Conservation Study of Imprinted Genes in Maize Triparental Heterozygotic Kernels

Genomic imprinting is a classic epigenetic phenomenon related to the uniparental expression of genes. Imprinting variability exists in seeds and can contribute to observed parent-of-origin effects on seed development. Here, we conducted allelic expression of the embryo and endosperm from four crosses at 11 days after pollination (DAP). First, the F1 progeny of B73(♀) × Mo17(♂) and the inducer line CAU5 were used as parents to obtain reciprocal crosses of BM-C/C-BM. Additionally, the F1 progeny of Mo17(♀) × B73(♂) and CAU5 were used as parents to obtain reciprocal crosses of MB-C/C-MB. In total, 192 and 181 imprinted genes were identified in the BM-C/C-BM and MB-C/C-MB crosses, respectively. Then, by comparing the allelic expression of these imprinted genes in the reciprocal crosses of B73 and CAU5 (BC/CB), fifty-one Mo17-added non-conserved genes were identified as exhibiting imprinting variability. Fifty-one B73-added non-conserved genes were also identified by comparing the allelic expression of imprinted genes identified in BM-C/C-BM, MB-C/C-MB and MC/CM crosses. Specific Gene Ontology (GO) terms were not enriched in B73-added/Mo17-added non-conserved genes. Interestingly, the imprinting status of these genes was less conserved across other species. The cis-element distribution, tissue expression and subcellular location were similar between the B73-added/Mo17-added conserved and B73-added/Mo17-added non-conserved imprinted genes. Finally, genotypic and phenotypic analysis of one non-conserved gene showed that the mutation and overexpression of this gene may affect embryo and kernel size, which indicates that these non-conserved genes may also play an important role in kernel development. The findings of this study will be helpful for elucidating the imprinting mechanism of genes involved in maize kernel development.


Introduction
Imprinting is an epigenetic phenomenon in which the expression of a subset of genes is dependent on the parent of origin [1]. In 1970, R1 was discovered in maize as a maternally expressed gene (MEG) based on the observation that fully colored kernels were produced when both the R and B alleles were inherited from the maternal parent but not from the paternal parent [2][3][4]. With the wide application of high-throughput sequencing technology, the efficiency of identifying imprinted genes has been greatly improved. Transcriptomic analysis during plant development in hybrids is an effective approach to detecting imprinted genes in flowering plants [5]. Hundreds of imprinted genes have been

Analysis of Gene Expression and Identification of Imprinted Genes in Reciprocal Crosses
To understand the differences in gene expression in the early embryo and endosperm, we performed RNA sequencing (RNA-seq) analysis on the immature embryo and endosperm from reciprocal crosses between the inducer line CAU5 and hybrid lines (BM or Int. J. Mol. Sci. 2022, 23, 15424 3 of 14 MB) at 11 DAP. Two reciprocal crosses were used and were designated as BM-C/C-BM, representing (B73 × Mo17) × CAU5/CAU5 × (B73 × Mo17), and MB-C/C-MB, representing (Mo17 × B73) × CAU5/CAU5 × (Mo17 × B73) ( Figure S1A). Considering that haploids may have been produced in the hybrid crosses (BM-C and MB-C) when the CAU5 inducer line was used as the male parent, we not only selected the diploid embryo of each cross for further analysis but also extracted three replicates from multiple samples each for the diploid embryo and triploid endosperm. By combining correlation analysis data between these samples, we found that the correlation of the three biological replicate samples in each combination of each tissue was greater than 0.97, which indicated that the gene expression data were highly reproducible ( Figure S1B). Illumina sequencing generated an average of 13.84 M paired-end clean reads from each biological replicate (Table S1), which were aligned to the reference genome (see Section 4).
Since three inbred lines were used in the reciprocal cross tests, an SNP locus was selected in our plant materials to distinguish the parental read origin. This locus was not only different between B73 and CAU5 but was also the same in B73 and Mo17 ( Figure 1A). To identify genes that showed significant bias toward the maternal or paternal origin of transcription, we generated a scatter plot to show the relative transcription output from the maternal and paternal alleles of each gene, with more than 20 allelic reads in the embryo and endosperm transcriptome data (Figure 1B-E). As shown in Figure 1F,G, using the threshold of 2:1 in the embryo and endosperm (see methods), we found 35 (23 MEGs and 12 PEGs) and 178 (62 MEGs and 116 PEGs) imprinted genes in the embryo and endosperm from the BM-C/C-BM hybrid combination, respectively, and 31 (23 MEGs and 8 PEGs) and 161 (49 MEGs and 112 PEGs) imprinted genes in the embryo and endosperm from the MB-C/C-MB hybrid combination, respectively (Table S2).
We next examined the chromosomal location of the imprinted genes detected in all the pairs of reciprocal crosses ( Figure 1H) and found that chromosome 1 had the greatest number of imprinted genes (51 genes), while both chromosomes 6 and 8 had the lowest number (17 genes). We also scanned the genome for candidate clusters containing at least two adjacent imprinted transcripts within a region of 1 Mb; two imprinted genes fell into one cluster in the embryo, and two imprinted genes fell into one cluster in the endosperm (Table S3). To further analyze the tissue-specific genes between the embryo and endosperm in the three crosses, we counted the number of imprinted genes detected in each cross. We found that 192 and 181 imprinted genes were detected in BM-C/C-BM and MB-C/C-MB, respectively, and the majority were specifically expressed in these tissues ( Figure 1I,J).

