Phenotypic analysis of different barley cultivars after bikinin application
To assess the function of the Glycogen Synthase Kinase 3 (GSK3) family in barley development, we applied three different concentrations of the bikinin, to two unrelated barley cultivars with distinct sensitivities to exogenous Brassinosteroids (BRs): Golden Promise (GP) and more BR-sensitive Haruna Nijo (HN) (Groszyk and Szechyńska-Hebda 2021b). Phenotypic differences between both barley genotypes were examined during the stage before tillering, up to the beginning of the 5th leaf development stage (in the development phase BBCH20), as previously described (Groszyk and Przyborowski 2022). In this study, we focus on the long-term response of two barley cultivar seedlings treated with 10 µM bikinin (BK10), 50 µM bikinin (BK50), and 100 µM bikinin (BK100). The eight-day-old GP seedlings and nine-day-old HN seedlings (BBCH12) were sprayed with bikinin and 0.11% DMSO (CK). After ten days science spraying, the plants were phenotypically characterized, including measurement of the fresh and dry weight of shoot and root; length of shoot and root; length and area of the 1st, 2nd, 3rd, and 4th leaf; relative water content; chlorophyll fluorescence; HvBZR1 and HvGSK2.1 protein activity. Additionally, for selected samples, a transcriptome analysis was performed.
Similarly, to the previous study using these two cultivars, after applications of bikinin concentrations, we observed genotype- and treatment-dependent reactions determined by barley phenotypic traits (Table 1). The application of various concentrations of bikinin led to genotype- (p < 0.000) and treatment-dependent (p < 0.002) reactions, resulting in phenotypic changes in both barley cultivars (Fig. 1a, b). For GP, the fresh weight of the shoot increased in BK10 to 116.6% and decreased in BK100 to 93.9%, with no significant changes in BK50 plants (Fig. 1c). All bikinin concentrations inhibited the accumulation of fresh weight in the HN shoot, resulting in 97.8%, 99.6%, and 80.5% weight for BK10, BK50, and BK100 barley, respectively (Fig. 1c). The reaction of the barley genotypes to bikinin application was significantly reflected in the dry weight of the GP (BK10, 121.6%; BK50, 108.6%; BK100, 101.7%; r = 0.967, p ≤ 0.000) and HN (BK10, 94.1%; BK50, 91.5%; BK100, 75.7%%; r = 0.849, p ≤ 0.000) shoot (Fig. 1d). Compared to the weight parameters, changes after consecutive doses of bikinin were observed during the shoot length analysis (Fig. 1e). The lower concentration of bikinin led to the development of shorter shoots than the highest concentrations of bikinin in GP (BK10, 106.0%; BK50, 107.5%; BK100, 120.0%). In HN, a longer shoot developed in BK10 and BK50 plants (105.3% and 113.1%, respectively), while in BK100 led to results similar to CK plants (99.7%). The response of barley genotypes to bikinin treatment significantly differed for each genotype and bikinin concentration when analyzing the length and area of consecutive leaves (Fig. 2). The use of 50 µM and 100 µM bikinin significantly reduced the development of the first two leaves of HN. The length of the first and second leaf reduced to 80.7%-89.2%, while the area decreased to 82.7%-89.0% compared to control plants (Fig. 2a, b, e, f). Analysis of the first two leaves of GP after all concentrations of bikinin and BK10 HN resulted in parameters comparable to CK. Subsequently, newly unfolded leaves in HN showed negligible changes in the 3rd leaf length and area (Fig. c, g); slightly larger differences were observed in the 4th leaf, which was longer (108.3%) and had a larger area (108.8%) in BK10 (Fig. 2d, h). Significant changes were observed in BK50 HN, where application resulted in an increase in 4th leaf length (115.6%) and area (124.6%), as well as after the application of 100 µM bikinin that inhibited both parameters to 95.8% and 82.3%, respectively (Fig. 2d, h). In this cultivar, the length of the 4th leaf determined shoot length (r = 0.716, p ≤ 0.000). A greater variation in response to different concentrations of bikinin was observed in sequentially developing GP leaves. While the following has been applied, the length and area of the 3rd and 4th leaf were treatment-dependent (Table 1). 10 µM bikinin application resulted in negligible changes 3rd leaf (Fig. 2c) and longer 4th leaf with a greater area (135.4% and 158.0%, respectively) (Fig. 2d, e). 50 µM bikinin resulted in a significantly longer 3rd leaf (110.4%) about a larger area (106.3%) (Fig. 2c, g), with slightly reduced parameters of the 4th leaf, i.e. length to 95.3% and area to 89.5% (Fig. d, h). The greatest changes in the 3rd and 4th leaves were detected in BK100 plants (Fig. 2c, d, g, h). The 3rd and 4th leaves were significantly longer (132.4% and 122.3%, respectively), with the area of 3rd leaf increased to 108.2%, but the 4th leaf reduced to 91.7%. The data indicate that the length of the 3rd GP leaf significantly influences the length of the GP shoot (r = 0.812, p ≤ 0.000). Relative water content indicates that all plants present good conditions (Supplemental Fig. S1).
