**1. Introduction**

Bread wheat (*Triticum aestivum*) is the most important cereal crop in the temperate climate and provides a staple food for more than a third of the world's population [1]. It belongs to the Triticeae tribe, which also includes barley, rye, and triticale. Among the species, wheat has the largest and most complex hexaploid genome (2n = 6x = 42), which consists of the three homoeologous subgenomes A, B, and D. Each gene present as homologues A, B, and D could retain its original function or, as a result of independent evolution, develop heterogeneous expression, and/or one or two copies may be silenced or deleted [2,3].

Cytokinins (CKs) perform a basic role in the growth, development, and productivity of any plant species, including wheat [4]. Their content in developing spikes of wheat is correlated with grain yield, grain number and weight, TGW, chlorophyll content in

**Citation:** Szala, K.; Dmochowska-Boguta, M.; Bocian, J.; Orczyk, W.; Nadolska-Orczyk, A. Transgenerational Paternal Inheritance of *TaCKX* GFMs Expression Patterns Indicate a Way to Select Wheat Lines with Better Parameters for Yield-Related Traits. *Int. J. Mol. Sci.* **2023**, *24*, 8196. https://doi.org/10.3390/ijms24098196

Academic Editors: Jian Zhang and Zhiyong Li

Received: 30 March 2023 Revised: 18 April 2023 Accepted: 24 April 2023 Published: 3 May 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

flag leaves, and seedling root weight [5–9]. There are two types of cytokinins, isoprenoid and aromatic. The most important and widely occurring are isoprenoids, *cis*-zeatin (cZ), *trans*-zeatin (tZ), isopentenyl adenine (iP), and dihydrozeatin (DZ), and fewer aromatic forms, e.g., benzylaminopurine (BA). The content of active forms in plant tissues and organs depends on metabolic processes, such as biosynthesis, degradation, inactivation, and reactivation. Active forms and their ribosides can also be transported throughout the plant [10]. Enzymes for all metabolic processes are encoded by members of the gene family (GFMs) [11–15]. One of the most important processes is the irreversible degradation of cytokinins by *TaCKX* GFMs. There are 13 basic *TaCKX* GFMs, 11 of which have homoeologs in subgenomes A, B, and D. Two of them, *TaCKX2.2.2* and *TaCKX2.2.3*, are located only in the D subgenome [11]. The *CKX* GFMs encode the cytokinin oxidase/dehydrogenase enzyme. Their role in the regulation of yield-related traits in wheat [5–7,11] and other species were already shown [4,16–18]. Decreased expression of the *TaCKX1* or *TaCKX2* genes significantly influenced TGW, seed number, and chlorophyl content in flag leaves, and the effect was dependent on the silent gene and/or genotype [5–7]. The *TaCKX4* copy number affected grain yield and chlorophyl content in flag leaves [19]. Haplotype variants of *TaCKX6-D1* (actually *TaCKX2.2.1-3D*) were associated with TGW [20], and the allelic variant of *TaCKX6a02* (annotated as *TACKX2.1*) influenced grain size, filling rate, and weight [21]. In addition to yield, some *CKX* GFMs influence other pleiotropic traits, including root growth, nutrient accumulation, and abiotic stress responses [18,22,23]. These genes could also be regulated at the transcriptional level by transcription factors (TFs), especially those belonging to the NAC family; however, knowledge about their function is still very limited [24]. A promising NAC-encoding candidate with a role in yield-related traits is *TaNAC2-5A* [25,26]. As reported, overexpression of the gene delayed leaf senescence and increased nitrate uptake and concentration, root growth, and grain yield under field conditions. It is interesting to note that in a controlled environment, *TaNAC2-5A* was negatively correlated with the activity of the CKX enzyme in seedling roots and the number of tillers [8].

