UBC19 is a new interacting protein of ORANGE for its nuclear localization in Arabidopsis thaliana

ABSTRACT

ORANGE (OR) is a member of the DnaJ-like zinc finger domain-containing protein family, of which all orthologs share a highly conserved quadruple repeat of the CxxCxxxG signatures at their C-termini. Dual subcellular localization and different interacting partner proteins have been reported for OR. In plastids, OR interacts with phytoene synthase, the entry enzyme for carotenoid biosynthesis, to promote chromoplast biogenesis and carotenoid accumulation in non-pigmented tissues. In the nucleus, OR interacts with the eukaryotic release factor eRF1-2 to regulate cell elongation in the petiole, and with the transcription factor TCP14 to repress the expression of Early Light-Induced Proteins (ELIPs) and chloroplast biogenesis in de-etiolating cotyledons. In this study, we demonstrated the E2 ubiquitin-conjugating enzyme UBC19 as a new interacting partner of OR. The lysine⁵⁸ of OR was found to be ubiquitinated, and OR lost its nuclear localization and the capability in repressing ELIPs when lysine⁵⁸ was substituted by alanine. Our findings raised the possibility that the ubiquitination by UBC19 is essential for the nuclear localization of OR.

The DnaJ-like zinc finger domain-containing protein ORANGE (OR) was initially identified from a cauliflower mutant, which accumulated a large amount of carotenoids in its curd.^1,2 Subsequent studies demonstrated that OR and its orthologs from different higher plants are highly conserved, all sharing a C-terminal quadruple repeat of the CxxCxxxG signatures.^2–4 A distinct function of OR and its orthologs is to trigger the development of plastids from the non-pigmented forms, such as amyloplasts and leucoplasts, into chromoplasts, which facilitates and enhances the biosynthesis and sequestration of carotenoids [5–7]. Such a function largely relies on the protein–protein interaction between OR and phytoene synthase (PSY), the entry enzyme for carotenoid metabolic pathway, in plastids.^7,8 In addition to plastids, OR was also found in the nucleus.^9–12 OR was demonstrated to interact with the eukaryotic release factor eRF1-2 in the nucleus to regulate the elongation of petiole cells.¹² Recently, we further identified the physical and genetic interactions between OR and the transcription factor TCP14 in the nucleus, through which OR represses the expression of Early Light-Induced Protein1/2 (ELIP1/2) and the development of chloroplasts in de-etiolating cotyledons.¹⁰ However, the molecular mechanism underlying the nuclear localization of OR was unknown.

To address this question, we first performed yeast two-hybrid (Y2H) screening. The full-length OR (OR^FL) was used as a bait to screen a normalized library prepared from mixed Arabidopsis tissues. After screening, positive clones harboring the coding region for the E2 ubiquitin-conjugating enzyme UBC19 (At3g20060) were identified. Because OR possesses an N-terminal chloroplast transit peptide (cTP) and a C-terminal zinc finger domain (ZF), separated by two transmembrane domains (TMs), we individually tested OR^FL and its truncated peptides before and after the TMs (OR^N and OR^C, respectively) for their interactions with UBC19 by pairwise Y2H assays (Figure 1a). Our result revealed that both OR^FL and OR^C interacted with UBC19, whereas OR^N did not show a positive interaction (Figure 1b). In A. thaliana, OR and UBC19 share the highest sequence similarities with OR-Like (AT5G06130) and UBC20 (AT1G50490), respectively.^7,13 By pairwise Y2H test, we found that OR-Like and UBC19 did not interact, neither did OR and UBC20 (Figure S1). Therefore, OR-Like and UBC20 might not function redundantly in our study.

Figure 1.