Table 1. Results of two-way ANOVA for two barley genotypes Golden Promise and Haruna Nijo treated with bikinin in three different concentrations, i.e., 10 µM, 50 µM, and 100 µM compared to plants treated with 0.11% DMSO calculated for aboveground and below-ground organ traits.
Parameters
|
Genotype
|
Treatment
|
Genotype * Treatment
|
MS
|
F
|
p
|
MS
|
F
|
p
|
MS
|
F
|
p
|
Shoot FW
|
4.5993
|
55.53
|
0.0000
|
0.3926
|
4.74
|
0.0024
|
0.1225
|
1.54
|
0.2054
|
Shoot DW
|
0.0996
|
132.22
|
0.0000
|
0.0048
|
6.31
|
0.0003
|
0.0037
|
4.91
|
0.0020
|
Shoot length
|
192.6042
|
29.04
|
0.0000
|
26.8413
|
4.05
|
0.0061
|
41.1229
|
10.62
|
0.0000
|
1st leaf length
|
517.9635
|
365.71
|
0.0000
|
3.8981
|
2.75
|
0.0372
|
6.4365
|
6.34
|
0.0003
|
2nd leaf length
|
294.1653
|
166.37
|
0.0000
|
2.8450
|
1.61
|
0.1866
|
7.8889
|
4.46
|
0.0037
|
3rd leaf length
|
4.7540
|
1.67
|
0.2020
|
28.2101
|
9.92
|
0.0000
|
24.3518
|
8.56
|
0.0000
|
4th leaf length
|
722.6969
|
97.78
|
0.0000
|
28.2535
|
3.82
|
0.0083
|
43.9528
|
9.84
|
0.0000
|
1st leaf area
|
524.7965
|
554.07
|
0.0000
|
2.6711
|
2.82
|
0.0346
|
2.8786
|
3.04
|
0.0255
|
2nd leaf area
|
546.1494
|
273.25
|
0.0000
|
4.8476
|
2.77
|
0.0374
|
5.0683
|
2.89
|
0.0314
|
3rd leaf area
|
30.6664
|
16.23
|
0.0002
|
0.4590
|
0.24
|
0.9127
|
5.0400
|
3.08
|
0.0242
|
4th leaf area
|
520.7232
|
105.84
|
0.0000
|
68.6702
|
13.96
|
0.0000
|
52.0185
|
10.57
|
0.0000
|
Root FW
|
0.0005
|
0.03
|
0.8537
|
0.1423
|
10.77
|
0.0000
|
0.0168
|
1.30
|
0.2820
|
Root DW
|
0.0064
|
50.31
|
0.0000
|
0.0015
|
11.52
|
0.0000
|
0.0001
|
0.60
|
0.6664
|
Root length
|
8.8935
|
0.38
|
0.5407
|
20.9898
|
0.90
|
0.4738
|
37.4568
|
1.60
|
0.1895
|
RWC
|
9.0561
|
8.30
|
0.0057
|
4.2324
|
3.88
|
0.0077
|
1.4891
|
1.41
|
0.2454
|
DWF4
|
0.0000
|
2.15
|
0.1577
|
0.0000
|
1.72
|
0.1856
|
0.0000
|
3.49
|
0.0255
|
BRI1
|
0.0003
|
4.43
|
0.0460
|
0.0001
|
0.69
|
0.6042
|
0.0001
|
1.16
|
0.3576
|
BAK1
|
0.0006
|
330.42
|
0.0000
|
0.0000
|
0.71
|
0.5919
|
0.0000
|
1.51
|
0.2385
|
BSU1
|
0.0000
|
0.00
|
0.9447
|
0.0001
|
0.44
|
0.7781
|
0.0001
|
0.42
|
0.7916
|
GSK1.1
|
0.0028
