3.1 IBA significantly promoted AR formation in various blueberry cuttings
To clarify the effect of the lignification level on blueberry AR formation, the cuttings of non-lignified and semi-lignified branches of ‘EMD’ trees were analyzed. The RR was calculated at 60 days post-cutting. Rooting was detected for both cutting types in response to the treatment with 2.0 mg·L−1 IBA, but the RR of the semi-lignified cuttings was significantly greater than that of the non-lignified cuttings (Figure 1A). For both cutting types, the IBA treatment induced AR formation, although this induction was greater for the semi-lignified cuttings. The IBA treatment also significantly increased the ARN for the non-lignified and semi-lignified cuttings (Figure 1B).
To further evaluate the AR formation capacity of different blueberry cultivars (e.g., ‘TFB’, ‘GDB’, ‘WDD’, ‘SPH’, ‘CLM’, ‘PRM’, ‘BRW’, ‘PDB’, ‘ONL’, and ‘STR’), the semi-lignified cuttings were cultivated in substrates containing 2.0 mg·L−1 IBA. The AR formation capacity at 60 days post-cutting varied among the cultivars. Considering the RR, ARN, and ARL (Figure A-D), the rank order of the RIs (high to low) was ‘BRW’, ‘CLM’, ‘STR’, ‘ONL’, ‘SPH’, ‘TFB’, ‘PRM’, ‘WDD’, ‘GDB’, and ‘PDB’ (Figure 2E, F).
Because tissue culture methods can produce high seedling yields, they are commonly used to propagate blueberry seedlings. In this study, 4 cm long TGCs were cultivated in substrate containing 2.0 mg·L−1 IBA. At approximately 9 days post-cutting, a callus was detectable at the bottom of the stems. At 12 days post-cutting, ARs were detected. The CR, RR, ARN, ARL, and TARSA gradually increased as the incubation time increased (Figure3A-F). Moreover, the RI of the TGCs was 54.5%, which was significantly higher than the RI (20.8%) of the FGCs. These results indicate that ARs formed easily and quickly in the TGCs(Figure3G).
3.2 Increases in the number of pericyclic cells in tissue culture-grown semi-lignified cuttings likely enhance AR formation
Microstructural examinations of cuttings were performed to clarify why the rooting abilities varied among cuttings. The non-lignified and semi-lignified cuttings contained a similar number of pericycle cells. However, there were significant differences in the phloem cells between the two cutting types. Compared with the non-lignified cuttings, the semi-lignified cuttings had more layers of xylem cells, especially secondary xylem cells (Figure 4A, B). These results suggest that xylem cells significantly influence AR formation when there are no differences in the pericyclic cells.
Because the ARs of TGCs formed easily and quickly, the microstructures of the semi-lignified cuttings and TGCs were analyzed further. There was no noticeable difference in the proportion of secondary xylem cells in these two cutting types, implying the degree of lignification was similar (Figure 4B, C). However, significant differences were observed in the pericyclic cells. There were more pericyclic cell layers in the TGCs than in the semi-lignified cuttings, implying that an increase in the number of pericyclic cells promotes AR formation when the degree of lignification is appropriate.
3.3 Identification and classification of VcWOX transcription factors in blueberry
Arabidopsis WOX protein sequences were downloaded and used to identify WOX genes in the blueberry genome database. Twenty-nine putative WOX genes (516–1,335 bp long) with conserved homeodomains were identified. The deduced WOX protein sequences comprised 171–444 amino acids. To clarify the phylogenetic relationships among WOX genes, an unrooted phylogenetic tree consisting of WOX proteins from V. corymbosum, Arabidopsis, Oryza sativa, Actinidia chinensis, and Populus trichocarpa was constructed. The WOX transcription factors were classified into the following three subfamilies (Figure 5A): ancient clade, modern/WUS clade, and intermediate clade. The modern/WUS clade was the largest group, with 21 members from V. corymbosum, 8 members from Arabidopsis, 7 members from O. sativa, 11 members from A. chinensis, and 11 members from P. trichocarpa (Figure 5B). The intermediate and ancient clades included only six and two members from V. corymbosum, respectively. Moreover, paralogous and orthologous relationships between the WOX gene families of Arabidopsis and blueberry were revealed. Therefore, on the basis of the phylogenetic relationships, some of the putative blueberry WOX genes were named as follows: VcWUSa/VcWUSb, VcWox1a/VcWox1b/VcWox1c, VcWox2a/VcWox2b/VcWox2c/VcWox2d, VcWox3a/VcWox3b/VcWox3c/VcWox3d, VcWox4a/VcWox4b/VcWox4c/VcWox4d, VcWox5/7a/VcWox5/7b/VcWox5/7c/VcWox5/7d, VcWox8/9a/VcWox8/9b, VcWox11/12a/VcWox11b/12/VcWox11/12c/VcWox11/12d, and VcWox13a/VcWox13b. The increase in the number of genes was the result of gene family evolution. Additionally, analyses of physicochemical properties indicated that the VcWOX protein molecular weight ranged from 13.96 to 71.82 kDa, with isoelectric points between 5.08 and 9.58. The subcellular localization analysis suggested that all 29 VcWOX proteins are nuclear proteins (Supplementary Table S3).
