Skip to main content
NCBI home page
Plant Physiology logoLink to Plant Physiology
. 2022 Feb 26;189(2):611–627. doi: 10.1093/plphys/kiac086

Pentatricopeptide repeat protein CNS1 regulates maize mitochondrial complex III assembly and seed development

Shuai Ma 1, Wenzhu Yang 1, Xiaoqing Liu 1, Suzhen Li 1, Ye Li 1,2, Jiameng Zhu 3, Chunyi Zhang 1,4, Xiaoduo Lu 5, Xiaojin Zhou 1,✉,†, Rumei Chen 1,✉,†
PMCID: PMC9157079  PMID: 35218364

Abstract

Mitochondrial function relies on the assembly of electron transport chain complexes, which requires coordination between proteins encoded by the mitochondrion and those of the nucleus. Here, we cloned a maize (Zea mays) cytochrome c maturation FN stabilizer1 (CNS1) and found it encodes a pentatricopeptide repeat (PPR) protein. Members of the PPR family are widely distributed in plants and are associated with RNA metabolism in organelles. P-type PPR proteins play essential roles in stabilizing the 3′-end of RNA in mitochondria; whether a similar process exists for stabilizing the 5′-terminus of mitochondrial RNA remains unclear. The kernels of cns1 exhibited arrested embryo and endosperm development, whereas neither conventional splicing deficiency nor RNA editing difference in mitochondrial genes was observed. Instead, most of the ccmFN transcripts isolated from cns1 mutant plants were 5′-truncated and therefore lacked the start codon. Biochemical and molecular data demonstrated that CNS1 is a P-type PPR protein encoded by nuclear DNA and that it localizes to the mitochondrion. Also, one binding site of CNS1 located upstream of the start codon in the ccmFN transcript. Moreover, abnormal mitochondrial morphology and dramatic upregulation of alternative oxidase genes were observed in the mutant. Together, these results indicate that CNS1 is essential for reaching a suitable level of intact ccmFN transcripts through binding to the 5′-UTR of the RNAs and maintaining 5′-integrity, which is crucial for sustaining mitochondrial complex III function to ensure mitochondrial biogenesis and seed development in maize.

A newly identified PPR protein CNS1 regulates 5′-terminal integrity of ccmFN transcripts and affects the function of mitochondrial complex III and seed development.

Introduction

The mitochondrion is a semi-autonomous organelle distributed throughout eukaryotic cells. As mitochondria are responsible for oxidative energy metabolism, they are essential for the survival of eukaryotic organisms (Peretó, 2015). Defects in mitochondrial function will cause a wide range of human diseases (Schapira, 2006) and create multiple impediments to plant development (Linke et al., 2003; Liu et al., 2013). The mitochondrial electron transport chain (ETC) contains five complexes, including complex I (NADH dehydrogenase), complex II (succinate dehydrogenase), complex III (cytochrome c [Cyt c] oxidoreductase), complex IV (Cyt c oxidase [cox]), and complex V (ATP synthase). In land plants, the mitochondrial genome encodes approximately 60 proteins with known functions, most of which participate in the ETC on the inner mitochondrial membrane (Clifton et al., 2004; Kubo and Newton, 2008). In plants, insufficient energy supply from the mitochondria may lead to slow vegetative growth (Marienfeld and Newton, 1994) and abnormal leaf formation (Gu et al., 1993).

Pentatricopeptide repeat (PPR) proteins possess RNA-binding capacity. These proteins localize mainly to the mitochondria or chloroplasts (Lurin et al., 2004), where they are associated with numerous aspects of gene expression in those organelles (Small and Peeters, 2000; Qiang et al., 2012). The PPR family consists of a large number of genes in land plants with 450 members in Arabidopsis (Arabidopsis thaliana) (Lurin et al., 2004) and 521 isoforms in maize (Zea mays) (Wei and Han, 2016). PPR proteins are characterized by 2–30 tandem repeats and can be categorized as P-type or PLS-type according to the C-terminal motifs. P-type PPR proteins contain solely canonical 35-amino-acid tandem repeats, while PLS-type members harbor L (long) and S (short) motifs with additional C-terminal domains, which are classified into the E, E+, and DYW subgroups (Barkan and Small, 2014). Generally, PLS-type PPR proteins assist in RNA editing at specific sites from cytidine (C) to uridine (U) during posttranscriptional processes in plant organelles (Kotera et al., 2005; Sosso et al., 2012). To date P-type PPR proteins are well known for two predominant functions: one is to facilitate intron splicing for mRNA maturation in organelles and another is maintaining the dynamic equilibrium state of transcripts. The maize NB mitochondrial genome encodes 22 group II introns derived from eight transcripts including ccmFC, cox2, NADH dehydrogenase subunit 1 (nad1), nad2, nad4, nad5, nad7, and rps3 (Clifton et al., 2004). As the group II introns encoded by the mitochondrial genome are unable to perform self-splicing, they need assistance from other molecules such as PPR proteins that could act as chaperones (De Longevialle et al., 2010). To date, several P-subgroup PPR proteins have been found to participate in splicing of maize mitochondrial transcripts. For instance, EMP10, EMP11, EMP12, and EMP16 affect the splicing of NAD mRNAs and thus are essential for the function of complex I (Xiu et al., 2016; Ren et al., 2017; Wang et al., 2017a, 2017b; Sun et al., 2019). Additionally, the chloroplast-localized PPR proteins exhibit similar functions as those in the mitochondria, as described above. OsWSL4 is required for the splicing of atpF, ndhA, rpl2, and rps12 in chloroplasts (Wang et al., 2017a¸ 2017b). However, a particular domain named chloroplast RNA splicing and ribosome maturation was found in the proteins, which is associated with the metabolism of group II introns (Asakura and Barkan, 2007). These results indicate there are different mechanisms of mRNA maturation processes in mitochondria and chloroplasts.

The dynamic equilibrium state between transcription and degradation is essential for maintaining a sufficient level of functional mRNA transcripts. Therefore, RNA stability is vital to determining the steady-state abundance of transcripts, which is a fundamental step in regulating the quality and quantity of organellar RNA (Giegé et al., 2000; Haïli et al., 2013). Based on these principles, a major function of the P-type PPR protein may be as a barrier, binding to the 5′ or 3′ termini of mRNAs to protect transcripts from degradation of exoribonucleases and, thereby increasing the stability of functional RNA molecules in chloroplasts (Prikryl et al., 2011). ZmPPR10 binds to similar sequences in the psaJ–rpl33 or atpI–atpH intercistronic region to block exoribonucleases from either the 5′ or 3′ terminus (Pfalz et al., 2009). Dysfunction of PPR53 abolishes the 5′ upstream sequences of ndhA and rrn23, leading to repressed expression. Further study shows that PPR53 can bind to the 5′-end of rrn23, suggesting that PPR53 acts as a protector of the 5′-proximal region of RNA in chloroplast (Zoschke et al., 2016).

A slightly different mechanism was reported for mitochondria-localized P-type PPR proteins in plants, which promote RNA cleavage at the 5′-end and RNA stabilization at the 3′-end, respectively. Arabidopsis mitochondrial stability factor 1 (AtMTSF1) was identified as a P-type PPR protein that is essential for the accumulation of mature nad4 mRNA in mitochondria. Further investigation revealed that AtMTSF1 binds to the 3′-terminus of nad4 mRNA, indicating that it acts as an RNA stabilizer (Nawel et al., 2013). In various ecotypes of Arabidopsis, the mitochondria-targeted P-class PPR protein, AtRPF4, could generate extra 5′-ends of mature ccmB transcripts in an ecotype-specific manner, depending on differences in the N-terminal repeats of RNA PROCESSING FACTOR 4 (RPF4) (Stoll et al., 2017). P-type PPR proteins play essential roles in stabilizing the 3′-end of RNA in mitochondria, whether a similar process exists for stabilizing the 5′ terminus of mitochondrial RNA remains largely unclear.

Maturation of c-type cytochromes is a crucial step in the biogenesis of complex III, which is composed of Cyt c, an electron shuttle between complex III and complex IV, and Cyt c1, a membrane protein localized to complex III. In plants, the maturation process of Cyt c is described as system I, which involves at least eight proteins encoded in both the nuclear (CCMA, CCME, and CCMH) and mitochondrial (CcmB, CcmC, CcmFN, and CcmFC) genomes (Giegé et al., 2008). Thus far, ccmFN has been identified as an essential factor for complex III, while whether the possible role of PPR protein in mediating the integrity of ccmFN to support complex III function remains undetermined.

Here, we described the maize kernel delayed-growth mutant cns1 with defects in mitochondrial complex III, embryogenesis, and endosperm development. CNS1 encodes a P-type PPR protein localized to the mitochondria, and neither splicing deficiency nor RNA editing differences in mitochondrial genes were observed in the mutant plants. Instead, 5′-truncated transcripts of ccmFN were isolated from cns1. Furthermore, we demonstrated that CNS1 bound to ccmFN at the 317 nt upstream of the start codon, which was essential for maintaining the level of intact ccmFN transcripts and thereby ensuring the assembly of mitochondrial complex III.

