The rubber tree clone CATAS7-33-97 was planted at the experimental farm of Danzhou, Hainan Province, China. At the flowering stage, samples of leaves, bark, latex, and flowers were collected from the same plants simultaneously. Tissue culture-generated seedlings were used for different stress and phytohormone treatments according to our reported method (Qin et al., 2015). Briefly, plants were subjected to drought stress treatment by continuously withholding water for 10 days, and the control plants were well-watered, with samples collected at the same time points. Wounds were made on the leaves by pinching with forceps, and untreated plants were used as controls. For chemical treatments, whole plants were sprayed with 2% (v/v) H₂O₂, 1.0% (v/v) ET, 200 mM methyl jasmonate (MeJA), 3 mM gibberellic acid 3 (GA₃), 200 μM ABA, and 5 mM SA, respectively. All the chemicals were diluted in distilled water that included 0.05% (v/v) ethanol, and the control plants were sprayed with distilled water that included 0.05% (v/v) ethanol. Six plants were included in each treatment, and leaves from the six plants were harvested at each time point. Three different biological replicates were analyzed. The harvested samples were immediately frozen in liquid nitrogen and stored at −80°C until analysis.

The AtMYB44 (AT5G67300) sequence was subjected to tBLASTn searches of the transcriptome shotgun assembly (TSA) and genome database of H. brasiliensis. According to the obtained sequence, primers (MYB44-F1 and MYB44-R1, Table 1) were designed to amplify the full-length open reading frame (ORF) of HbMYB44 from the rubber tree.

The National Center for Biotechnology Information (NCBI) conserved domain searches¹ and SMART² were used to analyze the structure of HbMYB44. Sequences of HblMYBs identified from the laticifer transcriptome database and the genome database of H. brasiliensis (Wang et al., 2017), and Arabidopsis MYBs obtained from the phytozome,³ were used for the phylogenetic analysis. Information on H. brasiliensis and Arabidopsis MYBs is listed in Supplementary Table 1. Sequence alignments were conducted using the DNAMAN version 8.0 and ClustalX (Thompson et al., 2002). The phylogenetic analysis was performed using the MEGA version X based on the neighbor-joining method with a Poisson correction model and a bootstrap test of 1,000 replicates (Tamura et al., 2013).

Total RNA was extracted from different tissues using a Total RNA Isolation kit (OMEGA Bio-tek, Norcross, GA, United States). cDNA was synthesized with ReverTra Ace qPCR RT Master Mix with gDNA Remover (TOYOBO, Shanghai, China) according to the manufacturer’s instructions. Quantitative real-time PCR (qRT-PCR) was performed on a CFX96™ Real-Time System (Bio-Rad Laboratories, Hercules, CA, United States) using the AceQ Universal SYBR qPCR Master Mix (Vazyme, Nanjing, China) according to the manufacturers’ instructions. Specific primers MYB44-F2 and MYB44-R2 were used for amplifying HbMYB44 (Table 1). HbActin (GenBank accession number: HO004792) and AtActin2 (GenBank accession number: NM_001338358.1) were used as housekeeping genes in the rubber tree and Arabidopsis, respectively. The relative expression levels of target genes were calculated using the 2^–ΔΔCT method (Nolan et al., 2006; Schmittgen and Livak, 2008). All the data were analyzed by one-way ANOVA, and multiple comparisons were performed by the Tukey’s test at the P < 0.01 level. Figures were constructed by OriginPro 2018 (OriginLab Corporation, Northampton, MA, United States).

To determine the subcellular localization of HbMYB44, a fluorescent GFP tag was fused to the C-terminus of HbMYB44. The ORF of HbMYB44 was amplified with primers (MYB44-EcoR I-F1 and MYB44-BamH I-R1, Table 1) and subsequently inserted in the EcoR I/BamH I restriction sites of the vector 35S: XXGFP (provided by Dr. Xingliang Hou). The recombinant plasmid 35S:HbMYB44:mGFP was then transformed into Agrobacterium tumefaciens strain GV3101, and then, the GV3101 strain harboring the recombinant plasmid was transiently expressed in tobacco (Nicotiana benthamiana) leaf epidermal cells according to the methods of our previous study (Qin et al., 2019). The fluorescent protein expression was examined by fluorescence imaging together with confocal microscopy (Leica, Wetzlar, Germany) after 1 day of dark/light cultivation post infiltration. The nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) as previously reported (Kapuscinski, 1995) and visualized by confocal laser scanning microscopy.

