OsCSN1 Regulates the Growth and Development of Rice Seedlings through the Degradation of SLR1 in the GA Signaling Pathway

Abstract

The constitutive photomorphogenesis 9 (COP9) signalosome (CSN) is involved in various regulations during plant development. The CSN is a highly conserved protein complex with nine subunits, and CSN1 acts in a network of signaling pathways critical for plant development. Although CSN1 has been widely studied in Arabidopsis thaliana, there have been few investigations on CSN1 in rice. In this paper, using the CRISPR/Cas9 system, CSN1 was edited from Oryza sativa subsp. japonica (rice). After screening out the OsCSN1 knockout mutant and OsCSN1 reduce mutant, the phenotype and protein expression were identified under different light conditions. Experiments showed that in OsCSN1 knockout mutant and OsCSN1 reduce mutant, the SLR1 protein was rapidly degraded at the rice seedling. In this study, the OsCSN1 acted as a negative regulator to affect seedling growth and development through CUL4-based E3 ligase, which is involved in the degradation of SLR1 in the GA signaling pathway. However, its direct target and mechanism of action are not clear.

1. Introduction

Light is one of the most needed environmental signals during plant development, participating in and regulating most plant developmental processes [1]. Plants rely on light to control cell differentiation, structural and functional changes, and eventually to assemble into tissues and organs [2]. The process of light-controlling plant development is called photomorphogenesis [2]. In studying the photomorphogenesis of Arabidopsis thaliana, a series of Arabidopsis photomorphogenesis mutants were obtained through genetic screening. The photomorphogenesis of these mutants was characterized by open apical hooks, short hypocotyls, expanded cotyledons, cell differentiation, chloroplast development, and anthocyanin accumulation, referred to as constitutive photomorphogenesis (COP) [3].

In higher eukaryotes (animals, plants, etc.), the COP9 signalosome (CSN) is a highly conserved protein complex [4]. CSN is acidic, with a molecular weight of about 560 kDa, and is a uniform spherical particle with a diameter of about 12 nm, mainly located in the nucleus [5]. The CSN regulates the plant response to light signals through in the protein degradation pathway, enabling plants to complete the transition from dark growth to light morphogenesis. CSN is considered to be an inhibitor of light regulation and development [6]. The CSN has two functional domains: the PCI domain and MPN domain, in which the PCI domain exists in CSN1-4, CSN7, and CSN8, and the MPN domain exists in CSN5 and CSN6 [4]. It was shown that the N-terminal helical repeat domains of PCI proteins (CSN1-4 and CSN7-8) radiate from the PCI ring at the bottom of the complex, where the most enormous helical bundle (CSN1-4) located on the PCI ring forms a prominent arm-like protrusion. Additionally, the heterodimers formed by the MPN domains of CSN5 and CSN6 are situated on the helical bundle [7]. The MPN dimer, helical pile, and PCI loop include a complex three-layer assembly. Since eukaryotic translation initiation factor 3 (eIF3), the 19S proteasome “lid” subcomplex, and CSN all contain two domains, PCI and MPN, and share this overall structure, it can be assumed that these three complexes are functionally similar and conserved [8].

The CSN can interact directly with the 26S proteasome, a proteasomal degradation pathway, by regulating the activity of ubiquitin E3 ligase [9]. The CSN is involved in and affects the aspects of the plant, regulating the nucleocytoplasmic distribution of COP1, regulating optical signal transduction, temperature signal transduction (cold response and heat stress), and hormone signal transduction in organic plants, etc. [10,11,12,13]. The CSN plays a vital role in cell division from the G₀ phase to the whole cell cycle. The knockout mutants of Arabidopsis CSN exhibit abnormal cell division, and individual subunits regulate cell cycle progression by controlling the activity of different cell cycle factors [14,15].