Results

The cns1 mutant displays delayed seed development phenotype

Maize cns1-ref mutant was isolated from the Maize Genetics Cooperation UniformMu stock which was generated by introgressing active Mutator (Mu) transposons into a W22 inbred line (McCarty et al., 2005). Self-crossed cns1-ref heterozygotes contained defective kernels (dek) at a frequency of ∼25% (118:409, χ2 test, P > 0.05), indicative of a nuclear recessive monogenic trait (Figure 1, A and B). Mutant kernels can be differentiated from normal kernels on the same ear as early as 8 d after pollination (DAP) based on their relatively small size and semi-transparent endosperm (Figure 1C). In addition, the abnormal kernels were more distinguishable at 15 DAP, appearing small, white, and collapsed (Figure 1, A and D). The growth of mutant endosperm and embryo was substantially delayed, whereas a growth trend was apparent for wild-type (WT) kernels at 8, 15, and 20 DAP (Figure 1E). At maturity, dehydration of the mutant seeds aggravated the phenotype, leading to collapsed and wrinkled pericarp (Figure 1B) and reduced sizes of the embryo and endosperm (Figure 1, F and G). As cns1-ref kernels could not germinate, the mutant allele is lethal and must be maintained in heterozygotes.

Figure 1.

Figure 1

Open in a new tab

Delayed development of cns1-ref mutant kernels. A, An immature self-pollinated ear from a heterozygous cns1-ref mutant. Arrows indicate mutant kernels. Scale bar = 1 cm. B, A mature self-pollinated ear from a heterozygous cns1-ref mutant. Arrows indicate mutant kernels. Scale bar = 1 cm. C–E, Defective (upper) and WT (lower) seeds at different developing stages of the cns1-ref mutant, including 8, 15, and 20 DAP, respectively. Scale bar = 0.5 cm. F, Germinal view of WT (left) and cns1-ref mutant (right) kernels. Scale bar = 1 mm. G, Sagittal hand sections of mature WT (left) and cns1-ref mutant (right) kernels. Scale bar = 1 mm.

Compared with the WT, mutant kernels not only showed smaller morphological sizes but also exhibited delayed development. To investigate the influence of cns1-ref mutation on the immature embryo and endosperm, paraffin sections were made from WT and mutant kernels taken from the same ear at 13 and 18 DAP. Embryogenesis of maize is a process of cell growth and differentiation that can be divided into four stages, including pro-embryo, transition, coleoptile, and initiation of kernel development (Vernoud et al., 2005; Doll et al., 2017). At 13 DAP, unfilled space between the pericarp and endosperm suggests that the mutant kernel was in the transition stage (Figure 2A), while the WT sibling developed at the coleoptile stage with differentiated shoot apical meristem (SAM) and root apical meristem (Figure 2B). The growth retardation became more severe at 18 DAP, as development of the mutant embryo was arrested at the transition stage, whereas the WT embryo continued to develop to further stages, with steady progression of both cell size and differentiation (Figure 2, C and D). In summary, both embryo and endosperm development of mutant kernels were blocked at the early development stages.

Figure 2.

Figure 2

Open in a new tab

Morphology of the developing embryo and endosperm between cns1-ref and WT. A and C, HE staining of longitudinal sections of cns1-ref at 13 and 18 DAP, respectively. Arrow indicates the embryo. Scale bar = 0.5 mm. B and D, HE staining of longitudinal sections of the WT at 13 and 18 DAP, respectively. Scale bar = 0.5 mm. E and F, BETL cells of WT (E) and cns1-ref kernels (F) at 18 DAP. Scale bar = 50 µm. Em, embryo; En, endosperm. The images were digitally extracted for comparison.

Basal endosperm transfer layer (BETL) cells absorb nutrients from the placenta-chalaza (Bihmidine et al., 2013). Substantial differences in BETL cells were found between WT and mutant kernels at 18 DAP, as fewer cells were observed to develop the highly specialized ingrowth wall toward the endosperm in mutant kernels (Figure 2, E and F). As impaired development of BETL transfer cells may be caused by insufficient energy from the ETC (Ren et al., 2017), arrested development of the BETL in cns1-ref may result from reduced energy supply.

Mapping of the CNS1 locus

The maize mutant UFMu-06493 was obtained from the UniformMu transposon-tagging population and designated the reference allele of cns1. The heterozygous mutant was pollinated to the B73 inbred background and F1cns1-ref/+ plants were self-crossed to create an F2 mapping population for bulk segregant RNA-seq (BSR-seq). WT and mutant individuals segregated from the same F2 ear were combined into two independent pools for high-throughput sequencing. A total of 32,449 filtered polymorphic markers were screened and the B73 genome (RefGen_v4) was selected as the reference for single-nucleotide polymorphism (SNP) index calculation (Jiao et al., 2017). We conducted frequency variance analysis of SNPs in two offspring and plotted a Manhattan map showing the whole-genome distribution of ΔSNP index values. The 95% confidence interval was selected as the threshold and a strong peak was observed at 20–25 Mb of chromosome 2, indicating that this region has a high probability of linkage with the cns1-ref locus (Figure 3A). Moreover, this locus was verified based on surrounding insertion–deletion (In–Del) markers used in 164 individuals in the F2 mapping population (Figure 3B).

Figure 3.

Figure 3

Open in a new tab

Cloning and identification of CNS1. A, Frequency variance analysis result from bulk segregant RNA-seq (BSR-seq). The 10 chromosomes are represented with different colors and the linked region was indicated by a red arrow. B, The cns1 locus was mapped to a region of 6 Mb on chromosome 2 between the two In–Del markers IDP668 and umc1555. C, The corresponding Mu element was identified using the Genome Walker assay. A band that co-segregated with cns1-ref mutants is indicated by a black arrowhead. D, Schematic structure of CNS1. The locations of two Mu insertions (cns1-ref and cns1-mu1) and an EMS mutant (stop-gain) are shown by black triangles. The EcoRV restriction site for Genome Walker assay is indicated with a red triangle. The coding regions for PPR motifs are marked with black boxes.

As the cns1 mutant was obtained from a Mu-tagging population, we speculated that the mutation was caused by a Mu insertion. Therefore, the Mu tagging element was isolated from both heterozygous cns1-ref/+ plants and WT siblings. One candidate band was found exclusively in heterozygotes, and the corresponding fragment was sequenced (Figure 3C). The Mu element was identified as a Mu7, which was inserted at +27 bp downstream of the predicted translation start site (ATG) of Zm00001d002771 (Figure 3D). Then we randomly genotyped 16 mutant kernels and confirmed that the Mu7 insertion was tightly linked with the mutant phenotype (Supplemental Figure S1), indicating that Zm00001d002771 is the candidate gene.

Allelic test and complementation of the cns1 mutant with a CNS1-overexpressing maize line

Two strategies, complementation of the cns1-ref mutant with CNS1 overexpression (OE) and allelic test with additional alleles, were employed to confirm that we had cloned CNS1. For complementation assays, the entire coding region of CNS1 was inserted into a binary vector under the control of a constitutive CaMV 35s promoter. Two T0 transgenic plants were obtained and substantial over-accumulation of exogenous CNS1 was confirmed through reverse transcription quantitative PCR (RT-qPCR; Figure 4A). We pollinated cns1-ref/+ heterozygotes using pollen from T0 transgenic plants and genotyped the generated F1 plants retaining individuals with the heterozygous cns1-allele and hemizygous transgene locus (Figure 4B). As shown in Figure 4C, self-pollinated ears of F1 plants were analyzed to determine the segregation ratio, which was 1:15 for dek: WT (31:399, χ2 test, P > 0.05). We further verified the result by genotyping analysis of individual kernels. Taken together, these results indicate that the CNS1-OE locus could complement the CNS1-dysfunction caused by the homozygous Mu insertion (Supplemental Figure S2).

Figure 4.

Figure 4

Open in a new tab

Transgenic complementation and allelic test of cns1-ref. A, Transcriptional level of CNS1 in two T0 OE plants. Data are shown as the mean ± sd; n = 4. B, Schematic illustration of the pipeline for complementary pollination. C, A mature self-pollinated ear from a heterozygous cns1/+ plant with the CNS1 transgene locus. White arrows represent defective kernels and the segregation ratio for WT:cns1 is 15:1 (399:31, χ2 test, P > 0.05). D, A 20 DAP ear generated through crossing of the heterozygous cns1-EMS1/+ plant (as the male parent) and heterozygous cns1-ref/+ plant (as the female parent) for an allelic test. Black arrows indicate mutant kernels. Scale bar = 1 cm. E, A 20 DAP ear generated by crossing of the heterozygous cns1-ref/+ plant (as male) and heterozygous cns1-EMS1/+ plant (as female) for the allelic test. Black arrows indicate mutant kernels. Scale bar = 1 cm.

