Nuclear Prohibitin3 Maintains Genome Integrity and Cell Proliferation in the Root Meristem through Minichromosome Maintenance 2

The nuclear localized PROHIBITIN3 acts as a transcriptional co-regulator that sustains genome integrity and cell proliferation by directly suppressing the expression of MINICHROMOSOME MAINTENANCE 2.

Abstract

The nucleo-mitochondrial dual-localized proteins can act as gene expression regulators; however, few instances of these proteins have been described in plants. Arabidopsis (Arabidopsis thaliana) PROHIBITIN 3 (PHB3) is involved in stress responses and developmental processes, but it is unknown how these roles are achieved at the molecular level in the nucleus. In this study, we show that nucleo-mitochondrial PHB3 plays an essential role in regulating genome stability and cell proliferation. PHB3 is up-regulated by DNA damage agents, and the stress-induced PHB3 proteins accumulate in the nucleus. Loss of function of PHB3 results in DNA damage and defective maintenance of the root stem cell niche. Subsequently, the expression patterns and levels of the root stem cell regulators are altered and down-regulated, respectively. In addition, the phb3 mutant shows aberrant cell division and altered expression of cell cycle–related genes, such as CycB1 and Cyclin dependent kinase 1. Moreover, the minichromosome maintenance (MCM) genes, e.g. MCM2, MCM3, MCM4, MCM5, MCM6, and MCM7, are up-regulated in the phb3 mutant. Reducing the MCM2 expression level substantially recovers the DNA damage in the phb3 mutant and partially rescues the altered cell proliferation and root deficiency of phb3 seedlings. PHB3 acts as a transcriptional coregulator that represses MCM2 expression by competitively binding to the promoter E2F–cis-acting elements with E2Fa so as to modulate primary root growth. Collectively, these findings indicate that nuclear-localized PHB3 acts as a transcriptional coregulator that suppresses MCM2 expression to sustain genome integrity and cell proliferation for stem cell niche maintenance in Arabidopsis.

Throughout the lifespan of a plant, stem cells continuously generate different meristems to sustain growth or to replace tissues. In angiosperms, root growth is maintained by coordinating cell proliferation and differentiation. Within root meristems, mitosis in the active quiescent center (QC), a structure that is essential for the maintenance of root growth (van den Berg et al., 1997), keeps the surrounding stem cells in an undifferentiated state and together constitutes the root stem cell niche that provides the source of cells for all tissues in roots (Scheres, 2007; Zhou et al., 2010).

The root stem cell niche identity maintenance depends on an auxin concentration gradient established in the root tip (Blilou et al., 2005; Grieneisen et al., 2007; Tian et al., 2013). Moreover, two sets of pathways, the SHORTROOT (SHR)/SCARECROW (SCR) and the PLETHORA (PLT), are required for maintaining the root stem cell niche (Di Laurenzio et al., 1996; Helariutta et al., 2000; Sabatini et al., 2003; Aida et al., 2004; Koizumi and Gallagher, 2013). The PIN auxin efflux carriers play an important role in modulating PLT1 and PLT2 expression in the distal root apical meristem (Blilou et al., 2005; Ding and Friml, 2010). Conversely, PLT1 and PLT2 are required for regulating root-related expression of PINs and polar localization of PINs (Blilou et al., 2005; Galinha et al., 2007; Pinon et al., 2013). There might be a positive-feedback loop between PLTs and PINs in stabilizing the auxin maximum at the root tip (Blilou et al., 2005). Thus, a positive-feedback loop among auxin homeostasis and PLT and PIN expression regulates root apical meristem maintenance.

Recent findings reveal that plant stem cells have a tight connection with genomic stability (Sablowski, 2011). Treatments with DNA damaging agents preferentially kill stem cells in the shoot and root meristem (Fulcher and Sablowski, 2009; Furukawa et al., 2010; Sablowski, 2011). In agreement with this, the accumulation of DNA damage in rad50 and mre11 mutants led to stem cell death and thus to developmental defects in growing plants. Moreover, a series of genomic integrity regulators, e.g. MDO1, MMS21, MAIN, JHS1, NRP1/2, and CAF-1, are required for the maintenance of stem cells through DNA damage. In plants, as in other organisms, activated Ataxia Telangiectasia Mutated (ATM) or ATM- and RAD3-related (ATR) transmits DNA damage signals to many downstream effectors, eventually arresting cell cycle progression and initiating DNA repair. Elements that arrest the cell cycle include the suppressor of gamma response 1 (SOG1) transcription factor, and the cell cycle inhibitory WEE1 kinase as well as SIAMESE-RELATED cyclin-dependent kinase inhibitors (SMR5 and SMR7; De Schutter et al., 2007; Yi et al., 2014). Thus, these data indicate that the protection of genomic stability and cell cycle is an important feature of the plant stem cell niche.

The root meristem cell exit from division into differentiation has been thought to coincide with a switch from the mitotic cycle to an endoreduplication cycle (De Veylder et al., 2011). By far, the molecular components and mechanisms that connect the root meristem regulators to the cell cycle machinery have not been identified distinctly. Like mammals, plants control the entry into the S-phase of the cell cycle by the E2F-retinoblastoma pathway (Berckmans and De Veylder, 2009). Arabidopsis (Arabidopsis thaliana) E2Fa in association with DPa induces cell proliferation and increases ploidy levels (De Veylder et al., 2002). E2F target genes have a cell cycle-modulated G1- or S-phase expression profile. Genes likely to be directly regulated by E2Fa-Dpa in Arabidopsis have specific binding sites with a WTTSSCSS (where W = A or T and S = G or C) cis-acting consensus element in their promoter region, and more than 300 such genes have been identified, including the minichromosome maintenance (MCM) genes (Vandepoele et al., 2005; Naouar et al., 2009). The MCM complex, which includes MCM2-7, is one of the components of the prereplication complex, which assembles at DNA replication origins during the G1 phase of the cell cycle (Shultz et al., 2009; Herridge et al., 2014). The MCM complex may act as DNA helicases that open up the replication fork to serve as a licensing factor for DNA replication (Ni et al., 2009; Tuteja et al., 2011). Recently, MCM2 has been reported to act as an essential regulator for embryo development and root meristem function (Ni et al., 2009).

Prohibitins (PHBs) are composed of a conserved protein family in eukaryotic cells and primitively discovered as a tumor suppressor factor in mammalian cells (McClung et al., 1989). Prohibitins play an essential role in a series of cellular processes, including receptor-mediated signaling at the cell surface (Rajalingam et al., 2005), cell cycle regulation (Wang et al., 2002b; Gamble et al., 2004), mitochondrial respiration (Nijtmans et al., 2000; Coates et al., 2001), cell death (Fusaro et al., 2003), and aging (Coates et al., 1997). PHB functional diversity is also reflected in its various subcellular location, including the plasma membrane (Terashima et al., 1994; Rajalingam et al., 2005), nucleus (Wang et al., 2002b; Fusaro et al., 2003), and mitochondria (Ikonen et al., 1995; Coates et al., 1997). Phylogenetic analysis showed that prohibitins in all species can be classified into two types depending on the phylogenetic relationships with yeast PHB1 and PHB2. In Arabidopsis, prohibitins were divided into type-I PHB3 and PHB4, and type-II PHB1, PHB2 and PHB6. All prohibitin genes are chiefly expressed in both shoot and root actively proliferating tissues and dividing cells in Arabidopsis (Ahn et al., 2006; Van Aken et al., 2007, 2010). A previous study showed that plant prohibitins are needed for plant defense (Nadimpalli et al., 2000) and in response to oxidative stress (Ahn et al., 2006), ethylene signaling (Christians and Larsen, 2007), and nitric oxide accumulation (Wang et al., 2010) as well as play a crucial role in cell division, senescence, and root hair elongation (Chen et al., 2005; Wen et al., 2005; Ahn et al., 2006; Van Aken et al., 2007). Although plant prohibitins have been implicated in the regulation of plant development, there is currently no information regarding the exact role and functional mechanism of plant prohibitins in regulating genome integrity and cell proliferation.

In this study, we functionally characterized the Arabidopsis PROHIBITIN 3 (PHB3) on genomic stability and cell proliferation for root stem cell development. The deficiency of PHB3 results in DNA damage and cell cycle arrest phenotypes as well as leads to defective root stem cell niche maintenance. Phenotypic analysis, genetic complementation studies, and cellular and molecular characterization of PHB3 demonstrated that the nuclear localized PHB3 is an important regulator for cellular proliferation-related gene expression, which is involved in DNA damage and root stem cell development.

RESULTS

PHB3 Is Required for Meristem Development and Root Stem Cell Niche Maintenance

To identify the role of Arabidopsis PHB3 in plant development in detail, first, the PHB3 gene expression pattern was analyzed using reverse transcription-quantitative PCR (RT-qPCR). The results revealed that PHB3 transcripts were more abundant in the root and leaf than in the flower, stem, and silique (Fig. 1A). In addition, the PHB3pro::GUS staining results showed that PHB3 mainly expressed in the active growth tissues such as the shoot apical and root meristem (Fig. 1B; Supplemental Fig. S1A). These expression pattern results were consistent in that PHB3 plays essential roles in root and shoot meristem development and cell proliferation (Fig. 1C; Supplemental Fig. S1, C, D, and E). The short root phenotype in both of the phb3-ko and phb3-3 mutant seedlings was observed (Fig. 1, C and D). Moreover, the meristem cell number of the phb3 root was reduced significantly (Fig. 1E), the arrangement of meristem cells was disorganized, and the phb3 seedlings showed obviously smaller meristem sizes (Fig. 1F). Intriguingly, a great part of the phb3 meristem cell size was larger than the wild type (Fig. 1F; Supplemental Fig. S2A). However, the maturation zone cell of the phb3 root has no obvious difference with that of the wild type root (Supplemental Fig. S2A). After transforming with the CDS of PHB3, driven by the promoter of PHB3 or the 35S promoter, respectively, the root lengths of the phb3 mutants were completely rescued (Fig. 1G; Supplemental Fig. S2C). Taken together, these results indicate that PHB3 plays an essential role in maintaining the root apical meristem size and the root meristematic activity.

Figure 1.

