Changes in chlorophyll content and chlorophyll fluorescence
Under the plain environment, the chlorophyll content increased significantly on the first and fifth days. Although the control level was restored on day 3, it was lower than that of the control on day 7 (Fig. 1a). Chlorophyll fluorescence (Fv/Fm) was significantly higher on the 5th day and lower on the 7th day than on the day before relocation (Fig. 1b).
Changes in the number of stomata and in the anatomical structure of leaves
Compared with the control, the number of stomatal openings decreased significantly on the first day, increased significantly on the third day, and reached a relatively stable state on the fifth and seventh days (Fig. 2 a-f). Common vetch is a double-sided leaf with palisade tissue and spongy tissue. On the first day of returning to the plain, the arrangement of upper and lower epidermal cells, palisade cells and spongy cells was loose, and the intercellular space was larger than that of the control. In the next few days, the intercellular space and spongy tissue gradually recovered (Fig. 3 a-e). Therefore, the changes in stomata and anatomical structure showed that the change in common vetch was obvious after it was transported back to the plain, and it was difficult to return to the control state.
Physiological response of common vetch to the plain environment
Compared with common vetch in the suitable plateau environment (J0), common vetch transported back to the plain showed higher MDA and soluble sugar contents; these contents decreased significantly on the first day (J1) and then increased significantly. The content of soluble protein decreased significantly and then gradually recovered to the same level as on the third day (J3). The levels of H2O2 and O2- did not change significantly but increased. The activities of POD and GSH decreased significantly on the 5th day (J5) and the 1st day (J1), respectively, but remained the same as the control. The activity of SOD increased significantly; although it decreased significantly on the 5th day (J5), it was still higher than that of the control. In summary, these results show that the damage to the cell membrane and osmotic regulation system of high-altitude adapted species is serious in the plain environment, but it has no significant effect on ROS levels.
RNA sequencing and de novo assembly
After the seedlings were transported to the plain, the photosynthetic pigment changed significantly on the third day and gradually recovered to stability on the seventh day. Therefore, to obtain the gene expression profile under plain and plateau conditions, the transcriptomes of the leaves of plateau control samples (J0) and plain samples at 3 and 7 days (J3, J7) were analyzed via an Illumina platform. From de dovo assembly, a total of 129,209 transcripts and 38,702 unigenes were obtained with an average length of 1552 bp. We recorded the N50 value of 2394 bp (Tab. 1). The unigene length ranged from 301 to 16,715 bp. Among the unigenes, 10973 (28.35%) were more than 2000 bp in length, and there was no single gene shorter than 300 bp. A total of 8789 (22.71%) were 300-500 bp in length, 9585 (24.77%) were 500 bp-1000 bp in length, and 9355 (24.17%) were 1000 bp-2000 bp in length (Tab. 2).
Table 1 Illumina RNA-seq reads and de novo assembly statistics of common vetch.
Total number of raw reads
|
247,896,340
|
Total number of clean reads
|
60,060,559
|
Mean length of reads(bp)
|
1919
|
Number of total unigenes
|
38,702
|
Mean length of unigenes (bp)
|
1552
|
Minimum unigene length (bp)
|
301
|
Maximum unigene length (bp)
|
16,715
|
N50 (bp)
|
2394
|
Table 2 Length distribution of the unigenes.
Length (nt)
|
Total
|
< 300
|
300–500
|
500-1000
|
1000-2000
|
≥2000
|
Number of Unigenes
|
38702
|
0
|
8789
|
9585
|
9355
|
10973
|
%
|
100%
|
0%
|
22.71%
|
24.77%
|
24.17%
|
28.35%
|
Functional annotation
Among the 38,702 unigenes, a total of 29,384 unigenes were annotated in at least one database, including 25,225 unigenes annotated in the NR database, 25,201 unigenes in NT, 18,771 unigenes in PFAM, 18,771 unigenes in GO, and 6042 unigenes in the KOG database (Fig. 5a).
