-
To identify the BGAL genes in longan, we scanned the D. longan genome database using the known BGAL members in Arabidopsis by BLASTP and TBLASTN. After the removal of incomplete and redundant sequences, a total of 20 putative BGAL genes were identified in the D. longan genome. For the annotation of 20 DlBGAL genes, the Citrus nomenclature system was pursued with numbers representing the highest sequence similarity with the corresponding CsBGAL orthologous (Table 1). The lengths of amino acids (aa) of 20 DlBGALs range from 444 aa (BGAL13-1) to 889 aa (BGAL9). The MW of DlBGALs varies from 50.15 kDa (BGAL13-1) to 99.79 kDa (BGAL9) with an average of 86.85 kDa. The pI ranged from 5.4 (BGAL8-1) to 9.4 (BGAL13-1). Except for BGAL16-1, the remaining DlBGALs were classified as stable proteins because their instability index was less than 40 (Table 1). The predicted aliphatic index ranged from 68.75 to 85.3 in DlBGAL proteins. Due to the GRAVY value (< 0), all DlBGAL proteins were predicted to be hydrophilic.
Table 1. Basic parameter analysis of DlBGAL family.
Gene ID Gene name Size (aa) Molecular weight/kD PI Instability coefficient Hydrophilicity Dlo000174 BGAL1-1 844 93,311.98 8.25 37.76 −0.287 Dlo000176 BGAL1-2 839 92,852.64 7.96 39.56 −0.267 Dlo000290 BGAL15-1 825 92,233.81 5.56 35.42 −0.444 Dlo002805 BGAL9 889 99,793.27 6.35 35.27 −0.271 Dlo006281 BGAL17 710 79,011.06 6.25 35.45 −0.211 Dlo006300 BGAL1-3 844 93,575.84 8.4 38.83 −0.196 Dlo013472 BGAL5 734 82,104.46 8.8 36.87 −0.276 Dlo014416 BGAL16-1 716 80,274.25 9.34 41.46 −0.262 Dlo014417 BGAL16-2 832 93,191.23 8.62 36.02 −0.416 Dlo015024 BGAL16-3 838 93,942.65 6.32 39.86 −0.208 Dlo015487 BGAL13-1 444 50,149.01 9.4 38.7 −0.233 Dlo015584 BGAL3-1 843 93,502.26 7.25 38.92 −0.221 Dlo016393 BGAL6 718 81,101.64 5.46 36.14 −0.226 Dlo016396 BGAL16-4 846 94,734.74 6.92 40.98 −0.33 Dlo021540 BGAL8-1 842 93,433.6 5.4 39.83 −0.16 Dlo023146 BGAL13-2 864 96,671.78 8.37 36.43 −0.214 Dlo023292 BGAL3-2 660 73,921.74 7.3 34.86 −0.29 Dlo026229 BGAL8-2 853 92,871.32 8.07 39.94 −0.192 Dlo026230 BGAL8-3 612 67,240.37 7.56 36.87 −0.325 Dlo029560 BGAL15-2 828 93,079.31 6.56 34.06 −0.305 The phylogenetic tree of the BGAL family contains BGAL gene members from four species: D. longan, O. sativa, C. clementina and A. thaliana. To determine their evolutionary relationship, a phylogenetic tree was constructed using the neighbour-joining (NJ) method and tested by 1,000 bootstrap replicates (Fig. 1). According to phylogenetic relationships, the BGALs were clustered into seven groups (Groups I–VII), each containing 4, 2, 6, 1, 3, 3 and 1 DlBGALs, respectively. In O. sativa, A. thaliana and C. clementina, BGAL genes had been identified in cluster forms. Thus, BGAL members may be evolutionarily close in their respective species. Interestingly, BGAL proteins of O. sativa and A. thaliana and C. clementina were clustered into distinct clades (Fig. 1), suggesting that BGAL proteins were evolved after monocots and dicots divergence. Among them, most of the BGAL genes in D. longan were closely clustered with C. clementina in the groups of the phylogenetic tree, supporting the hypothesis that D. longan and C. clementina shared close and similar ancestors[41].
