Arabidopsis NOTCHLESS plays an important role in root and embryo development

ABSTRACT

Ribosome biogenesis is a fundamental process in eukaryotic cells. NOTCHLESS (NLE) is involved in 60S ribosome biogenesis in yeast, but its role in Arabidopsis (A. thaliana) remains exclusive. Here, we found that Arabidopsis NLE (AtNLE) is highly conservative in phylogeny, which encoding a WD40-repeat protein. AtNLE is expressed in actively dividing tissues. AtNLE-GFP is localized in the nucleus. AtNLE physically interacts with the MIDAS domain of AtMDN1, a protein involved in the biogenesis of the 60S ribosomal subunit in Arabidopsis. The underexpressing mutant nle-2 shows short roots and reduced cell number in the root meristem. In addition, the null mutant nle-1 is embryo lethal, and defective embryos are arrested at the early globular stage. This work suggests that AtNLE interacts with AtMDN1, and AtNLE functions in root and embryo development.

KEYWORDS:

• Arabidopsis
• NLE
• MDN1
• root
• embryo

Introduction

The complicated process of ribosomal subunit assembly is initiated by transcription of a large pre-rRNA (45S in plants), which is subsequently modified and processed by a series of ribosomal biogenesis factors (RBFs), and assembled with the ribosomal proteins (RPs) to form a 90S ribosomal particle¹. After co-transcriptional cleavage of the 35S pre-rRNA, the 60S and 40S ribosomal subunits are in separate biogenesis processes and nuclear export patterns^2–6. The 40S small ribosomal subunit consists 18S rRNA and approximately 33 ribosomal proteins of small subunit (RPSs) and plays a role in mRNA decoding⁷. The 60S large ribosomal subunit consists 25S/28S, 5.8S, and 5S rRNA and approximately 47 ribosomal proteins of the large subunit (RPLs) and plays a role in peptide bond formation through the peptidyl transferase reaction⁷. The Cryo-EM structures of pre-ribosomes and ribosomes in prokaryotes and eukaryotes (especially in S. cerevisiae) facilitate our understanding of ribosome assembly and function, but the crystal structure and detailed assembly process of plant ribosomes are lacking^8–13.

A previous study in yeast suggests that more than 250 RBFs are involved in the modification and processing of rRNA, and the incorporation of RPs¹. The studies on RP and RBF genes suggest that the ribosome also plays a regulatory role in plant development^8,14–17. Despite the significant role of RBFs, few plant RBFs have been identified¹⁸. The AAA-ATPase MDN1 (also called Rea1 in yeast) consists of a AAA-ATPase domain, a linker domain, a D/E-rich domain, and a MIDAS (metal ion-dependent adhesion site) domain, and has been identified as an essential component of pre-60S particles^19–22. In yeast, Rea1 uses the MIDAS domain to contact with Ytm1 in the nucleolus before it contacts with Rsa4 (also called NLE in Arabidopsis) in the nucleoplasm^22–24. A mutation in AtMDN1 leads to the accumulation of 35S, 33S, 27SB, and 7S pre-rRNAs²⁵. Previous work has also shown that the UBL-MIDAS interaction is conversated in S. chacoense, S. cerevisiae and H. sapiens^9,26–28, however, their interaction and the function of NLE remain exclusive in Arabidopsis.

Depletion of NLE in yeast shows a strong slow-growth phenotype^9,29. Decrease AtNLE transcript levels using RNA interference (RNAi) method leads to semisterility defects which are caused by defects in female gametophyte development³⁰. Lower expression of S. chacoense NLE causes pleiotropic phenotype, such as a reduction in organ size and number, late flowering, and an increase in stomatal index²⁶. However, the role of AtNLE in root and embryo development has never been reported.