UBC19 interacts with OR

(a) Structures of OR and its N- and C-terminal truncates (OR^N and OR^C, respectively). cTP, chloroplast transit peptide; TM, transmembrane domain; ZF, zinc finger domain. (b) Yeast two-hybrid assay. OR or its truncates was cloned into pGBK-T7, and UBC19 was cloned into pGAD-T7. Yeast cells co-transformed with a combination of the indicated plasmids were plated in series dilution. EV, empty vector; DBD, DNA binding domain; AD, activation domain. DDO, double dropout (/-Leu/-Trp); QDO, quadruple dropout (/-Leu/-Trp/-His/-Ade). (c) Pull-down assay. Heterologously expressed and affinity-purified GST-UBC19 fusion protein was incubated with MBP-OR and glutathione particles. The bound protein was eluted, resolved by SDS-PAGE, blotted, and probed with the antibodies against the GST and MBP tags. GST and OR^N were used as negative controls. (d) BiFC observation showing that OR and UBC19 bind each other. A C-terminal fragment of yellow fluorescent protein (cYFP) was used as a negative control. The signal from the nuclear marker VirD2NLS-mCherry indicates the nucleus. Bar = 10 μm.

To verify the protein–protein interaction between OR and UBC19, we heterologously expressed UBC19 with a glutathione S-transferase (GST) fusion tag (GST-UBC19) and OR^FL with a maltose-binding protein (MBP) tag (MBP-OR^FL). A pull-down assay was performed using the affinity-purified fusion proteins. Our immunoblot analysis showed that OR^FL was captured by GST-UBC19, while no positive interactions were detected between OR^N and GST-UBC19, nor between OR^FL and GST (Figure 1c). These results demonstrated an in vitro interaction between OR and UBC19. We also performed a bimolecular fluorescence complemention (BiFC) assay to test the in planta interaction between OR and UBC19. We transiently expressed OR and UBC19 as fusion proteins with the N- and C-halves of enhanced yellow fluorescent protein (EYFP), respectively, in tobacco leaves by infiltration. After a 3-d growth in the dark, the reconstituted EYFP protein was mainly observed in the cytoplasm and the nucleus, agreeing with the localization of UBC19 as previously reported, and that of OR, which is translated in the cytoplasm before being targeted to either chloroplasts or the nucleus (Figure 1d).^11,13 Taken together, our results demonstrated UBC19 as a novel interacting partner of OR, while OR is also the first known interacting protein of UBC19.

To further resolve the nuclear localization of OR, we precipitated OR from 2-week-old OR-overexpression seedlings using the antibody against OR, and subjected to LC/MS-MS analysis after trypsin digestion. A ubiquitination modification at lysine⁵⁸ (K⁵⁸) of OR was identified (Table S1). We then generated a mutant of OR by substituting K⁵⁸ with alanine (OR^K58A), and individually fused OR and OR^K58A with EYFP (OR-EYFP and OR^K58A-EYFP, respectively). When OR-EYFP or OR^K58A-EYFP was co-expressed with the cyan fluorescent protein (CFP) in tobacco leaves, both OR-EYFP and OR^K58A-EYFP were observed in chloroplasts (Figure 2a). However, when co-expressed with a UBC19-CFP fusion protein, OR-EYFP was observed in both cytoplasm and the nucleus but not in chloroplasts, whereas OR^K58A-EYFP showed only chloroplast localization (Figure 2a). This result suggested the possibility that OR might be ubiquitinated at K⁵⁸ by UBC19 for its nuclear localization.

Figure 2.

Lys⁵⁸ is essential for the nuclear localization of OR

(a) Confocal observation of tobacco leaves infiltrated to co-express OR or OR^K58A (as EYFP fusion proteins) with UBC19 (as CFP fusion protein). CFP was used as a negative control. (b, c) Transcript abundances of ELIP1/2 in seedlings with different UBC19 (b) or OR background (c) post illumination. Transcript abundance was normalized against ACTIN2. Data are means ± SE, * P < .05, **P < .01 or better (Student’s t test, n = 3).

Because we recently reported that the nucleus-localized OR negatively regulates the expression of ELIP1/2, which are transiently induced during de-etiolation, we tested whether UBC19 was involved in such a regulation.^10,14 The transcript abundance of UBC19 itself was about constant during de-etiolation (Figure S2). However, significantly higher expression levels of both ELIP1 and ELIP2 were found in the ubc19 lines, especially at 3 and 6 h post illumination (Figure 2b, Figure S3). This result resembled the expression of ELIPs in the OR-silencing lines as previously reported,¹⁰ and indicated a de-repression of ELIPs by silencing UBC19. Although the overexpression of OR severely repressed ELIP1/2,¹⁰ the overexpression of UBC19 did not affect the expression of ELIP1/2 (Figure 2b). It is reasonable to postulate that a WT level of UBC19 was sufficient to target OR to the nucleus in de-etiolating cotyledons.