Figure 1.
Phylogenetic tree analysis of β-galactosidase (BGAL) proteins from longan, Citrus, Arabidopsis, and Oryza.
Gene structure and motif composition of longan BGAL genes
-
To further understand the evolutionary relationship of the 20 DlBGAL genes, we identified their structural characteristics and motifs (Supplemental Fig. S1). All DlBGAL genes had at least 12 exons, with 13 of the 20 genes having more than 18 exons (Supplemental Fig. S1b). The numbers of introns and exons in the genomic sequences were relatively similar in each group. In addition, ten conserved motifs of DlBGAL proteins were identified through MEME analysis (Supplemental Fig. S1c). Motifs 2 and 4, which were components of the DlBGAL domain, were found in all DlBGAL proteins. Motifs 6, 7 and 10 were detected in all DlBGAL proteins except DlBGAL13-1. For group III, IV and VI, all of the DlBGALs possessed 10 motifs (Supplemental Fig. S1c).
Chromosomal location and gene duplication events of DlBGAL genes
-
The chromosome information of DlBGAL genes was extracted and the chromosomal positions indicated that DlBGAL genes were distributed on eight of the 15 chromosomes of longan (Fig. 2a). Among them, Chr1, 6 and 7 had four DlBGAL genes. Chr3, 11 and 12 had two DlBGAL genes, while the Chr10 and 14 had the smallest numbers of DlBGAL genes (one gene each). The distribution of DlBGAL genes was uneven.
Figure 2.
Chromosomal distribution, synteny and cis-elements analysis of DlBGAL genes. (a) Distributions and duplications of DlBGAL genes along longan chromosomes. The gray and blue lines indicate chromosome locations of DlBGALs and segmentally duplicated genes, respectively. (b) The homologous relationships of BGAL genes in longan, Oryza and Arabidopsis. The green and orange lines indicate the homologous relationships of BGAL genes with Oryza and Arabidopsis, respectively. (c) Analysis of cis-elements in the promoter of DlBGAL genes. Heatmap of the numbers of cis-elements in the promoters of DlBGAL genes.
The results of collinearity analysis showed that there were six pairs of segmental duplication events in DlBGALs, corresponding to DlBGAL1-1/1-3, DlBGAL5/3-1, DlBGAL5/3-2, DlBGAL16-1/6, DlBGAL13-1/13-2 and DlBGAL3-1/3-2 (Fig. 2a). The paralogous gene pairs were distributed on different chromosomes, which may be beneficial for the evolution and expansion of the DlBGAL gene family. The expansion of the DlBGAL gene family was mainly due to segmental duplication, rather than tandem duplication. To verification the evolutionary rates and selective pressures between the DlBGAL genes and their duplicated genes, we used Ka (nonsynonymous), Ks (synonymous), and the Ka/Ks ratio for estimations. As shown in Supplemental Table S2, the Ka/Ks values of all the DlBGAL gene pairs were less than 1, indicating that the DlBGAL gene family underwent strong purifying selection. These results indicated that DlBGAL family was conservative during the domestication process of longan.
The collinearity analysis between longan and Arabidopsis thaliana or Oryza sativa showed that three DlBGAL homologue genes appeared in the three chromosomes of Oryza sativa, but 16 DlBGALs had corresponding paralogous genes on five chromosomes in Arabidopsis thaliana (Fig. 2b). The relationship between longan and Arabidopsis thaliana was closer than that between longan and Oryza sativa.
Analysis of cis-elements in the promoter region of DlBGAL genes
-
The cis-elements are involved in the regulation of gene expression, which are an important part of the response regulation network for plant growth and development, hormone response and stress response[42,43]. We isolated the 2-kb upstream promoter region of the DlBGAL genes to identify the potential cis-elements. Among them, many DlBGAL genes exhibit elements specifically related to light signaling, hormone, stress and development.