Compared with the annotation results of the NR database, it can be seen that the most homologous species with common vetch is Medicago truncatula, with 8391 species (33.3%). In addition, 18.4% of unigenes were similar to those of Trifolium pratense (Additional file 1: Figure S1.). In accordance with GO analysis, 105146 single genes were classified according to the three major categories of GO. There were 25 groups of biological processes (BPs), 20 groups of cellular components (CCs) and 10 groups of molecular functions (MFs) (Fig. 5b, additional file 2: Table S1). The most abundant GO terms in the unigenes of common vetch were “metabolic process” (10131), “cellular process” (11068) and “single-organism process” (8056) for BP; “cell part” (6127) and “cell” (6127) for CC; and “catalytic activity” (8350) and “binding” (10,690) for MF.
After KEGG annotation, 8185 DEGs were mapped into 130 KEGG metabolic pathways. The top three KEGG pathways were "metabolism" (4870), "genetic information processing" (2099) and "environmental information processing" (412). The proportion of genes in the total number of annotated genes was ribosome (ko03010, 510 single genes), plant pathogen interaction (ko04626, 304 single genes), carbon metabolism (ko01200, 3011 single genes) and plant hormone signal transduction (ko04075, 274 single genes) (Fig. 5c, Additional file 3: Table S2.).
Overall analysis of DEGs
In this experiment, we used deseq software to analyze the differential expression among the sample groups, with FDR < 0.05. Compared with the control (J0), samples collected on the third day (J3) had 3251 upregulated genes and 3317 downregulated genes, and 1359 upregulated genes and 1648 downregulated genes were observed on the seventh day (Fig. 6a). A total of 443 genes showed the same change in three comparisons when comparing the DEGs on different days. Compared with the control (J0), there were 1491 DEGs on the 3rd and 7th days (Fig. 6b). To confirm this result, 8 genes with different transcription abundances were verified by real-time RT-PCR. The gene expression levels obtained with the two methods were consistent.
In the J3 vs J0 comparison, the top seven positions of the KEGG enrichment pathway are shown in Fig. 7a, Additional file 4: Table S3 (Q value < 0.01), namely, 28 in photosynthesis (ko00195), 15 in photosynthetic antenna protein (ko00196), 25 in photosynthetic biological carbon fixation (ko00710), 16 in phenylalanine metabolism (ko00360), 15 in porphyrin and chlorophyll metabolism (ko00860), and 14 in tyrosine metabolism (ko00350).
In the J3 vs J0 comparison, the top four positions of the KEGG enrichment pathway are shown in Fig. 7a, Additional file 5: Table S4 (Q value < 0.01), 206 in ribosomes (ko03010), 36 in photosynthesis (ko00195), 43 in ribosome biogenesis in eukaryotes (ko03008), and 21 in photosynthesis-antenna proteins (ko00196).
The pathways photosynthesis (ko00195) and photosynthesis-antenna proteins (ko00196) were identified in the J3 vs J0 and J7 vs J0 comparisons. The pathways identified in only the J3 vs J0 comparison were carbon fixation in photosynthetic organisms (ko00710), phenylalanine metabolism (ko00360), porphyrin and chlorophyll metabolism (ko00860), and tyrosine metabolism (ko00350). Pathways specifically in J7 vs J0 are ribosome (ko03010) and ribosome biogenesis in eukaryotes (ko03008).
Differential expression of photosynthesis-related genes
To study the adaptability of plateau plants to the plain environment, genes related to environmental effects were screened out from the filtered DEGs for further study. In J3 and J7, DEGs related to the reaction center of the PSI and PSII, photosynthetic antenna protein, electron transfer and CO2 fixation were identified. The main pathways were photosynthesis (ko00195), photosynthesis antenna protein (ko00196) and photosynthesis biological carbon fixation (ko00710). When the plants were transported to the plain environment, the carbon fixation of photosynthetic organisms was inhibited. The main feature is that the core cytochrome b6/f (k02636), which connects PSII and PSI, is downregulated during electron transfer. During photophosphorylation, the genes encoding FD-NADP+ reductase (k02641) and ATP synthase (k02113, k02109) were downregulated (Fig. 8b).
