Dear Editor,

Maize (Zea mays L.) is one of the most important crops, providing calories and biofuel for humans. Over the past decades, the dramatic increase in corn yield has benefited from increased planting densities (Mansfield and Mumm 2014). Short stature in maize enhances grain yield by increasing lodging resistance and planting density. Although dwarf and semi-dwarf breeding has been successful in rice and wheat, it has not been effectively applied to maize, due to the yield compromise caused by dwarf plants. Although many genes regulating plant height have been identified in maize, the number of cloned genes is still limited compared to the complexity of plant height regulation (Wang et al. 2023). Therefore, identifying new plant height regulators and uncovering their underlying molecular mechanisms could facilitate the breeding of high-density-tolerant and semi-dwarf maize varieties.

Homeobox proteins can be divided into two superclasses: the homeodomain leucine zipper (HD-Zip) superclass and the three-amino-acid loop extension (TALE) superclass (Mukherjee et al. 2009). The Knotted1-like homeobox (KNOX) genes, which encode homeodomain-containing transcription factors, belong to the TALE superclass and act as master regulators in plant development (Hay and Tsiantis 2010). KN1, the first cloned member of the KNOX family, functions in plant meristems (Hake et al. 1995). It is also involved in Gibberellin acid (GA) signaling, directly regulating GA2ox1 (Bolduc and Hake 2009). The role of KNOX genes in plant development has been exhaustively reviewed (Hake et al. 2004; Hay and Tsiantis 2010).

Calcium-dependent protein kinases (CPKs) are involved in plant growth regulation, signal transduction, and responses to both abiotic and biotic stresses (Boudsocq and Sheen 2013; Schulz et al. 2013). A notable example is AtCPK28, which has been shown to regulate stem elongation and vascular development in Arabidopsis (Matschi et al. 2013). Remarkably, the function of AtCPK28 in development is strictly growth phase-dependent and is associated with tissue-specific alterations in Jasmonic acid (JA) levels (Matschi et al. 2015). However, the role of CPKs in regulating plant development in maize remains largely unexplored.

In this study, we conducted a phylogenic analysis of calcium-dependent protein kinases (CPKs) in maize, together with AtCPK28 in Arabidopsis. Among the 41 CPKs identified in the maize genome, ZmCPK39 showed the highest homology with AtCPK28 (Fig. S1). To test whether ZmCPK39 is involved in regulating plant height, we engineered three ZmCPK39-knockout mutants (ZmCPK39-KO) using CRISPR/Cas9 technology. As expected, these mutants exhibited significantly reduced plant and ear heights compared to the WT line ND101 (Fig. 1A–C). Compared to ND101, the internodes of these mutants, particularly the top four to eight internodes, were severely shortened (Fig. 1D). Additionally, two of the three mutants had a reduced number of nodes (Fig. 1E). In summary, the knockout of ZmCPK39 causes a severe growth retardation in stem elongation.

Fig. 1
figure 1

Interaction of ZmCPK39 with ZmKnox2 and their collective regulation of plant height in maize. A Gross morphologies of ND101 and ZmCPK39-KO lines. Bar, 19 cm. B, C Statistical analysis of plant height (B) and ear height (C). D Measurement of internode length. E Statistical analysis of internode number. F interaction of ZmCPK39 with ZmKnox2 in yeast cells. G, H Split-Luc (G) and BiFC (H) assays in N. benthamiana leaves to verify the interaction between ZmCPK39 and ZmKnox2. Bar, 20 μm. I Gene structure of ZmKnox2 and the Mutator insertion site in the UTR. J PCR assay to verify the Mutator insertion. TIR, P1 and P2 are primers used to detect the Mutator insertion. TIR is Mutator specific primer, and the location of P1 and P2 are present in I. K ZmKnox2-MU showed reduced plant height. L ZmKnxo2-MU exhibited ear height comparable to WT

Full size image

To elucidate the components involved in ZmCPK39-regulated plant growth, we screened for ZmCPK39-interacting proteins using a yeast two-hybrid (Y2H) assay. This screening identified a knotted-related homeobox protein, named ZmKnox2. To further verify the interaction between ZmCPK39 and ZmKnox2, we performed targeted Y2H assays in yeast. The results confirmed that ZmCPK39 indeed interacts with ZmKnox2 (Fig. 1F). Additionally, we divided ZmKnox2 into two segments: the N-terminus1−170 containing two KNOX domains, and the C-terminus171−293 comprising ELK and HOX domains. The targeted Y2H assays clearly demonstrated that ZmCPK39 interacts exclusively with the N-terminus1−170 of ZmKnox2 (Fig. 1F). Furthermore, the interaction between ZmCPK39 and ZmKnox2 was confirmed by split-luciferase complementation (SLC) and bimolecular fluorescence complementation (BiFC) assays (Fig. 1G, H). Additionally, this interaction between ZmCPK39 and ZmKnox2 was observed at the plasma membrane (Fig. 1H). These results demonstrated that ZmCPK39 interacts with ZmKnox2, in vivo.