PHB3 deficiency leads to stunted root growth and reduced root meristem size. A, RT-qPCR analysis of PHB3 expression in Arabidopsis seedlings, root, stem, leaf, flower, and silique. UBQ10 was used as the internal control gene. Data are means ± sd of three biological repeats. B, PHB3pro::GUS was expressed in the shoot and root meristem. C, Phenotypic comparison of 7-d-old wild-type (WT), phb3-ko, and phb3-3 seedlings. Bar = 1 cm. D, Primary root length of wild-type, phb3-ko, and phb3-3 seedlings from 0 to 11 days after germination (DAG). Data shown are means ± sd (n = 30). Asterisks denote significant differences by Student’s t test compared with wild type (**P < 0.01). E, Root meristem cell number in wild type, phb3-ko, and phb3-3 plants from 0 to 9 DAG. The data shown are means ± sd (n = 30). Asterisks denote significant differences by Student’s t test compared with wild type (**P < 0.01). F, Photograph of the root meristematic zone in 7-d-old wild type, phb3-ko, and phb3-3 plants. Red vertical lines represent the length of the meristem. Bar = 100 μm. G, Phenotypic comparison of 7-d-old wild-type (Col-0), complemented, and mutant (phb3-ko) seedlings. Bars = 1 cm.

Because maintenance of the stem cell niche is critical for regulating root meristem activity and primary root growth (Aida et al., 2004; Zhou et al., 2010; Della Rovere et al., 2013), the stem cell characteristic of the phb3 root apical meristem was investigated. The pattern of QC cells in the phb3 root apical meristem was indistinct (Fig. 2A), and the expression levels of three independent QC-specific markers (QC25, QC46, and QC184) in phb3 were either highly reduced or completely missing (Fig. 2B). The deficiency of QC cells in phb3 was further confirmed by an altered expression pattern and lower expression level of WOX5pro::GFP (green fluorescent protein), which is another QC-specific marker (Blilou et al., 2005; Ji et al., 2015; Zhang et al., 2015), compared with that in the wild type (Fig. 2C). Moreover, the expression patterns of the columella stem cells marker J2341::GFP and the cortical and endodermal cell-specific marker J0571::GFP (Petersson et al., 2015) were abnormal in phb3 mutants (Fig. 2D; Supplemental Fig. S3). However, the expression pattern of J1092::GFP in the phb3 root cap and epidermal initial cells showed no remarkable changes (Supplemental Fig. S3). Collectively, these data demonstrate that PHB3 is required for proper stem cell niche maintenance.

Figure 2.

phb3 mutants show defective root stem cell niche maintenance. A, Cellular organization of 7-d-old wild type (WT) and phb3 root tips using PI staining. The white arrow shows the QC cells. Bars = 25 μm. B, The expression of QC25, QC46, and QC184 GUS markers in the 7-d-old wild type and phb3 root tips. The white arrow shows the QC cells, and the asterisk shows the columella stem cells. Bars = 100 μm. C, The expression pattern of the WOX5pro:GFP marker was examined using the same confocal settings in 7-d-old phb3 and wild-type root tips. Bars = 25 μm. D, The expression pattern of the J2341:GFP marker was examined using the same confocal settings in 7-d-old phb3 and wild-type root tips. Bars = 25 μm.

Given that root stem cell niche maintenance was modulated by the SHR/SCR and PLT pathways (Cui et al., 2007; Galinha et al., 2007), the expression of these transcription factors in phb3 mutants was evaluated. The results showed that the expression levels of SHR/SCR (Supplemental Fig. S4A–D) and PLT1/PLT2 (Supplemental Fig. S5A–C) were obviously reduced in the phb3 root apical meristem. In addition, the primary root lengths of shr-1 phb3 and scr-1 phb3 double mutants are shorter than those of phb3, shr-1, and scr-1 single mutants (Supplemental Fig. S4E), supporting that PHB3 regulates the root stem cell integrity in parallel with the SHR/SCR pathway but affects SHR/SCR expression indirectly. Moreover, the meristem cell number and meristem size of the phb3 plt1-4 plt2-2 triple mutant are similar to the plt1-4plt2-2 double mutant (Supplemental Fig. S5, E and F). And the inducible expression construct 35Spro:PLT2:GR (Galinha et al., 2007) narrowed the gap of root meristem cell number between the wild type and phb3 mutant (Supplemental Fig. S5, G–I). However, PHB3 could not bind to PLT1 and PLT2 (Supplemental Fig. S6). These results indicated that PHB3 affects the PLT pathway indirectly.

Given that the root stem cell niche identity maintenance depends on an auxin concentration gradient (Blilou et al., 2005; Grieneisen et al., 2007; Tian et al., 2013), the specific auxin responsive markers DR5::GFP and DR5::GUS (Ulmasov et al., 1997; Blilou et al., 2005) in the phb3 mutant were analyzed. The results showed that the expression levels of DR5::GFP and DR5::GUS were significantly reduced in the phb3 root apical meristem (Fig. 3, A–C), suggesting that PHB3 is required for auxin accumulation in the root tip. Thus, the auxin efflux transporter PIN genes, which are required for stem cell niche maintenance and maximum auxin accumulation in root tips (Friml et al., 2002; Benková et al., 2003; Blilou et al., 2005; Wiśniewska et al., 2006), were analyzed. The results showed that the expression levels of PIN1 and PIN2 were markedly reduced in the phb3 root apical meristem (Fig. 3, D–G). Moreover, the expression of the auxin biosynthetic gene ASA1 that encodes a rate-limiting enzyme for auxin biosynthesis (Stepanova et al., 2005) was markedly reduced in the phb3 root tip (Supplemental Fig. S7). However, the phb3 primary root length could not be rescued by exogenous auxin (Supplemental Fig. S8), implying that the altered auxin distribution and transport is not directly responsible for the phb3 root deficiency, and they may be secondary effects resulting from other changes in phb3 mutant.

Figure 3.

Mutations of PHB3 affect the contents of auxin and the expression of PIN genes. A, The expression pattern of the DR5pro:GFP marker was examined using the same confocal settings in the 7-d-old wild type (WT) and phb3 root tips. Bars = 25 μm. B, Quantification of the fluorescence of the DR5pro:GFP marker in the wild type and phb3 root tips. Data shown are means ± sd (n = 20). Asterisks denote Student’s t test significant difference between wild type and phb3 (**P < 0.01). C, Expression pattern of the DR5pro:GUS marker in the 7-d-old wild type and phb3 root tips. Bars = 50 μm. D and E, PIN1pro:PIN1:GFP and PIN2pro:PIN2:GFP expression pattern in the 7-d-old wild type and phb3 root. Bars = 50 μm. F, Quantification of the fluorescence of the PIN1pro:PIN1:GFP and PIN2pro:PIN2:GFP markers in the wild type and phb3 root. Data shown are means ± sd (n = 20). Asterisks denote Student’s t test significant difference between wild type and phb3 (**P < 0.01). G, RT-qPCR analysis of the expression of PIN genes in wild-type and phb3 roots. The total RNAs were isolated from roots of the 7-d-old wild type and phb3 plants, and UBQ10 was used as a control. Data are means ± sd of three biological repeats. Asterisks denote Student’s t test significant difference between wild type and phb3 (*P < 0.05; **P < 0.01).

PHB3 Is Involved in DNA Damage Independent of Reactive Oxygen Species

Interestingly, we noticed that the cytoplasm of several cells in the root meristem of phb3 were stained with propidium iodide (PI; Fig. 2, A, C, and D). PI stains the walls of living plant cells and is also used as a marker for cell death due to loss of membrane integrity (Truernit and Haseloff, 2008). In phb3 root tips, cell death was observed in QC cells, surrounding stem cells, and vascular cells, which was not detected in the wild type plants (Fig. 2, A, C, and D). Diffusion and expansion of WOX5 expression has been reported in the root meristem after QC ablation and after treatment with DNA-damaging agents, and the cells in plant stem cell niches were selectively killed in response to DNA damage (Fulcher and Sablowski, 2009). Intriguingly, in phb3 mutants, the WOX5pro::GFP construct showed a similar diffusion and expansion around the QC cells (Fig. 2C), suggesting that the dead cells were in the stem cell niche of the phb3 mutant (Fig. 2C). Given that phb3 mutants showed spontaneous cell death even in the absence of DNA-damaging treatment, it is hypothesized that phb3 has naturally increased levels of DNA damage, which was sufficient to activate cell death, and PHB3 plays a role in preventing stem cell death in Arabidopsis. Therefore, absence of PHB3 triggers the death of root stem cells and vascular cells, which is probably caused by DNA damage, implying that PHB3 may be sensitive to DNA damaging agents.

To better characterize the function of PHB3 in DNA damage repair, we detected the expression of PHB3 in response to various concentrations of methyl methanesulfonate (MMS), the DNA-damaging reagent that produces both mutagenic and replication blocking lesions (Fu et al., 2012). The RT-qPCR analysis indicated that the expression level of PHB3 was associated with DNA damage response (Fig. 4A). Because PHB3 responds to MMS at the mRNA level, PHB3 protein accumulation was observed using PHB3pro:PHB3-GFP, which functionally complements the phb3 mutation and carries the 2.3-kb promoter and the PHB3 protein-coding region. Immunoblotting analysis of the total protein from seedlings showed increased accumulation of PHB3-GFP, after MMS treatment (Fig. 4B). Given that the cell death could be found during the PI staining process of phb3 mutant roots, the DNA damage levels in the wild type and phb3 mutants were measured by the percentage of DNA in the tail of the comet in the neutral comet assay (Menke et al., 2001). Higher DNA damage accumulation, as expected, was observed in phb3 than in wild type plants under normal conditions (Fig. 4, C and D). Furthermore, to examine whether the phb3 roots have constitutive activation of DNA damage responses, the expression of the DNA damage-related genes ATM, ATR, RAD51, BREAST CANCER SUSCEPTIBILITY1 (BRCA1), SOG1, WEE1, or PARP2 was measured by RT-qPCR (Culligan et al., 2006; Inagaki et al., 2006; Sakamoto et al., 2011). The transcript levels of all seven genes were up-regulated significantly in the phb3 seedlings compared with that in the wild type plants under normal growth conditions (Fig. 4E). In addition, stronger WEE1:GUS staining was observed in the phb3 mutant compared with that in the wild type seedlings (Supplemental Fig. S9). These data suggested that the phb3 cells are exposed to DNA damage and showed constitutively activated DNA damage responses, even without additional genotoxic stress. Taken together, the above observations indicate that PHB3 is involved in DNA damage.

Figure 4.

PHB3 is involved in the DNA damage response and genome integrity. A, The effects of MMS on PHB3 transcript. The 7-d-old wild type (WT) seedlings were treated with varying concentrations of MMS for 30 min, and the transcript levels were detected by RT-qPCR. B, The effects of MMS on PHB3 protein level. The 7-d-old PHB3pro:PHB3-GFP seedlings were treated with 15 ppm MMS for varying times from 0 to 90 min. Total extracted proteins were subjected to immunoblotting using antibodies against GFP, and Rubisco was used as a control. C, The DNA damage level in the phb3 mutant. The DNA damage status was observed via a comet assay. The results are representative of three independent biological experiments. Bar = 50 µm. D, DNA damage levels were measured by the percentage of DNA in comet tails of nuclei. Data shown are means (± sd) of three independent replications. E, The expression of DNA damage response genes in the phb3 mutant. RT-qPCR analysis of gene expression associated with DNA damage in 7-d-old seedlings. The RT-qPCR data are means ± se from triplicate experiments. Asterisks denote Student’s t test significant difference (*P < 0.05; **P < 0.01).