Crossbreeding and selection are basic steps in crop improvement, and the only potential limitation is too narrow genetic variability. Therefore, it is important to know how yield-related traits are inherited. Moreover, traits, including stably integrated transgenes and edited genes, are inherited according to Mendelian rules [27–30]. An exception to Mendel's principles that encompass both groups of genes is epigenetic inheritance [31] or the polygenic nature of genomic architecture for the linked traits, which can be regulated by transcription factors [32–34] or other gene regulatory networks [35].

The pattern of gene expression could be considered a trait with its own type of inheritance.

This concept, reviewed by Yoo et al. [2], called parental expression additivity, is defined as the arithmetic average of the expression of the parental genes. Expression additivity of parental genes is observed in the offspring of the diploid species. The deviation of additivity called parental non-additive expression is mainly found in the offspring of polyploid species. A bias when the expression of the offspring is similar to that of one of the parents is called expression-level dominance. If total offspring expression is lower or higher than in both parents, the phenomenon is called transgressive expression, and when the contribution of the parental homeologs to the total gene expression is unequal, it is named homeolog expression bias. All this deviation from additivity can be explained as a result of different factors, such as the influence of one of the parental genomes, epigenetic regulation, balance of gene dosage, and *cis-* and/or *trans*-regulatory elements [2].

Expression-level dominance, which is of uniparental origin, is also called genomic imprinting [36,37]. This phenomenon of epigenetic origin is the result of the asymmetries of DNA and histone methylation between maternal and paternal plants. Male and female genotypes are multicellular in origin; therefore, primary gene imprinting can occur in egg cells, central cells, and sperm and, subsequently, in the triploid endosperm or, less frequently, in the diploid embryo [37]. Generally, conservation of imprinting is limited across other crops; however, these genes that show conserved imprinting in cereals showed positive selection and were suggested to perform a dose-dependent function in the regulation of seed development [38]. However, as recently documented by Rodrigues et al. [39], most genes imprinted in the endosperm of seeds were imprinted across cultivars, extending their functions to chromatin and transcriptional regulation, development, and signaling. Only 4% to 11% of the imprinted genes showed divergent imprinting.

Imprinted gene expression affects mostly single genes or groups of genes. Most of them are maternally expressed and inherited [36,40,41]. The best recognized is the maternal effect of genes during embryo development [42]. Early development in Arabidopsis is coordinated by the supply of auxin from the mother integuments of the ovule, which is required for the correct embryo development of embryos [40]. The genomic imprinting of the cereal endosperm influences the timing of endosperm cellularization [43]. An example of imprinted maternally expressed genes in cereals is a polycomb group, which is important for the cellularization of endosperm in rice [44]. Reciprocal crosses between tetraploid and hexaploid wheats showed that imprinted genes were identified in endosperm and embryo tissue, supporting the predominant maternal effect on early grain development [45]. Paternally expressed imprinted genes were associated with hybrid seed lethality in Capsella [46]. In maize, the *Dosage-effect defective1* (*ded1*) locus that contributes to seed size was found to be paternally imprinted [47]. The gene encodes a transcription factor that is specifically expressed during early endosperm development. There is also evidence that small RNAs might determine the paternal methylome by silencing transposons [48]. In addition to these reports, it is very difficult to find examples of paternally inherited genes, especially in cereals. Many studies have indicated dynamic changes in the epigenetic state, including DNA methylation, chromatin modifications, and small RNAs, which are observed during the reproductive development of plants [49,50]. The spatiotemporal pattern of gene expression, imprinting, and seed development in Arabidopsis endosperm is predominantly regulated by small maternal RNAs; however, they also originate from the paternal genome and the seed coat [51]. As reported by Tuteja et al. [52], imprinted paternally expressed genes, but not maternally expressed genes, in Arabidopsis evolve under positive Darwinian selection. These genes were involved in seed development processes, such as auxin biosynthesis and epigenetic regulation. Imprinted paternally expressed genes are mainly associated with hypomethylated maternal DNA alleles, which can be repressed by small genic RNAs and rarer with transposable elements [49,53]. Epigenetic changes can be developmentally regulated (developmental epigenetics). The state in which changes in DNA methylation are stable between generations and heritable is called transgenerational epigenetics [54].