|
0.83
|
0.3702
|
0.0009
|
0.28
|
0.8908
|
0.0025
|
0.70
|
0.5994
|
GSK1.2
|
0.0021
|
0.91
|
0.3486
|
0.0010
|
0.42
|
0.7932
|
0.0034
|
1.62
|
0.2072
|
GSK1.3
|
0.0020
|
4.55
|
0.0433
|
0.0001
|
0.19
|
0.9394
|
0.0004
|
0.87
|
0.4991
|
GSK2.1
|
0.0273
|
8.06
|
0.0091
|
0.0007
|
0.21
|
0.9302
|
0.0029
|
0.84
|
0.5135
|
GSK2.2
|
0.0017
|
10.86
|
0.0030
|
0.0001
|
0.86
|
0.5026
|
0.0002
|
1.54
|
0.2292
|
GSK3.1
|
0.0010
|
2.80
|
0.1072
|
0.0001
|
0.41
|
0.8030
|
0.0005
|
1.57
|
0.2222
|
GSK4.1
|
0.0074
|
3.07
|
0.0952
|
0.0021
|
0.89
|
0.4883
|
0.0044
|
1.81
|
0.1670
|
BZR1
|
0.0000
|
1.02
|
0.3243
|
0.0000
|
0.47
|
0.7537
|
0.0000
|
0.59
|
0.6713
|
FV/FM
|
0.0008
|
34.26
|
0.0000
|
0.0001
|
4.78
|
0.0030
|
0.0000
|
0.86
|
0.4937
|
φP0
|
0.0008
|
34.26
|
0.0000
|
0.0001
|
4.78
|
0.0030
|
0.0000
|
0.86
|
0.4937
|
Ψ0
|
0.0175
|
94.85
|
0.0000
|
0.0007
|
3.85
|
0.0098
|
0.0007
|
3.55
|
0.0143
|
φE0
|
0.0149
|
106.25
|
0.0000
|
0.0007
|
4.70
|
0.0033
|
0.0005
|
3.48
|
0.0158
|
φD0
|
0.0008
|
34.26
|
0.0000
|
0.0001
|
4.78
|
0.0030
|
0.0000
|
0.86
|
0.4937
|
PIABS
|
4.3672
|
78.27
|
0.0000
|
0.2853
|
5.11
|
0.0020
|
0.0760
|
1.36
|
0.2640
|
ABS/RC
|
0.0274
|
11.25
|
0.0018
|
0.0175
|
7.18
|
0.0002
|
0.0079
|
3.24
|
0.0215
|
TR0/RC
|
0.0063
|
4.80
|
0.0343
|
0.0081
|
6.14
|
0.0006
|
0.0061
|
4.64
|
0.0036
|
ET0/RC
|
0.0235
|
53.66
|
0.0000
|
0.0023
|
5.31
|
0.0016
|
0.0051
|
11.67
|
0.0000
|
DI0/RC
|
0.0073
|
27.96
|
0.0000
|
0.0020
|
7.79
|
0.0001
|
0.0001
|
0.50
|
0.7334
|
Two-way ANOVA indicates that depending on bikinin concentrations, shoot weight influenced root weight (p ≤ 0.001) (Table 1). The application of 10 µM bikinin in long-term response resulted in nonsignificant changes; however, 50 µM and 100 µM bikinin led to a reduction in fresh weight (BK50: 85.7% and 68.6%; BK100: 67.4% and 50.3%) and dry weight (BK50: 89.0% and 79.7%; BK100: 60.2% and 66.8%) in GP and HN, respectively (Fig. 1f, g). The length of the roots in GP and HN with consecutive concentrations of bikinin was not significant, with one exception; HN had longer roots in BK10 (108.3%, p ≤ 0.05) (Fig. 1h).