Given that PHB3 is a mitochondrial localization protein (Van Aken et al., 2007), mitochondria disruption always results in overproduction of reactive oxygen species (ROS) and subsequently plant growth inhibition (Zhang et al., 2014). Thus, we observed whether the root meristem shortage of phb3 seedlings is caused by ROS. As predicted, the ROS level was significantly elevated in the phb3 mutants, and PHB3 expression could decrease the ROS level in phb3 plants (Fig. 5, A and B). Based on the importance of reduced glutathione (GSH) in maintaining and controlling cellular redox status (Rouhier et al., 2008), the phb3 seedlings were treated with GSH. The high ROS level of the phb3 mutant was eliminated effectively by exogenous GSH treatment (Fig. 5, A and B). However, the retarded root growth of phb3 mutants was not rescued by the decreased ROS (Fig. 5, C and D), indicating that the loss of function of PHB3 induced growth inhibition did not result from ROS.

Figure 5.

The phb3 mutation-induced DNA damage is independent of ROS. A, The effects of exogenous GSH on O²⁻ accumulation in phb3-ko and phb3-3 plants. The 7-d-old seedlings were stained by NBT. Bar = 100 μm. WT, wild type. B, The effects of exogenous GSH on H₂O₂ accumulation in phb3-ko and phb3-3 plants. The 7-d-old seedlings were stained by DAB. Bar = 100 μm. Red vertical lines in (A and B) represent the length of the meristem. C, The effects of exogenous GSH on the phb3 root phenotype. Phenotype of wild type, phb3-ko, and phb3-3 seedlings treated with or without GSH. Photographs are of 7-d-old seedlings that germinated and were cultured on MS medium with or without 500 μm GSH. Bar = 1 cm. D, Quantitative analysis of root length after GSH treatment. Three biological replicates were done with similar results, and the data shown are means ± sd (n = 30). NSD indicates no significant difference (two-way ANOVA testing). E, The cell death in the phb3 mutant root after ROS elimination. The 7-d-old roots were stained by PI. Bar = 50 μm. F, The transcript levels of constitutive activated DNA damage response genes in the phb3 mutant after ROS elimination. The gene transcript levels were detected by RT-qPCR, and UBQ10 was used as a control. Asterisks denote Student’s t test significant difference between wild-type and phb3 (*P < 0.05; **P < 0.01).

ROS generated endogenously either in the mitochondria or chloroplast to counter abiotic and biotic stresses can induce many types of DNA damage and thereby exerts genotoxic stress. Thus, we tested whether the DNA damage in the phb3 mutant resulted from ROS overproduction. Unexpectedly, the results showed that the phb3 root meristem cell death was not recovered or reduced by eliminating ROS accumulation (Fig. 5E). Moreover, the DNA damage response gene expression levels in the phb3 mutant were detected after GSH treatment. The results showed that all the DNA damage response genes also maintained higher expression levels in the phb3 mutants after ROS was eliminated by GSH (Fig. 5F). Intriguingly, the exogenous ROS could induce PHB3 expression, and the ROS-induced PHB3 accumulation in the cell nucleus rather than in the cytoplasm (Supplemental Fig. S10), suggesting that the PHB3-induced DNA damage and growth retardation is independent of ROS, and the function of PHB3 on meristem development may not be related to its mitochondrial localization.

The PHB3 Mutation Alters Cell Cycle Progression

In response to DNA damage, cells activate checkpoints to arrest cell cycle progression and allow cells enough time to repair DNA. In the phb3 mutant, the cell files in the root meristems were partially disordered with the occurrence of noncanonical periclinal cell divisions (Fig. 2). This result, combined with that PHB3 is involved in DNA damage, prompted us to investigate whether the cell cycle progression in the phb3 was altered.

First, to uncover the role of PHB3 in cell cycle regulation, the ploidy level of the phb3 mutant was analyzed by flow cytometry. In the 7-d-old plant cotyledons, phb3 had fewer nuclei with 2C DNA contents and more nuclei with higher DNA contents, compared with the wild type and complementation plants (Fig. 6A). This result suggests that endoreduplication is enhanced by the absence of PHB3, implying the important role of PHB3 on the plant cell cycle. Thus, the expression level of mitotic cyclin CycB1;1 and CycB1;2, which are the markers for the G2/M phase of the cell cycle (Ferreira et al., 1994; Colón-Carmona et al., 1999; Yang et al., 2015), were analyzed in the phb3 mutant. The histochemical staining result showed that the CycB1;1-GUS activity was stronger in the phb3 mutant root apical meristem compared with that in wild type plants (Fig. 6B). Furthermore, the transcript level detection also showed that the CycB1;1 expression level in the phb3 mutant was increased significantly (Fig. 6D), whereas the CycB1;2 expression level was decreased significantly in the phb3 mutant (Fig. 6D). These results indicate that the cell cycle in the phb3 root apical meristem was retarded. In addition, the expression content of the plant-specific cell cycle kinase gene Cyclin dependent kinase 1 (CDKA1), which is expressed throughout the cell cycle (Menges et al., 2002; Vandepoele et al., 2002), was detected. The histochemical staining results showed that the expression of CDKA;1 was reduced in the phb3 mutant (Fig. 6C), and the reduced CDKA;1 level in the phb3 mutant was verified by RT-qPCR (Fig. 6D). Furthermore, the expression levels of several cell cycle–related genes in the root of phb3 and wild type seedlings were detected. Among these genes, the expression of E2Fa and RBR1, which function at the G1/S transition, was obviously decreased in phb3, whereas the expression of DPa, which functions with E2Fa in cell cycle progression, had no significant change (Fig. 6D). In contrast, the CDK inhibitor and cell cycle negative regulator genes, e.g. SMR1, SMR4, SMR5, and SMR7, were increased significantly in the phb3 mutant (Fig. 6E). Intriguingly, the members of MCM complex genes MCM2, MCM3, MCM4, MCM5, MCM6, and MCM7 were up-regulated obviously in the phb3 mutant (Fig. 6F). Thus, these results indicate that the cell cycle progression was altered in the phb3 plant.

Figure 6.

Mutation of PHB3 affects cell cycle progression. A, The endoreduplication levels in wild type (WT), phb3, and PHB3pro::PHB3-phb3 plants were quantitatively analyzed by flow cytometry. The cotyledons of 7-d-old plants were collected for analysis. The data are means ± sd from at least three biological replicates. DNA content indicated by 2C, 4C, 8C, 16C, and 32C. B, CycB1;1:GUS expression in 7-d-old wild type and phb3 at root tips. Bars = 50 μm. C, CDKA1:GUS expression in 7-d-old wild type and phb3 root tips. Bars = 50 μm. D, Cell cycle-related gene expression in phb3 mutant seedlings. E, The CDK inhibitor and cell cycle negative regulator gene expression in phb3 mutant seedlings. F, Expression of MCM genes in phb3 mutant seedlings. The gene transcript levels were detected by RT-qPCR in the 7-d-old seedlings, and UBQ10 was used as a control. The RT-qPCR data are means ± se from triplicate experiments. Asterisks denote Student’s t test significant difference between wild type and phb3 (*P < 0.05; **P < 0.01).

PHB3 Regulates Cell Proliferation through MCM2

The essential role of PHB3 in root development and the cell cycle prompted us to further explore the potential mechanism of its action. The previous data showed that the functions of prohibitin proteins are conserved (Snedden and Fromm, 1997; Ahn et al., 2006; Van Aken et al., 2007; Gehl and Sweetlove, 2014), and the homo PHB1, which is the closest homology of PHB3 in Arabidopsis, is involved in modulating the cell cycle via physically interacting with MCM complex members MCM2, MCM5, and MCM7 in vitro (Rizwani et al., 2009). Intriguingly, MCM2 has been reported to play an essential role on embryo and root meristem development, because disruption of MCM2 is lethal at the very early embryogenesis stage and its over-expression results in reduced root growth (Ni et al., 2009), implying a possible role of MCM2 in modulating primary root development in the phb3 mutant.

To further investigate the possibility that MCM2 is involved in primary root development and whether the defective root phenotype of the phb3 mutant is caused by ectopic MCM2 expression, the MCM2 expression in the phb3 mutant was knocked down by RNA interference (RNAi). A series of MCM2 RNAi lines in phb3 mutants, named phb3 MCM2 RNAi-1, -2, and -3, were obtained based on their different root lengths (Supplemental Fig. S11). Interestingly, their root lengths and root meristem integrities are correlated inversely with MCM2 expression (Fig. 7, A–D; Supplemental Fig. S11). Thus, the decreased MCM2 expression resulted in longer primary root and meristem length as expected in both of the phb3 mutants and wild type plant. Previously, it is reported that MCM2 affects cell division in Arabidopsis roots (Ni et al., 2009). Then, the cell proliferation patterns of different PHB3 genotype plants were observed. The results showed that the loss of function of PHB3 resulted in arrested cell proliferation; however, the abnormal cell proliferation phenotype could be rescued by MCM2-RNAi in the phb3 plants (Fig. 7E), demonstrating that the root meristem deficiency and cell proliferation abnormality of the phb3 mutant is partially restored by decreasing the MCM2 expression level.

Figure 7.

Down-regulation of MCM2 partially rescues the root phenotypes of phb3 mutants. A, The phenotype of 7-d-old wild type (WT), phb3, MCM2-OE, MCM2-RNAi/wild type, and MCM2-RNAi/phb3 were observed. Bar = 1 cm. B, Quantitative analysis of root length in different genotypic seedlings. Data are means ± sd from three independent experiments. Black asterisks indicate significant differences (Student’s t test) relative to wild type seedling controls; red asterisks indicate significant differences (two-way ANOVA) between MCM2-RNAi seedlings relative to the corresponding background (*P < 0.05, **P < 0.01). C, RT-qPCR analysis of MCM2 gene expression level in 7-d-old wild type, phb3, MCM2-OE, MCM2-RNAi/wild type, and MCM2-RNAi/phb3 root tips. UBQ10 was used as a control, and the asterisks denote Student’s t test significant difference between wild-type and phb3 (*P < 0.05; **P < 0.01). D, The effects of MCM2 down-regulation on the root meristem development in phb3 mutant roots. The root meristem of different 7-d-old PHB3 genotype plants was observed by a confocal microscope. White vertical lines in the root tip represent the length of root meristem. Bars = 25 μm. E, The effects of MCM2 down-regulation on the cell division of the phb3 mutant roots. The root tip of wild type, phb3, PHB3pro::PHB3/phb3, MCM2-RNAi/phb3, and MCM2-RNAi/wild type treated with 1 μm 5-Ethynyl-2-deoxyyridine (EdU) for 24 h at 4 DAG, and cell division was observed by a confocal microscope. Bar = 25 μm.