Several *TaCKX* GFMs and *TaNAC2-5A* (*NAC2*) were previously selected as important regulators of yield-related traits. To determine how the expression patterns of selected genes are inherited in the developing spikes and seedling roots of the parents and the F<sup>2</sup> generation, we used a reciprocal crossing strategy. The research hypothesis assumed that knowledge of inheritance of gene expression patterns that regulated yield-related traits indicated the way of selection of genotypes in wheat breeding. There is a research gap in documenting the inheritance of expression patterns for yield-related genes. We found that most of the genes in the F<sup>2</sup> generation were expressed in a pater-of-origin-specific manner, which shed new light on the ways of selecting wheat lines and the breeding strategy.

#### **2. Results**

### *2.1. Reciprocal Crosses Indicate That the Expression Patterns of Most of the TaCKX GFM and Yield-Related Traits Are Mainly Inherited from the Male Parent*

Relative values (related to the female parent = 1.0) of the expression profiles of *TaCKX* GFM and *NAC2* in 7 DAP spikes, seedling roots, and phenotypic traits in the female parent, male parent, and their six F<sup>2</sup> progeny from one reciprocal cross of S12B × S6C (C1) and S6C × S12B (C2) are presented in Figure 1. The same data obtained in reciprocal crosses of D16 × KOH7 (C3) and KOH7 × D16 (C4); D19 × D16 (C5) and D16 × D19 (C6); D19 × KOH7 (C7) and KOH7 × D19 (C8) are visualized in Figure S1. S6C, the

paternal parent of the S12B × S6C cross (C1), showed higher expression of *TaCKX1* and *NAC2* and lower expression of *TaCKX5* and *TaCKX10* in spikes than the maternal parent (S12B). The expression of *TaCKX1*, *NAC2*, *TaCKX5,* and *TaCKX11* in the spikes of the F<sup>2</sup> progeny of this cross was higher (Figure 1A). In the reverse cross (S6C × S12B), when S12B was a paternal component (Figure 1B), *TaCKX1* and *NAC2* were expressed at low levels, *TaCKX5* and *10* were highly expressed in spikes compared to the maternal parent (S6C), and in F2, *TaCKX1* and *NAC2* were expressed at low levels, and *TaCKX9*, *10*, *11*, and *5* were upregulated. In seedling roots, the paternal component of S12B × S6C (Figure 1A) showed high expression of *TaCKX5* and *NAC2* and low expression of *TaCKX10* and *11* in the parental parent compared to the maternal parent. In the roots of the F<sup>2</sup> progeny of this cross, *TaCKX5* and *NAC2* were highly expressed, and *TaCKX11* was downregulated. The expression data in the parents of the reverse cross, in which S12B was the paternal component, were opposite. Their F<sup>2</sup> progeny showed a strong upregulation of *TaCKX8*, *10*, and *11* and a downregulation of *TaCKX1*, *TaCKX3*, and *NAC2* in seedling roots. The total grain yield and the number of seeds in F<sup>2</sup> of S12B × S6C were low, similar to the male parent (Figure 1A). The root weight in the F<sup>2</sup> progeny of the same cross (S12B × S6C) was lower than that in the parents. The same yield components in the opposite cross (S6C × S12B) were high in one F<sup>2</sup> sibling, comparable to the maternal parent in two progeny and lower than in the parents in three of them (Figure 1B). Interestingly, the root mass in F<sup>2</sup> was higher than that in both parents. *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