To illustrate the role of ZmKnox2 in regulating maize plant height, we acquired the Mutator (Mu) transposon-inserted line, MU2006, which is in the background of the maize inbred line W22, from MaizeGDB (https://www.maizegdb.org/). The presence of Mu insertion in the 5’-untranslated region (UTR) of ZmKnox2 was confirmed by polymerase chain reaction (PCR) (Fig. 1I, J). To eliminate the effects of other Mu insertions or genomic variations, as well as genetic background differences between the wild-type W22 and MU2006, we crossed these lines to generate an F2 segregation population. In this F2 population, individuals homozygous for the Mu insertion (MU) exhibited significantly reduced plant height compared to their siblings heterozygous for the Mu insertion and to the plants without the Mu insertion (Fig. 1K). Furthermore, plants heterozygous for the Mu insertion were selfed to produce F3 families, in which plants homozygous for the Mu insertion at ZmKnox2 also showed a marked reduction in plant height (Fig. 1K). Additionally, we produced a pair of near-isogenic lines (NIL) with or without homozygous Mu insertion. Lines with Mu showed lower plant height than lines without Mu insertion (Fig. 1K). However, plants with Mutator-induced ZmKnox2 mutations exhibited ear heights comparable to those of the WT plants (Fig. 1L). These observations supported the notion that ZmKnox2 is involved in the regulation of plant height in maize.

We conducted a transcriptomic analysis to illustrate the potential molecular mechanism underlying the regulation of plant height by the ZmCPK39-ZmKnox2 module in maize. In total, we identified 5937 down-regulated and 6792 up-regulated differentially expressed genes (DEGs) in ZmCPK39-KO mutants compared to the ND101 line (Fig. 2A). When comparing ZmKnox2-MU to WT plants, 3721 DEGs were identified, with 1882 being down-regulated and 1839 up-regulated (Fig. 2A). Surprisingly, a majority of the KNOX genes (8 out of 10) were down-regulated in ZmCPK39-KO compared to ND101 (Fig. 2B). However, in ZmKnox2-MU mutants, only two genes, ZmKnox2 and ZmKnox10-2, were down-regulated compared to WT (Fig. 2B). In contrast, two other genes, KN1 and ZmKnox6, were up-regulated in ZmKnox2-MU (Fig. 2B). These findings might explain why ZmCPK39-KO mutants exhibited a 40% reduction in plant height, whereas ZmKnox2-MU mutants showed only a 10% reduction.

Fig. 2
figure 2

Transcriptome analysis revealed the potential mechanism underlying ZmCPK39- regulated plant height in maize. A Venn diagram shows DEGs in the transcriptome analysis. B Expression levels of the KNOTTED family members. C KEGG analysis for the identified DEGs. D DEGs involved in plant hormone signal transduction. E Auxin signaling-related DEGs between ZmCPK39-KO and ND101. F Expression of auxin transport genes. G Root length of ND101, ZmCPK39-KO, WT, and ZmKnox2-MU seedlings. Bar, 1 cm. H, I Contents of IAA (H) and ICAld (I) in ND101, ZmCPK39-KO, WT and ZmKnox2-MU. Statistical significance is indicated by asterisks: *P < 0.05, **P < 0.01, determined by t-test

Full size image

Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis revealed that the down-regulated genes in the ZmCPK39-KO mutants were mainly categorized into “plant hormone signal transduction”, “photosynthesis”, “phenylpropanoid biosynthesis”, and “carbon metabolism”; while up-regulated genes were involved in “alcoholism”, “fatty acid metabolism”, “DNA replication”, and “cell cycle” (Fig. 2C). Comparative analysis indicated that genes involved in “photosynthesis” and “carbon metabolism” were consistently down-regulated in both ZmCPK39-KO and ZmKnox2-MU mutants (Fig. 2C). Hence, we concluded that the ZmCPK39-ZmKnox2 module is likely regulating plant growth, via photosynthesis and carbon fixation/metabolism pathways.