The CSN1 is the largest molecular weight of all subunits, and plays a vital part in the structural integrity of CSN [16]. The PCI domain of CSN1 is located at the C- terminal of the amino acid sequence. The CSN1 interacts with other subunits to promote the complex formation and maintain stability [16]. The N- terminal of CSN1 is not involved in the composition of CSN, but has an essential physiological function in plant development and is required for the nuclear localization of COP1 [17]. The CSN1 can interact with TSK-associating protein 1 (TSA1). Expression of TSA1 is significantly reduced in csn1 null mutants and regulates Arabidopsis hypocotyl development under dark conditions [18]. When a specific upstream signal is detected and the N-terminal end of the CSN is unable to bind to the complex, its cellular localization will translocate to the nucleus [19]. In the studies of Arabidopsis, it was found that CSN cannot be assembled, and the stability of each subunit would be destroyed when CSN1 was lacking. Thus, it can be seen the accumulation of CSN1 requires the assembly of a complete complex [20]. At the same time, the csn1 null mutant showed defective skotomorphogenesis and seedling growth stagnation, the mutant lacked Rub1/Nedd8 deconjugate activity on CULIINs. The phenotype of the mutant mainly depended on the function of the N-terminal of CSN1. Loss of CSN1 function can also lead to abnormal floral organ development, low fecundity, defects in the product of shoots, and inflorescences in Arabidopsis [21].

In plants, gibberellin (GA) and abscisic acid (ABA) in seed germination and dormancy are involved in the regulation, and these two hormones have antagonistic effects on seed germination. Seed germination was delayed in Arabidopsis csn5a-1 and csn1-10 mutants, and the two germination inhibitors expression RGL2 and ABI5 was increased, but the functional deletion of the CSN1 subunit in the csn1-10 mutant affected the RGL2 degradation [13]. The RGL2 excessive accumulation promoted the synthesis of ABA and activated the expression of ABI5, while CSN5A regulates the proliferation of ABI5 through another mechanism unrelated to CSN1 [13,22]. RT-qPCR was used to analyze the expression of eight OsCSN genes in 14-day-old rice seedlings. The results showed that the expression of these eight OsCSN genes was significantly reduced after GA treatment, and the transcript level of OsCSN1 was slightly increased after ABA treatment [23]. These results suggest that OsCSN genes may respond to plant hormones and perform different functions [23]. The length of the CSN1 homologous gene in rice is the same as that in Arabidopsis, with a total 441 amino acids of CSN1, and the amino acid sequence homology with AtCSN1 is 67% [24]. A previous study showed that rice homologous genes can ultimately rescue the defects of the AtCSN1 mutant, and can replace AtCSN1 to assemble CSN [25]. OsCSN1 overexpression mutants can make seedlings have longer hypocotyls and less anthocyanin pigmentation, and lead to enhanced CSN function. The abovementioned studies show that CSN1 plays a significant role in the germination process of plants.

The CSN1 is critical for structural integrity and functionality, and it is the largest-molecular-weight subunit in the CSN complex [19,26]. In rice, CSN1 research is still limited. Therefore, there is an urgent need to conduct studies on phenotypic changes in OsCSN1 mutants under different light conditions to elucidate the mechanism of action further. Because of the CSN complex and its highly conserved nature, this study used CRISPR/Cas9 technology to edit the CSN1 subunit of Oryza sativa subsp. japonica (rice). In addition, OsCSN1 knockout mutant and OsCSN1 reduce mutant were screened. Under various light conditions, phenotypic changes in the mutants were observed, and the OsCSN1 function was investigated. Therefore, it is necessary to further elucidate the growth and development mechanism of the OsCSN1 mutant in the seedling stage.

2. Materials and Methods

2.1. Cultivation Conditions of Plant Materials

O. sativa subsp. japonica (rice) was utilized as the no-treatment control in this paper. Use the CRISPR/Cas9 system to edit and screen out two homozygous mutants. During the reproductive cycle, it was sterilized with ethanol and sodium hypochlorite at O. sativa subsp. japonica (wild type) and oscsn1 mutants, washed with sterile distilled water, and soaked. Seeds on 0.8% (w/v) solid agar medium were germinated, and grown in glass tubes at 28 °C under various light conditions for 10 days. That the 10 days seedling mutants measured with the height of the plant, the length of the radicle, the length of the coleoptile, and the length of the first and second leaves. Plant samples were subjected to quantitative real-time PCR (qRT-PCR) and western blot (WB). All experimental materials were grown in a greenhouse at Jilin Agricultural University in Changchun, Jilin Province, China.

2.2. Vector Construction and Transformation of Rice

The knockout mutants of OsCSN1 (LOC_Os03g02540) by the CRISPR/Cas9-based genome editing method were constructed. Three targets guide RNA (sgRNA, Figure 1A) (http://cbi.hzau.edu.cn/cgi-bin/CRISPR, accessed on 4 December 2018) were selected according to the method of CRISPR/Cas9 Plant Gene Knockout Vector (Genloci Biotechnologies Inc., Nanjing, China) (Figure 1B). These sgRNAs using primers in

Table 1. All the primers that were used in this study.