During the Calvin cycle (Fig. 8a), the genes encoding Rubisco (k01601) in the pathway of 3-phosphoglycerate (PGA) synthesis were downregulated. The genes encoding phosphoglycerate kinase (k00927) and glyceraldehyde-3-phosphate dehydrogenase (k05298) were downregulated in the synthesis of 3-phosphoglyceraldehyde (GAP). Many enzymes involved in the regeneration of ribulose-1,5-disphosphate (RuBP) were downregulated. During the C4 cycle, oxaloacetic acid (OAA) is reduced to malic acid; the gene encoding malate dehydrogenase (k00025) is downregulated. Regarding regeneration, the genes encoding pyruvate transaminase (k00814) were downregulated. However, the genes encoding aspartate transaminase (k14454), involved in transamination, were upregulated in the J3 vs J0 comparison (Fig. 8d). Lhca2, Lhca4, Lhcb1, Lhcb2, Lhcb3, Lhcb4, Lhcb5 and Lhcb6 in the photosynthetic antenna protein pathway, components of the complementary light complexes LCHI and LCHII, were downregulated. In addition, the genes encoding proteins in the reaction center of photosynthesis were downregulated, including PSI reaction center subunits III-VI (PsaF, PsaE, PsaG, PsaH), PSI reaction center subunits X and XI (PsaK, PsaL), PSI reaction center subunits PsaN and PsaO, and PSII reaction center PsbK, PsbO, PsbQ, PsbW, and Psb27. In conclusion, the downregulation of these three pathways inhibited the growth of common vetch in the plain environment.
Differential expression of encoding photosynthetic pigment-related genes
The adaptability of plants to light is closely related to the photosynthetic pigment production pathway (Fig. 8c). In this study, the DEGs were mainly in the porphyrin and chlorophyll metabolism (ko0860) pathway. The genes encoding glutamine tRNA reductase (k02492), hydroxymethyldialane synthase (k01749), uroporphyrinogen decarboxylase (k01599), Mg-chelatase subunit H (k03403), magnesium-protoporphyrin O-xymethyltransferase (k03428), magnesium-protoporphyrin IX monomethyl ester (oxidative) cyclase (k04035), protochlorophyllide reductase (k00218), chlorophyll An oxygenase (k13600) and chlorophyll a synthase (k04040) were downregulated in J3 compared with the control.
RNA-seq validation: qRT-PCR analysis
We used qRT-PCR to detect the consistency of the expression patterns of 8 randomly selected genes. GAPA (Cluster-7705.13598, Cluster-7705.13381, Cluster-7705.13418), rbcS (Cluster-7705.13113), ATPF0B (Cluster-7705.13726), ATPF1D Cluster-7705.13438), petH (Cluster-7705.13612), and petC (Cluster-7705.14151) were found to be downregulated in J3 vs J0 and J7 vs J0 by qRT-PCR. The qRT-PCR results for all genes were consistent with those of RNA-Seq, which provided strong credibility for our RNA-Seq data. (Fig 9)
DEGs encoding TFs
There were more TFs in J7 than in J3. In J3, 109 upregulated TF genes were divided into 39 families, mainly NAC (11.01%), Orphans (7.34%), and HB (7.34%), and 94 downregulated genes were divided into 32 families, including MYB (15.96%), AP2-EREBP (10.64%), and orphans (8.51%) (Fig. 10a, Additional file 6: Table S5.). However, 206 upregulated TF genes in J7 were divided into 52 families, mainly AP2-EREBP (12.62%), mTERF (8.25%), MYB (7.77%), and Orphans (5.34%), and 238 downregulated genes were divided into 40 families, including MYB (13.45%), AP2-EREBP (9.66%), and NAC (8.40%) (Fig. 10b, Additional file 7: Table S6.). These results indicate that among all of these differentially expressed TFs, a large number of TFs of the MYB, NAC, AP2-EREBP and Orphans families were differentially expressed regardless of the number of days of adaptation after the move to the plain environment. These TFs may play an important role in the acclimation of common vetch to the plain environment.