Barley transcriptome analysis after bikinin treatment
To elucidate the molecular mechanisms of barley after BK treatment, 3rd and 4th leaves were collected from GP and HN plants following CK, BK10, and BK100 variants. Each biological samples underwent three replicate treatments, resulting in a total of 18 libraires constructed for transcriptome analysis. Utilizing DNB sequencing technology (BGI Genomics Co., Ltd.), each sample generated over 60 million raw pair-end reads with a size 100 bp (Table 2). Following quality control, each sample yielded a minimum of 60,187,800 (GP BK100_1) and a maximum 73,790,912 (GP BK100_3) clean reads, with an average Q20 of approximately 96% (Table 2), indicative of high-quality sequencing data in the RNA-seq experiments. After the removal of adaptor sequences and low-quality reads, all clean reads were mapped to the barley reference genome GPv1 (Schreiber et al. 2020) using HISAT2 software (Kim et al. 2019). The results demonstrated that, on average, 32.89 million reads were mapped to the reference genome, with approximately 30.94 million reads being uniquely mapped. We identified 58,815 coding type genes, 3,126 nontranslating CDS, and 127 identified as a pseudogene (Supplemental Fig. S2). Principal component analysis (PCA) was performed on the RNA-seq dataset of 18 samples. The control samples of barley cultivars (HN and GP) and two doses of bikinin, 10 µM (BK10) and 100 µM bikinin (BK100) were distinctly differentiated by the first principal component (PC1), explaining 50.35% of the total variation, and PC2, which accounted for 11.89% of the total variation (Fig. 3a).
Table 2
Summary of RNA sequencing results.
Genotype | Treatment | Clean reads | Clean beases | Clean bases (GB) | Q20 (%) | GC (%) |
GP | CK | 68538918 | 6853891800 | 5.36 | 95.91% | 54.68% |
GP | CK | 70336000 | 7033600000 | 5.48 | 95.94% | 54.34% |
GP | CK | 72421912 | 7242191200 | 5.61 | 96.11% | 53.92% |
GP | BK10 | 64398886 | 6439888600 | 5.02 | 95.81% | 54.43% |
GP | BK10 | 65148758 | 6514875800 | 5.07 | 96.12% | 54.58% |
GP | BK10 | 73043922 | 7304392200 | 5.66 | 96.22% | 53.94% |
GP | BK100 | 60187800 | 6018780000 | 4.67 | 95.90% | 54.62% |
GP | BK100 | 65715288 | 6571528800 | 5.07 | 96.24% | 54.47% |
GP | BK100 | 73790912 | 7379091200 | 5.72 | 96.15% | 54.11% |
HN | CK | 64174626 | 6417462600 | 4.92 | 96.28% | 53.78% |
HN | CK | 64348198 | 6434819800 | 4.95 | 96.19% | 54.10% |
HN | CK | 71423906 | 7142390600 | 5.53 | 96.12% | 54.83% |
HN | BK10 | 67858502 | 6785850200 | 5.22 | 96.24% | 53.40% |
HN | BK10 | 68091566 | 6809156600 | 5.28 | 96.13% | 54.08% |
HN | BK10 | 67886934 | 6788693400 | 5.26 | 96.09% | 54.44% |
HN | BK100 | 73010038 | 7301003800 | 5.60 | 96.25% | 53.73% |
HN | BK100 | 71828166 | 7182816600 | 5.55 | 96.32% | 54.52% |
HN | BK100 | 71812666 | 7181266600 | 5.54 | 96.23% | 53.98% |
Identification of DEGs regulated by bikinin
Differentially expressed genes (DEGs) between the combinations GP_BK10-GP_CK, GP_BK100-GP_CK, GP_BK100-GP_BK10, HN_BK10-HN_CK, HN_BK100-HN_CK, HN_BK100-HN_BK10 were identified for FDR < 0.01 and log2FC| > 2) (Fig. 3b). The DEGs were identified by comparing the expression levels of genes in GP and HN barley in CK, BK10 and BK100 variants, thereby elucidating genes associated with the long response of barley to bikinin treatment. Among the total 272 significant DEGs, 160 genes were upregulated (Fig. 3c), and 112 genes were downregulated (Fig. 3d) in GP, while in HN, 37 genes were upregulated (Fig. 3e), and 49 genes were downregulated (Fig. 3f) after bikinin treatment. When DEGs were compared between all the above bikinin variants, 2 upregulated and 2 downregulated genes in GP were common for BK10 and BK100 (Fig. 3c, d). In contrast to the common DEGs, 13 