Primer	Sequence (5′–3′)	Purpose
RG1	GCTATTTCTAGCTCTAAAAC-TGCGCGCTGGGATCGAAGGCT–TGCCACGGATCATCTGC	Plasmid construction
RG2	GCTATTTCTAGCTCTAAAAC-CTTGATCTCGTCATACGCCAT–TGCCACGGATCATCTGC	Plasmid construction
RG3	GCTATTTCTAGCTCTAAAACGAAATACGCGCTCGACCAGGT–TGCCACGGATCATCTGC
U3p3-F	CAGGAAACAGCTATGACCATATTCAAGGGATCTTTAAAC	Plasmid construction
JD-F3	CGTCTCGTCTCGCACTCTCGCATCG	Mutant detection
JD-R508	CCTGTAGCCATTGAGCTCGCTCTCG	Mutant detection
Hygjc2-F	GTCCGTCAGGACATTGTTGGAGCC	Mutant detection
Hygjc2-R	GTCTCCGACCTGATGCAGCTCTCGG	Mutant detection
RTCas9-F	AAGCCCATCAGAGAGCAGG	Mutant detection
RTCas9-R	TGTCGCCTCCCAGCTGAG	Mutant detection
G1	TGCGCGCTGGGATCGAAGGCT	sgRNA
G2	CTTGATCTCGTCATACGCCA	sgRNA
G3	GAAATACGCGCTCGACCAGG	sgRNA
GAPDHF	AAGCCAGCATCCTATGATCAGATT	q RT-PCR
GAPDHR	CGTAACCCAGAATACCCTTGAGTTT	q RT-PCR
Dye-ABI5F	TGGGATCTGGCATGGTCAAC	q RT-PCR
Dye-ABI5R	TACATGGCGTTTACCGGTCC	q RT-PCR
Dye-SLR1F	CATGCTTTCCGAGCTCAACG	q RT-PCR
Dye-SLR1R	TGACAGTGGACGAGGTGGAA	q RT-PCR
Dye-CUL4F	AGGACAGACAGTATCAGGTGGATGC	q RT-PCR
Dye-CUL4R	TCCGATGGCTTGATTGGGAACTTG	q RT-PCR
Dye-CSN2F	GAGCAGCTCTTGGTCTCACTCATTC	q RT-PCR
Dye-CSN2R	CGACCTGTCACCACGTTCTAGTAAC	q RT-PCR

were assembled into the vector pP1C.3 (RG1, RG2, RG3, and U3p3-F) (Table 1). Transformation of Agrobacterium tumefaciens strain EHA105 with the correctly sequenced vector [27]. Calluses were used to induce wild type mature seeds. Rice was transformed by Agrobacterium-mediated transformation using three constructed vectors (pP1C.3-sgRNA1, pP1C.3-sgRNA2, and pP1C.3-sgRNA3). Table 1 shows all the primers used in this study.

2.3. Detection of Mutations in T₀ Plants

The modified cetyltrimethylammonium bromide method was used to extract genome-wide transgenic DNA from leaves. The use of the Hygjc2-F and Hygjc2-R paired primers PCR confirmed the hygromycin phosphotransferase (HPT) gene presence, and the plants received another pair of primers, RTCas9-F and RTCas9-R (Table 1). This is a PCR method to identify the introduction and mutation, ensurung that all T₀ plants are positively transgenic.

2.4. Transgene-Free Mutant Lines Development

To obtain homozygous transgene-free mutants, DNA and proteins of T₁ seedlings were extracted for testing. At least five T₁ plants from each independent T₀ plant were selected for further analysis of site-specific mutations. We used a pair of primers to analyze T₁ genotype knockout mutants around the OsCSN1 target region by PCR amplification. Sequencing using PCR products generated by 2× Taq StarMix PCR using a loading die (GenStar, China) was performed to detect the OsCSN1 mutation sites. Two transgene-free mutant strains were successfully identified: oscsn1-580 and oscsn1-191 (Figure 2). The T₂ generation plants and their harvested seeds were used in subsequent experiments.