Conservation and GO Analysis of the Non-Conserved Genes
To study whether the imprinting status of some genes changed when another parent was added to the hybridization crosses, we compared the difference in imprinting status among the BC and CB crosses (designations representing B73 × CAU5 and CAU5 × B73), the MC and CM crosses (representing Mo17 × CAU5 and CAU5 × Mo17), the BM-C/C-BM crosses and the MB-C/C-MB crosses, and we denoted them as non-conserved genes.
In this comparison, two groups of genes were used to analyze whether the imprinting status difference changed. The first was the "Mo17-added" group, which represented six sets of genes ( Figure 2A): I: imprinted in BM-C/C-BM crosses but not imprinted in BC/CB crosses; II: imprinted in BC/CB crosses but not imprinted in BM-C/C-BM crosses; III: imprinted in both BM-C/C-BM crosses and BC/CB crosses; IV: imprinted in MB-C/C-MB crosses but not imprinted in BC/CB crosses; V: imprinted in BC/CB crosses but not imprinted in MB-C/C-MB crosses; and VI: imprinted in both MB-C/C-MB crosses and BC/CB crosses. Among the 153 genes in sets I, II, IV and V, 51 genes overlapped between these four sets, and we identified the 51 genes as "Mo17-added non-conserved genes" ( Figure 2B). Then, in the "B73-added" group, genes were also divided into six sets, and 51 of the 163 genes overlapped between sets VII, VIII, X and XI. Similarly, we identified these genes as "B73-added non-conserved genes" ( Figure 2B). We further analyzed the interspecific conservation of Mo17-added non-conserved genes   (A) Classification of the genes belonging to the B73-added group and Mo17-added group. The Mo17-added group was divided into six sets: I, II, III, IV, V and VI. The B73-added group was divided into six sets: VII, VIII, IX, X, XI and XII. (B) Overlap analysis of the Mo17-added non-conserved genes and B73-added non-conserved genes. I, II, IV and V refer to four sets of Mo17-added groups in Figure 2A. VII, VIII, XI and X refer to the four sets of B73-added groups in Figure 2A. (C) Overlap analysis of the Mo17-added conserved genes and B73-added conserved genes. III and VI refer to the two sets of Mo17-added groups in Figure 2A. IX and XII refer to the two sets of B73added groups in Figure 2A