For DlBGAL genes, 34 main types of cis-elements were identified in their promoter sequences, which included 11 light responsive, 10 hormone responsive, seven stress responsive, and six development elements (Fig. 2c). The major light responsive elements were Box4 and G-box and the most common motifs were the JA-responsive elements (MYC), abscisic acid (ABA)-responsive element (ABRE), and ethylene-responsive element (ERE) of the scanned hormone responsive motifs. MYB and ARE were the most cis-elements of the stress responsive. These results suggested that DlBGAL genes were participated in complex light, hormone regulation and stress responsive networks that may involve multiple regulatory modalities.
Expression analysis of DlBGAL genes during early SE and under different temperature stress conditions
-
The expression profiles of DlBGAL genes during early SE and under temperature treatments were illustrated using our previous RNA-seq dataset. Results showed that nine DlBGALs (DlBGAL8-3/15-1/13-2/16-2/15-2/13-1/8-1/6/3-2) were almost not expressed (FPKM < 1) in the early SE RNA-seq dataset (Fig. 3a). About the remaining 11 genes, there were four genes (DlBGAL17/16-3/1-2/16-1) that showed high expression levels at EC stage and six genes (DlBGAL1-1/16-4/9/8-2/1-3/3-1) had high expression levels at GE stage. Furthermore, only DlBGAL5 was presented high expression at ICpEC stage. These results suggested that the DlBGAL genes may be involved in promoting early SE. The differentially expressed date of DlBGAL genes under different temperature conditions were derived from the RNA-seq (Fig. 3b). From the results, the expression levels of some DlBGAL genes varied obviously under high-temperature and low-temperature treatments. Compared to normal culture temperature (25 °C), seven DlBGAL genes (DlBGAL3-2/1-2/16-2/17/5/1-1/16-3) were upregulated under the high-temperature and two DlBGAL genes (DlBGAL5/1-1) were upregulated at low-temperatures.
Figure 3.
Expression profile and multi-omics analysis of DlBGAL genes. (a), (b) Expression analyses of DlBGAL genes during early longan SE and different temperature treatments. (c) Alluvial diagram of DlBGAL genes was divided into eight types (types I–VII and other) based on ATAC-seq, ChIP-seq, and RNA-seq data. The width of different colored bars represents the number of genes, and the genes included in each type were listed in the right panel. EC vs GE, the differential expression of EC vs GE; EC vs GE & ICpEC vs GE, the differential expression of EC vs GE and ICpEC vs GE; EC vs ICpEC vs GE, the differential expression of EC vs ICpEC, ICpEC vs GE and EC vs GE; ICpEC/GE, ATAC-seq or ChIP-seq (H3K4me1) peaks were detected in ICpEC/GE; EC & ICpEC, ATAC-seq or ChIP-seq (H3K4me1) peaks were detected in EC and ICpEC; EC & ICpEC & GE, ATAC-seq or ChIP-seq (H3K4me1) peaks were detected in EC, ICpEC and GE; Na, undetectable peak of ATAC-seq or ChIP-seq (H3K4me1); Ns, no difference in transcripts among the samples.
The change of chromatin accessibility affects the binding of transcription factors and other protein molecules to chromatin, thereby regulating gene transcription[44]. ATAC-seq analysis revealed that chromatin accessibility of other DlBGAL genes, except DlBGAL17 and DlBGAL8-3, was open during the early SE (Fig. 3c). Histone methylation modification can activate or inhibit gene transcription by the state of methylation, in which H3K4me1 is a marker associated with enhancers[45]. Except for DlBGAL6, H3K4me1 binding peaks were detected in other DlBGAL genes during early SE, and most of them had binding peaks at three stages. Combined with transcription, chromatin accessibility, and H3K4me1 modification results, most DlBGAL genes differentially expressed during early SE could be detected in the ATAC-seq and ChIP-seq data. In summary, multi-omics analysis has implications for the potential mechanism of BGAL family involvement in early SE.
