Cadmium (Cd) is toxic to animals and plants. Extensive Cd contamination in soils can lead to cadmium accumulation in edible parts of crops, especially rice, threatening the food safety of rice consumption (Uraguchi and Fujiwara, 2012). To minimize the risk of soil Cd entering the food chain, understanding plants’ molecular response network to Cd and in-depth exploring the molecular mechanism of Cd stress have become an important direction in agriculture and environment studies (Pena et al., 2012; Wang et al., 2016; Chang et al., 2019).

The development of multicellular organisms relies on the orchestrated spatiotemporal regulation of cell division, including mitotic cell cycle and cell expansion. Therefore, cell cycles must be integrated into a complex histogenesis and organogenesis system (Mironov et al., 1997; Meijer and Murray, 2001; Stals and Inzé, 2001; Veylder et al., 2003). The plant cell cycle consists of four distinct phases: postmitotic interphase G1, DNA synthesis phase S, pre-mitotic interphase G2, and mitosis/cytokinesis phase M (Qi and Zhang, 2019). During the G1/S transition, the A-type cyclin-dependent kinase/Cyclin D (CDKA/CYCD) complex, via phosphorylating retinoblastoma-related (RBR) proteins, activates the S-phase E2 promoter binding factor (E2F). Activated E2F then regulates the expression of genes involved in DNA replication, cell cycle progression, and chromatin dynamics transition, thereby promoting the G1/S transition (del Pozo et al., 2006).

Both typical and atypical E2F transcription factors are key components of the cyclin D/RB/E2F pathway. Typical E2F proteins possess a homologous DNA-binding domain and form heterodimers with Dimerization part (DP) proteins to bind DNA via the leucine zipper in their dimerization region. In contrast, atypical E2F/DEL factors do not interact with DPs due to lacking a dimerization region. However, they have a homologous DNA-binding domain for replication, enabling them to bind DNA autonomously (Perrotta et al., 2021). In addition, typical E2F proteins also possess a conserved C-terminal transactivation domain that is absent in DPs and in atypical E2F/DEL proteins, allowing transcriptional activation of their target genes (Dimova and Dyson, 2005). Arabidopsis E2F/DP family includes three typical E2Fs (AtE2Fa, AtE2Fb, and AtE2Fc), three atypical E2Fs (AtE2Fd/DEL2, AtE2Fe/DEL1, and AtE2Ff/DEL3) and two DPs (Mariconti et al., 2002). Carrot E2F/DP family consists of four typical E2Fs, three atypical E2Fs, and three DPs (Perrotta et al., 2021). Yeast, which is widely used as an in vivo model to explore cell cycle response mechanisms, also utilizes E2F homologs MCB-binding factor (MBF) and SCB-binding factor (SBF) proteins to regulate G1/S transition and cell proliferation (Morshed et al., 2020). Previous studies have identified four E2Fs, three DPs, and two DP-E2F-like (DELs) in the rice genome, but little is known about their taxonomy and functions (Guo et al., 2007).

Proper cell division and cycling requires tight regulation of key cell cycle-related genes. RBR proteins act as negative regulators of E2F transcription factors and are crucial for E2F to function properly (Weintraub et al., 1992). The CDKA/CYCD complex regulates E2F/DP family activity via RBR proteins, which mediate G1/S transition. CDKA forms complexes with A, B, or D-type cyclins to drive G2/M transition (Gutierrez, 2009). The dramatic repression of Mitotic arrest deficient protein 2 (MAD2) expression may mediate G2/M arrest through a dual mechanism to regulate chromosome segregation (Cao et al., 2018). ATM and Rad3-related (ATR)is a key component of the G2/M checkpoint in plant cells and is activated by the mitogen-activated protein kinase (MAPK) signaling pathway via phosphorylation (Yoshioka et al., 2006).

Many replication- and mismatch repair-related genes in Arabidopsis contain conserved E2F binding sequences in their predicted promoter regions (Kosugi and Ohashi (2002); Vandepoele et al., 2005). Previous work on the Arabidopsis genome at E2F binding sites has identified over 180 potential E2F target genes associated with transcription, stress, DNA damage, plant defense, and signaling transduction, in addition to the cell cycle (Ramirez-Parra, 2003). Among them are genes related to mismatch repair (MMR), such as Mutated S homologue 3 (MSH3), Mutated S homologue 6 (MSH6), and Postmeiotic segregation increased 2 (PMS2). In addition, the putative rice homologs are listed.