Cis-Element Distribution, Expression Pattern and Subcellular Location of Proteins from the Conserved and Non-Conserved Imprinted Genes
To elucidate the conserved and non-conserved imprinted gene regulatory mechanism in response to abiotic or biotic stress in maize, the genomic sequence of the 1.5 kb upstream promoters of the non-conserved and conserved imprinted genes was used to query the Plant Care database to search for cis-regulatory elements (CREs) (Figures 3A and S2-S4). Thirteen classes of CREs related to hormones (ABRE, CGTCA-motif, TGACGmotif, ERE, TGA-element and TCA-element), light reactions (G-box, GT1-motif, TCT-motif and Box-4) and stress (ARE, W-box and LTR) were detected. The six hormone-related CREs had the largest proportion and were distributed in 51 (51), 48 (44) and 48 (44) nonconserved imprinted genes in the B73-added (Mo17-added) groups ( Figure S5). Each predicted CRE existed in at least 20 non-conserved genes ( Figure S5), and each gene has at least six CRE distributions ( Figure S6, Table S5). We also compared the difference in the (A) Classification of the genes belonging to the B73-added group and Mo17-added group. The Mo17-added group was divided into six sets: I, II, III, IV, V and VI. The B73-added group was divided into six sets: VII, VIII, IX, X, XI and XII. (B) Overlap analysis of the Mo17-added non-conserved genes and B73-added non-conserved genes. I, II, IV and V refer to four sets of Mo17-added groups in Figure 2A. VII, VIII, XI and X refer to the four sets of B73-added groups in Figure 2A. (C) Overlap analysis of the Mo17-added conserved genes and B73-added conserved genes. III and VI refer to the two sets of Mo17-added groups in Figure 2A. IX and XII refer to the two sets of B73-added groups in Figure  In contrast, we also compared the genes whose imprinting status did not change when another parent was added to the hybridization crosses and denoted them as conserved genes. As a result, 106 Mo17-added conserved genes (genes overlapped between sets III and VI) and 109 B73-added conserved genes (genes overlapped between sets IX and XII) were identified. Surprisingly, nearly 90% of the Mo17-added conserved genes were overlapped with B73-added conserved genes ( Figure 2C). Furthermore, these genes were also more conserved within species. For Mo17-added conserved genes, 49% of the genes were conserved within sorghum (28%) and rice (25%). Similarly, for B73-added conserved genes, 45% of the genes were conserved in sorghum and rice, and more genes showed conserved imprinting in sorghum (27%) and rice (19%) (Table S4).
To study the potential function of the above-conserved and non-conserved imprinted genes in maize development, especially in kernel development, GO analysis was performed ( Figure 2D,E). First, we focused on the pathways enriched in the B73-added conserved group and Mo17-added conserved group, and nearly 90% of the enriched pathways in the two groups were identical. Signal transduction and protein modification processes, especially histone modification in the biological process category, were both highly enriched in the two groups. In the molecular function category, kinase activity and signal transducer activity were highly enriched. However, neither category showed significant enrichment in the genes of B73-added or Mo17-added non-conserved groups.

Cis-Element Distribution, Expression Pattern and Subcellular Location of Proteins from the Conserved and Non-Conserved Imprinted Genes
To elucidate the conserved and non-conserved imprinted gene regulatory mechanism in response to abiotic or biotic stress in maize, the genomic sequence of the 1.5 kb upstream promoters of the non-conserved and conserved imprinted genes was used to query the Plant Care database to search for cis-regulatory elements (CREs) ( Figure 3A and Figure S2, Figure  S3 and Figure S4). Thirteen classes of CREs related to hormones (ABRE, CGTCA-motif, TGACG-motif, ERE, TGA-element and TCA-element), light reactions (G-box, GT1-motif, TCT-motif and Box-4) and stress (ARE, W-box and LTR) were detected. The six hormonerelated CREs had the largest proportion and were distributed in 51 (51), 48 (44) and 48 (44) non-conserved imprinted genes in the B73-added (Mo17-added) groups ( Figure S5). Each predicted CRE existed in at least 20 non-conserved genes ( Figure S5), and each gene has at least six CRE distributions ( Figure S6, Table S5). We also compared the difference in the cis-element ratios of the non-conserved and conserved imprinted genes in the B73-added group and the Mo17-added group ( Figure 3B and Figure S7), and we found that they had nearly the same ratio, which may indicate that these genes share a similar regulatory mechanism. Furthermore, we also compared the difference between the non-conserved and conserved imprinted genes in the B73-added group and the Mo17-added group, and both groups showed differences in G-box elements ( Figure S8).
Then, we compared the gene expression levels of the B73-added and Mo17-added conserved and non-conserved genes. More constitutively expressed genes were detected in the conserved group ( Figure 3C and Figure S9). The subcellular locations of proteins encoded by non-conserved and conserved imprinted genes in the B73-added and Mo17-added groups were analyzed on the website for GenScript-PSORT II (https://www.genscript.com/ psort.html?src=leftbar, accessed on 20 August 2022). The 249 B73-added conserved imprinted genes, 251 Mo17-added conserved imprinted genes, 51 B73-added non-conserved imprinted genes and 51 Mo17-added non-conserved imprinted genes were separated into various subcellular locations. In these four types of genes, more than 40% of the genes were located in the nucleus ( Figure 3D,E), which is consistent with the results of previous studies that showed that most imprinted genes were located in the nucleus; moreover, some of them were transcription factors that may play an important role in the kernel development process [41]. In addition, in the B73-added non-conserved genes and Mo17-added non-conserved genes, 20% and 17% of the genes were located in the mitochondria and cytoplasm, respectively. Unlike the above-mentioned non-conserved group of genes, the B73-added and Mo17-added conserved genes were present in all organelles ( Figure 3E).