Transcription factors target the DlBGAL9 promoters and activate its transcription
-
Transcription factors (TFs) can inhibit or enhance gene expression, and promote SE formation by directly activating early embryonic model genes[24]. To understand the regulatory network of transcription factors on DlBGAL genes, the upstream 2-kb sequences of DlBGAL genes were used as TF binding regions. In total, 11 DlBGALs were predicted as targets for 10 TFs (Fig. 4a). Based on multi-omics analysis, the functional role of DlBGAL9 in early longan SE attracted further interest. Previous studies have identified an important somatic embryo marker gene, Dlo004817_AGL80, which was involved the regulatory network associated with the cell wall of longan[29]. Based on TFs and DlBGALs regulatory network, Dlo006037_DlAGL61 and Dlo004817_DlAGL80 were selected as candidate TFs for regulating DlBGAL9 (Fig. 4a).
Figure 4.
Transcription factors (TFs) regulate the expression of DlBGAL9. (a) Network prediction of DlBGAL gene regulated by TFs. The red font represents the key genes. (b) The relative expression of DlBGAL9 in the TFs DlAGL61/80 transiently transformed. Beta-actin (ACTB) was used as a reference gene to normalize gene expression data. Each treatment was conducted in triplicate with three technical repeats. (c) Schematic diagrams of the effector vectors and reporter vectors in luciferase assays. (d) The interactions of DlAGL61 or DlAGL80 with the DlBGAL9 promoter that were detected in leaves of N. benthamiana through a dual luciferase reporter system. (e) Determination of the luciferase enzyme activity in N. benthamiana. Values were mean ± SD. **indicates a statistically significant difference, **p < 0.01.
To investigate the regulatory effect of DlAGL61/80 on DlBGAL9 transcription, we verified the expression of DlBGAL9 in longan EC that overexpressed TFs DlAGL61/80. QRT-PCR results showed that DlBGAL9 expression was significantly increased in longan EC with overexpressing of DlAGL61 (3.19-fold) and DlAGL80 (8.87-fold) (Fig. 4b). Thus, TFs DlAGL61/80 may enhance DlBGAL9 transcription by targeting its promoters. To further verify whether TFs DlAGL61/80 can activate the expression of DlBGAL9, a luciferase (LUC) reporter assay was performed in Nicotiana benthamiana leaves. Two constructs, 35S::DlAGL61 and 35S::DlAGL80 were used as effector, and DlBGAL9pro:LUC was used as reporter (Fig. 4c). Compared with control check (CK), the relative DlAGL61/80-induced firefly LUC/Renilla luciferase (REN) activity driven by the promoters of DlBGAL9 increased 1.41-fold and 1.86-fold, respectively (Fig. 4d, e). These results indicated that TFs DlAGL61/80 may target the DlBGAL9 promoters and activate their transcription.
Transcription factors and DlBGAL9 responses to heat stress and changes of β-galactosidase activity and pectin content during early longan somatic embryogenesis
-
Based on the RNA-seq data of longan EC under temperature treatments, DlBGAL genes expression detected in early longan SE responded to heat stress (Fig. 3b). Previous studies have found that heat stress affects the process of early longan SE, and EC cannot develop to GE under 35 °C treatment (heat stress). To further investigate the response of DlBGAL9 and TFs DlAGL61/80 to heat stress, we detected their expression of longan EC treated at 35 °C. The results showed that DlBGAL9 was significantly upregulated under heat stress at 6 d, and the TF DlAGL61 showed a significant downregulation trend under heat stress at 9 d (Fig. 5a). Notably, DlBGAL9 and TFs DlAGL61/80 have the highest level under heat stress at 12 d (Fig. 5a). The DlBGAL9 and TFs DlAGL61/80 showed a similar trend under heat stress, indicating that the DlAGL61/80-DlBGAL9 regulatory network may be involved in longan's response to heat stress.
Figure 5.
Expression pattern analysis of DlBGAL9 and TFs DlAGL61/80 in longan EC under heat stress. (a) Expression of DlBGAL9 and TFs DlAGL61/80 in longan EC under different days of heat stress. Beta-actin (ACTB) was used as a reference gene to normalize gene expression data. (b) The β-GAL activity during early longan somatic embryogenesis (EC, ICpEC, and GE). (c) The pectin content during early longan somatic embryogenesis (EC, ICpEC, and GE). (d) The β-GAL activity in longan EC under different days of heat stress. Somatic embryos were cultured on MS medium and treated at 25 °C (normal) and 35 °C (heat stress). Samples were harvested after 6, 9, and 12 d to detect the gene expression patterns and β-GAL activity. MS, Murashige and Skoog medium. EC, embryogenic callus; ICpEC, incomplete compact pro-embryogenic culture; GE, globular embryo. Each treatment was conducted in triplicate with three technical repeats. Values are mean ± SD. * Indicates a statistically significant difference, *p < 0.05, **p < 0.01.