The most important role of the MMR system is to identify and correct mispaired or unpaired bases (Schofield and Hsieh, 2003; Iyer et al., 2006). Eukaryotic cells rely on high-fidelity DNA replication to maintain genomic integrity and depend on DNA MMR to ensure proofreading of incorrect pairings. Thus, any deletion in MMR genes can lead to spontaneous mutations in organisms (Zhang et al., 2005). In addition to correcting base-base mismatches, MMR genes are involved in suppressing mutations and inducing protective responses to various types of DNA damage. MMR plays multiple roles in response to various DNA damage inducers, such as nucleotide methylation, oxidative DNA damage, and UV-induced DNA damage, among other degenerative damages (Ijsselsteijn et al., 2020). Interestingly, cadmium interacts with DNA repair systems, cell cycle checkpoints, apoptosis-related epigenetic factors, and factors controlling gene expression. Cadmium can bind directly to DNA at very low concentrations, inducing various DNA damages such as base-base mismatches, insertion/deletion loops, strand cross-links, and breaks (Filipic, 2012). DNAs under stress can induce both complex and precise repair mechanisms and signal transduction pathways in eukaryotic cells and a stage-specific arrest of the cell cycle process (Wang et al., 2013; Hu et al., 2016; Xiang et al., 2017). Mutated S α (MutSα), Mutated S β (MutSβ), and Mutated S γ (MutSγ) are three different MutS-associated protein heterodimer complexes consisting of MSH2/MSH6, MSH2/MSH3, and MSH2/MSH7, respectively. These complexes have different mismatch recognition properties and abilities to support MMR (Wang et al., 2020). In plants, Knockdown of OsMSH2 does not cause gene variants at other locations in the genome (Karthika et al., 2021). Mutated S homologue 2 (MSH2) preferentially activates ATR to trigger DNA damage responses (DDR), including regulation of cell cycle, endoreplication, cell death, and recruitment of other DNA repair, thereby enhancing plant tolerance to cadmium (Wang et al., 2020). OsMSH6 affects rice microsatellite stability and homologous recombination and plays an important role in ensuring genome stability and genetic (Jiang et al., 2020). MSH7 is a plant-specific protein similar to MSH6, and in rice OsMSH7 is able to interact with Meiotic chromosome association 1 (MEICA1) and play a role in meiotic recombination (Hu et al., 2017). AtMSH2 and AtMSH6 are involved in G2/M or G1/S transitions in Arabidopsis and soybeans. MSH2 and MSH6 may be the direct sensors of cadmium-mediated DNA damage. Expression of DNA mismatch repair-related genes AtMSH2, AtMSH3, AtMSH6, and AtMSH7 can be used as potential biomarkers for evaluating cadmium exposure in Arabidopsis seedlings (Liu et al., 2009).

In this study, we systematically classified E2Fs in rice, determined its relationship with downstream MSH target genes, and then explored the phenotype of rice E2F and MSH under cadmium stress. These conclusions combined with transcriptome and quantitative results, established a model hypothesis of E2Fs-MSHs.The study revealed that E2Fs co-participate in responses to cadmium stress by binding to MSHs during the cell cycle.

The E2F transcription factor is a key component of the RB/E2F pathway, which is controlled by cyclin-dependent kinases and regulates cell-cycle progression in plants and animals (Perrotta et al., 2021). Plant homologues of E2F have been tentatively identified in rice and Arabidopsis (Kosugi and Ohashi (2002); Mariconti et al., 2002), In this report, we extend the information on E2F factor in rice and conclusively confirmed the previous prediction (Ramirez-Parra, 2003; Vandepoele et al., 2005) that E2Fs can target and bind to MSHs, MutS is a key link in the MMR system. Plants lacking MutS will bypass MMR-mediated DDR, thereby reducing the tolerance of plants to cadmium. This has been shown in soybeans and Arabidopsis (Cui et al., 2017; Cao et al., 2018; Zhao et al., 2020), our study further revealed that E2F improves rice’s tolerance to cadmium stress by binding to MSH, indicating that E2F together with MSHs responds to cadmium stress in the cell cycle. To the best of our knowledge, this is the first study on the relationship between E2F/MSH and cadmium stress in rice. The model developed here provides a preliminary explanation of the response mechanism of the E2F/MSHs pathway to cadmium stress in rice and a theoretical basis for creating cadmium-tolerant rice varieties.

Through the evolutionary tree analysis and amino acid sequence comparison of rice and Arabidopsis E2F transcription factors, we found that rice OsE2Fa-1, OsE2Fa-2, OsE2Fa-3, and OsE2Fc are functional transcription factors belonging to typical E2Fs ( Figure 1 ), which, like AtE2Fs, have conserved C-terminal trans-activation domains and can specifically recognize E2F cis-elements and activate downstream target genes. By contrast, rice OsE2Fe-1 and OsE2Fe-2 are atypical E2Fs that possess DNA binding sites but lack transcriptional activation domains and cannot activate downstream gene expression. This further illustrates the complex regulatory mechanism of E2F transcription factors in rice cells (Rossi and Varotto, 2002).