Extremely Non-Conserved Imprinted Genes Detected in Different Species
To study whether the imprinted status of some non-conserved genes was stable in both the B73-added and Mo17-added groups, we performed an overlap analysis of 102 non-conserved genes in the two groups (51 genes in the B73-added non-conserved group and 51 genes in the Mo17-added non-conserved group). Eleven genes were analyzed further. We first detected the imprinted status of these 11 genes in our results and found that 1 MEG and 1 PEG were imprinted in the embryo, and that 2 MEGs and 7 PEGs were imprinted in the endosperm. Then, we analyzed the interspecific conservation of these 11 genes in sorghum, rice and Arabidopsis ( Figure 4A). Two MEGs and two PEGs were also imprinted in rice (one MEG and two PEG), sorghum (one PEG) and Arabidopsis (one MEG and one PEG).
cis-element ratios of the non-conserved and conserved imprinted genes in the B73-added group and the Mo17-added group (Figures 3B and S7), and we found that they had nearly the same ratio, which may indicate that these genes share a similar regulatory mechanism. Furthermore, we also compared the difference between the non-conserved and conserved imprinted genes in the B73-added group and the Mo17-added group, and both groups showed differences in G-box elements ( Figure S8).    Table S6); Lanes 3 and 4 represent samples using TIR6 as the forward primer and the same reverse primers as Lanes 1 and 2 (Supplemental Table S6).   Table S6); Lanes 3 and 4 represent samples using TIR6 as the forward primer and the same reverse primers as Lanes 1 and 2 (Supplemental Table S6

Phenotype Analysis of the Non-Conserved PEG Zm00001d020769 Transgenic Line
To explore the function of these 11 non-conserved imprinted genes during embryo development in maize, we used gene mutation and overexpression lines for further genotype analysis ( Figure 4B,C). First, we searched for mutants of 11 genes in the Maize Genetics COOP Stock Center, and one of them, Zm00001d020769 (Zm769), had a Mu insertion, UFMu-05983, with the mu1049153 insertion in the first exon ( Figure 4B). The Mu insertion sequences were verified with two sets of PCRs with gene-specific primers and a Mu-specific primer ( Figure 4D, Table S6). Zm769 was a PEG detected in the embryo and highly expressed in the kernel, which encoded a ubiquitin-specific protease ( Figure 4E). Therefore, we focused on comparing the embryo and kernel phenotypes of the mutant lines with the background line to determine whether any abnormal kernel phenotypes occurred during development. The embryos and kernels in mutant lines both showed significant differences compared to those in the wild W22 inbred line (p value < 0.01, Student's test), suggesting an important role for this gene in maize embryo development ( Figure 4F-I). Then, we used transgene technology to create an overexpression line (Zm769-OE) to further analyze the function of Zm769. As shown in Figure 4J-M, at 15 days after self-pollination, the embryo and kernel areas of the Zm769 overexpression lines were significantly larger than those of the transgenic receptor, which indicated that Zm769 may participate in the kernel development process.