For further confirmation of this result, two more cns1 mutants were selected for the allelic test. The cns1-EMS1 mutant was obtained from an ethyl methanesulfonate (EMS)-induced mutation population (Lu et al., 2018) with a G/A transition causing a stop-gained effect at +477 bp downstream of the start codon (Figure 3D). In addition, cns1-mu1 was isolated from UniformMu stock carrying a Mu insertion at −67 bp upstream of the start codon corresponding to a putative promoter region (UFMu-11274; Figure 3D). Allelic test via reciprocal crosses between cns1-ref/+ and cns1-EMS1/+ heterozygotes plants produced ears containing defective kernels at a 1:3 ratio (dek: WT = 84:275 χ2 test, P > 0.05 and dek: WT = 101:289 χ2 test, P > 0.05, respectively), confirming that these mutations could not complement each other and Zm00001d002771 is the causative gene for cns1 (Figure 4, D and E). Differentially expressed genes (DEGs) analysis of BSR data showed that CNS1 is expressed at a similar level between cns1-ref mutant and WT (Supplemental Figure S3A). Therefore, we examined the accumulation of CNS1 in the mutant kernels. RNA was extracted from kernels of 15 DAP cns1-ref and WT plants, respectively, and the expression level was detected using RT-qPCR. Interestingly, Mu-inserted CNS1 was slightly increased in the cns1-ref mutant (Supplemental Figure S3B). Since the insertion of Mu7 produces a premature stop codon in the cns1-ref mutant and generates a protein that is truncated before the first PPR motif, we speculated that the Mu insertion impairs the function of CNS1 (Supplemental Figure S4). Interestingly, no mutant kernels were present in the selfed ears of cns1-mu1/+ heterozygotes and we also identified cns1-mu1 homozygous individuals, suggesting that cns1-mu1 is not lethal for kernel development. Therefore, the expression of CNS1 was measured in cns1-mu1 homozygotes and WT plants. As a result, we found that the abundance of CNS1 mRNA was significantly reduced in the mutant but not abolished (Supplemental Figure S3C), confirming that cns1-mu1 is not a null allele and indicating that the remaining amount of CNS1 is sufficient for seed development.

CNS1 is constitutively expressed in various tissues and encodes a P-type PPR protein targeted to the mitochondrion

CNS1 is an intronless gene with a 1,650 bp coding region, which encodes a 549-amino acid PPR protein. The conserved motifs of CNS1 were identified using PPR protein analyzer TPRpred (Karpenahalli et al., 2007) and a plant PPR protein database (Cheng et al., 2016). Fourteen P-repeat domains and no E, E+, or DYW motifs were found in CNS1, suggesting that it is a P-type PPR protein (Supplemental Table S1).

We examined the expression patterns of CNS1 in various tissues of maize using RT-qPCR. CNS1 was constitutively expressed in all tissues with relatively low levels in developing kernels (Figure 5A). Based on the prediction obtained from TargetP (http://www.cbs.dtu.dk/services/TargetP/), CNS1 contains a putative mitochondrial localization signal. Thus, to determine the subcellular localization of CNS1, its full-length open-reading frame (ORF) was C-terminal fused with green fluorescent protein (GFP) and transiently expressed in maize mesophyll protoplasts. In addition, the fusion protein EMP16-mCherry was co-transformed as a maker of mitochondrial targeting (Xiu et al., 2016). The GFP signals co-localized with the EMP16-mCherry marker (Figure 5B), confirming the mitochondrial localization of CNS1.

Figure 5.

Figure 5

Open in a new tab

Gene expression pattern of CNS1 and subcellular localization of CNS1 protein. A, Expression profile of CNS1 in various tissues and developing seeds (8, 10, 12, 14, 16, 18, and 20 DAP). ZmActin1 (Zm00001d010159) was used as an endogenous control. The relative expression level of CNS1 is calculated from three biological repeats. Bars represent standard deviations. B, Subcellular localization of CNS1–GFP fusion protein in mitochondria. Maize mesophyll protoplasts were co-transformed with CNS1-GFP and a mitochondria-targeted protein (EMP16-mCherry), and the fluorescence signals were detected through confocal microscopy. Scale bar = 10 μm.

cns1 mutation causes abnormal mitochondrial morphology

As strong developmental defects and histological changes were observed in cns1 and WT kernels, transmission electron microscopy (TEM) was used to detect differences in the subcellular structures of 18 DAP kernels. Compared with the WT, fewer amyloplasts and oil bodies were present in the cns1 mutant (Figure 6, A and D), suggesting that the development of endosperm cells is delayed in mutant kernels. Moreover, in contrast to the well-defined regular structures of mitochondria observed in WT, mitochondria of the cns1-ref mutant exhibited several types of morphological defects, including disordered inner membrane structures, abnormal shapes, and enlargement with empty internal space (Figure 6, B, C, E, and F). These results indicate that mutation of CNS1 severely impairs mitochondrial biogenesis, which may affect mitochondrial function and leads to defects in seed development.

Figure 6.

Figure 6

Open in a new tab

Ultrastructure of WT and cns1-ref mutant kernels at 15 DAP. TEM images of WT (A–C) and cns1-ref mutant (D–F) kernels were obtained at 15 DAP. A and D, TEM images of endosperm cells at low magnification. Scale bar = 5 μm. B and C, Regular mitochondrial structures observed in WT seeds at high magnification. Scale bar = 500 nm. E and F, Abnormal mitochondrial morphologies detected in mutant kernels at high magnification. Scale bar = 500 nm. Mt, mitochondrion; Pb, protein body; Am, amyloplast; Al, aleurone; N, nucleus; Ob, oil body.

Mutation of CNS1 did not affect intron splicing or RNA editing efficiency

PPR is a large protein family with roles in RNA metabolism in the mitochondria and chloroplasts (Fujii and Small, 2011). Previous studies have indicated that PPR proteins participate in intron-splicing, RNA editing, stability, and processing of corresponding RNA precursors (Qi et al., 2017; Dai et al., 2018). Since CNS1 is classified as a P-type PPR protein, we first explored whether the splicing efficiency was altered in the cns1-ref mutant. To minimize differences in growth conditions, total RNA was extracted from WT and mutant kernels segregated from the same ear at 15 DAP. Primers were designed to amplify the full length of the ORF. The transcript abundance and integrity of 35 mitochondrial protein-coding genes were analyzed in mutant and WT kernels through RT-PCR. Surprisingly, in contrast to mutations of conventional P-type PPR proteins reported previously, we observe no substantial differences in RNA splicing between WT and cns1-ref (Figure 7A).

Figure 7.

Figure 7

Open in a new tab

Splicing efficiency of mitochondrial transcripts in the cns1-ref mutant and phylogenetic analysis of CNS1. A, The splicing efficiency of 35 protein-coding mitochondrial transcripts was measured in WT (left) and cns1-ref (mutant, right) by RT-PCR. RNA was extracted from kernels taken from the same ear with the pericarp removed and primers were designed for the sequences flanking the 5′ or 3′ cDNA termini. The expression of ZmActin1 (Zm00001d010159) was used as an internal control. B, The phylogenetic relationship between CNS1 and functionally identified PPR proteins. The phylogenetic tree was constructed using CNS1 and PPR proteins associated with various posttranscriptional functions, including RNA stability (ZmPPR78, AtMTSF1, and AtRPF3), intron splicing (ZmEMP16, ZmPPR4, AtABO5, and AtOTP43), RNA editing (AtMEF19, ZmEMP5, ZmPPR2263, ZmEMP9, and ZmPPR10), and RNA processing (AtCRR2, ZmPPR6, and AtRPF4). Green indicates PPR proteins with RNA editing functions, while orange indicates proteins with putative RNA stability functions. CNS1 is marked with an asterisk.

To explore the function of CNS1, we performed RNA-sequencing (RNA-seq) to compare both gene expression levels and RNA editing processes between mitochondria of WT and cns1-ref. RNA was extracted from mitochondria, isolated from WT and cns1-ref kernels at 15 DAP. Through evaluation of the error rate distribution with reads count, determination of base contents, and filtering of raw reads, we qualified the datasets as suitable for subsequent analysis (Supplemental Table S2). Total clean reads from the two samples were mapped to the mitochondrial genome of maize (B73 RefGen_V4). The coverage of the total reads was 96.97% and 97.56% for cns1 and WT, respectively.

For RNA editing efficiency screening, at least 10 counts of reads for each SNP variant were used as the threshold for validated sequencing depth. As a result, 270 SNP sites in the protein-coding region were detected in between cns1-ref and WT mitochondria transcripts. The most substantial changes in editing efficiency were found in orf107-c, with the level of 50.7% in cns1and 76.2% in the WT (Supplemental Table S3). It was reported that the mutation in PPR proteins led to a significant reduction in RNA editing efficiency, for example 4.4% in dek39 compared with 64% in WT (Li et al., 2017). Considering that the change in editing efficiency of orf107-c in cns1 was not as substantial as previous observations, we conclude that CNS1 may not mediate RNA editing.

Then, a phylogenetic tree was constructed using PPR proteins with various functions. Interestingly, we found that CNS1 was grouped in a subclade (colored in yellow) related to RNA stabilization functions, while another branch of PPR proteins (colored in green) assisted in RNA editing (Figure 7B). This phylogenetic relationship suggests that CNS1 may play roles in mediating mRNA stability.