Our LUC results showed that the tandem sequence consisting of cis-elements could bind to four typical transcription factors (OsE2Fa-1, OsE2Fa-2, OsE2Fa-3, and OsE2Fc) and atypical OsE2Fe-1, which also confirms that atypical E2F does not affect DNA binding in the absence of the transcriptional activation domain. Our activation assays showed that other rice E2F can directly or indirectly activate s OsMSH2, OsMSH3, OsMSH6, and OsMSH7, except OsE2Fe-2. These results confirmed the prediction that the cis-acting element of each gene could bind to 1-3 transcription factors and validated previous microarray analysis of the predicted target loci. In addition to the previously predicted OsMSH3 and OsMSH6, our results suggest that OsMSH2 and OsMSH7 are also the targets of E2F (Vandepoele et al., 2005).

OsE2Fa-1 activates to OsMSH2, OsMSH6, and OsMSH7 ( Figure 2 ), where OsMSH2 and OsMSH6 form MutSα heterodimers. MutSα recognizes single base mismatches such as polymerase errors, small insertion/deletion loops, and oxidative mismatches and methylation mismatches, which may be caused by translocation synthesis (TLS), (Zhang et al., 2002; Iyer et al., 2006; Campregher et al., 2008; Lippard, 2009). MSH2 and MSH7 form MutSγ heterodimers, which mainly recognize single base mismatches, including G/G, G/A, A/A, and C/A mismatches (Culligan and Hays, 2000; Wu et al., 2003; Gomez and Spampinato, 2013). In addition, MSH2 and MSH3 form MutSβ heterodimers, which sense insertion/deletion cycles and inter-strand cross-links (ICLs) (Marti et al., 2010). All of these suggest that OsE2Fa-1 is a key transcription factor that activates MMR-related genes in canonical E2F. Other canonical E2F genes such as OsE2Fa-2 activation of OsMSH3 and OsE2Fc activation of OsMSH7 also fully demonstrate that canonical E2F does not act alone and needs to cooperate with other MMR genes in MutSα, MutSβ, and MutSγ heterodimers.

The cadmium stress tests of the mutant and wild-type rice showed that the height of plants without cadmium stress and under 2 mg/L cadmium stress differed the most, possibly because rice leaves were most sensitive to 2 mg/L CdCl₂. At CdCl₂ > 2 mg/L, plants’ defense system against cadmium stress was damaged, leading to no significant difference ( Figure 3 ). The differences in plant height between wild-type and mutant rice under different cadmium stresses reflected the important roles of OsMSH2, OsMSH3, OsMSH6, and OsMSH7 in rice response to cadmium stress, indicating that MutS is involved in rice cadmium stress response. In Arabidopsis, the root growth of MSH2 or MSH6-deficient seedlings under Cd stress (1.25-4.0 mg/L) was much more inhibited than that of wild-type seedlings, which was consistent with our results (Cao et al., 2018).

Yeast heterologous expression assays of E2F genes showed significant differences in the growth of E2F-expressing yeast strains under 5 μM and 10 μM cadmium treatments, with yeast strains expressing OsE2Fa-1, OsE2Fa-2 and OsE2Fc growing slightly better than those expressing other genes, possibly because they are typical E2Fs with transcriptional activation domains that bind to downstream target genes ( Figure 5 ). The results of LUC assay of E2F binding to the downstream MSHs gene and the phenotype of each msh mutant seedling under cadmium stress further indicate that E2F responds to cadmium stress in yeast by mobilizing its downstream MSH components.

Cadmium disturbs epigenetic modification and induces DNA damage in mouse preimplantation embryos and soybeans (Zhao et al., 2020; Cheng et al., 2021; Zhu et al., 2021). Cadmium exposure can affect oxidative stress, cell cycle arrest, DNA damage, and apoptosis in green crabs (Scylla paramamosain). Once cadmium-induced DNA damage is detected in plants, DDR is triggered. DDR is a cellular response to DNA damage, including regulation of DNA damage recognition and recruitment of DNA repair factors. Inhibition of cell cycle regulatory genes during DDR contributes to cell proliferation arrest (Jackson and Bartek, 2009; Schade et al., 2019). Therefore, genes related to mismatch repair and cell cycle regulation are, in principle, located in the nucleus. Our experiments demonstrated that OsMSH2/3/6/7 are localized in the nucleus and function together with cell cycle regulatory genes in response to cadmium stress ( Figure 6 ).