Results

Sequence analysis of AtNLE

The NLE homologs of Arabidopsis were obtained from the National Center for Biotechnology Information, including S. cerevisiae, D. melanogaster, H. sapiens, Z. mays, O. sativa, B. napus, R. sativus, C. sativa, G. hirsutum, C. sativus, C. annuum, I. nil, N. tabacum, S. lycopersicum, P. patens, and C. reinhardtii. A phylogenetic tree was constructed based on protein sequence homologies (Figure 1). This phylogenetic tree showed that the NLE gene is highly conservative in different species (Figure 1). NLE genes encode a protein that containing a WD40-repeat domain and a ubiquitin-like (UBL)domain (Figure 1). NLE proteins in S. cerevisiae, D. melanogaster, H. sapiens, Z. mays, O. sativa, B. napus, R. sativus, G. hirsutum, C. sativus, C. annuum, I. nil, N. tabacum, S. lycopersicum, and P. patens species contain eight WD40 domains, while NLE proteins in C. sativa and C. reinhardtii species have seven WD40 domains (Figures 1 and 2). Multiple sequence alignment revealed that the protein sequences of NLE in the different species are highly conservative (Figure 2).

Figure 1. Phylogenetic analysis of NLE. phylogenetic tree of NLE homologs from representative organisms and schematic comparison of NLE homologs. The orange boxes indicate UBL (ubiquitin like) domains and green boxes indicate WD40-repeat domains.

Figure 2. Multisequence alignment of NLE. alignment of NLE protein sequences from representative organisms labeled on the left. The UBL domains and WD40-repeat domains are labled on the sequences. A conserved E (E114 in yeast) residue which is critical for contacting with the MIDAS domain is highlighted using a asterisk.

Expression pattern of AtNLE

We examined the expression pattern of AtNLE in different tissues by qRT-PCR. The results showed that AtNLE is highly expressed in actively dividing tissues, such as root tip and shoot apex, and in dry seeds (Figure 3a).

Figure 3. Expression analysis of NLE. (a) Relative expression levels of NLE in the indicated tissues. Total RNA was isolated from the 10-d seedlings (for root tips, shoot apex and shoot), the rosette leaves of 30-d plants, the stems of 40-d plants, the flowers and siliques of 60-d plants, and mature seeds (10 days after pollination). actin-2 and actin7 were used as control for qRT-PCR and the relative expression of genes was calculated using the 2ΔΔCt method from three biological replicates. (b) distribution of the NLE-GFP protein. Scale bars, 100 μm.

A previous study showed that transient expression of ScNLE-GFP in tobacco protoplasts is located both in the cytoplasm and the nucleus²⁶. A bipartite nuclear localization signals (NLS) sequence was predicted for the AtNLE protein using an online prediction tool (http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi). To test the subcellular localization of AtNLE, we examined the distribution of AtNLE-GFP fusion protein driven by the CaMV 35S promoter in the root of 4 DAG (days after germination) seedlings. The AtNLE-GFP fusion proteins were restricted to the nucleus (Figure 3b).

Physical interaction between AtNLE and AtMDN1

The crystal structure shows that the MIDAS domain contacts with the UBL domain of Rsa4 in Chaetomium thermophilum (PDBID: 6QTA) (Figure 4a)³¹. To test the interaction between AtNLE and AtMDN1, full-length coding sequence (CDS) of AtNLE and the MIDAS domain of AtMDN1 were cotransformed in yeast to perform a two-hybrid (Y2H) assay. Increased β-galactosidase activity was observed, suggesting that AtNLE could directly interact with AtMDN1(Figure 4b). We also constructed the Nle domain (NLE-UBL) and the WD40-repeat domain (NLE-WD40) and tested their binding capability to AtMDN1. We found that the UBL domain was essential for AtNLE binding to AtMDN1 (Figure 4b). In addition, the bimolecular fluorescence complementation (BiFC) result also suggested that AtNLE might interact with AtMDN1 (Figure 4c). Taken together, our results suggested that Arabidopsis NLE might play a role in 60S ribosomal subunit biogenesis.