The complete ORF of HbMYB44 (amplified by MYBY44-EcoR I-F1 and MYBY44-BamH I-R2) and fragments of HbMYB44 encoding the N-terminus [amino acids (aa) 1-200, amplified by MYBY44-EcoR I-F1 and MYBY44-▲N-BamH I-R3] or the C-terminus (aa 201-345, amplified by MYBY44-▲C-EcoR I-F2 and MYBY44-BamH I-R2), were amplified and cloned into the EcoR I/BamH II restriction sites of the vector pGBKT7,⁴ respectively. The target fragment was fused to the 3′ end of the GAL4 DNA-binding domain (DNA-BD) of the vector pGBKT7 between the EcoR I and BamH I restriction sites of MCS, such that a GAL4 DNA-BD-HbMYB44 fusion protein would be produced when the yeast cells were transformed. The pGBKT7-HbMYB44, pGBKT7-▲N-HbMYB44, and pGBKT7-▲ C-HbMYB44 constructs; the negative control pGBKT7; and the positive control pGBKT7-HbNAC24 were used (HbNAC24 is an NAC TF of the rubber tree and has been shown to have transcriptional activation activity) (Kang et al., 2014). These constructs were individually transformed into yeast strain Y2H Gold using a Yeastmaker™ Yeast Transformation System 2 (Clontech, Mountain View, CA, United States) according to the manufacturer’s instructions. If the fusion protein had transactivation activity, the colonies containing the pGBKT7+fusion protein construct in Y2H Gold would grow on SD/-Trp/-His/-Ade (TDO) media and become blue on TDO supplemented with X-α-Gal (TDO/X) media. All the yeast transformants were grown at 30°C for 2–3 days on SD media lacking tryptophan (SD/-Trp). Then, the positive clones were transferred to TDO media or TDO/X media and cultured at 30°C for 2–3 days (Fujita et al., 2004; Zhang et al., 2014). The experiment was repeated three times.

GV3101 cells harboring the recombinant plasmid 35S:HbMYB44:mGFP were used in Arabidopsis transformation for the overexpression of HbMYB44 (OE). Arabidopsis thaliana Col-0 wild type (WT) was transformed via the floral-dip method as previously described (Guan et al., 2014). All the transgenic plants were selected on Murashige and Skoog (MS) media that included 10 mg L^–1 Basta. Then, the resistant plants of the T2 generation were tested with the primers MYB44-EcoR I-F1 and MYB44-BamH I-R1 of HbMYB44 and used for the following studies.

The expression of HbMYB44, its homologous genes (including AtMYB44, AtMYB73, AtMYB70, and AtMYB77), and genes (including AtWRKY70, AtPYL8, and AtPYL9) encoding the interacting proteins of AtMYB44 in OE and WT plants were analyzed using qRT-PCR according to the method mentioned above. The primers used are listed in Table 1.

The seeds of the T3 generation and WT seeds were sterilized with 10% bleach and 0.001% Triton X-100 for 10 min and washed five times with sterile water. The sterilized seeds were sown on MS media, stratified at 4°C in darkness for 2 days, and then transferred to a plant growth chamber (MMM/Climacell 707, Germany) set at 22°C with a 16 h light/8 h dark photoperiod (light intensity of 13,000 lux) for germination for 1–2 days. Then, the germinated seeds were transferred to MS media supplemented with 100 mM NaCl, 250 mM mannitol, 15% (w/v) PEG4000, 0.3 μM ET, 10 μM ABA, 20 μM MeJA, 50 μM GA₃, and 10 μM SA. The plates were placed vertically in the illumination incubator under the same conditions used for germination, and the root length of the seedlings was measured after 10–13 days of treatments. For each treatment, at least six plants were tested in triplicate, and three different biological replicates were performed. The final value was the average of three batches of measurement data.