**Figure 1.** Example of *TaCKX* GFM and *NAC2* expression profiles in 7 DAP spikes, seedling roots, and phenotypic traits in the maternal parent, paternal parent, and their six F<sup>2</sup> progeny, from reciprocal cresses of S12B × S6C, C1 (**A**) and S6C × S12B, C2 (**B**). The data represent mean values with standard deviation. Black and red asterisks indicate statistical significance compared to the maternal parent or paternal parent, respectively (\* 0.05 > *p* ≥0.01, \*\* 0.01 > *p* ≥ 0.001, \*\*\* *p* < 0.001) using the ANOVA test followed by the LSD post hoc test (STATISTICA 10, StatSoft). **Table 1.** *TaCKX* GFMs and *NAC2* with high (↑), very high (↑↑), low (↓) or very low (↓↓) expression **Figure 1.** Example of *TaCKX* GFM and *NAC2* expression profiles in 7 DAP spikes, seedling roots, and phenotypic traits in the maternal parent, paternal parent, and their six F<sup>2</sup> progeny, from reciprocal cresses of S12B × S6C, C1 (**A**) and S6C × S12B, C2 (**B**). The data represent mean values with standard deviation. Black and red asterisks indicate statistical significance compared to the maternal parent or paternal parent, respectively (\* 0.05 > *p* ≥ 0.01, \*\* 0.01 > *p* ≥ 0.001, \*\*\* *p* < 0.001) using the ANOVA test followed by the LSD post hoc test (STATISTICA 10, StatSoft).

levels in 7 DAP spikes and seedling roots of the maternal parent (M), the paternal parent (P) and their F<sup>2</sup> progeny from cresses of D16 × KOH7 (C3) and KOH7 × D16 (C4); D19 × D16 (C5) and D16 × D19 (C6); D19 × KOH7 (C7) and KOH7 × D19 (C8), and high (↑) and low (↓) parameters of yield and

**M P F2**

*NAC2*↓ *NAC2*↑ *CKX5*, *NAC2*↑

*CKX11*, *10*↑ *CKX11*, *10*↓ *CKX11*↓

CKX act. spike= CKX act. spike= CKX act. spike↓↓ root = ↓ root=↑ root = ↓

*CKX5*, *9*↑ *CKX5*, *9*↓ CKX expression root *CKX5*, *NAC2*↓↓ *CKX5*↑↑ *CKX5*↑↑↑, *NAC2*↑ *CKX3*↓ *CKX3*↑, *NAC2*↑

CKX expression 7 DAP *CKX1*↓ *CKX1*↑ *CKX1*, *11*↑↑

yield-related traits yield↑ yield↓ yield↓↓

**C1 = S12B × S6C** 

As presented in Table 1 by colours, most of the expression patterns tested for *TaCKX* GFM and *NAC2* are inherited from the male parent (red). For example, up-regulated in 7 DAP spikes of the paternal parent *TaCKX1* and *NAC2* compared to the maternal parent is up-regulated in F<sup>2</sup> as well. The upregulated *TaCKX5* and *NAC2* and downregulated *TaCKX11* in seedling roots of the paternal parents are similarly expressed in an F2. To summarize, the expression levels of all tested *TaCKX* GFMs and *NAC2* in 7 DAP spikes, in addition to being represented in different crosses, showed similar expression patterns to the paternal parents and were independent of the cross path. Among the *TaCKX* GFMs in 7 DAP spikes, which showed the paternal expression patterns were *TaCKX1*, *2.1*, *2.2.2*, *5*, *9*, *10*, and *11*. In the seedling roots, there were *TaCKX5*, *11*, *NAC2*; *TaCKX10*, *11*, *NAC2*; *TaCKX1*, *11*, *NAC2*; *TaCKX10*; *1*, *3*, *5*, *8*, *10*, *11*, *NAC2*; *TaCKX1*, *8*, *10*, *NAC2*; and *3*, *5*, *8*, *11*; *NAC2* (all tested but represented in different crosses). The only exceptions are *TaCKX5* in 7 DAP spikes of S12B × S6C (C1) and *TaCKX3* in seedling roots of KOH7 × D16 (C4), whose expression level is similar to that of the maternal parent (green). **Figure 1.** Example of *TaCKX* GFM and *NAC2* expression profiles in 7 DAP spikes, seedling roots, and phenotypic traits in the maternal parent, paternal parent, and their six F<sup>2</sup> progeny, from reciprocal cresses of S12B × S6C, C1 (**A**) and S6C × S12B, C2 (**B**). The data represent mean values with standard deviation. Black and red asterisks indicate statistical significance compared to the maternal parent or paternal parent, respectively (\* 0.05 > *p* ≥0.01, \*\* 0.01 > *p* ≥ 0.001, \*\*\* *p* < 0.001) using the ANOVA test followed by the LSD post hoc test (STATISTICA 10, StatSoft). *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 6 of 26 CKX act. root↓↓ CKX act. root↑↑ CKX act. root↑↑ **C2 = S6C × S12B** CKX expression 7 DAP *CKX5*, *9*↓ *CKX5*, *9*↑ *CKX9*, *10*, *11*, *5*↑ *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