Discussion
Cross direction plays an important role in heterotic and kernel development in maize by affecting gene expression in hybrid offspring [11,12,34,40,42,43]. For example, the size of the 301 × 005 embryo was significantly larger than that of the 005 × 301 embryo at 6 DAP [44]. Genomic imprinting refers to the parental-specific expression of a number of genes after fertilization that occurs in the embryo and endosperm. However, the effects of hybridization on genomic imprinting and its influence on kernel development remain elusive. In our results, the imprinted status of 51 genes was changed when introducing a new parent in both the B73-added group and the Mo17-added group. In view of the factors besides environmental impact that affect the imprinting state of genes, different parental backgrounds may be the key factor of the imprinted status difference. As in the crosses BM-C/C-BM and MB-C/C-MB, the B73, Mo17 and CAU5 parents used in the two reciprocal crosses were the same, so the different imprinted statuses of these non-conserved genes may be substantially affected by the hybridization direction of the two parents (BM and MB). Interestingly, the overlap ratio of the imprinted genes detected in BM-C/C-BM and MB-C/C-MB was more than 80%, which not only confirmed the accuracy of our results but also verified the stability of the conserved imprinted genes. Of course, the cross direction seems to have an effect on the allelic expression of imprinted genes. When BM or MB was used as one parent to cross with CAU5, approximately 25% of imprinted genes exhibited differential allelic expression ( Figure 1I-J). Hence, by systematically investigating imprinted genes in four crosses, BC/CB, MC/CM, BM-C/C-BM and MB-C/C-MB, we found that the process of hybridization and cross direction have effects on genomic imprinting. Recently, global epigenetic variations and their potential association with altered gene expression in hybrids have been extensively discussed [45][46][47]. Therefore, in future work, there will be a need to comprehensively determine epigenetic variations during hybridization and their association with imprinting variations.
In addition to the conserved imprinted genes identified previously in plants [8,9,12,34,36], we identified 137 and 135 conserved imprinted genes in our results. GO term analysis revealed that signal transduction, protein modification processes and, especially, histone modification in the biological process category were highly enriched in the identified genes. Additionally, the molecular function category was highly enriched in genes related to kinase activity and signal transducer activity ( Figure 2). Unfortunately, none of the categories was significantly enriched in either the 51 B73-added non-conserved group of genes nor the 51 Mo17-added non-conserved group of genes. To explore whether non-conserved imprinted genes are a byproduct of gametogenesis epigenetic reprogramming, genotypic and phenotypic analyses of one non-conserved PEG were performed. The mutation of this gene affects embryo and kernel size, which indicates that these non-conserved genes may also play an important role in kernel development. Therefore, further investigations of the extent of parental bias related to the function of B73-added/Mo17-added non-conserved imprinted genes should be conducted in the future.

Plant Materials
The hybrid lines B73(♀) × Mo17(♂) and Mo17(♀) × B73(♂) were obtained from the inbred lines B73 and Mo17 in the summer of 2020 at the experimental station of Shenyang Agriculture University in Shenyang, Liaoning. Then, in the summer of 2021, the inducer line CAU5 and these two hybrid lines were used as the parental lines for reciprocal crosses. The ears and tassels of the three lines were bagged with kraft paper one day prior to pollination. The next day, each paper bag was patted to collect pollen from one parent, which was used to pollinate the ear of the other parent. After 11 days, the ears of four reciprocal crosses (BM-C/C-BM, MB-C/C-MB) were collected. In this study, BM-C/C-BM represents the crosses between (B73 × Mo17) × CAU5 and CAU5 × (B73 × Mo17), and MB-C/C-MB represents the reciprocal crosses between (Mo17 × B73) × CAU5 and CAU5 × (Mo17 × B73).

Library Construction for RNA-Seq
The embryo and endosperm samples were isolated using a Quick RNA Isolation Kit (Huayueyang Biotechnology Company, Beijing, China). mRNA library construction and sequencing were conducted following the Illumina manufacturer instructions. Total RNA was extracted as input material for the RNA sample preparations. The NEB Next ® Ultra TM RNA Library Prep Kit from Illumina ® (New England Biolabs (NEB), Ipswich, MA, USA) was used to generate mRNA libraries. High-throughput mRNA sequencing was performed using the Illumina NovaSeq 6000 platform, and 150 bp paired-end reads were generated for each library. An average of 4.2 Gb of data for each replicate was obtained and used for the following analyses, providing sufficient sequencing depth for the imprinting analysis.