Impaired complex III and complex I+III2 assembly in cns1-ref

As multiple defects in mitochondrial morphology were observed, we suspected that the respiration activity of mitochondria was affected. Mitochondria were extracted from mutant and normal kernels. Then, the respiratory complexes were separated through blue-native polyacrylamide gel electrophoresis (BN-PAGE), and the activity of complex I was detected via in-gel staining of NADH dehydrogenase activity. Unlike mutants of P-type PPR proteins reported in maize, neither the abundance nor activity of mono-complex I exhibited substantial differences between cns1-ref and WT (Figure 8A). On the other hand, we found that the accumulation and activity of complex III and complex I+III2 were dramatically reduced in cns1 kernels compared to WT (Figure 8, B and C). These results suggest that the function and biogenesis of mitochondrial complex III are reduced in cns1-ref kernels.

Figure 8.

Figure 8

Open in a new tab

BN-PAGE and activity detection of complex I and III. A, BN-PAGE analysis of mitochondrial complexes. Mitochondrial protein extracts were separated on 4%–16% polyacrylamide gel and stained with CBB. The positions of various respiratory complexes are indicated. B, In gel detection of mitochondrial complex I activity. The positions of complex I and super complex I+III2 are indicated. C, Complex III activity determined through DAB staining. The position of the respiratory complex III is indicated.

Truncated 5′ extremity of ccmFN was observed in the cns1 mutant

A conserved function of P-type PPR proteins is protection of RNA from degradation, thereby maintaining RNA integrity. CNS1 was classified in a clade with RNA stabilization functions, and we verified that CNS1 was not involved in splicing or editing of mitochondrial RNA. Therefore, we hypothesized that CNS1 plays a role in mediating mRNA stability. As the assembly and activity of complex III were impaired in the cns1 mutant, we tested whether CNS1 regulates the stability or homogeneity of the 5′ and 3′ termini of ccmC, ccm B, ccmFC, and ccmFN, four transcripts required for the assembly of complex III. Interestingly, in circular RT-PCR (cRT-PCR), the only difference in the amplification length of ccmFN was observed between cns1 and WT (Figure 9A). The corresponding bands from cns1 and WT samples were subcloned, sequenced, and mapped to the mRNA of ccmFN. As a result, sequences truncated at the 5′ extremities of ccmFN were found in the cns1 mutant enriched relative to the CDS region, as the 5′ termini of ccmFN located at 317 nt upstream of the start codon in WT kernels (29 of 35 clones) (Figure 9C). Unlike the heterogeneity observed in the 5′ terminus of ccmFN, the 3′ ends of ccmFN were consistent between cns1 and WT. Moreover, the 5′-UTR (including the 5′ terminal region in cns1 mutant), CDS, and 3′-UTR of ccmFN were amplified by RT-qPCR, demonstrating that the amount of intact 5′-UTR was significantly reduced in cns1 kernels, verifying the truncation of the 5′ extremities of ccmFN in the mutant (Figure 9B). These results suggest that CNS1 may bind to the 5′ UTR of ccmFN (mainly at 317 nt upstream of the start codon to protect ccmFN from degradation).

Figure 9.

Figure 9

Open in a new tab

Transcript end mapping of cns1 and binding affinity measurement between CNS1 and its ligand oligonucleotide. A, cRT-PCR identification of the 5′ and 3′ extremities of four transcripts associated with mitochondrial complex III synthesis, respectively. Bands indicated with asterisks were purified and sequenced. B, RT-qPCR analysis results for three different regions of ccmFN transcripts. RNA was isolated from WT and cns1-ref, and data are shown as means ± sd of three biological replicates (ns, not significant; *P-value < 0.05; ***P-value < 0.001; Student’s t test). C, Schematic diagram of the 5′-region of ccmFN. The start codon (ATG) of ccmFN mRNA was set to 1. Vertical lines indicate the 5′ termini identified by sequencing of cRT-PCR amplification products and numbers in the bottom row represent amounts of clones. D, A Coomassie blue-stained gel of purified His6-CNS1 protein expressed in E. coli. M, marker of protein weight. E, His6-CNS1 bound to the candidate site of ccmFN. The candidate and control sequences of the CNS1 binding site (upper panel) are indicated in the mRNA-seq of ccmFN (C). A fixed concentration of various Cy5-labeled RNA oligonucleotides was incubated with increasing amounts of His6-CNS1 protein, respectively. The plot shows the relationship between the normalized fraction of bound RNA and the concentration of the protein. The dissociation constant (KD) of each tested combination is shown on the diagram. Data from three biological measurements were analyzed by MST and curves were fitted using Prism version 8.0 software. Bars represent means ± sd. F, The result of EMSA shows that CNS1 binds to the 317 nt upstream of the putative start codon of ccmFN transcript. The Cy5-labeled and competitor RNA probes are indicated in (C).

CNS1 protein binds to the 5′ UTR of ccmFN in vitro

To verify that CNS1 protein binds to the 5′ end of mitochondrial ccmFN mRNA, the full-length CNS1 protein was expressed in Escherichia coli and purified using nickel-nitrilotriacetic acid (Ni2+-NTA) affinity chromatography (Figure 9D). The candidate binding site (317 nt upstream of the start codon of ccmFN mRNA) and the control site (randomly selected in 5′ UTR upstream of the candidate site) were synthesized as ssRNA oligonucleotides with 5′ terminal Cy5 labeling (Figure 9E). Binding efficiency was determined using microscale thermophoresis (MST), which is a widely used technique to precisely quantify biomolecular interactions along a temperature gradient by measuring the binding dissociation constant (KD) (Wienken et al., 2010; Seidel et al., 2013; Parker and Newstead, 2014). The result showed that CNS1 was able to bind to the candidate site (KD = 22.4 μM), while a substantially weaker interaction (KD = 86.1 μM) was observed at the control site (Figure 9E). Further, electrophoretic mobility shift assay (EMSA) was used to verify that CNS1 was able to directly bind the 5′ end of ccmFN transcript (Figure 9F). Taken together, these results confirm that CNS1 binds to the 5′ UTR of ccmFN in vitro, and that the binding of CNS1 may contribute to maintaining the integrity of the 5′-UTR.

Discussion

In this study, we isolated the maize Mu insertion mutant cns1-ref with a delayed kernel growth phenotype and abnormal mitochondrial morphology. CNS1 was identified as a P-type PPR protein containing 14 PPR motifs that is essential for kernel development in maize. Truncation of the 5′-terminus of ccmFN was detected in cns1, which resulted in loss of the start codon of ccmFN in the mutant. Meanwhile, a close phylogenetic relationship was found between CNS1 and PPR proteins with reported functions in stabilizing mitochondrial gene transcripts (Figure 7B). Therefore, we speculated that loss of CNS1 may reduce the level of intact ccmFN transcripts, thereby disrupting the assembly of complex III and super-complex I+III2. Moreover, nuclear transcriptome analysis revealed 5,852 DEGs, which were annotated using the Gene Ontology (GO) database. As a result, the enriched GO classes, RNA biosynthetic process (GO:0032774), and RNA metabolic process (GO:0016070) are closely related to mitochondrial function (Figure 10A).

Figure 10.

Figure 10

Open in a new tab

Nuclear transcriptome analysis of cns1-ref and the WT. A, The most significant GO terms among DEGs in the nuclear transcriptome. The number of genes in each GO accession category is shown according to the level of significance. B, Read counts of genes encoded in the nucleus and associated with mitochondrial function are shown. Data were obtained through nuclear transcriptome analysis. C, RT-qPCR confirmation of the genes listed in (B). Gray and black indicate cns1-ref and WT, respectively.

CNS1 is involved in maintaining the integrity of ccmFN mRNA in maize

PPR proteins have been reported to associate with a series of organelle posttranscriptional modifications in various organisms (Gothandam et al., 2005; Klein et al., 2005; Uchida et al., 2011). To date, several mitochondria-localized PPR proteins have been identified in maize, the reported functions mostly related to splicing group II introns of mitochondrial encoded transcripts or RNA editing from C to U. In maize, P-subgroup PPR proteins generally function in intron splicing of genes encoding subunits of complex I, as they are enriched in group II introns. For instance, EMP11 and DEK2 are required for splicing of several introns in nad1 (Qi et al., 2017; Ren et al., 2017), while DEK35 is responsible for cis-splicing of the first intron in nad4 (Chen et al., 2017). Additionally, EMP8 affects both cis- and trans-splicing of multiple introns in nad1, nad2, and nad4 (Sun et al., 2018). In contrast, PLS-type PPR proteins generally participate in RNA editing of organelle-encoded gene transcripts, a process that transforms certain C residues on target RNA into U residues, depending on the E or DYW domain at the C-terminus of the PPR protein (Zehrmann et al., 2011).

In mitochondria, the integrity of the 3′-terminal region is crucial to mRNA stability. MTSF1 binds to the 3′ end of nad4 mRNA and prevents the progression of 3′ to 5′ exoribonucleases (Haïli et al., 2013). Also, MTSF2, a P-type PPR protein, plays a role in defining the 3′ end of nad1 exons 2–3, thus affecting RNA stability, in addition to exhibiting similar functions to MTSF1(Wang et al., 2017a, 2017b). However, different from 3′ termini, the 5′ definitions are usually created by endonucleolytic cleavage, mediated by RPFs and several PPR proteins in mitochondria (Jonietz et al., 2010, 2011; Hauler et al., 2013). The possible involvement of PPR proteins in the 5′-terminal protection of mitochondrial mRNA remains unknown.