Figure 4. Physical interaction between MDN1 and NLE. (a) The crystal structure of NLE bound to the MDN1, PDBID: 6QTA. (b) yeast-2-hybrid assays on protein interaction between MIDAS and NLE, the UBL-domain of NLE (NLE-UBL), or the WD40-domain of NLE (NLE-WD40). (c) BiFC assay on protein interaction between MIDAS and NLE.

AtNLE is essential for embryo development

To characterize the function of NLE in Arabidopsis, two independent T-DNA insertion lines were obtained (Figure 5a). PCR-based genotyping was carried out to identify homozygous T-DNA insertion mutants. The homozygous nle-2 mutant was obtained, whereas no homozygous nle-1 mutant was obtained. The roots of nle-2 were shorter than that of the wild type at 7 DAG (Figure 5b). Besides, the number of meristem cells in nle-2 was also reduced (Figure 5b). Interestingly, white shriveled ovules were observed in the self-pollinated nle-1/+ siliques (Figure 5d). The embryos in the white shriveled ovules were arrested at the early globular stage, whereas the embryos in wild-type ovules were developed into the cotyledon stage (Figure 5e), suggesting that homozygous nle-1 was embryo-lethal. These results suggest that AtNLE plays an important role in root and embryo development.

Figure 5. Phenotypes analysis of NLE mutants. (a) The gene structure and the T-DNA inserted positions of the mutants. (b) the relative expression level of NLE in the nle-2 mutant. Values shown are averages ± standard errors (three biological replicates). (c) seedlings (scale bars, 1 mm) and RAM (scale bars, 20 μm) phenotypes at 7 DAG. The RAM was stained using PI (propidium iodide). White arrowheads indicate the cortex transition zones. (d) quantitative measurement of root length and the number of root meristem cells. Values shown are averages ± standard errors (n = 15). (***P < 0.001, Student’s t-test). (e) the phenotype of wild type (WT) and nle-1/+ siliques. Red arrowheads indicate abnormal seeds. (f) the embryo in the abnormal seeds. The yellow outline indicates the embryo. Scale bars = 50 mm.

Discussion

In S. cerevisiae, NLE is identified as an RBF that functions in 60S ribosome biogenesis³². The process of ribosome biogenesis is mainly studied in yeast, but most of the RBFs are well conserved in eukaryotes, such as MDN1/Rea1 and NLE/Rsa4³⁰. Therefore, the process of ribosome biogenesis is expected to be similar throughout yeasts, human, and plants⁵. About 170 plants RBFs orthologs of 255 yeast RBFs have been identified in 14 plant species³³. However, only a small part of these RBFs has been characterized in plants and the general information of ribosome biogenesis in plants is less known^18,34. In this study, we characterized the NLE gene in Arabidopsis.

The AtNLE gene encodes a protein comprising a WD40-repeat domain and a Nle domain (Figures 1 and 2). Phylogenetic analysis and multiple sequence alignment of NLE from different organisms further suggest that NLE is well conserved in eukaryotes (Figures 1 and 2). The N terminal of MDN1 contains a six AAA-ATPase domain, which is involved in the interaction with the pre-60S ribosomal subunit⁹. The C terminal of MDN1 is a MIDAS domain, which involves in the NLE-MDN1 interaction in yeast and S. chacoense⁹. Our previous work shows that MIDAS interacts with the UBL (ubiquitin-like) domain of AtPES2, a yeast Ytm1 homolog²⁵. Here, we show that MIDAS interacts with the UBL domain of AtNLE (Figure 4). Besides, AtNLE shows a similar expression pattern with that of AtMDN1 (Figure 3a)²⁷. AtNLE-GFP localizes in the nucleus, which is consistent with the AtMDN1 (Figure 3b)²⁵. These results suggest that AtNLE might also function in ribosome biogenesis.