Adaptation to changes in cellular nutrition and external environments is a fundamental cellular property. This adaptation requires the coordination of multiple networks among metabolic, growth, and cell cycle regulators, including the CDKfamily, members of the RBfamily, and E2F transcription factors (Huber et al., 2021). In response to external growth stimuli, including plant hormones, the abundance of specific G1 cell cycle proteins increases (Riou-Khamlichi et al., 2000). The CYCD-CDKA;1 complex phosphorylates RBR at multiple conserved sites, releasing activator E2F from bound RBR, thereby inducing the expression of downstream cell cycle genes (Nakagami et al., 2002; Magyar et al., 2012). Our qRT-PCR further confirmed the RNA-seq results, showing significant upregulation of CDKA;1 and CYCD as the front end of G1/S activation, significant downregulation of RBR as the response to being phosphorylated, and significant upregulation of OsE2Fa-1 and OsE2Fa-2 due to being released from RBR ( Figures S2B, C ). All the results of LUC assay, yeast spot plate assay, and qRT-PCR indicated that the mobilization of E2F transcription factors is the key factor responsible for significant upregulation of OsMSH2, OsMSH3, OsMSH6, or OsMSH7 under cadmium stress. OsMSH2, OsMSH3, OsMSH6, or OsMSH7 were significantly downregulated in rice leaves in response to cadmium stress, contrary to their expression patterns in Arabidopsis roots. This discrepancy is possibly because cadmium stress induces different damages in rice leaves and Arabidopsis roots or because rice and Arabidopsis have different regulation patterns in response to cadmium stress. As shown in Figure S2A , the rice MMR system was not damaged, and MSH2/3/6/7 was upregulated to repair DNA damage. A fully functional MMR system can regulate the G2/M phase by upregulating G2/M regulatory proteins and/or by activating p53, ATM, and ATR signaling pathways in human cells in response to exogenous and endogenous stresses. Therefore, we selected key G2/M genes (CDKB and MAD2) and other DNA stress sensors that coordinate stress responses with cell cycle checkpoint ATR genes (Wang et al., 2013). We found that both CDKB and MAD2 were significantly downregulated after the knockdown of either MSH genes ( Figures 4L-M ) and significantly upregulated under cadmium stress, suggesting that MSH genes are also involved in the cell cycle regulation of G2/M. OsMSH2 is significantly downregulated in Osmsh3, Osmsh6, and Osmsh7 mutant rice, indicating an important role of OsMSH2 in MutSα, MutSβ, and MutSγ complexes. OsMSH3, OsMSH6, and OsMSH7 expression levels are altered in all mutant rice, indicating that the MutS system is a comprehensive complex involving the whole body ( Figures 4B-E ). Interestingly, CycD5;2 expression was significantly upregulated in wild-type rice under cadmium stress and significantly downregulated in all mutants. Moreover, their expression was further significantly downregulated in the leaves after exposure of the mutants to cadmium stress. The expression of CYCD-CDKA genes was also changed under cadmium stress, as shown in transcriptome and qRT-PCR analyses. Therefore, we speculate that CYCD-CDKA may directly target and regulate MSH genes in G1/S for DNA damage repair in addition to activating downstream RBR phosphorylation ( Figures 4B-E ). However, this speculation needs to be proven by additional tests at a later stage. Based on our findings, we propose that the CDKA-RBR-E2F-MSH pathway may be the primary mechanism for enhancing cadmium stress tolerance in rice by mobilizing cell cycle factors of downstream key mismatch repair genes ( Figure 7 ).

In conclusion, we systematically categorized and analyzed rice E2F genes by bioinformatics and demonstrated that rice E2F transcription factor can activate MSHs target genes. The cadmium stress experiments in rice seedlings and the yeast heterologous expression system demonstrated that E2F transcription factors and MSHs can respond to cadmium stress and elucidated the specific link between E2F and MSHs from transcriptomic and quantitative perspectives. Furthermore, our study provided a preliminary theoretical basis for revealing the intrinsic mechanism of E2F transcription factors responding to cadmium stress by binding to MSH mismatch repair genes E2F in the cell cycle. Next, we will take this as a breakthrough to further explore the relationship between E2F factor and cadmium stress in rice, and study the relationship between E2F and MSH in rice and other crops from multiple perspectives and layers. Our work paved a novel way for expanding the theoretical basis of the plant cell cycle and provided a theoretical reference for the study of the mechanism responding to cadmium stress in rice.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: NCBI Sequence Read Archive under the BioProject identification number PRJNA890021.

B-GM and B-RZ designed the experiments. W-JZ carried out most of the experiments. W-QL, YP, YS, LT, C-TL, DZ, L-JZ and J-HL. assisted in phenotypic identification and protein interaction tests. W-QL and Z-CY assisted in field management. W-JZ wrote the manuscript. B-GM and B-RZ revised and approved the final version of the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported by National Natural Science Foundation of China (32172042), Natural Science Foundation of Hunan Province (2021JJ30487), the earmarked fund for China Agriculture Research System. The authors would like to thank TopEdit (www.topeditsci.com) for its linguistic assistance during the preparation of this manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.