Read Mapping, Gene Expression Analysis and SNP Calling
First, clean reads were mapped to the B73 reference genome (Version 4) using HISAT2 software with default parameters [48]. Cufflinks software (V2.2.1) was used to estimate the normalized gene expression values (FPKM) [49]. The calculated log 2 (FPKM + 1) values were used to analyze the correlation coefficient between replicates. Hierarchical clustering analysis was performed on the relative expression value by setting the parameters average linkage and Euclidean distance using MeV (http://www.tm4.org/mev.html, accessed on 20 July 2021). For a gene in a special sample, the relative expression value was the FPKM normalized by the maximum FPKM value of the gene over all samples. Based on the clustering results, MEGs (or PEGs) primarily expressed in endosperm as a subgroup were obtained, and the rest of the MEGs (or PEGs) were assigned to the constitutively expressed subgroup.
Resequencing reads of B73, Mo17 and CAU5 inbred lines were downloaded from NCBI (SRR12415217, SRR12415218 and SRR3124079). Reads were mapped using BWA with default parameters [50]. Samtools were used to exclude reads that were not uniquely mapped with the -q 20 parameter [51]. SNPs between B73, Mo17 and CAU5 inbred lines were called using Bcftools with default parameters [51]. Finally, we identified 508,700 SNPs covering 11,959 genes in the BM-C/C-BM and MB-C/C-MB crosses that could be used to distinguish parental alleles.

Measuring Allelic Expression and Identification of Imprinted Genes
To avoid bias, SNP sites were converted to CAU5 nucleotides to obtain the SNPsubstituted CAU5 genome. All clean reads from three biological replicates of each sample were mapped to the B73 (Version 4) and SNP-substituted CAU5 genomes using HISAT2 with default parameters [48]. Samtools were used to exclude reads that were not uniquely mapped with the -q 20 parameter [51]. Three replicates from each sample were merged for the further identification of imprinted genes. According to the SNP information, the reads aligned at the SNP site were split into maternal or paternal alleles using Samtools mpileup. The maternal and paternal read counts of each gene were summed. If the summed read counts of annotated genes at all SNP sites were ≥20, the imprinting status of the gene could be analyzed. The ratio of maternal to paternal alleles of the analyzed genes was determined using the χ 2 test to detect the deviation of the maternal: paternal ratio from the theoretically suggested 1:1 ratio in the embryo and 2:1 ratio in the endosperm. Finally, MEGs/PEGs were identified in the embryo if significant allelic bias (p value < 0.05; χ 2 test) was detected and if >66% of the transcripts were derived from the maternal or paternal allele. In the endosperm, imprinted genes were identified as having significant allelic bias (p value < 0.05; χ 2 test) if >80% of reads were from the maternal allele for MEGs or >50% of the reads were from the paternal allele for PEGs.

GO Term Enrichment and Functional Category Analysis
GO analysis of the imprinted genes was performed using Agri GO v2.0 [52]. Only GO terms among cell components, molecular functions and biological processes with significant (p value < 0.05) enrichment compared to all genes are displayed.

Validation of Imprinted Genes
We randomly tested the status of four SNPs in four imprinted genes detected in our study using a PCR sequencing method (Zm00001d024959, Zm00001d020769, Zm00001d030305 and Zm00001d032148). Each gene fragment was amplified by different primers in eight 11-DAP embryo or endosperm cDNA samples: BB (self-cross of B73), MM (self-cross of Mo17), CC (self-cross of CAU5), BC and CB, MC and CM, BM-C and C-BM and MB-C and C-MB ( Figure S10). The primer information is listed in Table S6.

Genetic Transformation of Maize
We prepared overexpression constructs for the genetic transformation of the nonconserved gene Zm00001d020769 (Zm769). The full-length CDS (without the stop codon) of Zm769 was amplified from Zm769 cDNA and cloned into the binary vector pBCXUN-MYC to generate the pOE Zm769-MYC construct driven by the ubiquitin promoter. Transformations using the overexpression construct were introduced into the maize receptor line KN5585 via Agrobacterium-mediated transformation [53]. Independent positive transgenic lines were obtained and self-pollinated to generate homozygous progenies for kernel phenotype analysis.

Primers
All primers used in this study are listed in Table S6.

Conclusions
Imprinting variability exists in maize kernels and can contribute to observed parentof-origin effects on kernel development. In this study, among the 192 and 181 imprinted genes identified in the BM-C/C-BM and MB-C/C-MB crosses, respectively, 51 genes were non-conserved, and specific GO terms were not enriched in these genes. Moreover, these genes were less conserved across other species. In addition, 11 high-confidence examples of allele-specific imprinting were discovered. Gene phenotypic analysis of one non-conserved PEG verified that the mutation of this gene made the embryo and kernel smaller than those of the wild type, and the overexpression of this gene enlarged the embryo and kernel area. Therefore, we concluded that non-conserved genes may also play an important role in kernel development, warranting more detailed functional analysis and further investigation of the mechanism of the non-conserved imprinted genes.