2.5. Protein Extraction and Western Blot/Antibody Analysis

Approximately fresh seedling tissue 0.5 g, 2.5 mM EDTA, 10 mM MgCl₂, 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 0.1% NP-40, 1 mM DTT, 1 mM PMSF, and 1× complete protease inhibitor cocktail (Roche, China) were used. We transferred the supernatant to a new tube after centrifugation for 10 min, added the same amount of total protein to each sample, and added sample buffer 5× SDS-PAGE (Genstar, China) for mixing. For western blot, boil the sample was mixed for 10 min and loaded onto 10% SDS-PAGE gel. The antibodies employed in this study were all purchased from Wuhan ABclonal Biotechnology Co., Ltd. These include primary rabbit polyclonal antibody OsCSN2, rabbit polyclonal antibody OsABI5, rabbit polyclonal antibody OsCUL4, and rabbit polyclonal antibody OsSLR1, anti-plant actin, and secondary antibody IgG HRP goat anti-rabbit antibody. Universal Hood III (731BR03292, BIO-RAD) was used to visualize the signal.

2.6. Real-Time RNA Separation and Quantitative Polymerase Chain Reaction

We used the Spectrum Plant Total RNA kit (Sigma-Aldrich, Darmstadt, Germany) to extract total RNA from seedlings. Reverse transcription was performed using StarScript II First Strand cDNA Synthetic Mix and gDNA Remover (GenStar, Beijing, China) using total RNA 0.5 mg as a template. We used 2× RealStar Green Fast Mixture with ROX (GenStar, China) to complete a rapid quantitative polymerase chain reaction on the StepOnePlusTM real-time PCR device (Applied Biosystems by Thermo Fisher Scientific, Shanghai, China). We used the blank control (O. sativa subsp. Japonica) GAPDH gene (JN848809) to quantify relative mRNA levels calculated by averaging three replicates. The details of the qPCR and RT-PCR gene-specific primers are shown in Table 1.

2.7. Analysis of Statistical

We used at least three biological replicates in each control and treatment. Fisher’s least significant difference (LSD) test was employed to compare the average value and SPSS 24.0 software for statistical data analysis.

3. Results

3.1. Overview of the Structure and Function of CSN1

The AtCSN1 included the RPN7 domain and PCI domain, and its crystal structure consisted of a PCI domain (PCID), a linker helix (LH), and two helical repeat domains (HR-I and HR-II) (Figure 1C) [16]. The PCI domain of AtCSN1 can be divided into two subdomains, the winged-helix (WH) and the helix bundle (HB) subdomain in the PCID. Additionally, the HR-I and HR-II are composed in the AtCSN1 RPN7 domain (Figure 1C). The LH, HR-I, and HR-II are respectively tandem arrays of two and three helix-turn-helix units. AtCSN1 and OsCSN1 molecules are compared to the tertiary structure to understand OsCSN1 in this study better. OsCSN1 has subdomains WH and HB in the PCI domain, but the AtCSN1 is longer than the OsCSN1 HB subdomain. However, in the RPN7 domain, the AtCSN1 is shorter than OsCSN1 because of has HR-II and LH structures in OsCSN1. In addition, it allows the protein to integrate into the COP9 signalosome in the C-terminal part of subunit 1 (CSN1-C) (Figure 1C). Therefore, in the present study, we focused on the functional analysis of CSN1 in the rice COP9 signalosome.

3.2. OsCSN1 Mutation Homozygous Transgene-Free Mutant Lines Development in T₀ Plants

In this study, three target guide RNA (sgRNA) (sgRNA1, sgRNA 2 and sgRNA 3, Figure 1A) sequences were designed in the exon1 of the OsCSN1 gene using the CRISPR/Cas9 system. One of them was chosen at the beginning of the coding region of the OsCSN1 gene (Figure 1A), and the others were chosen behind the beginning. This was predicted to raise mutations in the coding region, thereby inactivating the OsCSN1 protein. At T₀ generation, the transformation of the CRISPR/Cas9 vector with sgRNA1, sgRNA2, and sgRNA3 yielded 78, 67, and 53 independent transgenic plants, respectively. However, Cas9 and HPT implants identified approximately 150 double positives in T₀ plants (Figure 2A). Seeds from 87 mutated T₀ plants were collected and bred from CRISPR/Cas9 vector sgRNA1 and sgRNA2 transformations and grown as T₁ plant lines. Interestingly, mutant T₀ plants grown and transformed from the CRISPR/Cas9 vector sgRNA3 had no seeds.