To further analyze the role of β-GAL and pectin during early longan SE, the changes in β-GAL activity and pectin content were measured (Fig. 5b, c). The β-GAL activity was significantly upregulated during early longan SE (Fig. 5b). Pectin content was opposite to β-GAL activity, and the highest pectin content was found in the EC stage (Fig. 5c). After that, the β-GAL activity of longan ECs treated at heat stress for different durations was determined. Compared with the normal temperature (25 °C), the β-GAL activity was significantly downregulated at 6 and 12 d of heat stress (35 °C) and significantly increased at 9 d (Fig. 5d). It was suggested that β-GAL can not only participate in somatic embryonic process by negatively regulating pectin content, but also can respond to heat stress.
Overexpression of DlBGAL9/DlAGL80 thickens the root cell wall by affecting β-GAL activity
-
As the embryonic callus is composed of parenchyma cells, it poses certain obstacles to the study of cell walls. Hairy roots have the characteristics of rapid growth and hereditary stability, providing a good material for studying the impact of genes on cell walls. In this study, the constructed p1302: DlBGAL9: GFP was introduced into the Agrobacterium rhizogenes strains and transferred into longan seedlings. The GFP-signals and PCR results both confirmed that DlBGAL9 had been transferred into longan seedlings and successfully induced the hairy roots (Fig. 6a). Compared with the wild-type (WT), transgenic roots overexpressing DlBGAL9 (DlBGAL9-OE) showed a 24.20-fold increase in DlBGAL9 expression (Fig. 6a). We examined the β-GAL activity and pectin content in the transgenic hairy roots (Fig. 6b, c). The β-GAL activity of DlBGAL9-OE root was higher, while the pectin content was lower. These results suggested that DlBGAL9 influences pectin content by regulating β-GAL activity in transgenic hairy roots.
Figure 6.
Functional analysis of DlBGAL9 and TF DlAGL80 in longan hairy root. (a) Left: The bright field and GFP-fluorescent DlBGAL9-OE transgenic hairy roots. Right: Molecular identification DlBGAL9-OE transgenic hairy roots. Scale bar = 2 cm. (b) The β-GAL activity in longan hairy root of DlBGAL9-OE transgenic. (c) The pectin content in longan hairy root of DlBGAL9-OE transgenic. (d) Left: The bright field and GFP-fluorescent longan hairy root of the DlAGL80-OE transgenic. Right: Molecular identification longan hairy root of the DlAGL80-OE transgenic. Scale bar = 2 cm. (e) Expression of DlBGAL9 in longan hairy roots of DlAGL80-OE transgenic. (f) The β-GAL activity in longan hairy roots of DlAGL80-OE transgenic. (g) Anatomical and morphological observations of WT, DlBGAL9-OE and DlAGL80-OE longan hairy roots. Arrows indicate the position of cell wall thickening. Scale bar = 20 μm. (*p < 0.05, **p < 0.01).
As DlAGL80 target the DlBGAL9 promoters and activate its transcription, we tested whether the expression of DlBGAL9 was affected by DlAGL80 transgenic roots. The overexpressing DlAGL80 (DLAGL80-OE) transgenic hairy roots was confirmed by GFP-signals and PCR results, and the expression of DlAGL80 increased by 9.61-fold (Fig. 6d). In DlAGL80-OE roots, DlBGAL9 expression was dramatically increased, while β-GAL activity was increased (Fig. 6e, f). Transverse sections of DlBGAL9 and DlAGL80 transgenic hairy roots showed thickening of cell walls in both genotypes compared to WT (Fig. 6g). In summary, DlAGL80 may regulate the expression of DlBGAL9 by regulating β-GAL activity and participate in the cell wall modification processes.