Moreover, we overexpressed the OR^K58A-EYFP fusion protein in the OR knockout mutant (OR-Cas) to test whether the repression of ELIPs could be regained (Figure S4). Different from the OR-GFP protein, which we previously proved to localize in both chloroplasts and the nucleus,¹¹ OR^K58A-EYFP was only observed in chloroplasts (Figure S4e). No significant variation in the expression of UBC19 was identified among the WT and OR-overexpressing, silencing, and complementation lines (Figure S4f). However, transcript abundances of ELIP1/2 in the complementation lines were similar to their corresponding levels in OR-Cas, which were significantly higher than the WT levels during de-etiolation (Figure 2c). Protein abundances of ELIP1/2 also showed similar variations among these lines (Figure S5). These results demonstrated that OR^K58A is incapable of repressing ELIPs during de-etiolation, which is different from the function of OR.¹⁰

In the past decade, OR has been reported to regulate various developmental processes through its protein–protein interactions with different partner proteins. Here, we added UBC19 as its novel interacting partner. OR has a characteristic C-terminal DnaJ-like zinc finger domain with 8 conserved cysteine residues, which might facilitate such protein–protein interactions.^2,3 This postulation was first demonstrated in this study, as our Y2H assay showed the specific interaction between OR^C and UBC19 (Figure 1b). It is possible that OR interacts with various pattern proteins under specific spatiotemporal circumstances. It would be interesting to explore its other interacting proteins, which might help to decipher a full plethora regulation mechanisms of OR.

Ubiquitination is an essential post-translational modification in eukaryotes.^15,16 There have been numerous reports on its regulation of plant development and acclimation.^17–19 During de-etiolation, the E3 ubiquitin-ligase CONSTITUTIVELY PHOTOMORPHOGENIC 1 (COP1) targets transcription factors such as HY5 and HYH for ubiquitination and degradation in the dark, and thus plays an essential role in the regulation of photomorphogenesis.^20–22 In this study, our results suggested that, during de-etiolation, the E2 enzyme UBC19 might be a new member of the regulatory machinery throught its interaction with OR, and also raised the possibility that the ubiquitination by UBC19 is essential for the nuclear localization of OR.

Materials and methods

Plant materials and growth conditions

Arabidopsis thaliana plants used in this study were all in the Columbia-0 wild-type (WT) background. In general, seeds were stratified in the dark at 4°C for 3 d and then germinated on Murashige-Skoog (MS) plates at 22°C under a light intensity of 120 μmol photons m⁻² s⁻¹ with a 16 h/8 h light/dark photoperiod. One-week-old seedlings were transferred to grow in the soil (a mixture of peat moss, vermiculite, and perlite at 1:1:1) under the same conditions.

Antibodies for immunoblot assays

Antibodies against ACTIN, GFP, GST, and MBP were from Sangon (Shanghai, China). Polyclonal antibodies against OR, ELIP1, and ELIP2 were prepared in our previous study.¹⁰ Horseradish peroxidase-conjugated secondary antibodies against rabbit/mouse IgG from Promega (Madison, WI, USA) and BeyoECL Star Western Blotting Substrate (Beyotime, Shanghai, China) were used for immunodetection. Standard protocols²³ and the manufacturer’s manuals for SDS-PAGE, semi-dry blotting, and immunodetection were followed.

Molecular manipulation and gene expression quantification

Genomic DNA was extracted from leaves using the cetyltrimethylammonium bromide method.²⁴ Total RNA was isolated using the RNAiso reagent (TaKaRa, Shiga, Japan), and cDNA was synthesized with a PrimeScript Double Strand cDNA Synthesis Kit (TaKaRa) following the manufacturer’s instructions. Gene expression levels were quantified by quantitative real-time PCR (qPCR) using TB Green Premix Ex Taq II (Tli RNaseH Plus) with a Thermal Cycler Dice Real-Time System TP800 (TaKaRa) following the manufacturer’s instructions. ACTIN2 (At3g18780) was used as a reference. The expression values were calculated according to the comparative C_T method.²⁵ For each sample, three biological replicates were analyzed, each with three repeats. All primers used in this study are listed in Supplemental Table 2.