The rubber tree MYB44 homolog HbMYB44 (GenBank accession number: AZZ09361.1) was cloned using RT-PCR. The ORF of HbMYB44 is 1,038 bp in length and encodes a protein of 345 aa. No introns were found in the genomic sequence of HbMYB44. The deduced HbMYB44 protein contains 2 SANT domains (one is from the 11th to 60th aa, and the other is from the 63rd to 111th aa; Figures 1A,B), which indicated that HbMYB44 is an R2R3-MYB TF. A BLASTp search of the NCBI non-redundant (NR) protein database revealed that HbMYB44 was highly similar to MYB44 homologs from Ricinus communis, Populus euphratica, Glycine max, and A. thaliana with 72, 71, 61, and 55% identities, respectively (Figure 1B).

Multiple sequence alignments between rubber tree and A. thaliana MYB family members indicated that HbMYB44 shared the highest similarity with rubber tree HblMYB7 (with 82% identity) and Arabidopsis AtMYB44 and AtMYB73 (with 55% identity). The phylogenetic analysis demonstrated that HbMYB44 clustered into subgroup 22 together with AtMYB44, AtMYB70, AtMYB73, AtMYB77, rubber tree HblMYB7, HblMYB9, HblMYB10, and HblMYB35 (Figure 2). Moreover, two specific motifs of subgroup 22, namely, motif 22.1 (TGLYMSPxSP) and motif 22.2 (GxFMxVVQEMIxxEVRSYM) were found to be present in HbMYB44 (Figure 1B, indicated by red boxes).

To determine the subcellular localization of HbMYB44, a fusion protein HbMYB44-GFP driven by the 35S promoter was generated and transformed into tobacco epidermal cells. The nucleus dyed with DAPI was blue, and HbMYB44-GFP fusion protein was green, both of which were cyan after being merged. The results demonstrated that HbMYB44-GFP was exclusively localized in the nucleus of tobacco epidermal cells, while the control empty vector free-GFP was distributed throughout the cells (Figure 3A). These results demonstrated that HbMYB44 is a nuclear protein.

Since MYB proteins function as TFs (Li et al., 2006), we tested the transactivation activity of HbMYB44. GAL4 DNA-BD-HbMYB44 fusion proteins were expressed in yeast, and their ability to activate transcription from the GAL4 upstream activation sequence and to promote yeast growth in defective media was analyzed. The full-length ORF of HbMYB44 (HbMYB44), the N-terminal (encoding the 1st–200th aa, HbMYB44-▲N) and the C-terminal (encoding the 201st–345th aa, HbMYB44-▲C) fragments of HbMYB44 were individually fused to the GAL4 DNA-BD within the pGBKT7 vector (Figure 3B) and transformed into yeast strain Y2H Gold for transactivation activity analysis as previously reported (Zhao T. et al., 2016). As shown in Figure 3B, all constructs (including pGBKT7-HbMYB44, pGBKT7-HbMYB44-▲N, pGBKT7-HbMYB44-▲C, negative control pGBKT7, and positive control pGBKT7-HbNAC24) grew on the SD/-Trp media, but only pGBKT7-HbNAC24, pGBKT7-HbMYB44, and pGBKT7-HbMYB44-▲C grew on the TDO media and appeared blue on TDO/X media. The full-length ORF of HbMYB44 fused with the DNA-BD of GAL4 can self-activate the reporter genes in yeast cells. There was no effect on its transactivation activity when 200 aa within the N-terminus was deleted. However, truncation of 145 aa within the C-terminus resulted in the loss of transactivation activity. These results showed that HbMYB44 is a nuclear protein whose last 145 aa of the C-terminus have transactivation activity.