DlBGAL9 and TF DlAGL80 overexpression enhances the resistance to heat stress
-
The expression of DlBGAL9 and TF DlAGL80 in roots under heat stress were assessed to investigate whether DlBGAL9 and TF DlAGL80 were involved in the resistance of longan roots to heat stress. QRT-PCR results showed that heat stress induced DlBGAL9 and TF DlAGL80 expression, which reached the peak value at 6 d (Fig. 7b). Further results showed that the hairy root cell wall of longan was thickened under heat stress (Fig. 7a). The above results indicated that DlBGAL9 and TF DlAGL80 may participate in the mechanism of longan resistance to heat stress. When plants were subjected to biotic or abiotic stress, the concentration of reactive oxygen species (ROS) in cells increases and the homeostasis of cells was destroyed. The activity of ROS scavengers (SOD, POD and CAT) will increase accordingly to maintain the homeostasis of cells. Fe-SOD (FSD) and Mn-SOD (MSD) belong to the super oxide dimutese (SOD) gene family. The expression of SOD, peroxidase (POD) and catalase (CAT) related genes in DlBGAL9-OE/DlAGL80-OE roots were significantly upregulated (Fig. 7c). In summary, DlBGAL9 and TF DlAGL80 may reduce the damage caused by heat stress to longan hairy roots by regulating the content of SOD, POD and CAT.
Figure 7.
Expression pattern analysis of DlBGAL9 and DlAGL80 under heat stress in longan hairy root. (a) Anatomical and morphological observations of longan hairy root under heat stress. Arrows indicate the position of cell wall discrepancy. (b) Expression pattern analysis of DlAGL80 and DlBGAL9 under different days of heat stress in longan hairy root. (c) Expression profiles of ROS pathway related genes in WT, DlBGAL9-OE and DlAGL80-OE longan hairy roots. (*p < 0.05, **p < 0.01).
-
All data generated or analyzed during this study are included in this published article and its supplementary information files.
-
About this article
Cite this article
Chen Y, Ma X, Ma W, Xu L, Lai C, et al. 2024. AGL61/80 regulates BGAL9, which participates in early somatic embryogenesis and increases longan resistance to heat stress. Fruit Research 4: e013 doi: 10.48130/frures-0024-0005
AGL61/80 regulates BGAL9, which participates in early somatic embryogenesis and increases longan resistance to heat stress
- Received: 17 November 2023
- Accepted: 04 January 2024
- Published online: 02 April 2024
Abstract: β-galactosidase (BGAL) is a glycosyl hydrolase that participates in cell wall modification, playing an essential role in plant growth, development and environmental stress adaptation. However, the function of BGALs in longan remains unclear. In this study, a total of 20 BGAL genes were identified from the longan (Dimocarpus longan Lour.) genome, which were distributed to eight chromosomes. Transcript profiles revealed that the majority of 11 detected DlBGAL genes were highly expressed in the embryogenic callus (EC) (four) and globular embryos (GE) (six), and responsive to heat stress. Combined with RNA-seq, ATAC-seq, and ChIP-seq results, most DlBGAL genes differentially expressed during early somatic embryogenesis (SE) were related with chromatin accessibility and high levels of H3K4me1. DlBGAL9 was selected for further analysis. Dual-luciferase assays and transient transformation indicated that the transcription factor (TF) DlAGL61 and DlAGL80 might bind to the DlBGAL9 promoters to activate DlBGAL9 transcription. Overexpression of DlBGAL9 and TF DlAGL80 induces longan hairy roots β-GAL activity and thickening of cell walls. The expression levels of DlBGAL9 and TF DlAGL80 in longan hairy roots were significantly increased under heat stress, and ROS scavengers related genes were significantly upregulated in overexpressing DlBGAL9 and TF DlAGL80 hairy roots. This study proposes the significance of the regulatory network composed of DlBGAL9 and TF DlAGL80 in regulating the early longan SE and heat stress response.
-
Key words:
- D. longan /
- β-galactosidase (BGAL) /
- Transcriptional factor /
- Somatic embryogenesis /
- Heat stress