In our study, neither defects in the splicing of group II introns nor substantial RNA editing differences were observed in the cns1 mutant. Meanwhile, CNS1 falls into a group consisting of PPR members that function as stabilizing factors, suggesting that CNS1 belongs to a restricted class of PPR proteins which may be required for maintaining mRNA integrity. Consistent with this hypothesis, the 5′ terminal truncation was detected for ccmFN in the cns1 mutant, and the MST and EMSA assay further verified that binding occurs between CNS1 and the truncation site, supporting our suggestion that CNS1 participates in maintaining the integrity of the 5′-UTR of ccmFN.

ccmFN is essential for the maturation of Cyt c and, thus, to the assembly of complex III

Cyt c, located on the inner membrane of mitochondria, functions as an electron shuttle from complex III to complex IV, and its activity is necessary for all living cells. The Cyt c maturation process differs between plants and animals, with plant cells sharing a putative heme delivery pathway known as a system I (Kranz et al., 1998). In brief, an ATP-binding cassette (ABC) transporter is predicted to be essential for the Cyt c maturation, although its substrates and precise functions in heme c biogenesis remain obscure (Thöny-Meyer et al., 1995). Interestingly, the ABC transporter consists of two domains encoded by separate genomes: CcmB, a transmembrane domain encoded in mitochondrial sequences, and CCMA, an ATP-binding domain originating from the nuclear genome (Faivrenitschke et al., 2001; Rayapuram et al., 2007). Moreover, the putative heme transporter CcmC and two enzymes (CcmFN and CcmFC) catalyzing apocytochrome are also encoded by mitochondria genome and involved in the Cyt c maturation pathway (Bonnard and Grienenberger, 1995; Clifton et al., 2004). In addition, Cyt c1, another c-type cytochrome localized to the inner membrane of mitochondria, shares the maturation pathway of Cyt c. Thus, the expression level of ccmFN also affects the function of Cyt c1 and the assembly and activity of complex III, with resulting impacts on electron transport between complex III and complex IV.

The essential role of CcmFN in the biogenesis of complex III has been reported in maize. EMP7 encodes an E-subgroup PPR protein targeted to the mitochondrial. Loss of EMP7 strongly reduced the C to U editing efficiency at the ccmFN-1,553 and thereby impaired the assembly of complex III (Sun et al., 2015). Likewise, the absence of CcmFC, another catalytic enzyme involved in the maturation of Cyt c, led to the retarded growth of the Arabidopsis mutant wtf9 (Kroeger et al., 2012). In the cns1-ref mutant, we identified a truncated 5′-terminal sequence of ccmFN, lacking its original start codon, which presumably impaired protein translation. In addition, the abundance of mitochondrial complex III was almost abolished and its activity was impaired. Taken together, our results support the hypothesis that CNS1 acts as an mRNA stabilizer to maintain the integrity of ccmFN, which is essential for Cyt c maturation as well as the function and assembly of complex III.

Besides, the expression of AOX2, a core gene in the alternative oxidation pathway, was substantially induced in cns1, as revealed through RNA transcriptome analysis (Figure 10B; Supplemental Table S4) and RT-qPCR (Figure 10C), indicating that deficiency of complex III leads to activation of additional pathways for obtaining energy to support kernel development.

Materials and methods

Plant materials

The maize (Z. mays) cns1-ref mutant was isolated from the UniformMu insertion population with stock number UFMu06493 (McCarty et al., 2005). The cns1-ref heterozygotes were grown under field condition and self-crossed. The F2 ear exhibited a 3:1 segregation ratio of normal and defective kernels. The heterozygous cns1-ref/+ plants were crossed into the B73 inbred line, and the progenies were self-crossed to obtain the F2 population for positional cloning. For the allelic test, an additional mutant allele cns1-EMS1, which contained a premature stop gain mutation was obtained from the Maize EMS induced Mutant Database (Lu et al., 2018).

Cytological sections

Immature WT and mutant seeds from the same ear were collected at 13 and 18 DAP, respectively. The seeds were fixed in Carnoy’s solution under vacuum at room temperature for 2 h. The fixed tissues were dehydrated in an ethanol gradient series (70%, 75%, 80%, 85%, 90%, 95%, 100%, and 100% [v/v], for 2 h at each step) and then dealcoholized in a gradient of xylene (25%, 50%, 70%, 100%, and 100% [v/v]). The samples were immersed in liquid paraffin with 50% (v/v) xylene at 42°C overnight and the paraffin infiltration was performed six times in pure paraffin at 59°C. The tissues were embedded in paraffin and sectioned with a Leica microtome RM 2235. The sections were stained with hematoxylin–eosin (HE) and images were captured with a Leica M165 FC stereomicroscope.

Positional cloning of CNS1

The cns1-ref allele was crossed into B73, and selfed F2 populations were used for BSR-seq. WT and mutant kernels were collected from the same ear on 12 DAP. The pericarps were removed and 60 kernels of either WT or mutant phenotype were pooled for BSR analysis. Total RNA was extracted from each pool using TransZol Plant (TransGen, Beijing, China) and subjected to sequencing by Anoroad Gene Technology (Beijing, China). The data were analyzed by Allwegene Technologies (Beijing, China). In brief, whole chromosomes were scanned using the mean value of SNP-index derived from a 1-Mb window and scanning was performed with a step size of 1 kb. For digestion–ligation–amplification analysis, genomic DNA was isolated from the leaves of WT and heterozygous plants, obtained through self-pollination to identify their phenotypes. Mu-tagged sequences were obtained using the Genome Walker Kit (Clontech, Mountain View, CA, USA), and the Mu-flanking sequences linked with defective kernel phenotype were sequenced and analyzed.

Complementary test using CNS1 OE plants

The ORF of CNS1 was constructed into the pCAMBIA3301 binary expression vector, which contains a double CaMV 35S promoter. The vector was transformed into Hi-II immature embryos following a previously reported method with some modifications (Frame et al., 2002). Transgenic T0 seedlings were selected based on Basta tolerance and high expression of CNS1 compared with the WT. Positive lines were crossed with heterozygote cns1-ref/+ plants. F1 plants heterozygous for both cns1 mutation and CNS1 OE were self-pollinated for determination of the segregation ratio between WT and mutant kernels.

Subcellular localization of CNS1

The full-length CDS of CNS1 except the stop codon was cloned into a pRTL2 expression vector with GFP fused to the C-terminus. To generate a mitochondrial localization marker, an EMP16-mCherry fusion construct was obtained using vector pTF101.1 (Xiu et al., 2016). These two constructs were co-transformed into maize mesophyll protoplast cells according to previous reports (Sang-Dong et al., 2007). The fluorescent signals were detected using a Zeiss confocal microscope LSM700 (40× oil objective). GFP was excited at 488 nm and emission was detected at 493–551 nm. The mCherry signal was excited at 555 nm and emission was detected at 610 nm.

RNA splicing efficiency analysis and RT-qPCR

To assess the spatial and temporal expression patterns of CNS1, various tissues were sampled from W22 inbred lines. To identify DEGs and RNA splicing efficiencies, WT and mutant kernels were collected from the same ear. The tissues were ground in liquid nitrogen into fine powder and total RNA was extracted with TRIzol solution according to the user manual (Invitrogen, Waltham, MA, USA). The first-strand cDNA was synthesized using SuperScript VI Reverse Transcriptase (Invitrogen) with oligo(dT) for genes encoded by nucleus and random hexamers for mitochondrial transcripts, respectively. The RNA splicing efficiency was measured by RT-PCR with primers designed for the regions flanking the 5′ or 3′ terminus of the protein-coding sequences of mitochondrial genes. The RT-qPCR was performed with the TB Green Premix Ex Taq kit (TaKaRa, Shiga, Japan) using the ABI 7500 Real-Time PCR System according to standard protocol. Data were analyzed using ABI7500 software via the ΔΔCT method. ZmActin1 (Zm00001d010159) was used as an internal control for measuring the expression level of CNS1 and genes encoded in the nucleus, while 26s rRNA encoded in the mitochondrion was used as a control for the expression analysis of mitochondrial genes in cns1-ref and WT. Primers used for RT-qPCR and splicing efficiency analysis are listed in Supplemental Table S5. Representative results from three biological repeats are shown.

Transmission electron microscopy

Immature WT and mutant seeds were harvested at 15 DAP and fixed in 5% (v/v) glutaraldehyde under vacuum. TEM was performed at the Institute of Crop Science of the Chinese Academy of Agricultural Sciences using a Hitachi HT7700 transmission electron microscope.

Phylogenetic analysis

Phylogenetic analysis of CNS1 and different types of PPR proteins was performed using neighbor-joining methods by MEGA version 7 software. Gene IDs and corresponding annotations were listed in the Accession numbers section.