We observed that DEGs related to plant hormones were largely enriched in ZmCPK39-KO, but not in ZmKnox2-MU mutants (Fig. 2C). Given the crucial role of plant hormones in controlling plant height, we focused on DEGs involved in the plant hormone signal transduction pathway. Among the 84 down-regulated plant hormone-related genes in ZmCPK39-KO, 32 were involved in auxin signaling, 13 in abscisic acid (ABA) signaling, 12 in cytokinin (CK) signaling, and 12 in salicylic acid (SA) signaling. Additionally, eight DEGs were related to jasmonic acid (JA), four to ethylene (ETH) signaling, and three to brassinolide (BR) signaling (Fig. 2D). These results suggest that the knockout of ZmCPK39 disrupts multiple phytohormone pathways, particularly affecting signaling pathways for auxin, ABA, CK, and SA.

Among the 32 auxin signaling-related down-regulated DEGs in ZmCPK39-KO mutants, the majority were auxin/indoleacetic acid (Aux/IAAs) and small auxin-up RNA (SAURs) genes (Fig. 2E), known as early auxin-responsive genes (Abel and Theologis 1996). This indicates that ZmCPK39 plays a positive role in regulating auxin-responsive signaling. Notably, auxin transport genes, such as PIN1, PIN5, and PIN9, were down-regulated in ZmCPK39-KO compared to ND101, while no significant difference was observed between ZmKnox2-MU and WT (Fig. 2F). Correspondingly, root elongation was significantly inhibited in ZmCPK39-KO compared to ND101, but there was no noticeable difference between ZmKnox2-MU and WT (Fig. 2G). To further explore the role of ZmCPK39 in auxin signaling, we measured the auxin levels in both ZmCPK39-KO and ND101 lines, as well as in ZmKnox2-MU and WT plants. The ZmCPK39-KO line exhibited slightly lower, albeit not significantly different, indole-3-acetic acid (IAA) levels compared to ND101 (Fig. 2H). Additionally, the IAA-derived product indole-3-carboxaldehyde (ICAld) was significantly higher in ZmCPK39-KO than in ND101 (Fig. 2I). On the contrary, no difference in IAA and ICAld contents was detected between ZmKnox2-MU mutant and WT plants (Fig. 2H, I). Taken together, the ZmCPK39-KO mutant, but not the ZmKnox2-MU mutant, demonstrates disturbances in multiple phytohormone pathways, particularly in auxin signaling.

It is noteworthy that while ZmCPK39-KO showed a dramatic reduction in plant height, ZmKnox2-MU exhibited only a slight decrease. This disparity suggests that additional signaling pathways, such as auxin signaling, might also play a role in the regulation of plant height by ZmCPK39. Another possible explanation is the functional redundancy of other KNOXs, such as KN1, which may compensate for the ZmKnox2-MU. Additionally, the Mutator inserts into the UTR of ZmKnox2, rather than a complete knockout, could account for the observed effects. Considering the widespread use of hybrids in maize production, evaluating the breeding value of ZmCPK39 and ZmKnox2 in hybrids across various plant densities could be beneficial. Overall, our findings highlight ZmCPK39 and ZmKnox2 as two potential targets for the breeding of dwarf or semi-dwarf varieties using genome editing technology.

Materials and methods

Plant materials

The transgenic maize lines were generated using ND101 as the recipient line by the Center for Crop Functional Genomic and Molecular Breeding at China Agricultural University. A Mutator transposon-inserted mutant in the W22 genetic background was obtained from MaizeGDB (https://www.maizegdb.org/). This mutant was then crossed with the wild-type (WT) W22 to generate segregated populations.

Total RNA extraction and transcription assay

Total RNA was extracted using the EasyPure Plant RNA kit (TransGen, ER301), and first-strand cDNA was synthesized using the cDNA synthesis kit (TransGen, AE311), following the manufacturer’s protocols. Quantitative Real-time PCR (qRT-PCR) was performed using the TB Green Fast qPCR mix (TaKaRa). The transcript level of ZmGAPDH was used as an internal control.

Yeast two-hybrid assay

To verify the interaction between ZmCPK39 and ZmKnox2, the full-length cDNAs of ZmCPK39 and ZmKnox2 were cloned into the pGBKT7 and pGADT7 vectors, serving as bait and prey, respectively. The yeast strain Y2HGold was used for this assay. Both the bait and prey vectors were co-transformed into the Y2HGold strain, and the protein interactions were subsequently tested on a selective medium.

Split luciferase complementation (SLC) assay

The full-length cDNAs of ZmCPK39 and ZmKnox2 were cloned into the JW771-35S-nLuc and JW772-35S-cLuc vectors, respectively, to generate the ZmCPK39-nLuc and cLuc-ZmKnox2 fusion constructs. The SLC assay was performed as previously described (Zhong et al. 2024).