PHB3 Participates in DNA Damage Dependent on MCM2

DNA damage is also generated endogenously by errors during DNA replication, which is often referred to as replication stress. To evaluate whether the altered cell cycle progression in the phb3 mutant resulted from constitutive activation of the MCM2-dependent DNA damage response, the presence of DNA damage hallmarks in MCM2 over-expression plants was analyzed first. Indeed, the expression level of MCM2 was induced by the MMS treatment (Fig. 8A). In addition, the MCM2 over-expression plants showed enhanced sensitivity to the exogenous DNA stress. The survival rate of MCM2 over-expression plants was decreased to only 15% after MMS treatment for 28 h (Fig. 8, B and C; Supplemental Fig. S12), whereas almost all the wild type and MCM2-RNAi plants survived at the same condition (Fig. 8, B and C; Supplemental Fig. S12). Furthermore, MCM2 over-expression resulted in serious root cell death compared with the wild type and MCM2-RNAi roots, after low concentration DNA damage agent treatment (Fig. 8D). Moreover, a number of genes that are up-regulated in response to DNA damage such as RAD51, BRCA1, ATM, ATR, SOG1, WEE1, or POLY (ADP-RIBOSE) POLYAMERASE2 (PARP2; Culligan et al., 2006) were up-regulated in MCM2 over-expression plants (Fig. 8G). Meanwhile, the increased MMS sensitivity phenotype and elevated transcript levels of DNA damage response genes were observed in the phb3 mutants (Figs. 4E and 8, E and F; Supplemental Fig. S12). These results indicate that both MCM2 and PHB3 play important roles in the DNA damage response. Then, whether the phb3 mutant participates in DNA damage response via MCM2 was investigated. The results showed that down-regulated MCM2 expression levels in the phb3 not only reduced the sensitivity of phb3 seedlings to the DNA damage reagent but also recovered the expression levels of DNA damage response genes in phb3 to normal levels similar to wild type seedlings (Fig. 8, E–G). Combining the fact that MCM2-RNAi rescues the cell death in phb3 roots (Fig. 7D), these results indicate that the functional deficiency of PHB3 results in the DNA damage response through dysregulation of MCM2.

Figure 8.

Down-regulation of MCM2 rescues the DNA damage response of phb3 mutants. A, The effects of DNA damage agent on MCM2 expression level. The 7-d-old wild type (WT) seedlings treated with varying concentrations of MMS for 30 min. Data are means ± sd from three independent experiments. Asterisks denote Student’s t test significant difference compared with control (*P < 0.05; **P < 0.01). B, The sensitivity of different MCM2-related seedlings to DNA damage agent. The 7-d-old MCM2-related seedlings were transferred to MS medium with or without 50 ppm MMS for 28 h. C, The survival rates of MCM2-related seedlings after MMS treatment. Data are means ± sd from three independent experiments. Asterisks denote Student’s t test significant difference compared with wild type (*P < 0.05; **P < 0.01). D, The effects of low concentration of DNA damage agent on the root cell death in MCM2-related seedlings. The 7-d-old wild type, MCM2-RNAi/wild type and MCM2-OE seedlings were treated with 5 ppm MMS for 8 h. Bar = 50 μm. E, The sensitivity of phb3 to the DNA damage agent after MCM2 down-regulation. The 7-d-old PHB3-related seedlings were transferred to MS medium with or without 20 ppm MMS for 24 h. F, The survival rates of PHB3-related seedlings after MMS treatment. Data are means ± sd from three independent experiments. Asterisks denote Student’s t test significant difference compared with wild type (*P < 0.05; **P < 0.01). G, The expression of DNA damage response genes in the phb3 mutant after MCM2 down-regulation. The 7-d-old seedlings were used for RT-qPCR, and the data are means ± sd from triplicate experiments. Asterisks denote Student’s t test significant difference compared with wild type (*P < 0.05; **P < 0.01).

PHB3 Inhibits MCM2 Expression as a Transcriptional Repressor in the Nucleus

The above data prompted us to explore the potential relationship between PHB3 and MCM2. To investigate whether PHB3 has the potential to interact with MCM2 in Arabidopsis, we performed yeast two-hybrid, BiFC, Pull-down, and co-immunoprecipitation experiments. Unexpectedly, there was no direct physical interaction between MCM2 and PHB3 in vitro and in vivo (Supplemental Fig. S13). However, obvious self-activation of PHB3 was observed in yeast (Supplemental Fig. S13A).

The distinct self-activation of PHB3 in yeast prompted us to determine whether PHB3 acts as a transcriptional regulator to regulate gene expression and modulate primary development. First, the PHB3pro::PHB3-GFP and PHB3-YFP (yellow fluorescent protein) constructs were used for subcellular localization analysis in the Arabidopsis root and protoplasts, respectively. The PHB3pro::PHB3-GFP expression was observed in the nucleus and cytoplasm of the root meristem cell (Fig. 9A), and strong PHB3-YFP signals were detected in the nucleus (Fig. 9B), indicating that PHB3 displayed prominent nuclear localization. Most intriguingly, PHB3 was colocalization with the transcriptional adaptor HOMOLOG OF YEAST ADA (ADA2B) and BRCA1 (Fig. 9B), two DNA damage-related proteins, which localized in the nucleus (Lai et al., 2018). Meanwhile, PHB3-YFP signal was also observed in the mitochondria of the cytoplasm and surrounding the nucleus by labeling the mitochondria with Mito-Tracker dyes (Fig. 9B), and the rotenone-induced mitochondrial stress resulted in more nuclear localization of PHB3 (Supplemental Fig. S14). Consistent with this, immunoblot analysis of total protein from seedlings showed distinct accumulation of PHB3-GFP, especially in the cell nucleus, after rotenone treatment (Fig. 9C). Consequently, the nuclear localization of PHB3 suggests that PHB3 may have transcriptional activity.

Figure 9.

PHB3 acts as a negative transcriptional coregulator in the nucleus. A, The nuclear and cytoplasmic localization of PHB3pro::PHB3-GFP in Arabidopsis root meristem. B, The nuclear and mitochondrial localization of PHB3 protein. PHB3-YFP and other nucleus markers are expressed in the indicated protoplasts dyed with mito-Tracker. The PHB3-YFP signal (green), ADA2b-CFP and BRCA1-CFP signal (blue), mito-tracker (red), bright field (gray), and merged photos are shown from representative samples. The white arrows indicate the mitochondria. Bars = 10 µm. C, PHB3 accumulation in different cell components after mitochondrial stress agent rotenone treatment. Total (T), nuclear (N), and cytoplasmic (C) proteins were prepared from 7-d-old PHB3pro:PHB3-GFP seedlings treated with or without rotenone. The immunoblots were probed with anti-GFP antibody; Fru-1, 6-bisphostase (FBPase) was used as a cytoplasmic protein control; Histone H3 was used as total and nuclear protein control. D, Transient expression assays of GAL4-PHB3. Schematic representation of the constructs used for transient expression assays. The reporter construct consists of the CaMV 35S promoter, five repeats of the GAL4 binding sequence (5xGAL4BS), nopaline synthase terminator, and firefly luciferase (LUC) coding sequence. Effector constructs express the GAL4 DNA binding domain (GAL4DB)-fused protein under the control of the CaMV 35S promoter. Error bars indicate the SD of results from three replicates. Asterisks denote Student’s t test significant difference compared with control (*P < 0.05; **P < 0.01). E, Effects of PHB3 on MCM2 transcriptional regulation in the Arabidopsis protoplast transient expression assay. Schematic representation of the constructs used for transient expression assays. The reporter construct consists of the Firefly LUC driven by the MCM2 promoter and the REN luciferase driven by the CaMV35S promoter. Effector constructs express PHB3 protein driven by the CaMV35S promoter. Data are means ± sd (n = 3 experiments). Asterisks denote Student’s t test significant difference compared with control (*P < 0.05; **P < 0.01).

To investigate whether PHB3 possesses transcriptional activity in vivo, the dual-luciferase assay was performed in Arabidopsis protoplasts. The dual-luciferase reporter harbors five copies of the GAL4 DNA-binding element and CaMV 35S fused to the firefly luciferase (LUC) reporter, whereas a renilla luciferase (REN) reporter under the control of the 35S promoter was used as an internal control. The PHB3 effector, in which the protein coding region of PHB3 was fused to the yeast GAL4 DNA binding domain (GAL4-PHB3), repressed LUC expression (Fig. 9D), indicating that PHB3 has transcriptional repression activity. Then, the transcriptional activity of PHB3 on regulating MCM2 expression was examined. The dual-luciferase reporter harbors firefly LUC driven by the MCM2 promoter and REN driven by the CaMV35S promoter. The PHB3 coding sequence was introduced into the vector driven by the CaMV 35S promoter as an effector. In the transient expression assay, after cotransformation of the effector and reporter, the relative LUC activities (LUC/REN) decrease by ∼2.5-fold, compared with transformation of the reporter only (Fig. 9E). These results indicated that PHB3 may be a transcriptional repressor for the promoter of MCM2, consistent with the increased expression level of MCM2 in the phb3 mutant (Fig. 6F). These results suggest that PHB3 may play roles as a transcriptional repressor involved in the cell cycle and root meristem development through partially inhibiting MCM2.

PHB3 Regulates MCM2 Expression Antagonistically with E2Fa through E2F-Cis-Acting Elements

Prohibitin specifically represses E2F-mediated transcription and growth through inhibiting the E2F-target genes (Rastogi et al., 2006). In Arabidopsis, a previous study reported that MCM genes are putative E2F targets that contain the TTTCCCGC cis-acting element in their promoter region (Vandepoele et al., 2005). The E2F pathway controls the cell cycle by modulating the transcription of genes needed for DNA replication and cell cycle, and the E2F transcription factors are able to transactivate reporter genes that harbor the E2F consensus cis-acting element in their promoter region (Egelkrout et al., 2002; Vandepoele et al., 2005; Li et al., 2014). As expected, two E2F binding sites were found in the promoter of Arabidopsis MCM2 (Fig. 10A). Combined with the essential role of MCM2 in PHB3 regulated DNA damage and cell cycle, we speculated that PHB3 may act as a transcriptional repressor in regulating MCM2 expression through the conserved E2F-cis-acting element.

Figure 10.