Isolation of mitochondria from WT and mutant seeds

Mitochondria were isolated from 15 g samples of immature seeds (18 DAP) according to previously reported methods (Meyer et al., 2009; Chen et al., 2017) with some modifications. In brief, WT and cns1-ref seeds without the pericarp were ground individually in liquid nitrogen and the powder was resuspended in 25 mL extraction buffer (0.3 M sucrose, 5 mM tetrasodium pyrophosphate, 2 mM EDTA, 10 mM KH2PO4, 1% [w/v] polyvinylpyrrolidone 40, 1% [w/v] BSA, 5 mM cysteine and 20 mM ascorbic acid, pH 7.4). After filtration through two layers of Miracloth (Millipore, Burlington, MA, USA), the samples were centrifuged twice for 10 min at 3,000 g and the resulting crude mitochondrial extracts were precipitated via centrifugation of the clear supernatant for 10 min at 20,000 g. Then, the pellet was washed three times in mitochondrial washing buffer (0.3 M sucrose, 1 mM EGTA, and 10 mM MOPS, pH 7.2), and subjected to BN-PAGE gel analysis.

cRT-PCR analysis

The cRT-PCR assay was based on previous studies with some modifications (Slomovic and Schuster, 2013; Hang et al., 2015). RNA of WT and cns1 mutant kernels were extracted according to the procedure described before, and 10 μg total RNA was ligated from the 5′-phosphorylated (Epicentre, Paris, France) termini with T4 RNA ligase (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. After RNA purification (Tiangen, Beijing, China), the first-strand cDNA was generated using SuperScript IV reverse transcriptase (Invitrogen) with gene-specific primers. The primers used for nested PCR amplification reaction are listed in Supplemental Table S5 and the amplified products were separated using 1% agarose gel electrophoresis. Finally, the corresponding bands were cloned into the pEASY-Blunt vector (TransGen) and sequenced.

Prokaryotic expression and protein purification

Full-length CNS1 was constructed into pET32a vector (Invitrogen) to generate the plasmid pET32a-CNS1. For prokaryotic expression, the plasmid pET32a-CNS1 was transformed into host E. coli Rosetta (DE3) grown in TB liquid medium at 37°C to an optical density at 600 nm (OD600) of 0.6–0.8. Then, protein expression was induced by 0.5 mM IPTG at 16°C for 16–18 h. The recombinant CNS1 protein was eluted with elution buffer (20 mM Tris–HCl, 0.5 M NaCl, 10% [v/v] glycerol, pH 8.0 with 250 mM imidazole) on a Ni-NTA Agarose (Qiagen, Hilden, Germany). The resulting protein extract was used for subsequent analysis after concentration.

MST affinity quantification

MST was conducted to quantify molecular interactions. RNA probes were synthesized by the company AUGCT (Beijing, China) and labeled with fluorescent molecules (Cy5). For MST, a constant concentration (20 nM) of target RNA was used. A two-fold dilution series of non-labeled His-CNS1 titrant was prepared in the binding buffer (50 mM Tris–HCl [pH 8.0], 500 mM NaCl, 10 mM MgCl2, 1 mM DTT, 0.005% [v/v] Tween-20). For MST assay, 10 μL fluorescently labeled RNA was mixed with 10 μL serial dilutions of the titrant. MST measurements were performed on the Monolith.NT115 instrument (Nano Temper Technologies, Munich, Germany) at room temperature (40% laser power) with three biological replicates. The KD values were analyzed through fitting of the fraction RNA of bound versus CNS1 concentration relationship using Prism8 software.

Electrophoretic mobility shift assay

Cy5-labeled and unlabeled RNA probes were synthesized in the company AUGCT (Beijing, China). EMSA was performed using a Lightshift Chemiluminescent EMSA Kit (Thermo Fisher). The reactions were then loaded onto 8% 0.5× TBE PAGE. The fluorescence was captured using the Tanon5200 Multi Chemiluminescent Imaging System (Tanon Science and Technology, Shanghai, China).

BN-PAGE and mitochondrial complex I and complex III activity assay

BN-PAGE was performed according to the protocol for the Native PAGE Novex Bis-Tris Gel system (Invitrogen) reported in previous studies (Ren et al., 2019a, 2019b) with some modifications. In brief, the crude mitochondrial extract was suspended in wash buffer, divided into multiple tubes, and centrifuged, resulting in approximately 100 µg total protein content in each tube. Then, the pellet was resuspended in 25 µL solubilization solution (1% [w/v] dodecyl-d-maltoside, 1× Sample Buffer from Bis-Tris Gel system), mixed on ice for 40 min, and centrifuged at 20,000 g and 4°C for 10 min. The supernatant was subjected to BN-PAGE (4%–16%) using 1% Coomassie Brilliant Blue (CBB) G-250. Electrophoresis was performed at low temperature (4°C) at 100 V until the dye reached across one-third of the gel. Then, the dark blue cathode buffer was replaced with light blue buffer according to the user manual (Invitrogen) and the voltage was increased to 250 V for the remainder of the run. The gel was stained with CBB R-250. The in-gel activity assays of mitochondrial complex I (3.06 mM NBT, 0.14 mM NADH, 5 mM Tris–HCl, pH 7.4) and complex III (2.33 mM diaminobenzidine [DAB], 50 mM sodium phosphate, pH 7.2) were performed as described previously (Wittig et al., 2007).

RNA-seq analysis

For nuclear transcriptome analysis, RNA was isolated from immature cns1-ref and WT kernels with the pericarp removed at 15 DAP. Poly (A) RNA isolation for library construction was performed according to a standard protocol. For mitochondrial transcriptome analysis, RNA was extracted from mitochondria isolated from developing cns1-ref and WT kernels (15 DAP). Library construction was performed using the NEBNextUltra RNA Library Prep Kit. Library construction and sequencing were performed by Allwegene Technologies (Beijing, China). DEGs were identified based on fold change >2.0 or <−2.0 with P < 0.005.

Statistical analysis

The data were analyzed using GraphPad version 9.0 software and Student’s t test was employed to identify the significance.

Accession numbers

Sequence data for maize CNS1 can be found in the GenBank database under accession number (Zm00001d002771). The accession numbers of proteins used to construct the phylogenetic tree are shown below: AtMTSF1, At1g06710; AtRPF4, At1g62910; AtRPF3, At1g62930; AtOTP43, At1g74900; AtABO5, At1g51965; AtMEF19, At3g05240; AtCRR2, At3g46790; ZmPPR78, Zm00001d034428; ZmEMP5, Zm00001d042039; ZmPPR2263, Zm00001d045089; ZmEMP9, Zm00001d022480; ZmATP4, Zm00001d028221; ZmCRP1, Zm00001d021716; ZmPPR6, Zm00001d045550; ZmPPR10, Zm00001d036698; ZmPPR4, Zm00001d026654; ZmEMP16, Zm00001d011559. RNA-seq data are available from the National Center for Biotechnology Information Sequence Read Archive under accession number PRJNA800836.

Supplemental data

The following materials are available in the online version of this article.

Supplemental Figure S1. Linkage verification of Mu in cns1-ref kernels.

Supplemental Figure S2. Genotype analysis of kernels isolated from the same selfed ear of cns1-ref/+; CNS1/−.

Supplemental Figure S3. CNS1 expression levels in the WT and two cns1 mutant alleles.

Supplemental Figure S4. Sequence and domain composition of CNS1 in the WT and cns1 mutants.

Supplemental Table S1. PPR protein structure prediction of CNS1.

Supplemental Table S2. Quality control of mitochondrial transcriptome analysis.

Supplemental Table S3. Cytidine (C) to uridine (U) editing efficiency of transcripts encoded by the mitochondrial genome.

Supplemental Table S4. Transcriptome analysis of nuclear genes associated with the mitochondria.

Supplemental Table S5. Primers used in this study.

Supplementary Material

kiac086_Supplementary_Data
Click here for additional data file. (2.2MB, pdf)

Acknowledgments

We acknowledge Prof. Xiangyu Zhao (Shandong Agricultural University), Prof. Rentao Song (China Agricultural University), and Prof. Guoying Wang (Chinese Academy of Agricultural Sciences) for their supports in the BN-PAGE assay. We also acknowledge the Maize Genetics Cooperation Stock Center, Dr. Xiaoduo Lu, and Prof. Chunyi Zhang for distributing the EMS mutant in this work.

Funding

This work was supported by the National Key Research and Development Program of China to X.Z. (2021YFF1000304), the National Special Program for GMO Development of China to R.C. (2016ZX08003-002), and the National Natural Science Foundation of China to X.Z. (grant number 31771707).

Conflict of interest statement. None declared.

 

R.C. and X.Z. designed the experiments. S.M. carried out the experiments. W.Y. and J.Z. helped in microscale thermophoresis assay. X.L., S.L., and Y.L. generated the maize transgenic lines. C.Z. and XD.L. provided the EMS mutant line. R.C., X.Z., and S.M. analyzed the data and wrote the manuscript.

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (https://academic.oup.com/plphys/pages/general-instructions) is: Rumei Chen (chenrumei@caas.cn).