3.4 Detection of conserved residues and motifs in blueberry WOX proteins
To characterize putative conserved motifs in the VcWOX family, the 29 deduced VcWOX amino acid sequences were examined using MEME. Fifteen conserved motifs (1–8) were detected, ranging in size from 15 to 50 amino acids (Table 1). Motifs 1 and 2 were distributed in all 29 VcWOX proteins. In contrast, motif 3 was present only in the WUS clade proteins. Thus, the motifs in VcWOX proteins appear to be family-specific (Figure6A, B).
The featured domain sequences usually determine the primary function of a plant transcription factor. An analysis of the exon and intron structures revealed that VcWOX genes contain 2–4 exons and 1–3 introns (Figure6C). By comparing the amino acid sequences of VcWOX and AtWOX proteins, the conserved amino acid residues within the homeodomain were identified for each clade. Residues 4R, 5W, 7P, 11Q, 15L, 23G, 27P, 35I, 39L, 43G, 49N, 50V, 53W, 54F, 55Q, 56N, and 60R were conserved in the 29 VcWOX sequences (Figure7).
Table 1
Sequence of the conserved motif of VcWOX proteins
|
Conserved Motif
|
Motif sequence
|
|
Motif 1
|
ILEELYRGGMRTPTADQIQQITAQLSKYGKIEGKNVFYWFQNHKARERQK
|
|
Motif 2
|
SGEGGNSPGSSRWNPTPEQJR
|
|
Motif 3
|
RRIETLELFPJHPTG
|
|
Motif 4
|
YQPPGFMTVFINGVPTEVGRGPIDIKGMFGEDLVLVHSSGIPVPFNDYGF
|
|
Motif 5
|
GALDPLAPELPGGSSSACTAATADAAFDLKSFIAPESGPQGYLSNEQKTP
|
|
Motif 6
|
LRRKLMNKQMMHQQHVLYNPQQQQQQNHHHHSHLLHQFPGHYTPVGRGFL
|
|
Motif 7
|
PRASAASTITTISLDTRGQMEKEVEESPYKRKCRTWTFEGLEEEKRHCKD
|
|
Motif 8
|
CEYVDKSEPKTYPPHYLKMLEQGPTKP
|
3.5 Expression profiles of VcWOX genes in different blueberry organs
Gene expression patterns in different organs may provide critical information regarding gene functions related to organ development. The VcWOX expression data for various ‘Draper’ blueberry organs (i.e., root, salt-treated root, leaf, methyl jasmonate-treated leaf, flower bud, and fruit) were downloaded from the Vaccinium genome database (https://www.vaccinium.org). The heatmap of VcWOX expression generated using TBtools revealed the diversity in the expression levels among untreated and treated organs (Figure 8). More specifically, VcWOX4b, VcWOX5/7c, and VcWOX11/12c were highly expressed in the root, but their expression levels decreased in response to the salt treatment. Hence, these three genes might affect root development and responses to salt stress. The VcWOX4a, VcWOX4b, VcWOX4c, VcWOX4d, VcWOX13a, and VcWOX13b genes were highly expressed in the shoot. During fruit development, the VcWOX11b, VcWOX11d, and VcWOX13b expression levels were upregulated in the pink stage, but downregulated in the ripe stage, suggesting that these genes regulate blueberry fruit development. Both VcWOX1 and VcWOX3 were mainly expressed in the leaves. The VcWOX1a and VcWOX3a expression levels decreased at 1 h after the methyl jasmonate treatment, but then subsequently increased at 8 h before decreasing again at 24 h.
3.6 Expression level analysis of selected VcWOX genes during AR formation in blueberry
The root VcWOX expression levels determined on the basis of transcriptome data may not fully reveal the potential functions of these genes related to root formation and development. Therefore, the VcWOX4b, VcWOX5/7c, VcWOX8/9a, VcWOX11/12c, and VcWOX13b expression levels were verified by a qRT-PCR analysis (Figure 9). Total RNA was extracted from the IBA-treated stems, which were then harvested at 0, 3, 6, 9, 12, 15, 18, 21, and 24 days post-treatment. The VcWOX4b and VcWOX13b expression levels were downregulated continuously following the 2.0 mg L−1 IBA treatment (Figure 9A, B), indicating IBA decreased the expression of these genes during AR formation. Additionally, VcWOX8/9a expression was significantly upregulated (7.0 times) at 6 h, but was then downregulated (Figure 9C). The VcWOX11/12c expression levels were upregulated continuously, peaking at 18 h (Figure 9D). These results suggest that IBA activated the expression of VcWOX8/9a and VcWOX11/12c, which might be involved in AR formation. The VcWOX5/7a expression levels fluctuated after the IBA treatment (Figure 9E).