PHB3 directly binds to the MCM2 promoter. A, Schematic diagram of the MCM2 promoter from nucleotide -517 to the transcription start site +1. The two E2F binding consensus sites are shown in red font in black boxes. EB1, E2F binding cis-element 1; EB2, E2F binding cis-element 1; TSS, transcription start site. B, The effects of E2F binding cis-elements mutation in the MCM2 promoter on PHB3 regulating MCM2 transcript in the transient Arabidopsis protoplast assay. Normal, no mutation in E2F binding cis-elements; MEB1, mutation in E2F binding cis-element 1; MEB2, mutation in E2F binding cis-element 2; MEB1MEB2, mutation in E2F binding cis-element 1 and 2. Error bars indicate the SD of results from three replicates. Asterisks denote Student’s t test significant differences (*P < 0.05; **P < 0.01). C, Schematic diagram of MCM2. Black boxes, gray boxes, and black lines indicate the exon, untranslated region, and promoter and intron of indicated genes, respectively. The regions analyzed by ChIP-qPCR are indicated by p1-p8. D, ChIP-qPCR analysis of enrichment of PHB3pro:PHB3-GFP to the different regions of MCM2 in Col-0 and PHB3pro:PHB3-GFP/phb3 roots. An anti-GFP antibody was used for the immunoprecipitation. PHB3pro:PHB3-GFP/phb3 is a transgenic line expressing GFP-tagged PHB3 under the control of the PHB3 native promoter. TA3 and TUBULIN BETA CHAIN2 were used as negative controls. Data are mean values ± sd of three replicates. Similar results were obtained for at least two additional independent experiments. Asterisks denote Student’s t test significant difference between Col-0 (Col) and PHB3pro:PHB3-GFP/phb3 roots (*P < 0.05). E, The binding ability of PHB3 to the E2F-binding cis-element of the MCM2 promoter. The ability of PHB3 protein to bind to the MCM2 promoter was detected by an electrophoretic mobility shift assay (EMSA). Probe, E2F binding cis-element; Muted-probe, mutation in E2F binding cis-element.

To investigate whether the E2F-cis-acting elements in the MCM2 promoter are critical to the suppression of MCM2 expression by PHB3, the dual-luciferase assay was performed. The LUC reporter protein was driven by the MCM2 promoter that mutated in one single E2F-1 or E2F-2 motif mutation (MEB1 or MEB2) or both of the E2F-1 and E2F-2 motifs double mutation (MEB1MEB2). Compared with the normal MCM2 promoter, both the single mutation and double mutation in E2F binding sites resulted in increased LUC/REN levels, which implies enhanced activity of the MCM2 promoter (Fig. 10B), suggesting that E2F-cis-acting elements are required for PHB3 to repress the activity of the MCM2 promoter. Consistent with that, we observed markedly increased transcript levels of MCM2 in phb3 roots (Fig. 6E).

Next, we investigated whether the effect of PHB3 on the mRNA accumulation of MCM2 is direct or indirect. To test the binding of PHB3 on the MCM2 loci, the GFP-tagged PHB3 (PHB3pro:PHB3-GFP) that fully rescued the root meristem defects of the phb3 null mutants was used in chromatin immunoprecipitation (ChIP) assays. Roots of seedlings 7 DAG were selected to investigate the enrichment of PHB3 in different regions of MCM2 in the PHB3pro:PHB3-GFP plants. ChIP-qPCR was used to determine the regions enriched by ChIP with the anti-GFP antibody, and the amplicon lengths are shown in Supplemental Table S1. As shown in Figure 10, PHB3 bound to the MCM2 promoter regions with E2F binding elements (Fig. 10, C and D). This suggests that MCM2 is the direct target gene regulated by PHB3. Furthermore, the capacity of PHB3 to bind directly to the MCM2 promoter was shown using electrophoretic mobility shift assays (EMSA) in vitro. The results showed that PHB3 could directly bind to the E2F binding elements of the MCM2 promoter (Fig. 10E). These results suggest that the E2F-cis-acting elements are essential for MCM2 expression in PHB3 modulating cell cycle and primary root development.

It was previously reported that the MCM complex genes are the direct target genes of E2F in Arabidopsis (Vandepoele et al., 2005; Naouar et al., 2009). Consistently, the transcriptional activity of E2Fa on the MCM2 promoter without the E2F-binding motif (MEB1MEB2) was lost (Supplemental Fig. S15). Thus, the effect of E2Fa on the expression of PHB3-inhibited MCM2 was tested in the Arabidopsis transient expression protoplasts using the dual-luciferase assay. The results showed that E2Fa reversed the inhibition of PHB3 on MCM2 (Fig. 11A), indicating that PHB3 may have antagonistic roles with E2Fa in regulating MCM2 expression. Furthermore, the antagonistic effect of PHB3 and E2Fa on MCM2 transcripts was analyzed in the e2fa mutant. Although the negative regulation of PHB3 on MCM2 was not affected (Fig. 11B), the MCM2 transcript level in the phb3 mutant was reduced significantly by the loss of function of E2Fa (Fig. 11C). Subsequently, the root deficiency phenotype and DNA damage level of the phb3 mutant were rescued by the absence of E2Fa (Fig. 11, D–F). In conclusion, these data suggest that PHB3 antagonistically regulates MCM2 transcription with E2Fa through the conserved E2F-cis-acting elements.

Figure 11.

PHB3 regulates MCM2 expression antagonistically with E2Fa. A, The effects of PHB3 on E2Fa regulating MCM2 expression. Transient expression assay was analyzed in the wild-type (WT) plant protoplasts. The reporter construct consists of the Firefly LUC driven by the MCM2 promoter and the REN luciferase driven by the CaMV35S promoter. Effector constructs express PHB3, or E2Fa, or both PHB3 and E2Fa proteins driven by the CaMV35S promoter. Data are means ± sd (n = 3 experiments). Asterisks denote Student’s t test significant difference (*P < 0.05; **P < 0.01). B, The effects of E2F binding cis-elements on the PHB3 repression ability. Transient expression assay was analyzed in the e2fa mutant protoplasts. Normal, no mutation in E2F binding cis-elements; MEB2, mutation in E2F binding cis-element 2; MEB1MEB2, mutation in E2F binding cis-element 1 and 2. Data are means ± sd (n = 3 experiments). Asterisks denote Student’s t test significant difference (*P < 0.05; **P < 0.01). C, The MCM2 expression level in the phb3 mutant after loss of function of E2Fa. The transcript level was detected by RT-qPCR in the 7-d-old root tips, and UBQ10 was used as a control. Error bars indicate the SD of results from three replicates. Asterisks denote Student’s t test significant difference (*P < 0.05; **P < 0.01). D, The phb3 root length after loss of function of E2Fa. Photographs were taken of 7-d-old seedlings. Bar = 1 cm. E, Quantitative analysis of root length in different genotypic seedlings. Data are means ± sd from three independent experiments. Black asterisks indicate significant differences (Student’s t test) relative to wild type seedling controls; red asterisks indicate significant differences (two-way ANOVA) between e2fa seedlings relative to the corresponding background (*P < 0.05, **P < 0.01). F, The extent of DNA damage-induced cell death in the phb3 root meristem and the phb3 root deficiency after loss of function of E2Fa. The 7-d-old roots were stained by PI. Bars = 50 μm.

DISCUSSION

Plant prohibitins play roles in diverse biological processes related to development and stress responses. However, their functions in DNA damage and cell proliferation have not been elucidated clearly in plants as they have in yeast and mammals. In this study, we proposed that PHB3 regulates stem cell niche maintenance via DNA damage and cell proliferation through MCM2 during root development in Arabidopsis.

PHB3 Maintains Stem Cell Niche Activity Via Regulating Genome Integrity and the Cell Cycle

In the phb3 mutant primary root meristem, the pattern of stem cell organization is abnormal, and PHB3 is essential for maintaining proper identity of the root stem cell niche and activity of the QC (Fig. 2; Supplemental Fig. S2). The distribution and maximum level of auxin plays an essential role in determining the identity of the stem cell niche and differentiation of stem cells in the root meristem (Ding and Friml, 2010). Although the auxin polar transport from aerial parts to roots and auxin biosynthesis may be impaired in the phb3 mutant (Fig. 3; Supplemental Fig. S7), the phb3 root length was not rescued by exogenous auxin treatments (Supplemental Fig. S8), supporting that local auxin production may not be the cause of the defective root development in the phb3 mutant.

In the phb3 root tips, cell death was observed in QC cells, surrounding stem cells, and vascular cells, which were not detected in the wild-type plants (Fig. 2, A, C, and D). In addition, reduced and diffused expression of WOX5pro::GFP was observed in the phb3 mutants (Fig. 2). These phenotypes are consistent with the root meristem after the DNA-damaging reagent treatment (Fulcher and Sablowski, 2009), implying that the phb3 root deficiency resulted from DNA damage. In mammals, induction of apoptosis has been recognized as a possible outcome of severe DNA damage. It is reported that the mammalian PHBs are required for stem cell and cancer cell apoptosis. PHB1 and PHB2 deficiency could increase or decrease cell apoptosis. These reports show that PHBs may play important roles in the DNA damage response, although no direct evidence has elucidated the involvement of prohibitin proteins in DNA damage response. Because the phb3 mutants are sensitive to the DNA damage reagent (Fig. 8E), and the PHB3 transcript and protein levels are induced by the DNA damage reagent (Fig. 4, A and B), as well as the DNA damage response genes (e.g. SOG1, BRCA1, RAD51, PARP2, and WEE1) are up-regulated in the phb3 mutant (Fig. 7), the role of plant PHB3 on DNA damage response has been verified. Intriguingly, almost all the stress-induced PHB3 proteins are accumulated in the nucleus rather than in the cytoplasm (Fig. 9C; Supplemental Fig. S10B), hinting that the nuclear-localized PHB3 plays a role in DNA damage response. PHB is a shuttle protein that shuttles between subcellular compartments (Sripathi et al., 2011). The subcellular localization of PHB is affected by apoptotic signals, and inhibition of this translocation affects apoptosis. During the apoptosis of cancer cells induced by abrin or ESC-3, PHB1 translocates from the cytoplasm or mitochondria to the nucleus (Song et al., 2014). During the apoptosis of cancer cells induced by estradiol or capsaicin, PHB2 translocates from the cytoplasm or mitochondria to the nucleus (Kim et al., 2009; Kuramori et al., 2009). Similar to the previous observation (Van Aken et al., 2007), Arabidopsis PHB3 has been shown to localize both in the nucleus and mitochondria (Fig. 9, A and B), implying that PHB3 has the potential to translocate from the mitochondria to the nucleus after stimuli. In contrast with the importance of mitochondrial PHB3 as a regulator for plant development, very little is known about its function in the nucleus. It is debated whether PHB3 plays important roles to regulate gene expression directly in the nucleus for plant development or whether the mitochondrial-localized PHB3 affects the nuclear activity indirectly, henceforth named mitochondrial retrograde signaling, which depends on ROS production. Although it is very difficult to prove directly whether the nuclear- or the mitochondrial-localized PHB3 deficiency leads to the DNA damage response, three indirect evidences support that the nuclear-localized PHB3 is responsible for this process. First, PHB3 can colocalize with the DNA damage-related proteins BRCA1 and ADA2b (Fig. 9B). Second, the mitochondrial stress caused by rotenone or the ROS treatment resulted in more nuclear localization of PHB3 (Fig. 9C; Supplemental Fig. S10B). Finally, the PHB3 protein with the nuclear localization signal but not with the nuclear export signal in the N terminus can completely rescue the phb3 developmental deficiency (Supplemental Fig. S2, B and C). Previous research showed that the mitochondrial disruption results in ROS accumulation to retard plant growth and development (Zhang et al., 2014), and DNA damage can result from the action of endogenous ROS (Jackson and Bartek, 2009). However, the elimination of the ROS by GSH in the phb3 mutant rescues neither the DNA damage nor the root development deficiency (Fig. 5). Moreover, the inhibition of mitochondrial produced ROS has a weak effect on rescuing phb3 root development (Kong et al., 2018), suggesting that the major role of PHB3 on root meristem development dependents on its nuclear localization, and the PHB3 mutation-induced DNA damage does not result from endogenous ROS accumulation.