Mutants of RBFs normally display common defects phenotypes, including defects in embryogenesis and root growth^15,18. AtNLE-RNAi lines display a delay and/or an arrest in female gametophyte development³⁰. Here, we further found that the null mutant nle-1 causes embryo lethal at the early globular stage (Figure 5d). Underexpressing mutant nle-2 displays a short root phenotype (Figure 5b, c). We previously also showed that the embryo development of the null mutant of MDN1 (mdn1–2) is arrested at the early globular stage²⁵. These common defective phenotypes might be tightly associated with their roles in ribosome biogenesis.

In conclusion, this work suggests that Arabidopsis NLE might function as a 60S RBFs and plays an important role in root and embryo development.

Materials and methods

Plant materials and growth conditions

Arabidopsis ecotype Columbia (Col-0) was used in this study. The T-DNA insertion mutants of NLE (nle-1, SALK_032034, and nle-2, SALK_114276) were obtained from the Arabidopsis Biological Resource Center. Seeds were surface-sterilized using H₂O₂ (10%) and plated on solid 1/2 Murashige and Skoog (MS) medium containing 1% sucrose (pH 5.7). After 3d stratification, seeds were moved to the greenhouse (16 h light, 21°C; 8 h dark, 21°C).

Construction of transgenic plants and identification of T-DNA insertion mutants

For the 35S:NLE-GFP, the CDS of NLE was amplified by RT-PCR (NLE-GFP-F: GGTACCATGACCATGATGGATACAGACGAA, NLE-GFP-R: GGATCCACCCTTCCATAGCTTCAACACTCT) from total mRNA extracted from 7 DAG Col-0 Arabidopsis seedlings and cloned into a modified pCAMBIA2300 vector. A GFP sequence was inserted between the SalI and PstI. The construct was transformed by Agrobacterium tumefaciens (GV3101)-mediated floral dip method. For selection, 50 µg/mL Kanamycin was used. The T-DNA insertion mutants were identified by PCR-based genotyping with the primers (LBb1.3:ATTTTGCCGATTTCGGAAC, nle-1 LP: AAAGCAATCTACTGTCTGCGG, nle-1 RP: CCTCCATGTTTAAGCTCCATG, nle-2 LP: ATCAACCAATACACAGAGGCG, nle-2 RP: CGTTCAGGTTTTGCTGATAGC) obtained from T-DNA Primer Design (http://signal.salk.edu/tdnaprimers.2.html).

Protein interaction assay

The Y2H procedure and β-galactosidase activity assay were performed according to the Yeast Protocols Handbooks (Clontech). The C terminal sequence of MDN1 (MIDAS domain, MIDAS-Y2H-F: ggaattcCATATGATGACCAACATGGCCAACGG

TGAG, MIDAS-Y2H-R: cgcGGATCCTCAGTCCCGCGAGCTTTGCATCAG) was cloned into pGADT7 (Clontech). The full-length CDS sequence of NLE (NLE-Y2H-F: GGAATTCCATATGATGACCATGATGGATACAGACGAA, NLE-Y2H-R: cgcGGA

TCCTTAACCCTTCCATAGCTTCAACAC), the NLE N terminal sequence (UBL domain, UBL-Y2H-F: GGAATTCCATATGATGACCATGATGGATACAGACGAA, UBL-Y2H-R: CgcGGATCCTTAGTTAACAGGACGAATTCGAAA), and the NLE C terminal sequence (WD40-repeat domain, WD40-Y2H-F: GGAATTCCATATGATGC

GTTGCTCACAGACAATTGCT, WD40-Y2H-R: cgcGGATCCTTAACCCTTCCAT

AGCTTCAACAC) were cloned into pGADT7 (Clontech).