In addition, the mutational effects of all T₁ plant lines were investigated, and each T₁ line sequenced approximately 10 surviving plants for the target site. At last, nine mutations were identified by PCR and sequencing, from single nucleotide (nt) deletion to short fragments, maximum length of 123 nt (Figure 2C). However, these mutants have a change in the size of the OsCSN1 protein. The other mutants displayed many nucleotide deletions from the N-terminal, which deleted a start codon in the respective genes and displayed the most noticeable reductions in the steady-state level of CSN1. Especially, oscsn1-191, a OsCSN1 reduced mutant, had diminished expression of the OsCSN1 protein. Additionally, oscsn1-580 is a OsCSN1 knockout mutant, and the OsCSN1 protein was knocked-out in the oscsn1-580 mutant (Figure 2C,D). Seeds of the stably inherited oscsn1-580 and oscsn1-191 mutants were obtained from transgene-free plants and used in subsequent experiments.

3.3. Under Dark Conditions, Oscsn1 Mutants Exhibit Photomorphogenic Growth of Rice

Light is one of the most critical environmental influences in plant growth and development. Plants depend on light to control cell differentiation and structural and functional changes, which eventually converge into tissue and organ building. The process of controlling plant development by light is called photomorphogenesis. The CSN is a negative regulator of photomorphogenesis. The growth state of rice coleoptile is an essential feature of photomorphogenesis. The experiment measured phenotypic data of 10-day-old wild type seedlings to analyze the regulated function of OsCSN1 by measuring coleoptile length, radicle length, plant height, and the incomplete leaf length of each line in oscsn1-191 and oscsn1-580 mutants. All oscsn1 mutants exhibited short coleoptile phenotypes in darkness. Additionally, under the light, the coleoptile length was significantly longer in the wild type than in the oscsn1 mutants (Figure 3A,B). These results suggest that mutant coleoptile elongation was significantly inhibited under dark conditions and exhibited a photomorphogenesis phenotype. Thus, oscsn1 and atcsn1 mutants play an inhibiting role in photomorphogenesis.

Moreover, the analyzed data revealed that both oscn1-191 and oscsn1-580 mutants had radically longer radicle length and plant height than the wild type under light and dark conditions. The oscsn1-580 mutant was significantly higher in the dark. The wild type plant height was shorter than that of the oscsn1-191 mutant (Figure 3A,B). Under light, the wild type first and second leaves lengths were shorter than those of the mutants (Figure 3A,B). The wild type second leaf length was appreciably shorter than that of mutants under dark, and the first leaf length of the mutants was appreciably different from that of the wild type (Figure 3C).

This paper also investigated the effect of two oscsn1 mutants on seed germination under light and dark conditions. The oscsn1-191 and oscsn1-580 mutants grew faster compared with the wild type (Figure 3C). These results suggest that OsCSN1 has supplementary functions in rice germination that may inhibit rice photomorphogenesis, and may suppress rice height at the seedling stage.

3.4. Oscsn1 Mutants Phenotype Comparison under Different Light Conditions

To analyze the regulatory effects of OsCSN1 on the photomorphogenesis of rice seedlings. The oscsn1 mutants and wild type were incubated under different light conditions of red, far-red, and blue light for 10 days, and the plant height, radicle length, incomplete leaves length, and coleoptile length were measured. Under red light, the wild type coleoptile was longer than the two oscsn1 mutants. The radicle length, plant height, and incomplete leaves lengths among the mutants were similar to the wild type (Figure 4A,D). Under far-red light, the wild type plant height, radicle length, and coleoptile length were much longer than oscsn1-580 mutant, but all wild type phenotypic data were longer than oscsn1-191 mutant. There were no significant differences between wild type data and other data (Figure 4B,D). Under blue light, the wild type radicle length was longer than the oscsn1 mutant. Both plant height and coleoptile length were much shorter in the wild type than in the oscsn1 mutant. However, the incomplete leaf lengths of the wild type and mutants did not differ much (Figure 4C,D).

These results indicated that red light inhibited seedling elongation of the oscsn1-191 mutants. Under blue light, the growth of wild type and mutants were inhibited, but that of the oscsn1 mutants were significantly suppressed (Figure 4C,D). These observations suggest that OsCSN1 is an inhibitor of red, far-red, and blue light-mediated regulatory processes and that the inhibitory effect of OsCSN1 is stronger under far-red light and blue light conditions than under red light conditions.