DNA damage also occurs from stochastic errors during chromosomal replication except for those originating from endogenous ROS and exogenous factors (Jackson and Bartek, 2009). Intriguingly, the DNA replication required MCM complex genes (Blow and Dutta, 2005), e.g. MCM2, MCM3, MCM4, MCM5, MCM6, and MCM7, all are up-regulated significantly in the phb3 mutant. It was observed that reduced MCM transcription promotes the expression of genes involved in DNA damage response in the ATM- and ATR-dependent pathways (Herridge et al., 2014). Over-expressing the DNA damage agent-inducible MCM2 gene not only up-regulates the DNA damage-related gene transcripts but also increases the sensitivity of seedlings to the DNA damage agent (Fig. 8, A–D). Strikingly, knocking down the MCM2 expression level in the phb3 mutants not only down-regulates the DNA damage-related gene transcripts to the levels in the wild type seedling but also decreases the sensitivity of the phb3 mutant to the DNA damage reagent (Fig. 8, E–G). These intriguing results highlight the idea that the homeostasis of MCMs is very important for the correct replication, and their expression regulators, e.g. PHB3, are essential for maintaining genome integrity. Thus, the nuclear localized PHB3 is predicted to play roles in DNA damage through MCMs by affecting DNA replication.

By far, the function of the nuclear localized prohibitin proteins is unknown in plants. Multiple lines of evidence indicate that nuclear-localized mammalian PHB protein is a unique regulator of specific transcription factors and cell cycle-associated proteins. The transcriptional regulation function of PHB1 in the nucleus may provide a link between the proliferative and apoptotic pathways (Fusaro et al., 2003). DNA damage triggers cell cycle arrest and, in severe cases, cell death, so ensuring DNA repair (Jackson and Bartek, 2009). Upon DNA damage, the cell cycle progression is delayed or arrested at a critical stage before or during DNA replication and before cell division. The activation of the DNA damage checkpoint could result in the onset of endoreduplication in the damaged cells to prevent them from proceeding into mitosis (De Veylder et al., 2011). Expectedly, the endoreduplication level was increased notably in the phb3 mutant (Fig. 6A). It remains unknown why endoreduplication is induced upon DNA damage. One explanation is that the initiation of the endocycle program might be a mechanism to prevent the transmission of DNA lesions into the pool of meristematic cells by pushing the damaged cell into a nondividing state, in such way safeguarding the progeny from DNA mutations (De Veylder et al., 2011).

Activation of the DNA stress checkpoint results in the simultaneous induction of DNA-repair genes and inactivation of genes that are required for mitosis and cytokinesis (Chen et al., 2003; Culligan et al., 2006). This coordinated action ensures that cells repair their damaged genome before they proceed into mitosis (De Veylder et al., 2007). Among the activated DNA damage response genes in the phb3 mutant, SOG1 was shown to directly control several cell cycle–related and DNA repair-related genes. For example, SOG1 binds to the promoters of genes for B1-type cyclin CYCB1;1 (Weimer et al., 2016), CDK inhibitors SMR5 and SMR7 (Yi et al., 2014), and the DNA repair gene BRCA1 (Sjogren et al., 2015), and thereby up-regulates their expression. Some SOG1 targets such as SMR1, SMR4, SMR5, SMR7, and WEE1, are negative regulators of the cell cycle. These genes all are up-regulated significantly in the phb3 mutants (Figs. 4E and 6E). SMR genes encode CDK inhibitors that bind to CDK–cyclin complexes and inhibit kinase activity (Nakai et al., 2006; Van Leene et al., 2010; Guérinier et al., 2013; Yi et al., 2014). The WEE1 kinase phosphorylates and inactivates CDKs (De Schutter et al., 2007). Consistent with that, the CDKA protein is reduced in the phb3 mutants (Fig. 6C). Therefore, it is likely that DNA damage immediately reduces CDK activities in the phb3 mutant under the control of SOG1. Intriguingly, the mitotic cyclin CYCB1;1 is induced rather than repressed in the phb3 mutant (Fig. 6, B and D). This result is consistent to the CYCB1;1 expression pattern upon DNA damage or inhibition of DNA replication (Chen et al., 2003; Culligan et al., 2006). The reason for the induction and stabilization of CYCB1;1 during the DNA damage response is currently unclear. A tempting hypothesis is that the dramatic increase in CYCB1;1 abundance helps to achieve a G2 arrest by titrating out the cell-cycle components that are essential for proceeding into mitosis. Besides the DNA damage-induced expression of CYCB1;1 and SMRs, reduced CYCB1;2, E2Fa, and RBR1, whereas increased MCM levels, are observed in the phb3 mutant (Fig. 6D, 6F); together with the changed DNA ploidy levels (Fig. 6A), suggesting that the cell cycle in the phb3 mutant should be severely affected and might be arrested at the G1/S and G2/M phases.

Although PHB3 plays important roles in regulating genome integrity and the cell cycle, its relationship with the components of the DNA damage response, e.g. ATM, ATR, and SOG1, and how PHB3 regulates them, needs further research to elucidate.

PHB3 Regulates Cell Proliferation through Regulating MCM2 Expression

Prohibitin is an evolutionarily conserved and ubiquitously expressed protein, and has the potential to act as a tumor suppressor, antiproliferative protein, and regulator of cell-cycle progression in the mammalian nucleus. The mutation in PHB3 caused strikingly reduced root meristem size with blocked cell proliferation (Fig. 7E). In combination with that the disturbed cell cycle progression affecting both G1/S and G2/M transitions was observed in the phb3 root meristem (Fig. 6), thus, we concluded that PHB3 is required for proper control of cell proliferation in the root meristem.

The roles of mammalian prohibitin in cellular proliferation and apoptosis are accomplished by its interaction with E2F, RETINOBLASTOMA-RELATED (Rb), and p53 (Wang et al., 1999a; Fusaro et al., 2003). The transcriptional activator E2F1 has been shown to up-regulate the expression of many genes involved in differentiation, proliferation, and apoptosis, and it plays a major role in the G1/S transition and DNA synthesis (Bell and Ryan, 2004). E2F activity is essential for the expression of critical cellular genes required for progression into and through the DNA-synthetic S-phase of the cell cycle. Prohibitin participates in the regulation of E2F via interaction with E2Fs (Wang et al., 1999a, 1999b). However, in plants, researchers have failed to detect the interaction between prohibitin and E2Fa or RBR1 using the yeast two-hybrid system, BiFC, Pull-down, and co-immunoprecipitation (De Diego et al., 2007; Supplemental Fig. S13), suggesting that the plant prohibitin protein may regulate cell proliferation via a different mechanism from mammals.

On the other hand, it has been demonstrated that prohibitin specifically represses E2F-mediated transcription and growth through binding to the promoters of E2F-target genes (Wang et al., 1999a, 1999b; Rastogi et al., 2006). It is reported that the MCM complex genes are the direct targets of E2F (Vandepoele et al., 2005; Naouar et al., 2009). MCM 2–7 is a hexameric complex that recruits to the chromatin early in the G1-phase of the cell cycle to form part of the prereplicative complex at origins of DNA replication and is required for replication origin activation to occur as cells enter the S-phase (Blow and Dutta, 2005). In mammals, activation of the MCM complex by CDKs leads to the initiation of DNA synthesis, and MCM proteins also act as replicative helicase to unwind DNA at replication forks during DNA synthesis (Chuang et al., 2009; Moritani and Ishimi, 2013; Wei et al., 2013). MCM proteins are expressed in cycling cells but are down-regulated and dissociated from chromatin in quiescent cells (Wei et al., 2013). Among the MCM proteins, MCM2 has been studied in a wide range of human organs, and its overexpression has been identified in various types of tumors as well as tumor-like lesions of the oral mucosa, breast, ovary, kidney, and soft tissue (Bailis and Forsburg, 2004; Suzuki et al., 2012). In addition, MCM2 has a typical ‘‘licensing’’ behavior because it binds to chromatins during the G1 phase, dissociates from chromatins once the S-phase has started, and again binds to chromatins at the end of mitosis (Kato et al., 2003). In addition, previous data showed that prohibitin can physically interact with MCM2 in vitro (Rizwani et al., 2009). In plants, the MCM complex is involved in DNA replication origins during the G1 phase of the cell cycle (Shultz et al., 2009; Tuteja et al., 2011; Rizvi et al., 2016). MCM2 has been reported to act as an essential regulator for the cell cycle and root meristem function (Ni et al., 2009). Therefore, we concentrated on the expression and function of MCM2 in the phb3 mutant on root cell proliferation and differentiation. Indeed, the MCM2 expression level is increased in the phb3 mutant (Fig. 6E), and knock-down of MCM2 can rescue the root meristem and stem cell deficiency of phb3 mutants (Fig. 7; Supplemental Fig. S11). Despite no obvious physical interaction between PHB3 and MCM2 (Supplemental Fig. S13), interestingly, transcriptional activity of PHB3 was observed (Fig. 9). Thus, taking the nuclear localization of PHB3 into consideration, we speculated that PHB3 is likely to function as a coregulator of transcription to regulate the expression level of MCM2 in modulating primary root development and cell proliferation. This prediction was supported by the results that PHB3 inhibited LUC expression driven by the promoter of MCM2 (Fig. 7), and MCM2 RNA interference in phb3 seedlings resulted in different root lengths correlated inversely with MCM2 expression (Fig. 7; Supplemental Fig. S11). Moreover, the cell proliferation deficiency in the phb3 mutants was completely rescued by down-regulated expression of MCM2 (Fig. 7E). These data suggest that PHB3 functions as a coregulator of transcription in modulating primary root development through inhibiting MCM2 transcription. Previous research showed that PHB3 maintains root stem cell niche identity through ROS-responsive AP2/ERF transcription factors such as ERF109, ERF114, and ERF11; however, repressing the expression level of ERF109, ERF114, or ERF115 in the phb3 mutant only partially restored the short root of phb3 (Kong et al., 2018). These results suggest that PHB3 also regulates root development through other components except for MCM2. The root lengths of wild type and phb3 seedlings are negatively correlated to the MCM2 expression level (Fig. 7; Supplemental Fig. S11), indicating that PHB3 is not the sole regulator of MCM2, which regulates root development as an unknown dosage effect. Combined with the findings that MCM2 down-expression completely rescued the sensitivity of phb3 to DNA damage and the DNA damage phenotype in the phb3 root (Fig. 8; Supplemental Fig. S12), these results indicate that the function of MCM2 downstream of PHB3 is necessary; however, for the root length regulation, there are other MCM2 regulators that are parallel to PHB3.