References

  1. Asakura Y, Barkan A (2007) A CRM domain protein functions dually in group I and group II intron splicing in land plant chloroplasts. Plant Cell 19: 3864. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Barkan A, Small I (2014) Pentatricopeptide repeat proteins in plants. Annu Rev Plant Biol 65: 415–442 [DOI] [PubMed] [Google Scholar]
  3. Bihmidine S, Hunter CT, Johns CE, Koch KE, Braun DM (2013) Regulation of assimilate import into sink organs: Update on molecular drivers of sink strength. Front Plant Sci 4: 177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bonnard G, Grienenberger JM (1995) A gene proposed to encode a transmembrane domain of an ABC transporter is expressed in wheat mitochondria. Mol Gen Genet 246: 91–99 [DOI] [PubMed] [Google Scholar]
  5. Chen X, Feng F, Qi W, Xu LM, Yao D, Wang Q, Song R (2017) Dek35 encodes a PPR protein that affects cis-Splicing of mitochondrial nad4 Intron 1 and seed development in maize. Mol Plant 10: 427–441 [DOI] [PubMed] [Google Scholar]
  6. Cheng S, Gutmann B, Zhong X, Ye Y, Fisher MF, Bai F, Castleden I, Song Y, Song B, Huang J (2016) Redefining the structural motifs that determine RNA binding and RNA editing by pentatricopeptide repeat proteins in land plants. Plant J 85: 532–547 [DOI] [PubMed] [Google Scholar]
  7. Clifton SW, Minx P, Fauron CM, Gibson M, Allen JO, Sun H, Thompson M, Barbazuk WB, Kanuganti S, Tayloe C, et al. (2004) Sequence and comparative analysis of the maize NB mitochondrial genome. Plant Physiol 136: 3486–3503 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dai D, Luan S, Chen X, Wang Q, Feng Y, Zhu C, Qi W, Song R (2018) Maize dek37 encodes a P-type PPR Protein that affects cis-splicing of mitochondrial nad2 Intron 1 and seed development. Genetics 208: 1069–1082 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. De Longevialle AF, Small ID, Lurin C (2010) Nuclearly encoded splicing factors implicated in RNA splicing in higher plant organelles. Mol Plant 3: 691–705 [DOI] [PubMed] [Google Scholar]
  10. Doll NM, Depege-Fargeix N, Rogowsky PM, Widiez T (2017) Signaling in early maize kernel development. Mol Plant 10: 375–388 [DOI] [PubMed] [Google Scholar]
  11. Faivrenitschke SE, Nazoa P, Gualberto JM, Grienenberger JM, Bonnard G (2001) Wheat mitochondria ccmB encodes the membrane domain of a putative ABC transporter involved in cytochrome c biogenesis. Biochim Biophys Acta 1519: 199–208 [DOI] [PubMed] [Google Scholar]
  12. Frame BR, Shou H, Chikwamba RK, Zhang Z, Xiang C, Fonger TM, Pegg SEK, Li B, Nettleton DS, Pei D, et al. (2002) Agrobacterium tumefaciens-mediated transformation of maize embryos using a standard binary vector system. Plant Physiol 129: 13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Fujii S, Small I (2011) The evolution of RNA editing and pentatricopeptide repeat genes. New Phytol 191: 37–47 [DOI] [PubMed] [Google Scholar]
  14. Giegé P, Grienenberger JM, Bonnard G (2008) Cytochrome c biogenesis in mitochondria. Mitochondrion 8: 61–73 [DOI] [PubMed] [Google Scholar]
  15. Giegé P, Hoffmann M, Binder S, Brennicke A (2000) RNA degradation buffers asymmetries of transcription in Arabidopsis mitochondria. EMBO Rep 1: 164–170 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gothandam KM, Kim ES, Cho H, Chung YY (2005) OsPPR1, a pentatricopeptide repeat protein of rice is essential for the chloroplast biogenesis. Plant Mol Biol 58: 421–433 [DOI] [PubMed] [Google Scholar]
  17. Gu J, Miles D, Newton KJ (1993) Analysis of leaf sectors in the NCS6 mitochondrial mutant of maize. Plant Cell 5: 963–971 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Haïli N, Arnal N, Quadrado M, Amiar S, Tcherkez G, Dahan J, Briozzo P, des Francs-Small C C, Vrielynck N, Mireau H (2013) The pentatricopeptide repeat MTSF1 protein stabilizes the nad4 mRNA in Arabidopsis mitochondria. Nucleic Acids Res 41: 6650–6663 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hang R, Deng X, Liu C, Mo B, Cao X (2015) Circular RT-PCR assay using Arabidopsis samples. Bio-protocol 5: e1533 [Google Scholar]
  20. Hauler A, Jonietz C, Stoll B, Stoll K, Braun HP, Binder S (2013) RNA PROCESSING FACTOR 5 is required for efficient 5′ cleavage at a processing site conserved in RNAs of three different mitochondrial genes in Arabidopsis thaliana. Plant J 74: 593–604 [DOI] [PubMed] [Google Scholar]
  21. Jiao Y, Peluso P, Shi J, Liang T, Stitzer MC, Wang B, Campbell MS, Stein JC, Wei X, Chin CS, et al. (2017) Improved maize reference genome with single-molecule technologies. Nature 546: 524. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Jonietz C, Forner J, Hildebrandt T, Binder S (2011) RNA PROCESSING FACTOR3 is crucial for the accumulation of mature ccmC transcripts in mitochondria of Arabidopsis accession Columbia. Plant Physiol 157: 1430–1439 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jonietz C, Forner J, Hölzle A, Thuss S, Binder S (2010) RNA PROCESSING FACTOR2 is required for 5′ end processing of nad9 and cox3 mRNAs in mitochondria of Arabidopsis thaliana. Plant Cell 22: 443–453 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Karpenahalli MR, Lupas AN, Söding J (2007) TPRpred: A tool for prediction of TPR-, PPR- and SEL1-like repeats from protein sequences. BMC Bioinform 8: 2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Klein RR, Klein PE, Mullet JE, Minx P, Rooney WL, Schertz KF (2005) Fertility restorer locus Rf1 of sorghum (Sorghum bicolor L.) encodes a pentatricopeptide repeat protein not present in the collinear region of rice chromosome 12. Theor Appl Genet 111: 994–1012 [DOI] [PubMed] [Google Scholar]
  26. Kotera E, Tasaka M, Shikanai T (2005) A pentatricopeptide repeat protein is essential for RNA editing in chloroplasts. Nature 433: 326–330 [DOI] [PubMed] [Google Scholar]
  27. Kranz R, Lill R, Goldman B, Bonnard G, Merchant S (1998) Molecular mechanisms of cytochrome c biogenesis: Three distinct systems. Mol Microbiol 29: 383–396 [DOI] [PubMed] [Google Scholar]
  28. Kroeger T, Zmudjak M, Rahimi N, Small I, Barkan A (2012) A PORR domain protein required for rpl2 and ccmFC intron splicing and for the biogenesis of c-type cytochromes in Arabidopsis mitochondria. Plant J 69: 996–1005 [DOI] [PubMed] [Google Scholar]
  29. Kubo T, Newton KJ (2008) Angiosperm mitochondrial genomes and mutations. Mitochondrion 8: 5–14 [DOI] [PubMed] [Google Scholar]
  30. Li XJ, Gu W, Sun SL, Chen ZL, Chen J, Song W, Zhao H, Lai J, Li X, Gu W (2017) Defective Kernel 39 encodes a PPR protein required for seed development in maize. J Integr Plant Biol 60: 45–64 [DOI] [PubMed] [Google Scholar]
  31. Linke B, Nothnagel T, Börner T (2003) Flower development in carrot CMS plants: mitochondria affect the expression of MADS-box genes homologous to GLOBOSA and DEFICIENS. Plant J 34: 27–37 [DOI] [PubMed] [Google Scholar]
  32. Liu YJ, Xiu ZH, Meeley R, Tan BC (2013) Empty pericarp5 encodes a pentatricopeptide repeat protein that is required for mitochondrial RNA editing and seed development in maize. Plant Cell 25: 868–883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Lu X, Liu J, Ren W, Yang Q, Chai Z, Chen R, Wang L, Zhao J, Lang Z, Wang H, et al. (2018) Gene-indexed mutations in maize. Mol Plant 11: 496–504 [DOI] [PubMed] [Google Scholar]
  34. Lurin C, Andres C, Aubourg S, Bellaoui M, Bitton F, Bruyere C, Caboche M, Debast C, Gualberto J, Hoffmann B, et al. (2004) Genome-wide analysis of Arabidopsis pentatricopeptide repeat proteins reveals their essential role in organelle biogenesis. Plant Cell 16: 2089–2103 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Marienfeld JR, Newton KJ (1994) The maize NCS2 abnormal growth mutant has a chimeric nad4-nad7 mitochondrial gene and is associated with reduced complex I function. Genetics 138: 855–863 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. McCarty DR, Settles AM, Suzuki M, Tan BC, Latshaw S, Porch T, Robin K, Baier J, Avigne W, Lai J, et al. (2005) Steady-state transposon mutagenesis in inbred maize. Plant J 44: 52–61 [DOI] [PubMed] [Google Scholar]