3.5. GA Biosynthesis Characterization of Oscsn1 Null Mutants at the Seedling Stage

Gibberellin (GA) is a growth-regulating phytohormone that affects various developmental processes in plants. In Arabidopsis, the gibberellin response pathway is negatively regulated by five DELLA proteins: RGA-LIKE1 (RGL1), RGA-LIKE2 (RGL2), RGA-LIKE3 (RGL3) [12,13], REPRESSOR OF ga1-3 (RGA), and GA-INSENSITIVE (GAI). The SLENDER RICE1 (SLR1) of DELLA protein is only one protein in rice [28,29].

In seedling growth and development, the wild type develops more slowly than the oscsn1 mutants. In addition, the oscsn1 mutants may regulate the level of seed dormancy by responding to the GA signaling pathway and promote plant growth and development at the seedling stage by responding to GA. It is hypothesized that in rice, OsCSN1 may have a function in the seedling stage that is not present in AtCSN1.

To test this hypothesis, OsSLR1 and OsABI5 protein levels were examined by western blot in wildtype and oscsn1 mutants grown in the light for 10 days. The SLR1 protein in oscsn1-580 and oscsn1-191 mutants exhibited a decrease compared with the wild type, inconsistent with the abnormal accumulation of SLR1 protein in Arabidopsis. The amount of ABI5 protein was significantly reduced in oscsn1-580 and oscsn1-191 mutants compared with the wild type (Figure 5A). Under red, far-red, and blue light, SLR1 protein expression was significantly up-regulated in oscsn1 mutants compared with light (Figure 5B–D). The ABI5 expression was significantly down-regulated in wild type and mutants compared with light (Figure 5B–D). The SLR1 protein showed up-regulation in mutants under blue light, but the ABI5 protein showed down-regulation (Figure 5B–D). The resulting analysis showed that knockout or reduce of OsCSN1 in response to different light caused abnormal synthesis or degradation of SLR1 and ABI5 in the oscsn1 mutant, thus affecting the oscsn1 mutant seedlings to exhibit a fast growth phenotype. Based on the analysis of the results, it can be concluded that the rapid-growth phenotype of the oscsn1 mutants in seedling may be due to abnormal hormone metabolism and biosynthesis.

In this study, qRT-PCR was used to analyze the gene expression of ABI5 and SLR1 in OsCSN1 knockout mutant and OsCSN1 reduce mutant (Figure 5B). Under light, the expression of the SLR1 gene in oscsn1-580 and oscsn1-191 mutants showed significant down-regulation compared with the wild type (Figure 5E). Additionally, the expression of the ABI5 gene in mutants showed up-regulation compared with the wild type (Figure 5E). The SLR1 gene expression in the oscsn1-580 mutant showed significant down-regulation under red and far-red light, and the SLR1 gene expression in the oscsn1-191 mutant showed no significant change compared to the wild type (Figure 5F,G). The ABI5 gene showed considerable down-regulation in mutants under red light and far-red light (Figure 5F,G). Under blue light, the SLR1 gene showed considerable up-regulation in oscsn1 mutants, and the ABI5 gene expression was unchanged (Figure 5H). The results suggest that OsCSN1 may be degraded and play a role in rice seedling growth and development through the DELLA protein SLR1 in the GA signaling pathway.

3.6. OsCSN1 Affects Seedling Growth via CUL4-Based E3 Ligase

The CSN2 is a component of CSN and is stable only when present in the form of a complex. It was found that the CSN2 protein interacts strongly with the CSN1 protein and with the CSN complex in Arabidopsis [30]. The CSN accumulates steadily only when all subunits are present. CULLIN-4 (CUL4) is a member of the CULLIN family of related ubiquitin ligases that primarily mediates the ubiquitination and subsequent proteasomal degradation of target proteins. The CUL4 can exist as a ubiquitin E3 ligase complex, and CSN regulates plant photomorphogenesis by controlling the RUB/NEDD8 modification of CUL4-based E3 ligases. It can also interact with the CSN complex in response to DNA damage [4,31,32].