Intriguingly, PHB3 can directly bind to the MCM2 promoter regions with E2F binding cis-element (Fig. 10, D and E), and competitively regulate the MCM2 expression with E2Fa (Fig. 11). These results imply that PHB3 may be an important cell proliferation and cell cycle regulator similar to E2Fa. Although the transcriptional activity of PHB3 is not affected by the absence of E2Fa (Fig. 11B), and the MCM2 transcript level in the phb3 mutant is decreased by the absence of E2Fa (Fig. 11C), the mechanisms of PHB3 and E2Fa in regulating the expression of MCM2 and other E2Fa-target genes to modulate cell proliferation are complex. The E2Fa transcript level is decreased in the phb3 mutant (Fig. 6D), which implies that other relevance between PHB3 and E2Fa may exist. Except for binding on the E2F-binding cis-elements directly to repress MCM2 expression, another potential functional mechanism of PHB3 in gene regulation would be to recruit coeffectors to affect the E2Fa transcriptional activity. PHB3 exhibits repression activity in plants but activation activity in yeast (Fig. 9, D and E; Supplemental Fig. S13A), supporting the hypothesis that its transcriptional activity may be affected by its interacting proteins. Mammalian prohibitin represses the transcriptional activity of E2Fs1-5 by recruiting corepressors, including histone deacetylase 1, N-Co-repressors, and BrG1/Brm (Wang et al., 2002b, 2002a), whereas Rb represses the transcriptional activity of E2Fs1-3 by recruiting histone deacetylase 1, DNA methyltransferase, and BrG1/Brm (Luo et al., 1998; Magnaghi-Jaulin et al., 1998; Zhu, 2005). Prohibitin and Rb interact with distinct regions of E2F1 and respond to different upstream signals (Wang et al., 1999b). In addition, prohibitin interacts with RING finger protein 2 (RNF2), a member of the PcG (polycomb-group) family of proteins, and the two proteins regulate the activity of E2F1 via dual pathways: the direct, prohibitin-mediated pathway and the indirect, p16-mediated pathway of E2F1 transcriptional regulation (Choi et al., 2008). This leads us to speculate that the plant PHB3 may interact with some uncharacterized proteins to regulate the E2Fa transcriptional activity together. Thus, exploring the PHB3 interacting factors will be necessary to elucidate the functional mechanisms of PHB3 regulation of DNA replication, cell cycle, and cellular proliferation.

CONCLUSION

In conclusion, we propose that PHB3 functions as a negative coregulator of transcription to inhibit the expression of MCM2 so as to exert influence on the cell cycle and cell proliferation in regulating root meristem development in Arabidopsis. It is possible that the nucleo-mitochondrial dual localization of PHB3 may be induced and translocated into the nucleus, to regulate genome stability and cell proliferation under replication stress or in the presence of DNA damage agents and oxidative stress. In the nucleus, PHB3 binds directly on the target genes, e.g. MCM2, to regulate their homeostasis and guarantee accurate chromosome replication, in cooperation with E2Fa. Upon PHB3 loss of function, its repressed genes will be elevated out of control and result in replication stress. Subsequently, the genome integrity needs to be repaired through the DNA damage response and the cell cycle needs to be arrested. Consequently, the stem cell niche modulators are repressed or altered to prevent meristem cell proliferation (Fig. 12). Because the functional mechanisms differ between plant and mammalian nuclear prohibitin, some processes and important questions involved in plant PHB3 need to be identified. First, how is the replication stress sensed by and transduced to PHB3? Second, how does PHB3 orchestrate genome integrity and cell proliferation? Third, how does PHB3 translocate between the cytoplasm and nucleus? Finally, does PHB3 regulate other MCM and E2F-cis-acting element-containing genes directly to modulate DNA replication? Hence, these questions will provide more evidence to expound the activities of PHB3 in the nucleus and the communication between the mitochondria and nucleus. Consequently, the detailed regulation mechanisms for PHB3 regulation on genome integrity and cell proliferation need further study.

Figure 12.

Schematic representation of PHB3 regulation of genome integrity and cell proliferation. PHB3 can translocate into and accumulate in the nucleus and act as a transcriptional coregulator to regulate the expression of MCM2 by binding to the promoter regions that contain E2F binding cis-elements in Arabidopsis. Loss of function of PHB3 induces high expression levels of MCM2 and subsequently disturbs cell proliferation. Meanwhile, PHB3 functions in regulating genome integrity through the MCM2-dependent pathway. TRs (transcriptional regulators) represent the PHB3-interacting proteins such as transcription factors and other transcriptional coregulators; green arrow indicates promotion; red line indicates inhibition.

MATERIALS AND METHODS

Plant Materials and Growth Conditions

The Arabidopsis (Arabidopsis thaliana) T-DNA insert knockout mutant lines were obtained from the Arabidopsis Biological Resource Center (http://signal.salk.edu): phb3-ko (PHB3-knock out mutant, SALK_020707) and phb3-3 (a mutation that resulted in the conversion of Gly-37 to an Asp; Wang et al., 2010). The following types of marker lines and mutant were used: CycB1;1:GUS (Yang et al., 2015); CDKA;1:GUS (Menges et al., 2002; Vandepoele et al., 2002); WOX5pro: GFP (Blilou et al., 2005; Ji et al., 2015; Zhang et al., 2015); QC25, QC46, and QC18 (Sabatini et al., 2003); JO571:GFP (Petersson et al., 2015); J2341:GFP (Zhou et al., 2010); DR5pro:GFP (Ulmasov et al., 1997); DR5:GUS (Ulmasov et al., 1997); PIN1pro:PIN1:GFP (Benková et al., 2003); PIN2pro:PIN2:GFP (Blilou et al., 2005); shr-1 (Benfey et al., 1993); scr-1(Di Laurenzio et al., 1996); plt1-4 plt2-2 (Aida et al., 2004); SHRpro:SHR–GFP; SCRpro:SCR–GFP (Aida et al., 2004; Galinha et al., 2007); ASA1pro:GUS (Stepanova et al., 2005); and PLT1pro:PLT1:YFP and PLT2pro:PLT2:YFP (Galinha et al., 2007).

Seeds were surface sterilized for 2 min in 75% (v/v) ethanol followed by 5 min in 1% (w/v) NaClO solution with 0.1% (v/v) Triton X-100, rinsed five times with double-distilled water. For germination, seeds were plated on Murashige and Skoog (MS) medium with 1.5% (w/v) Suc and 0.8% (w/v) agar, and then vernalized at 4°C in the dark for 2 d before transferred to a growth chamber at 22°C. Plants were grown under long-day conditions (16 h of light/8 h of dark) at in a phytotron.

Root Meristem Size Analysis

Seeds were germinated and grown on vertically oriented plates from 1 to 14 d. Roots were examined at different DAG depending on the experiment requirement. Approximately 30 to 50 seedlings were examined in at least three independent experiments, which gave similar results. Roots were mounted in chloralhydrate, and then root meristem size was determined by counting the number of cortex cells in a file extending from the QC to the first elongated cell, which was excluded (Perilli and Sabatini, 2010).

Microscopy Observation and Auxin Treatments

Histochemical staining for GUS activity in homozygous transgenic plants was performed according to the described method with some modification. GUS stock solution [0.05 m NaPO₄ buffer (pH 7.0), 5 mm K₃Fe(CN)₆, 5 mm K₄Fe(CN)₆, and 10 mm X-glucuronide] was made according to described previously. Whole seedlings were stained in the GUS staining solution, and incubated at 37°C in the dark for 2 to 8 h depending on the experimental requirement, rinsed with 75% (v/v) ethanol, and mounted in HCG solution before microscopy analysis (Huang et al., 2009; Yang et al., 2015).

For homozygous transgeneic plant confocal microscopic analyses, 7-d-old seedlings grow in the MS medium were used for microscopy analysis. Cell walls of root were stained with 10 mg.ml⁻¹ PI for 5 min, washed once in distilled water, and mounted in water before confocal microscopy analysis as described (Truernit and Haseloff, 2008). Confocal images were taken using Zeiss LSM 710 laser scanning microscope with the following excitation/emission wavelengths: 561/591 nm to 635 nm for PI, 488 /505 nm to 530 nm for GFP, 514 /530 nm to 600 nm for YFP. For quantifying the fluorescence intensity, ImageJ software was used to analyze the median sections of the roots of images. Only the area with fluorescence was taken into account. The fluorescence intensities of the wild type were taken as 100%, and those samples that are needed to quantify are presented as relative values (Zhang et al., 2015). Three independent experiments were performed, and the statistical significance was evaluated by Student’s t test analysis.

For exogenous auxin treatment, 5-d-old seedlings were transferred from hormone-free medium to medium with the specified concentrations of indole-3-acetic acid for another 4 d according to the described method (Casson et al., 2002; Huang et al., 2009) with some modification.

Gene Expression Analyses

For total RNA extraction, 7 DAG roots or root tips were used to extract total RNA with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. Then, the total RNA was used to synthesize complementary DNA (cDNA) with oligo(dT) primer. RT of total RNA was performed following the manufacturer’s protocol. Total RNA (1 μg) with oligo(dT) primer was heated at 70°C for 10 min and then received cooling treatment immediately. The intermixture coupled with reverse-transcriptase MMLV-RT SPCL (Invitrogen) was heated at 42°C for 1 h, and UBQ10 (At4g05320) was used as a reference gene. The primers used for gene expression analysis by qPCR are listed in Supplemental Table S2. Each sample was performed at least three times.

For RT- PCR experiments, total RNA was extracted from the seedings roots, leaves, flowers, stems, and siliques of Arabidopsis with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. Then, the total RNA was used to synthesize cDNA with oligo(dT) primer. Next, synthesized cDNA was used as the PCR template for RT-PCR analysis. The expression levels of PHB3 in different tissue were analyzed by qPCR.