3.7 Overexpression of VcWOX4b resulted in multiple developmental defects in transgenic plants
The VcWOX4b gene was primarily expressed in the root and shoot. Moreover, its expression level decreased in response to the IBA treatment during AR formation. To further elucidate the regulatory function of VcWOX4b in blueberry, VcWOX4b was cloned and inserted into pMDC32 to produce the 35Spro:VcWOX4b recombinant plasmid. A total of 30 transgenic tobacco plants were produced via A. tumefaciens-mediated transformation (Figure10).
Some regenerating T0 plants had abnormal roots (Figure10A). Specifically, the number and length of ARs decreased significantly and the tips were swollen, indicating that VcWOX4b inhibited AR formation and growth (Figure10C). This abnormal root development caused most of these plants to die after they were transferred to soil. Only one severely defective T0 plant survived in soil, but it exhibited extreme dwarfism and was unable to produce flowers or seeds. The VcWOX4b expression levels varied among the transgenic lines, but some of the surviving transgenic lines with low VcWOX transcript levels grew to maturity and produced some T1 seeds. To investigate the effects of VcWOX4b on the root system, T1 transgenic and WT seeds were germinated on half-strength MS medium(Figure 10B). The primary root of the VcWOX4b-overexpressing transgenic seedlings was significantly shorter than that of the WT seedlings (Figure 10D), implying that VcWOX4b repressed primary root growth.
3.8 Overexpression of VcWOX4b altered vascular cell division and differentiation in transgenic tobacco
Because AR formation and growth was inhibited in the transgenic tobacco, the microstructures of the stem and root vascular tissues were characterized. There were significantly more vascular cells in the transgenic tobacco stem than in the WT tobacco stem, suggesting the overexpression of VcWOX4b promoted vascular cell division (Figure 11). Additionally, the vascular tissue of transgenic tobacco stems contained more cambial and primary xylem cells, but fewer secondary xylem cells, than the vascular tissue of WT tobacco stems (Figure 11). This indicated that the overexpression of VcWOX4b enhanced vascular cell differentiation. There were no noticeable differences in the pericycle cells between the WT and transgenic tobacco. Increases and decreases in the number of primary xylem cells and secondary xylem cells, respectively, may explain the decreased production of ARs by the transgenic tobacco.
The overexpression of VcWOX4b resulted in increased root thickness, decreased AR length, and swollen root tips. A microstructural analysis of the vascular tissue at the root base and tip revealed there were more cambial and xylem cells in the transgenic tobacco root base and tip than in the WT tobacco root base and tip. This observation suggests the overexpression of VcWOX4b accelerated vascular cell division and differentiation in the root, which altered the transgenic root phenotype (Figure 12).
3.9 VcWOX4b inhibited AR formation by modulating the IAA:CTK ratio
Hormone levels and proportions were analyzed in transgenic tobacco and WT cuttings at five time-points during AR formation. On the basis of the observations, tobacco AR formation was divided into the following three phases: induction (P1 to P2), initiation (P2 to P3), and extension (P3 to P5) . In transgenic tobacco, the ABA content decreased from P1 to P2, increased from P2 to P4, and decreased from P4 to P5. However, in WT tobacco, the ABA content increased from P1 to P2 and then decreased in the subsequent phases(Figure13A). The IAA content of transgenic tobacco was significantly lower than that of WT tobacco at P2, but it subsequently gradually increased and was significantly higher than that of WT tobacco at P3 and P5 (Figure 13B). The JA content of both WT and transgenic tobacco decreased to its lowest level at P3 and then gradually increased; however, the JA content was lower in transgenic tobacco than in WT tobacco. The JA-ILE content was significantly lower in transgenic tobacco than in WT tobacco at P2 and P5 (Figure 13D, E). The TZR, isopentenyl adenine (IP), and cis-anthocyanidin (CZ) contents were significantly higher in transgenic tobacco than in WT tobacco from P3 to P5 (Figure 13G–I). The TZ contents were substantially higher in transgenic tobacco than in WT tobacco at P3 and P4 (Figure 13J). The IPR content was significantly higher in transgenic tobacco than in WT tobacco from P2 to P5 (Figure 13K). In contrast, the CZR content in transgenic tobacco increased from P1 to P5. Moreover, the CZR content was higher in transgenic tobacco than in WT tobacco at P1 and P2, but the opposite pattern was observed at P3, P4, and P5 (Figure 13J). Plant growth and development are mediated by multiple hormones. The TZR, CZR, IPR, TZ, CZ, and IP contents increased as the CTK content increased(Figure13K). During AR formation, the IAA:CTK ratio exceeded 3.0 in WT tobacco, but it was significantly lower in transgenic tobacco (i.e., less than 1.5 from P2 to P5) (Figure 13L).