upregulated and 2 downregulated genes were specific for BK10 GP; 145 upregulated and 108 downregulated genes were specific for BK100 GP; 26 upregulated and 16 downregulated genes were specific for BK10 HN; and 11 upregulated and 33 downregulated were specific for BK100 HN (Fig. 3e, f). Moreover, by comparing the transcriptome profiles of barley cultivars in BK10 and BK100, a total of 78 upregulated and 162 downregulated genes were identified in GP, and 68 upregulated and 234 downregulated genes were identified in HN, suggested treatment-dependent reaction of barley to bikinin concentration. Furthermore, a total of 54 induced and 59 repressed genes in GP, and 5 induced and 23 repressed genes in HN were identified as common DEGs between barley treated with 100 µM bikinin, suggesting that both concentrations of bikinin lead to opposite reactions.
Functional annotation of bikinin-induced DEGs
To elucidate the function of bikinin-responsive genes in barley cultivars, an enrichment analysis of Gene Ontology (GO) function was conducted on DEGs. The results revealed that 5 DEGs specifically detected in BK10 GP, annotated to 10 GO terms in the molecular function category (Fig. 4a). Among the upregulated genes, annotations fell under 3 GO terms: intramolecular lyase activity, isomerase activity, and hydroxymethylglutaryl-CoA reductase (NADPH) activity. 1 of the 2 downregulated genes was annotated under 4 GO terms, including alpha-glucosidase activity, glucosidase activity, sucrose alpha-glucosidase activity, and beta-fructofuranosidase activity. A total of 9 DEGs in BK10 HN were annotated to 5 GO terms in the molecular function category (Fig. 4b). The upregulated genes were annotated to DNA-binding transcription factor activity, transcription regulator activity, and glutathione transferase activity. Moreover, 10 µM bikinin regulated the activities of carbon-oxygen lyase activity pathways in both barley genotypes but in specific parts. In GP, 1 gene regulated 3 GO terms: tryptophan synthase activity, hydro-lyase activity, and carbon-oxygen lyase activity. In HN, the gene regulated terpene synthase activity, and carbon-oxygen lyase activity acting on phosphates. After 100 µM bikinin treatment, a greater number of DEGs were identified in GP compared to HN. In GP after the application of 100 µM bikinin, 67 DEGs were detected and annotated to 41 GO terms across 3 categories: biological processes, molecular function, and cellular components (Fig. 4c). Under biological processes, 37 genes were related to the organic substance biosynthetic process, with 35 genes of them annotated to the biosynthetic process and cellular biosynthetic process. The most abundant terms were amine and peptide biosynthetic processes, translation, and cellular amide and peptide metabolic processes. In the molecular function category, 20 genes belonged to oxidoreductase activity, and 15 genes were associated with structural constituent of ribosome. In the cellular component category, the 24 genes mainly fell under intracellular non-membrane-bounded organelle, ribosome, and extracellular region. For BK100 HN, 3 genes were uniquely differentially expressed and functionally annotated. These DEGs were enriched in 8 GO terms across the biological processes and molecular functions categories (Fig. 4d). 1 gene was annotated to the biological process of terpenoid metabolic process, and the other 2 genes were associated with molecular functions, including inositol heptakisphosphate kinase activity and fatty-acyl-CoA reductase (alcohol-forming) activity. The overall distribution between GP and HN with bikinin exhibited differences, with the number of genes involved in each GO term and the number of GO terms being genotype- and treatment-dependent.