Generation of transgenic plants

Total RNA was extracted from A. thaliana seedlings and reverse transcribed into a cDNA pool as previously reported for cloning OR and UBC19.¹⁰ The full-length open reading frame (ORF) of UBC19 was amplified from the cDNA pool. The amplicon was first cloned after the 35S promoter in the pRTL2 vector, and then the entire cassette was digested from pRTL2 and cloned into the multiple cloning site of pCAMBIA1300 to form the pCAMBIA1300-35S:UBC19 construct, which was then used to generate UBC19 overexpression (UBC19-OE) lines.¹⁰

For generating UBC19 knockout lines (ubc19), a nucleotide fragment (GATGCTGCTTCCACCCCAATG) was designed to edit UBC19 using the CRISPR/Cas9 strategy. The fragment was cloned into the BbsI site of the pCAMBIA1300-Cas9-AtU6 vector (kindly provided by Dr. Honggui La from Nanjing Agriculture University) to produce the pCAMBIA1300-Cas9-UBC19 construct.

For OR, its overexpression line OE#2 from our previous study was used.¹⁰ To generate its knockout mutant, a nucleotide fragment (GATGCTGCTTCCACCCCAATG) was designed to produce the pCAMBIA1300-Cas9-OR construct, which was then used to edit OR using the CRISPR/Cas9 strategy as mentioned above. For gene complementation, we re-designed the OR^K58A coding sequence, of which the gDNA region was re-coded to avoid editing but still encoded the same amino acids as the WT, and the codon for Lys⁵⁸ (AAA) was substituted by that for Ala (GCA). The entire fragment was chemically synthesized (Convenience Biology, Changzhou, China) and fused to the 5ʹ-end of EYFP in pA7-EYFP (ABRC, Columbus, OH, USA). The entire expression cassette was amplified, and subsequently cloned to pCAMBIA1300/BAR, a vector in which we substituted the hygromycin-resistance gene of pCAMBIA1300 with the Basta-resistance gene from pFGC5941 (ABRC) through homologous recombination. This generated the pCAMBIA1300-ORK58A complementation construct.

For each of the constructs, Agrobacterium tumefaciens was used to transform A. thaliana using the floral dipping method.²⁶ In general, the WT seedlings were used for transformation. However, for OR, we used the previously reported or-1 line, which contains a T-DNA insertion in its 5ʹ-UTR,¹⁰ for pCAMBIA1300-Cas9-OR to generate the gene edited OR-Cas lines, and used one of its progeny lines (OR-Cas#1) for pCAMBIA1300-ORK58A to generate the OR^K58A complementation lines (OR-Cas/OR^K58A). The transformants were screened on MS plates containing 50 mg L⁻¹ hygromycin B, whereas those of the OR-Cas/OR^K58A lines were further screened for their resistance to Basta, to homozygous for subsequent studies.

Protein–protein interaction assays

For yeast two-hybrid (Y2H) screening, full-length ORF of OR was amplified from the cDNA pool and cloned into pGBK-T7 vector (TaKaRa) to screen a Mate & Plate Library (Universal Arabidopsis Normalized, TaKaRa). Positive clones were screened on quadruple dropout (QDO, SD/-Leu/-Trp/-His/-Ade) plates according to the manufacturer’s instructions.

For pairwise Y2H assay, ORF of UBC19 was cloned into pGAD-T7 (TaKaRa) as a prey construct. The sequences corresponding to the N- and C-terminal fragments of OR (OR^N and OR^C, respectively) were separately amplified using the full-length OR as a template. OR, OR^N and OR^C were individually cloned into pGBK-T7 as the bait constructs. Yeast (Saccharomyces cerevisiae strain AH109) cells transformed with both bait and prey constructs were screened on QDO plates supplemented with X-α-Gal. Empty vectors were used as negative controls. For assessing the redundancies between homolog proteins, coding regions for OR-Like (At5g06130) and UBC20 (At1g50490) were similarly cloned and tested in parallel.