In the study, the protein expression of CSN2 was significantly lower in oscsn1-580 and ocsn1-191 mutants under light than in the wild type (Figure 5A). Under red light, it resulted in a significant up-regulation of CSN2 protein expression in wild type and oscsn1-191 mutant compared with light (Figure 5B). Additionally, the CSN2 protein expression of wild type and mutant under far-red and blue light also showed some degree of up-regulation (Figure 5C,D). Under light, the CSN2 gene expression in oscsn1 mutants showed down-regulation compared with the wild type (Figure 5E). The OsCSN1 knockout mutant and OsCSN1 reduce mutant showed significant up-regulation of CSN2 gene expression under red, far-red, and blue light compared with light (Figure 5F–H). According to the analysis of the results, when OsCSN1 was knockout or had reduced expression, it resulted in a significant downregulation of CSN2 expression. Under light, the protein expression of CUL4 in the oscsn1-580 mutant showed significant down-regulation compared with the wild type. The protein expression of CUL4 in the oscsn1-191 mutant showed down-regulation compared with the wild type but to a lesser extent than that of oscsn1-580 mutant (Figure 5A). Under red light, the protein expression of CUL4 in wild type and mutants showed significant up-regulation compared with that under light (Figure 5B). The CUL4 protein expression in the wild type appeared up-regulated under both far-red and blue light compared to light (Figure 5C,D). The expression of CUL4 protein in the mutants under far-red light appeared up-regulated compared with light conditions. No significant changes were observed in the expression under blue light (Figure 5C,D). Under far-red light, OsCSN1 mutants showed significant up-regulation of CUL4 protein expression compared with the wild type. Among them, the CUL4 protein expression was significantly up-regulated in the oscsn1-580 mutant (Figure 5C). Under light, the CUL4 gene expression in the oscsn1-580 and oscsn1-191 mutants exhibited significant down-regulation compared with the wild type (Figure 5E). The expression of CUL4 gene showed significant up-regulation in mutants under light, far-red light and blue light (Figure 5F–H). Analysis of the results indicated that rice seedling growth may be regulated by the CSN complex controlling the CUL4 protein in the CULLIN-Ring family, which governs seedling growth or photomorphogenesis establishment in rice. Deletion of the OsCSN1 protein leads to some degree of abnormal expression of CUL4 protein.

4. Discussion

Previous papers have shown that the CSN is involved in seed germination and seedling growth, and promotes germination by regulating the degradation of the DELLA protein RGL2 and the ABA effector ABI5 in Arabidopsis. That showed apparent defects in the timely degradation of RGL2 in the atcsn1-10 mutant, and this mechanism does not exist in rice. The ABI5 protein is rapidly degraded during the seedling stage in the oscsn1 mutant. Furthermore, the oscsn1 mutant phenotype is significantly different from that of the atcsn1-10 mutant, showing a weak resting phenotype and a fast-growing phenotype. Yet, its direct molecular target and mechanism of action are still indistinct.

4.1. The Functional Difference between OsCSN1 and AtCSN1 Subunits in the Seedling Period

Since the function of CSN1 has been less studied in rice, OsCSN1 knockout mutant exhibit photomorphogenesis in the dark. In contrast, null mutants of CSN1 in Arabidopsis cause premature seedling death with a typical “fusca” phenotype [13,33]. Ineffective mutants of the OsCSN1 subunit exhibited a short germinal sheath phenotype in the dark, consistent with the role of AtCSN1. However, the ocsn1 mutant exhibited a fast-growth phenotype, which differed from the atcsn1-10 mutant [13,33]. These causes suggest that OsCSN1 and AtCSN1 have different regulatory mechanisms in germination and seedling growth and development, which they have additional functions in OsCSN1 seedling germination and growth.

However, the analyses showed that the phenotype of the oscsn1 mutant was different from the wild type in the following aspects. First, the oscsn1 mutant had a significantly shorter radicle sheath than the wild type under light and darkness (Figure 3A,B). Second, the phenotypes of the mutants differed under different lights (Figure 4). The wild type had a longer coleoptile length than the oscsn1 mutant under red conditions (Figure 4A,D). The oscsn1-191 and the oscsn1-580 mutant had different phenotypes: plant height, coleoptile length, and radicle length were suppressed in oscsn1-191 under far-red light conditions (Figure 4B,D). Seedling elongation was inhibited under blue light conditions in the oscsn1 mutant, to a much stronger extent than in the wild type (Figure 4C,D). The OsCSN1 may have an auxiliary function that acts differently, inhibiting photomorphogenesis and rice seedling height.