Photosystem II efficiency
A total of 2 genes uniquely differentially expressed in the GP treated with 100 µM bikinin were functionally annotated to biological processes under GO terms, specifically related to photosynthetic electron transport in photosystem II (PSII) (Fig. 4c). Analysis of normalized data counts showed that the transcript levels of these genes were upregulated after 100 µM bikinin application, reaching 141.61% and 123.43% in GP and 117.81% and 112.67% in HN, respectively (Fig. 6a). Evaluation of consecutive parameters of chlorophyl fluorescence (Fig. 6b) revealed that GP treated with BK100 exhibited higher light absorption (ABS/RC) compared to HN. Simultaneously, increased dissipation of light energy as a heat (DI0/RC, φD0) resulted in a reduced performed index (PIABS) in GP. Lower trapped energy flux (ET0/RC, TR0/RC) in HN contributed to a decreased photosynthetic electron transport (Fig. 5b, c). The quantum yield of the primary PSII photochemistry and electron transport (φP0, φE0, Ψ0) was negligible in the leaves of both barley cultivars. Conversely, changes in PSII efficiency under bikinin treatment in the fourth leaf were impaired, suggesting that plants were readjusting to homeostasis. Statistical analysis indicated that the parameters of the efficiency of PSII, such as quantum yield of the primary PSII photochemistry (FV/F0), efficiency and flow of energy (FV/FM, ψ0, φP0, φE0, φD0), and performance index (PIABS), depended on genotype, bikinin concentrations, and barley responses to bikinin treatment (Table 1).
Gene expression analysis and protein activity
Consecutive Rael-Time PCR analysis of few genes controlling the BR biosynthesis pathway, HvDWF4 (Fig. 6a), BR signaling pathway, HvBRI1, HvBAK1, HvBSU1, HvGSK1.1, HvGSK1.2, HvGSK1.3, HvGSK2.1, HvGSK2.2, HvGSK3.1, HvGSK4.1 (Fig. 6b-k), and transcription factor, HvBZR1 (Fig. 6L) indicates, that only five of them – HvBRI1, HvBAK1, HvGSK1.3, HvGSK2.1, and HvGSK2.2 – exhibit a genotype-dependent response (Table 1); however, none of them shows a reaction dependent on bikinin concentrations (Table 1).
To elucidate the role of bikinin in the regulation of HvGSK2.1 and HvBZR1 activity, Western blot analysis was conducted using the 3rd and 4th leaves from both GP and HN treated with 10 µM, 50 µM, and 100 µM bikinin. Similarly to results obtained in five-day-old barley shoots (Groszyk and Szechyńska-Hebda 2021a), in leaves, we detected HvGSK2.1 and phosphorylated form of HvBZR1 at two negligible different levels (Fig. 7). As in five-days-old barley seedlings, HvGSK2.1 was detected at a consistently level in all samples. Contrary to previous study, in HN leaves, the dephosphorylated form of HvBZR1 was undetected, even after bikinin treatment. However, in GP leaves, both 50 µM and 100 µM bikinin treatments resulted in a reduction of pBZR1 accumulation. The HvGSK2.1 levels remained comparable across all samples.