A serine/threonine protein kinase encoding gene KERNEL NUMBER PER ROW6 regulates maize grain yield

Abstract

Increasing grain yield of maize (Zea mays L.) is required to meet the rapidly expanding demands for maize-derived food, feed, and fuel. Breeders have enhanced grain productivity of maize hybrids by pyramiding desirable characteristics for larger ears. However, loci selected for improving grain productivity remain largely unclear. Here, we show that a serine/threonine protein kinase encoding gene KERNEL NUMBER PER ROW6 (KNR6) determines pistillate floret number and ear length. Overexpression of KNR6 or introgression of alleles lacking the insertions of two transposable elements in the regulatory region of KNR6 can significantly enhance grain yield. Further in vitro evidences indicate that KNR6 can interact with an Arf GTPase-activating protein (AGAP) and its phosphorylation by KNR6 may affect ear length and kernel number. This finding provides knowledge basis to enhance maize hybrids grain yield.

Fig. 1. Phenotype of two QTL parental lines and map-based cloning of qKNR6.

a Ear inflorescence meristems of the two parents. Scale bar = 500 μm. b–d Comparisons of floret number (b), EL (c), and KNR (d) between NIL^qknr6 and NIL^qKNR6 grown in the field at Sanya, China in 2015. The values in (b–d) are shown as the means ± s.d., and significance was estimated by the one-way ANOVA. The numeral on the bottom of each column is the number of ears examined. e Fine mapping of qKNR6. qKNR6 was located on chromosome 6, bin02. The refined 110_kb region at the qKNR6 locus contained two genes, Zm00001d036601 and Zm00001d036602. f Gene structure of two candidate genes and polymorphisms between the two parents. No polymorphism is identified in Zm00001d036601, but three SNPs in exons and a Harbinger-like TE (triangle) in the 5′-UTR intron of Zm00001d036602 were detected. g A structural diagram of the TE inserted in KNR6, predicted using CENSOR. h, i Expression of Zm00001d036601 (h) and Zm00001d036602 (i) in the Ims of ten recombinant lines and two parental lines. Gene-expression level is analyzed using quantitative PCR with three biological replicates, each with three technical replicates. The maize Actin gene (Zm00001d010159) is used as an internal control. The values in (h and i) are shown as the means ± s.d., and p value is estimated by the Duncan’s test. Source data underlying Fig. 1b–d are provided in a Source Data file.

Fig. 2. Function of KNR6 on kernel number per row and ear length.

a KNR6 RNAi transgenic lines (RNAi-1 and RNAi-2) have shorter ears than do non-transgenic sib lines (NTs). b, f KNR6 expression in two RNAi lines (b) and two KNR6-overexpressing lines (f). Gene expression level is analyzed using quantitative PCR with three biological replicates, and each with three technical replicates. The maize Actin gene (Zm00001d010159) is used as an internal control. p Value is estimated using the two-tailed Student’s t test. c, d Phenotypes of KNR (c) and EL (d) in two RNAi lines. e KNR6-overexpressing lines (OE3 and OE4) have longer ears than do non-transgenic sib lines (NTs). g, h Phenotypes of KNR (g) and EL (h) in two KNR6-overexpressing lines. The values in (c, d, g, h) are means ± s.d. Number on the bottom of each column is the sample size. Significance is estimated by the one-way ANOVA. Scale bar = 3 cm in (a) and (e). Source data underlying Fig. 2c, d are provided in a Source Data file.

Fig. 3. Association mapping and DNA methylation assay.

a Associations of 433 DNA polymorphisms at the KNR6 locus with KNR in 224 diverse maize inbred lines. Each dot represents a polymorphic site. Black dots, polymorphic sites within 10_kb flanking sequences of KNR6; purple dot, the Harbinger-like TE PAV; green dot, the LTR-TE PAV; gray dots, others. Black rectangles, exons; white rectangles, UTRs. The significance level of the associated site, p = 0.05/the number of polymorphic sites. b The pattern of linkage disequilibrium for polymorphic sites. Asterisks, two PAV sites in the LD block. c Pearson correlation between KNR6 expression levels and KNR. The p value is determined by the two-tailed Student’s t test. d Expression levels of KNR6 in 20 Hap1 lines and 85 Hap2 lines using qPCR, with three biological replicates in each case. e, f Box-and-whisker plots of KNR (e) and EL (f) for 48 Hap1 lines and 176 Hap2 lines. Each box represents the median and interquartile range. Whiskers extend to maximum and minimum values. Significance of difference is estimated by the one-way ANOVA. g A diagram of the constructs. TE⁺ construct includes a 5570_bp fragment containing the Harbinger-like TE; TE⁻ construct represents the 567_bp sequence lacking the TE. h Luciferase activity in maize leaf protoplasts. Luciferase activity is measured with three biological replicates, each with two technical replicates. Data are normalized with respect to the average value of the empty construct, and are shown as means ± s.d. p Value is estimated by the one-way ANOVA. i Visualization of the DNA methylation regions. DNA methylation levels were measured by bisulfite sequencing, and the methylation status was illustrated using the Integrated Genome Viewer. The upper and lower panels show the sense and antisense strands, respectively. j DNA methylation level of the three contexts in the long terminal repeat (LTR) retrotransposon and the Harbinger-like TE. k Methylation patterns of the flanking regions of the Harbinger-like TE in both parent lines. The methylation levels of CG, CHG, and CHH were scanned using a window size of 200 bp and a step size of 100 bp. Source data underlying Fig. 3e–h are provided in a Source Data file.