4.2. OsCSN1 Plays a Role in GA Signaling

By understanding the comparison, the OsCSN1 may have a different function than AtCSN1. It can decrease seed dormancy levels by responding to ABA in oscsn1 mutants, followed by seedling growth and development by responding to GA. This study showed that SLR1 and ABI5 proteins exhibited significant down-regulation in oscsn1 mutants (Figure 5A). The SLR1 protein expression in OsCSN1 knockout and reduce mutants showed significant up-regulation compared with light under red, far-red, and blue light. In contrast, ABI5 protein expression in wild type and mutants showed significant down-regulation compared with light (Figure 5B–D). At the mRNA level, the expression of the SLR1 gene showed considerable down-regulation in the two mutants under light. The expression of the ABI5 gene did not differ from the wild type (Figure 5E). Under red, far-red and blue light, the SLR1 gene expression in OsCSN1 knockout and reduced mutants showed significant up-regulation compared with light (Figure 5F–H). However, the ABI5 gene expression was up-regulated in OsCSN1 knockout and reduced mutants under red and far-red light (Figure 5F,G). There was no significant change in ABI5 gene expression under blue light (Figure 5G). The results indicated that OsCSN1 can directly or indirectly regulate the expression of the DELLA protein SLR1 in the GA signaling pathway and inhibit rice seedling development. The OsCSN1 may also inhibit ABI5 protein degradation at the seedling stage and cause ABI5 to be specifically modified in the OsCSN1 mutants.

The OsCSN1 knockout mutant and OsCSN1 reduce mutant showed significant down-regulation of CSN2 protein under light (Figure 5A). The CSN2 protein expression was upregulated in oscsn1 mutants under different light compared with light (Figure 5B–D). The changes in the CSN2 gene expression at the mRNA level in the two oscsn1 mutants under different light were consistent with the trend of protein expression. Combining the protein functions of CSN1 and CSN2, the analysis revealed that the presence of CSN1 affects the expression of CSN2 protein and even the assembly of the CSN complex. Red light can up-regulate the expression of CSN2. When CSN1 completely lost its function, there was no significant change in CSN2 expression. When in the OsCSN1 reduced mutant, the CSN2 protein expression could be restored to match that of the wild type. It is hypothesized that OsCSN1 can promote the expression of CSN2 in response to red light, and it will even promote the assembly of the CSN complex.

CULLIN-RING ligases (CRL) are the most prominent family of ubiquitin E3s regulated by the COP9 signalosome (CSN) [34]. The CUL4 protein is a member of the CULLIN-Ring family of proteins. CSN can control the RUB/NEDD8 modification of several CUL4-based E3 ligases, thereby regulating plant photomorphogenesis. The results showed that the OsCSN1 knockout resulted in a significant downregulation of CUL4 protein and CUL4 gene expression due to the complete loss of function of CSN1. In contrast, OsCSN1 reduction also resulted in down-regulation of CUL4 protein and CUL4 gene expression, but to a lesser extent than in the OsCSN1 knockout mutant. There was no significant change in the expression of the CUL4 gene at the mRNA level under red and blue light compared to that under light. Far-red light resulted in significant up-regulation of CUL4 gene expression in mutants. The changes in CUL4 protein expression in the two mutants under different light were consistent with the trend of gene expression. It is thus suggested that CSN1 may have some effect on rice seedling growth by regulating the expression of CUL4 protein. It may also have an impact on the photomorphogenesis of rice. We hypothesize that OsCSN1 may affect the degradation of SLR1 in the GA signaling pathway through CUL4-based E3 ligase, which affects rice seedling growth and development.

Based on the above results and analysis, the OsCSN1 may respond to different light and affect the photomorphogenesis of rice, which has some effects on the growth and development of rice seedlings. The OsCSN1 senses different light signals and exerts a crucial role in the seedling growth and development process through the GA signaling pathway. In this study, the OsCSN1 acted as a negative regulator to affect seedling growth and development through CUL4-based E3 ligase, which is involved in the degradation of SLR1 in the GA signaling pathway. However, the exact molecular mechanism of its action is still unknown, which will be the direction of future research. Studies on the function of OsCSN1 gene revealed that loss of OsCSN1 function resulted in rapid growth occurring in rice seedlings. The oscsn1 mutants exhibited significant fertility shortening. Therefore, it could be an important breakthrough affecting the shortening of fertility in rice. Based on the phenotype and related protein expression, the molecular regulatory mechanism of CSN1 affecting seedling growth was explored to improve grain production to meet the food demand of the growing population.