Yeast Two-Hybrid Analysis

Yeast two-hybrid analysis was performed as described previously (Matsui et al., 2004). In brief, Saccharomyces cerevisiae strain L40 was transformed by the LiAc/DNA/PEG method (Gietz et al., 1992). Cotransformed yeast cells were plated on complete synthetic yeast medium without Trp, His, and Leu to test for protein interactions and self-activation.

Transient Expression Assay

The rosette leaves were used for protoplast transformation performed as described previously protocol from Jen Sheen’s laboratory (Yoo et al., 2007; Lai et al., 2015). The transformed protoplasts were cultured for 48 h at 23°C in the opaque background, and the subcellular localization of YFP fusion protein was observed under Zeiss LSM 710 laser scanning microscope with 514 nm for excitation and 530–600 nm for emission. The chlorophyll autofluorescence was also recorded. The photos from YFP, chlorophyll, and bright field channels were merged.

For LUC assay, 5 μg reporter plasmid and 5 μg of each effector plasmid were used for protoplast transformation. The control was transformed with 5 μg reporter plasmid and 5 μg of empty effector plasmid. The transformed protoplasts were cultured for 24 h at 23°C in the darkness. Before the luciferase activity was quantified, the transformed protoplasts were treated using Dual-Luciferase Reporter Assay reagent (Invitrogen) according to the manufacturer’s instructions. At least three biological replicates were performed in each experiment.

Flow Cytometric Analysis

Plants were chopped with a razor blade in 400 mL nuclei extraction buffer (Partec Cystain UV Precise P), the supernatant was filtered over a 30-mm mesh, and 1.6 mL staining buffer (Partec Cystain UV Precise P) was added. The nuclei were analyzed by CyFlow Cube 8 with CyView software (Partec). At least 8000 nuclei were scored in triplicates for each sample.

EdU Assay

Root tips of 4 DAG Arabidopsis seedlings were submerged in liquid MS medium with 1 μM EdU for 24 h. They were then fixed for 30 min at room temperature in a 4% (w/v) formaldehyde solution in phosphate-buffered saline (PBS) with 0.1% (v/v) Triton X-100. The fixative was washed away with PBS for three 15-min washes, and the root tip sections were incubated in an EdU detection cocktail (RIBOBIO, Cell-Light EdU Apollo488 In Vitro Imaging Kit[100T]) for 30 min followed by three 15-min washes with PBS. The root tips were visualized using 545- to 600-nm wavelengths for EdU under an LSM 710 confocal microscope.

ChIP Assays

Roots (0.3 g) from 3 DAG seedlings grown on vertically oriented plates with MS medium were collected for ChIP assays (Gendrel et al., 2005; Liu et al., 2012). After fixation with formaldehyde, the chromatin was sheared to an average length of 500 bp by sonication and then immunoprecipitated with GFP-Trap_A agarose beads (ChromoTek). After cross-linking was reversed, the amount of precipitated DNA fragments and input DNA was detected by quantitative real-time PCR using specific primers listed in Supplemental Table S2. The percentage of input was calculated by determining 2^-ΔCt (= 2^{-[Ct(ChIP)-Ct(Input)]}). The exon region of retrotransposon TA3 (Han et al., 2012) and TUBULIN BETA CHAIN2 were used as negative control.

EMSA

EMSA were performed via the Light Shift Chemiluminescent EMSA Kit (Thermo Fisher Scientific). The DNA sequence (PHB domain from 28aa to 195aa) for the PHB3 protein was cloned in-frame to pMALC2X. The purified, bacterially expressed PHB3 protein was used to test the binding ability to the E2F-cis-acting elements with purified FLAG protein. For binding assays, the double-stranded oligonucleotides from the MCM2 promoter were used for EMSA analysis: 5′-GGT​TTT​AGG​TAT​TTC​CCG​CTA​TTC​GTT​TAA-3′ and 5′-TTA​AAC​GAA​TAG​CGG​GAA ATACCTAAAACC-3′. All of the oligonucleotides were labeled with biotin using the DNA 3′ End Biotinylation Kit (Thermo Fisher Scientific) and annealed to form double-stranded oligonucleotides. The unlabeled oligonucleotides were used as the specific competitor. The binding reaction and detection were carried out as follows: 10 fmol labeled double-stranded probe DNA was incubated with recombinant proteins (1–5 mg) in binding buffer (10 mm Tris-HCl [pH 8.0], 10 mM MgCl₂, 5 mm dithiothreitol, 10% (v/v) glycerol, and 50 ng/mL Poly [dI⋅dC] as the nonspecific competitor). After the reaction at room temperature for 30 min, samples were resolved by 4% native acrylamide gel in 0.5 X TBE and then were transferred to nylon membranes for chemiluminescent detection (Thermo Fisher Scientific). Three biological replicates each with three technical replicates were performed.

Diaminobenzidine and Nitrobluetetrazolium Staining

Using diaminobenzidine (DAB) staining to detect H₂O₂, the seedlings were incubated in 0.3 mg/mL DAB (Sigma-Aldrich) dissolved in 50 mm Tris-HCl (pH 5.0) for 8 h. For nitrobluetetrazolium (NBT) staining to detect superoxides, the seedlings were incubated in a reaction buffer containing 1 mm NBT (Sigma-Aldrich), 20 mm K-phosphate, and 0.1 m NaCl at pH 6.2 for 15 min. The seedlings stained by DAB or NBT were then washed three times with water. For clearing, seedlings were incubated in acidified methanol buffer (10 mL of methanol, 2 mL of HCl, 38 mL of water) at 57°C for 15 min and then in a basic solution (7% [w/v] NaOH in 60% [v/v] ethanol) for 15 min at room temperature. The seedlings were incubated 10 min at each step in the following series: 40% (v/v) ethanol, 20% (v/v) ethanol, 10% (v/v) ethanol, 5% (v/v) ethanol, and 25% (v/v) glycerol. The seedlings were then examined in 50% glycerol with an LEICA microscope.

MMS Treatment

For MMS treatment, 7-d-old seedlings on MS medium were transferred to MS medium with or without MMS (Sigma, CAT 129925) before being scored and photographed.

GSH Treatment

Seeds were plated on MS medium with or without 500 μm GSH, and then vernalized at 4°C in the dark for 2 d before transferred to a growth chamber at 22°C. Next, 7-d-old seedlings were scored and photographed before DAB and NBT staining.

Cytoplasm and Nuclear Separation from Plant

Seven-day-old seedlings on MS medium were collected for ∼1 g, frozen in liquid nitrogen, and then treated as shown in following steps: (1) Grind the tissue to a fine powder in liquid nitrogen using a cold mortar and pestle; (2) Collect the powder into a 15-mL conical tube; (3) Add 2 mL cold Lysis buffer (20 mm Tris-HCl [pH 7.4], 25% [v/v] glycerol, 20 mm KCl, 2 mm EDTA, 2.5 mm MgCl₂, 250 mm Suc, 30 mm beta-mercaptoethanol, and 1 mm phenylmethylsulfonyl fluoride) with dithiothreitol and protease inhibitor to a final concentration of 1 mm, respectively, into the powder and homogenize the mixture by gentle shaking or pipetting; (4) Filter the homogenate through a 100 μm and 40 μm nylon mesh sequentially. The filtered homogenate was centrifuged at 1,500 g under 4°C for 20 min to pellet the nuclei fraction; (5) Remove the supernatant to a new tube and centrifuge at max speed (> 10000 g) for 10 min at 4°. The supernatant is the cytoplasmic fraction. For the nuclei fraction, add 3 ml nuclei resuspension buffer (20 mm Tris-HCl [pH 7.4], 25% [v/v] glycerol, 2.5 mm MgCl₂, 0.2% [v/v] Triton X-100) to the above nuclei pellet fraction; and (6) Resuspend the nuclei by gentle pipetting or simply invert the tube several times.

Comet Assay

Protoplasts from 3-week-old seedlings were used to perform the comet assay using the Comet Assay Kit from Trevigen (CAT 4250-050-K). Comets were stained with SYBR Gold from Life Technologies, captured using a Zeiss LSM 710 confocal microscope with excitation/emission wavelengths of 488 nm/505–530 nm, and analyzed by CASP Comet Assay Software Project.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis Genome initiative or GenBank/EMBL databases under the following accession numbers: PHB3 (AT5G40770), CycB1;1(AT4G37490), CDKA;1 (AT3G48750), PIN1(AT1G73590), PIN2 (AT5G57090), PIN3 (AT1G70940), PIN4 (AT2G01420), PIN7 (AT1G23080), SCR (AT3G54220), SHR (AT4G37650), PLT1 (AT3G20840), PLT2 (AT1G51190), and WOX5 (AT3G11260).

Supplemental Data

The following supplemental materials are available.

• Supplemental Figure S1. The expression pattern of PHB3 in root and the function of PHB3 in shoot meristem development.

• Supplemental Figure S2. PHB3 with nuclear localization signal can rescue the short root phenotype of phb3 mutants.

• Supplemental Figure S3. The abnormal expression pattern of J0571:GFP and normal expression of J1092:GFP in the phb3 root tips.

• Supplemental Figure S4. SHR and SCR were affected in the phb3 mutant.

• Supplemental Figure S5. PLT1 and PLT2 were affected by the phb3 mutation and PLT2-inducible overexpression can partially rescue the phb3 meristem deficiency.

• Supplemental Figure S6. PHB3 does not bind to PLT1 and PLT2 directly.

• Supplemental Figure S7. The abnormal expression pattern of ASA1pro:GUS and down-regulated ASA1 expression level in the phb3 mutant.

• Supplemental Figure S8. Exogenous auxin can rescue the auxin level in the phb3 root tip rather than the root development deficiency of phb3 mutants.

• Supplemental Figure S9. The increased expression of WEE1pro:GUS in the phb3 seedling.

• Supplemental Figure S10. The expression of PHB3 and accumulation of PHB3 protein in the nucleus were induced by exogenous ROS.

• Supplemental Figure S11. Down-regulation of MCM2 promotes the primary root length of phb3 and wild type seedlings, and the root growth is opposite to the MCM2 expression level.

• Supplemental Figure S12. MCM2 down-regulation rescues the sensitivity of the phb3 mutant to the DNA damage agent.

• Supplemental Figure S13. PHB3 has self-activation activity in yeast but cannot interact with E2Fa.

• Supplemental Figure S14. The nucleus and mitochondrial localization of PHB3 under normal or rotenone stress conditions.

• Supplemental Figure S15. E2Fa regulates MCM2 expression dependent on E2F-binding cis-elements in the MCM2 promoter.

• Supplemental Table S1. Primers used for ChIP-qPCR.

• Supplemental Table S2. Primers used for qRT-PCR and so on.