For pull-down assay, coding regions for OR and OR^N were individually cloned into pMAL-C5X (NEB, Ipswich, MA, USA) to express their corresponding fusion proteins with N-terminal MBP tags (MPB-OR and MBP-OR^N, respectively). The coding region of UBC19 was cloned into pGEX-4T-1 (GE Healthcare, Pittsburgh, PA, USA) for the expression of a GST-UBC19 fusion protein. Recombinant proteins were heterologously expressed in Escherichia coli BL21(DE3) (Merck Millipore, Darmstadt, Germany) cells and affinity-purified following the manufacturer’s instructions. Pull-down assay was carried out following our previous report using the MagneGST glutathione particles (Promega).¹⁰ Proteins captured by the particles were separated by SDS-PAGE and examined by immunoblotting using antibodies against GST and MBP.

For BiFC assay, ORFs of UBC19 and OR were cloned into pSAT1A-cEYFP-N1 and pSAT4A-nEYFP-N1 (ABRC), respectively, through homogenous recombination. Expression cassettes of both constructs, together with the cassette from the pSAT6-mCherry-VirD2NLS vector (for expressing VirD2NLS as a nuclear marker), were cloned into the I-SceI, AscI, and PI-PspI sites, respectively, of pPZP-RCS2-Bar (ABRC).¹⁰ Tobacco leaves were infiltrated and observed using the FLUOVIEW FV1000 Laser Confocal Microscopy System (Olympus, Tokyo, Japan) as previously described.¹⁰

Subcellular localization analysis

For detecting the co-localization of UBC19 and OR in planta, we cloned UBC19 in pCNHP-CFP, which we constructed based on pCAMBIA1300 and harbored sequentially the enhanced Cauliflower mosaic virus (CaMV) 35S promoter, synthetic 5ʹ and 3ʹ untranslated regions of Cowpea mosaic virus RNA2 flanking the coding region fused in frame to the 5ʹ-end of the gene for CFP, and the Heat Shock Protein (HSP) terminator from A. thaliana, to generate pCHNP-UBC19-CFP.²⁷ OR was cloned in pA7-EYFP as pA7-OR-EYFP. For generating the OR^K58A mutant, site-directed mutagenesis was introduced using the megaprimer strategy.²⁸ The PCR product was cloned into pA7-EYFP as pA7-ORK58A-EYFP. Different combinations of constructs for expressing CFP and EYFP fusion proteins were co-infiltrated to tobacco leaves. Fluorescent signals were observed after 3 d using the FLUOVIEW FV1000 Laser Confocal Microscopy System as previously described.¹⁰

Mass spectrometry

To determine the ubiquitination modification of OR, 2-week-old OE#2 seedlings were collected, frozen in liquid nitrogen, ground into fine powder, and then suspended in lysis buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 10% glycerol, 5 mM MgCl₂, 0.5% NP-40, 1 mM PMSF, 1 mM DTT). After a 10-min incubation on ice, the mixture was centrifuged at 4,000 g for 10 min. The supernatant was collected and further centrifuged at 12,000 g for 20 min. The supernatant was incubated with a mixture of Protein G beads (Promega) and the antibody against OR for 5 h. After incubation, the beads were washed with the lysis buffer for 5 times. OR was eluted from the beads by 100 mM glycine (pH 2.5) and subjected to SDS-PAGE. The gel bands were excised, in-gel digested, and analyzed by a Nexera UHPLC LC-30A system (Shimadzu, Kyoto, Japan) and a TripleTOF 4600 mass spectrometer (AB Sciex, Framingham, MA, USA) at the State Key Laboratory of Pharmaceutical Biotechnology of Nanjing University.

Statistical analysis

GraphPad Prism 7 (GraphPad Software, San Diego, CA, USA) was used for statistical analysis. To determine statistical significance, we employed Student’s t test. Differences were considered significant at P < .05.

Funding Statement

This work was supported by Guangdong Basic and Applied Basic Research Foundation [2020B1515120023].

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website.

Author contributions

Conceptualization, W.-C. C., Q. W. and S. L; Investigation, W.-C. C, Q. W, and T.-J. C; Writing, W.-C.C. and S.L. All authors have read and agreed to the published version of the manuscript.

Supporting information

Additional supporting information may be found online in the Supporting Information section at the end of the article