Fig. 4. KNR6-interacting proteins and phenotypes of AGAP-knockout lines.

a mRNA in situ hybridization with antisense probes of KNR6 (left) and AGAP (right). Arrows point out the inflorescence meristem (IM), spikelet- paired meristem (SPM), and spikelet meristem (SM), respectively. b, c Confirmation of the KNR6-AGAP interaction by firefly luciferase complementation imaging assay (b) and pull-down assay (c). Recombinant KNR6-His protein was incubated with GST or GST-tagged AGAP, and then was bound to His resin, respectively. The eluates were resolved by SDS-PAGE and blotted using anti-GST and anti-His antibodies. MW, molecular weight. d, e Protein kinase activity of KNR6 and mutated proteins. KNR6 phosphorylated itself and myelin basic protein (MBP) (d) and AGAP (e) in vitro. f The kinase activity of mutated KNR6 proteins. The phosphorylation reaction was carried out using [³²P] γ-ATP, and the phosphorylated proteins were detected by autoradiography. CBB Coomassie brilliant blue, K₇₄R lysine 74 is substituted by arginine, D₁₇₂A asparagine 172 is substituted by alanine, S₁₇₆A serine 176 is substituted by alanine, and L₇₀V leucine 70 is substituted by valine. g, h Ear inflorescences (g) and ears (h) of the AGAP-knockout line and its non-transgenic sib line. i, j The length of the inflorescence meristem (i) and kernel number per row (j) are reduced in the AGAP-knockout line compared with its non-transgenic sib line (NT). The numeral on the bottom of each column is the number of inflorescences in (i), and the number of ears examined in (j). k A model to illustrate the regulatory pathways of KNR6 in the ear inflorescence. KNR6 may function in auxin-dependent inflorescence development by mediating AGAP phosphorylation. A plus sign (+) means that a given substrate has been added; a minus sign (−) indicates that a given substrate has not been added. Scale bars = 100 μm in (a), 1 cm in (b), 200 μm in (g), and 5 cm in (h). Source data underlying Fig. 4c–f, i, j are provided in a Source Data file.

Fig. 5. Genetic effect of KNR6 Haplotype2 on maize grain yield.

a Ears of inbred line Chang7-2 and its improved line Chang7-2^qKNR6. b Ears of the inbred line Zheng58 and its improved line Zheng58^qKNR6. c Ears of hybrids Zheng58 × Chang7-2, Zheng58 × Chang7-2^qKNR6, Zheng58^qKNR6 × Chang7-2, and Zheng58^qKNR6 × Chang7-2^qKNR6. d Kernel number per row in the original and improved lines. Both Zheng58 and Chang7-2 were improved by introducing the qKNR6 desirable allele using marker-assisted backcrossing. The values are means ± s.d.; p value is estimated by the one-way ANOVA. e–g Phenotypic evaluation of KNR (e), EL (f), and grain yield (g) in hybrids carrying one or two desirable alleles. Phenotypic evaluations of the original hybrid Zheng58/Chang7-2 and improved hybrid Zheng58^qKNR6/Chang7-2^qKNR6, were performed at Wuhan (30.60°N, 114.30°E) and Zhengzhou (ZZ, 34.75°N, 113.62°E) in 2017 spring with randomized block design, respectively. Different letters in (e, f, g) at the top of each column indicate a significant difference at p < 0.05 determined by the Tukey HSD test. n is the number of ears examined in (d) and the number of blocks in (e, f, g). C7-2, Chang7-2; iC7-2, improved line Chang7-2^qKNR6; ZH58, Zheng58; iZH58, improved line Zheng58^qKNR6. Scale bars = 2 cm in (a–c). Source data underlying Fig. 5d–g are provided in a Source Data file.