For BiFC assay, the MIDAS domain (MIDAS-BiFC-F: cgcggatccATGGAAAGCAATCAGGATAATCAGGAA, MIDAS-BiFC-R: ccgggtacc

GTCCCGCGAGCTTTGCATCAG) was cloned into pSPYNE vector and full-length NLE was cloned into pSPYCE vector (NLE-BiFC-F: cgcggatccATGACCATGATGGA

TACAGACGAA, NLE-BiFC-F: ccgctcgagACCCTTCCATAGCTTCAACACTCT). Agrobacterium tumefaciens (GV3101) containing the pSPYNE and pSPYCE construct were coinfiltrated into leaves of 20 DAG N. benthamiana. The fluorescence was observed 48 h after infiltration in the abaxial side of leaf epidermis using a confocal laser scanning microscope (OLYMPUS FV1200). Imaging of YFP and DAPI fluorescence was performed sequentially.

qRT-PCR

Total RNA was extracted from various tissues using RNAiso Plus Reagent (TAKARA) and DNase Kit (TAKARA). cDNA synthesis was performed using the PrimeScript^TM II 1st Strand cDNA Synthesis Kit (TAKARA). The qRT-PCR (qNLE-F: AGTGTCTGTGGAGAAGGT, qNLE-R: GTCTGTGAGCAACGGTTA) was performed using the SYBR Premix Ex Taq (TAKARA). Actin 7 was used as the control for normalization and ^△△CT method was used for the calculation of relative gene expression.

Bioinformatic analysis

Arabidopsis NLE protein sequence was obtained from TAIR10 (https://www.arabidopsis.org/) and was used for search NLE homologs using the protein basic local alignment search tool (BLASTp, http://blast.ncbi.nlm.nih.gov/BLAST/). Protein NLS sequence was predicted using NLS Mapper (http://nls-mapper.iab.keio.ac.jp/cgi-bin/NLS_Mapper_form.cgi). Multiple sequence alignment was performed using Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/) and Jalview.

Accession numbers

Protein sequence data for the NLE homologs used in the phylogenetic analysis with the following accessions numbers: NP_477294.2 (D. melanogaster), NP_009997.2 (S. cerevisiae), NP_060566.2 (H. sapiens), NP_001278731.1 (Z. mays), XP_015615305.1(O. sativa), XP_013680604.1 (B. napus), XP_018449712.1 (R. sativus), XP_019091290.1 (C. sativa), XP_016721780.1 (G. hirsutum), XP_004149723.1 (C. sativus), XP_016563668.1 (C. annuum), XP_019151006.1 (I. nil), XP_016508648.1 (N. tabacum), XP_004236242.1 (S. lycopersicum), XP_024372534.1 (P. patens), XP_001700359.1 (C. reinhardtii). Arabidopsis genes accession numbers were deposited in the TAIR10: NLE (AT5G52820) and MDN1 (AT1G67120).

Authors’ contributions

Ke Li and Pengfei Wang designed the study. Ke Li, Qingtian Zhang, Huiping Liu, Fengxia Wang, Ao Li, Tingting Ding, Qian Mu and Hongjun Zhao performed the research and analyzed the data. Ke Li and Pengfei Wang contributed to writing the article. All authors reviewed and approved the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (32100266), Shandong Natural Science Foundation (ZR2021QC022, ZR2020QC145 and ZR2022QC076), Shandong Academy of Agricultural Sciences, Introduction and Training of High-level Talents (CXGC2022E15, CXGC2022F15 and CXGC2023F15), Improved Variety Program of Shandong Province (2020LZGCQY019 and 2020LZGC008), the Key Research and Development Project of Shandong Province (2021LZGC025 and 2022CXGC010605), Scientific Research Guide Foundation of Shandong Academy of Grape (SDAG2021A01 and SDAG2021A02), Innovation Project of Shandong Academy of Agricultural Sciences (CXGC2022F38, CXGC2023A22, CXGC2023A41, CXGC2023C02 and CXGC2023G32) and Yantai Science and Technology Plan Project(2022XCZX086).