different between EMs and NECs (
We analyzed our results by SNEA, a type of enrichment analysis based on protein interaction networks using daily updated bibliographic databases, connecting significantly differentially expressed proteins and BPs. Our results show that proteins more strongly expressed in EMs were associated with cell division and differentiation, embryonic development, protein synthesis and carbon metabolism, while those more strongly expressed in NECs were associated with defense and stress responses, and to a lesser extent primary (carbon and nitrate) metabolism (
Cell divisions are usually more frequent in embryogenic lines, which grow more rapidly than non-embryogenic lines. Accordingly, microscopic comparison revealed large numbers of mitotic meristematic cells in our EMs (
The large number of BPs associated with numerous upregulated proteins and corresponding transcripts demonstrates the very strong orientation of EM tissues toward cell differentiation, which could be considered their main characteristic relative to NEC. This conclusion is new regarding the comparison studies between embryogenic and non-embryogenic tissues.
Active carbon and energy metabolism play key roles in embryogenicity by supporting cell division and modification. Hence, more than 30% of the significantly differentially expressed proteins detected in both this (
Protein synthesis and processing play key roles in all cellular development processes, and thus differ between non-embryogenic and EM tissues (
The importance of the number of proteins involved in protein metabolism, overexpressed in EM compared to NEC, suggest that these BPs are a key factor in embryogenic competence and somatic embryogenesis.
Defense responses against biotic or abiotic stress modulate embryogenic culture capacity or somatic embryogenesis (
Cell cultures as well as plants develop against pathogen attack various reactions involving notably pathogenesis-related (PR) proteins (
Reactive oxygen species are readily detectable in plants’ responses to abiotic and biotic stresses, which may be key triggers of somatic embryogenesis (
These stress defense reactions have been interpreted as responses of the tissues to
To summarize, in proteomic analysis the establishment of a network of interactions between significant proteins was proven very useful. In addition, our study confirms again the importance of working with modern proteomics techniques and with several unrelated genotypes to screen for proteins involved in embryogenic state.
Numerous differences between EM and NEC of Douglas-fir were observed at the cytological, biochemical and molecular levels. Key characteristics of Douglas-fir EMs may include cell multiplication and differentiation of embryogenic tissue, supported by enhancement of energy metabolism and protein recycling machinery, while upregulation of stress responses may be more characteristic of NECs. In addition, numerous differences between EMs and NECs of the three genotypes indicate that auxin, bioactive and transport forms of isoprenoid and aromatic CKs, as well as isoprenoid CK nucleotides, may be markers of EM formation. In contrast, NECs were characterized by high levels of ABA, ABA-GE and glycosylated CKs (both isoprenoid and aromatic). Finally, the study illustrates the value of comprehensive multi-level comparisons of isogenic embryogenic and non-embryogenic lines, with advanced methodology. Our innovative application of network analysis also contributed to our results, which provide the first report providing integrated insights into cellular, biochemical and molecular events involved in embryogenesis in conifers, and more specifically cellular embryogenic state in Douglas-fir.
FG participated in the acquisition of all the data, as well as somatic embryogenesis, and transcriptomic and protein analysis. PL participated in design of the study and transcriptomic analysis. KE performed histological and microscopical analyses. J-CL participated in Sub-Network Enrichment analysis. VM analyzed phytohormone data. NB performed carbohydrate analysis. ZV carried out somatic embryogenesis. JM performed LC-MS phytohormone analysis. AT pre-treated samples for phytohormone analysis. CL carried out somatic embryogenesis and collected the material. M-CL-D helped in the transcriptomic analysis. A-ML carried out mass spectrometric analysis. J-FT participated in design of the study. GC participated in design of the study. CT participated in design of the study, protein and Sub-Network Enrichment analysis. M-AL-W participated in design of the study, coordinated it and participated in somatic embryogenesis. FG, PL, KE, J-CL, VM, NB, ZV, M-CL-D, A-ML, J-FT, GC, CT, and M-AL-W also contributed to writing of the manuscript, and all authors read and approved the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The Supplementary Material for this article can be found online at:
abscisic acid
ABA-glucose ester
cytokinins
dry weight
embryonal mass
fresh weight
Gene Ontology
indole-3-acetic acid
liquid chromatography-tandem mass spectrometry
non-embryogenic callus
reactive oxygen species
somatic embryo
Sub-Network Enrichment Analysis