The expression pattern of HbMYB44 in rubber tree was measured using qRT-PCR. HbMYB44 was expressed in all the tested tissues, but its expression in flowers was higher than that in other tissues; the expression in the flowers was nearly 6.5-fold higher than that in the bark (Figure 4A). Since MYBs of subgroup 22 play important roles in regulating the stress response and phytohormone signaling in plants, HbMYB44 expression following different phytohormones and stress treatments was analyzed systematically. As shown in Figure 4B, HbMYB44 expression was rapidly induced by wounding, and it was upregulated by 45.9-fold at 2 h after treatment. Under drought stress, HbMYB44 expression was dramatically increased and peaked at 10 days after treatment, the level of which was 34.1-fold greater than that at 0 day (before treatment) (Figure 4C). After H₂O₂ treatment, HbMYB44 expression was increased by 12.7-fold at 6 h (Figure 4D). Furthermore, HbMYB44 expression was strongly induced in response to ET, MeJA, and ABA and peaked at 6 h, the levels of which were increased to 31.2-fold, 9.9-fold, and 28.4-fold, respectively (Figure 4E). In response to SA treatment, the HbMYB44 expression level was upregulated by 2.8-fold at 2 h (Figure 4E). Interestingly, HbMYB44 expression was rapidly triggered by GA₃ application. HbMYB44 expression peaked at 0.5 h, which was an increase of approximately 83-fold, but then quickly fell back at 2 h (Figure 4E). These results indicated that the HbMYB44 expression was induced by multiple stress conditions and exogenous phytohormones.

To verify the function of HbMYB44 in response to stress and phytohormones, transgenic Arabidopsis lines overexpressing HbMYB44 (OE) were generated. The OE plants of the T2 generation were tested using PCR with the specific primers of HbMYB44, the results of which indicated that all 10 OE lines contained the target gene (Figure 5A). Then, the expression level of HbMYB44 in OE lines was confirmed using qRT-PCR. There were differences in the expression abundance of HbMYB44 among different transgenic lines, and OE2 with the highest expression was used for further analysis (Figure 5B).

Since HbMYB44 functions as a TF with transactivation activity (Figure 3B), and its sequence is highly homologous to those of Arabidopsis MYBs of subgroup 22 (Figure 2), we analyzed whether HbMYB44 altered the expression of its homologous genes and their interacting proteins in transgenic Arabidopsis plants by qRT-PCR. As shown in Figure 6, except for that of AtMYB77, the expression of genes of subgroup S22, including AtMYB44, AtMYB70, and AtMYB73, was obviously increased in the HbMYB44-overexpressing plants compared with the WT plants. Moreover, the expression of AtWRKY70 and AtPYL8 but not AtPYL9 was also obviously increased in OE plants. These results indicated that HbMYB44 enhanced the expression of its homologous genes and interacting protein-encoding genes in transgenic Arabidopsis.

Under normal conditions, there was no marked difference in the morphology of WT and OE plants cultivated in MS media, but different phenotypes, especially in terms of root growth, were found under stress and treatments involving exogenous phytohormones (Figure 7A). As shown in Figure 7, the growth of leaves and roots of both in WT and OE2 plants was severely suppressed under stress and exogenous phytohormone conditions. In the media supplemented with exogenous ABA, MeJA, GA₃, and SA, the root length of OE plants significantly increased compared with that of the WT plants under the same conditions. However, the root length of OE plants was reduced compared with that of the WT plants in the media supplemented with ET (Figures 7A,B). These results demonstrated that the root growth of OE plants is less sensitive to ABA, MeJA, GA₃, and SA, but more sensitive to ET. Under NaCl-salt stress, mannitol-mediated osmotic stress, and PEG-simulated drought stress conditions, the root lengths were markedly increased in the OE plants compared with the WT; the roots especially increased more than 5-fold under NaCl conditions (Figures 7A,B). These results indicated that HbMYB44 enhanced tolerance to abiotic stress and promoted recovery from root growth inhibition in response to exogenous ABA, MeJA, GA3, and SA, suggesting that HbMYB44 is involved in a complex phytohormone signaling network.