  37. Meyer EH, Tomaz T, Carroll AJ, Estavillo G, Delannoy E, Tanz SK, Small ID, Pogson BJ, Millar AH (2009) Remodeled respiration in ndufs4 with low phosphorylation efficiency suppresses Arabidopsis germination and growth and alters control of metabolism at night. Plant Physiol 151: 603–619 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Nawel HL, Nadège A, Martine Q, Souad A, Guillaume T, Jennifer D, Pierre B, Catherine FS, Nathalie V, Hakim M (2013) The pentatricopeptide repeat MTSF1 protein stabilizes the nad4 mRNA in Arabidopsis mitochondria. Nucleic Acids Res 41: 6650–6663 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Parker JL, Newstead S (2014) Molecular basis of nitrate uptake by the plant nitrate transporter NRT1.1. Nature 507: 68–72 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Peretó J (2015) Mitochondrion. InGargaud M, Irvine WM, Amils R, Cleaves HJ, Pinti DL, Quintanilla JC, Rouan D, Spohn T, Tirard S, Viso M, eds, Encyclopedia of Astrobiology. Springer, Berlin, Heidelberg, p 1592 [Google Scholar]
  41. Pfalz J, Bayraktar OA, Prikryl J, Barkan A (2009) Site-specific binding of a PPR protein defines and stabilizes 5' and 3' mRNA termini in chloroplasts. EMBO J 28: 2042–2052 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Prikryl J, Rojas M, Schuster G, Barkan A (2011) Mechanism of RNA stabilization and translational activation by a pentatricopeptide repeat protein. Proc Natl Acad Sci USA 108: 415–420 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Qi WW, Yang Y, Feng X, Zhang M, Song R (2017) Mitochondrial function and maize kernel development requires Dek2, a pentatricopeptide repeat protein involved in nad1 mRNA splicing. Genetics 205: 239–249 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Qiang Z, Jasper D, Chunyi Z, Mizuki T, Kristina K, Christian C, Jan S, Michael K, Jurgen D, Janny P (2012) SLO2, a mitochondrial pentatricopeptide repeat protein affecting several RNA editing sites, is required for energy metabolism. Plant J 71: 836–849 [DOI] [PubMed] [Google Scholar]
  45. Rayapuram N, Hagenmuller J, Grienenberger JM, Giegé P, Bonnard G (2007) AtCCMA interacts with AtCcmB to form a novel mitochondrial ABC transporter involved in cytochrome c maturation in Arabidopsis. J Biol Chem 282: 21015–21023 [DOI] [PubMed] [Google Scholar]
  46. Ren RC, Wang LL, Zhang L, Zhao YJ, Wu JW, Wei YM, Zhang XS, Zhao XY (2019a) DEK43 is a P-type PPR protein responsible for the cis-splicing of nad4 in maize mitochondria. J Integr Plant Biol 62: 299–313 [DOI] [PubMed] [Google Scholar]
  47. Ren X, Pan Z, Zhao H, Zhao J, Cai M, Li J, Zhang Z, Qiu F (2017) EMPTY PERICARP11 serves as a factor for splicing of mitochondrial nad1 intron and is required to ensure proper seed development in maize. J Exp Bot 68: 4571–4581 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Ren Z, Fan K, Fang T, Zhang J, Yang L, Wang J, Wang G, Liu Y (2019b) Maize empty pericarp602 encodes a P-type PPR protein that is essential for seed development. Plant Cell Physiol 60: 1734–1746 [DOI] [PubMed] [Google Scholar]
  49. Sang-Dong Y, Young-Hee C, Jen S (2007) Arabidopsis mesophyll protoplasts: A versatile cell system for transient gene expression analysis. Nat Protocol 2: 1565–1572 [DOI] [PubMed] [Google Scholar]
  50. Schapira AHV (2006) Mitochondrial disease. Lancet 368: 70–82 [DOI] [PubMed] [Google Scholar]
  51. Seidel SA, Dijkman PM, Lea WA, van den Bogaart G, Jerabek-Willemsen M, Lazic A, Joseph JS, Srinivasan P, Baaske P, Simeonov A, et al. (2013) Microscale thermophoresis quantifies biomolecular interactions under previously challenging conditions. Methods 59: 301–315 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Slomovic S, Schuster G (2013) Chapter 13—circularized RT-PCR (cRT-PCR): Analysis of the 5′ ends, 3′ ends, and poly(A) tails of RNA. InLorsch J, ed, Methods in Enzymology, Vol 530. Academic Press, Cambridge, MA, pp 227–251 [DOI] [PubMed] [Google Scholar]
  53. Small ID, Peeters N (2000) The PPR motif—a TPR-related motif prevalent in plant organellar proteins. Trend Biochem Sci 25: 45–47 [DOI] [PubMed] [Google Scholar]
  54. Sosso D, Mbelo S, Vernoud V, Gendrot G, Dedieu A, Chambrier P, Dauzat M, Heurtevin L, Guyon V, Takenaka M, et al. (2012) PPR2263, a DYW-Subgroup pentatricopeptide repeat protein, is required for mitochondrial nad5 and cob transcript editing, mitochondrion biogenesis, and maize growth. Plant Cell 24: 676–691 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Stoll K, Jonietz C, Schleicher S, Francs-Small CCD, Small I, Binder S (2017) In Arabidopsis thaliana distinct alleles encoding mitochondrial RNA PROCESSING FACTOR 4 support the generation of additional 5' termini of ccmB transcripts. Plant Mol Biol 93: 659–668 [DOI] [PubMed] [Google Scholar]
  56. Sun F, Wang X, Bonnard G, Shen Y, Xiu Z, Li X, Gao D, Zhang Z, Tan BC (2015) Empty pericarp7 encodes a mitochondrial E-subgroup pentatricopeptide repeat protein that is required for ccmFN editing, mitochondrial function and seed development in maize. Plant J 84: 283–295 [DOI] [PubMed] [Google Scholar]
  57. Sun F, Xiu Z, Jiang R, Liu Y, Zhang X, Yang YZ, Li X, Zhang X, Wang Y, Tan BC (2019) The mitochondrial pentatricopeptide repeat protein EMP12 is involved in the splicing of three nad2 introns and seed development in maize. J Exp Bot 70: 963–972 [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Sun F, Zhang X, Shen Y, Wang H, Liu R, Wang X, Gao D, Yang YZ, Liu Y, Tan BC (2018) The pentatricopeptide repeat protein EMPTY PERICARP8 is required for the splicing of three mitochondrial introns and seed development in maize. Plant J 95: 919–932 [DOI] [PubMed] [Google Scholar]
  59. Thöny-Meyer L, Fischer F, Künzler P, Ritz D, Hennecke H (1995) Escherichia coli genes required for cytochrome c maturation. J Bacteriol 177: 4321. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Uchida M, Ohtani S, Ichinose M, Sugita C, Sugita M (2011) The PPR-DYW proteins are required for RNA editing of rps14, cox1 and nad5 transcripts in Physcomitrella patens mitochondria. FEBS Lett 585: 2367–2371 [DOI] [PubMed] [Google Scholar]
  61. Vernoud V, Hajduch M, Khaled AS, Depège N, Rogowsky PM (2005) Maize embryogenesis. Maydica 50: 469–483 [Google Scholar]
  62. Wang C, Aubé F, Planchard N, Quadrado M, Dargel-Graffin C, Nogué F, Mireau H (2017a) The pentatricopeptide repeat protein MTSF2 stabilizes a nad1 precursor transcript and defines the 3′ end of its 5′-half intron. Nucleic Acids Res 45: 6119–6134 [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Wang Y, Ren Y, Zhou K, Liu L, Wang J, Xu Y, Zhang H, Zhang L, Feng Z, Wang L, et al. (2017b) WHITE STRIPE LEAF4 encodes a novel P-Type PPR protein required for chloroplast biogenesis during early leaf development. Front Plant Sci 8: 1116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Wei KF, Han P (2016) Pentatricopeptide repeat proteins in maize. Mol Breed 36: 170 [Google Scholar]
  65. Wienken CJ, Baaske P, Rothbauer U, Braun D, Duhr S (2010) Protein-binding assays in biological liquids using microscale thermophoresis. Nat Commun 1: 100. [DOI] [PubMed] [Google Scholar]
  66. Wittig I, Karas M, Schägger H (2007) High resolution clear native electrophoresis for in-gel functional assays and fluorescence studies of membrane protein complexes. Mol Cell Proteomics 6: 1215. [DOI] [PubMed] [Google Scholar]
  67. Xiu Z, Sun F, Shen Y, Zhang X, Jiang R, Bonnard G, Zhang J, Tan BC (2016) EMPTY PERICARP16 is required for mitochondrial nad2 intron 4 cis-splicing, complex I assembly and seed development in maize. Plant J 85: 507–519 [DOI] [PubMed] [Google Scholar]
  68. Zehrmann A, Verbitskiy D, Härtel B, Brennicke A, Takenaka M (2011) PPR proteins network as site-specific RNA editing factors in plant organelles. RNA Biol 8: 67–70 [DOI] [PubMed] [Google Scholar]
  69. Zoschke R, Watkins KP, Miranda RG, Barkan A (2016) The PPR-SMR protein PPR53 enhances the stability and translation of specific chloroplast RNAs in maize. Plant J 85: 594–606 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

kiac086_Supplementary_Data
Click here for additional data file. (2.2MB, pdf)

Articles from Plant Physiology are provided here courtesy of Oxford University Press

RESOURCES