**Table 1.** *TaCKX* GFMs and *NAC2* with high (↑), very high (↑↑), low (↓) or very low (↓↓) expression levels in 7 DAP spikes and seedling roots of the maternal parent (M), the paternal parent (P) and their F<sup>2</sup> progeny from cresses of D16 × KOH7 (C3) and KOH7 × D16 (C4); D19 × D16 (C5) and D16 × D19 (C6); D19 × KOH7 (C7) and KOH7 × D19 (C8), and high (↑) and low (↓) parameters of yield and root mass. Character colours indicate similar patterns of gene expression and yield-related traits in F<sup>2</sup> and paternal parent (red) or in F<sup>2</sup> and maternal parent (green). **Table 1.** *TaCKX* GFMs and *NAC2* with high (↑), very high (↑↑), low (↓) or very low (↓↓) expression levels in 7 DAP spikes and seedling roots of the maternal parent (M), the paternal parent (P) and their F<sup>2</sup> progeny from cresses of D16 × KOH7 (C3) and KOH7 × D16 (C4); D19 × D16 (C5) and D16 × D19 (C6); D19 × KOH7 (C7) and KOH7 × D19 (C8), and high (↑) and low (↓) parameters of yield and root mass. Character colours indicate similar patterns of gene expression and yield-related traits in F<sup>2</sup> and paternal parent (red) or in F<sup>2</sup> and maternal parent (green). *CKX1*, *2.1*, *NAC2*↑ *CKX1*, *2.1*, *NAC2*↓ *CKX1*, *NAC2*↓ CKX expression root *CKX10*, *11*↓↓ *CKX10*, *11*↑↑ *CKX5, 8*, *10*, *11*↑↑ *CKX1, 3*, *5*, *NAC2*↑ *CKX1, 3*, *5*, *NAC2*↓ *CKX1*, *3*, *NAC2*↓ yield-related traits yield↓ yield↑ yield=↓ CKX act. spike= CKX act. spike= CKX act. Spike= root↑ root=↓ root↑↑ CKX act. root↑ CKX act. root↓ CKX act. Root= **Figure 1.** Example of *TaCKX* GFM and *NAC2* expression profiles in 7 DAP spikes, seedling roots, and phenotypic traits in the maternal parent, paternal parent, and their six F<sup>2</sup> progeny, from reciprocal cresses of S12B × S6C, C1 (**A**) and S6C × S12B, C2 (**B**). The data represent mean values with *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 6 of 26 CKX act. root↓↓ CKX act. root↑↑ CKX act. root↑↑


*CKX3*↓ *CKX3*↑, *NAC2*↑

CKX expression root *CKX 5*, *8*, *NAC2*↑ *CKX5*, *8*, *NAC2*↓↓ *CKX3*, *5*, *8*, *11*↓

yield-related traits yield↓ yield↑↑ yield↑

yield-related traits yield↓ yield↑ yield↑

CKX expression root *CKX3*, *8*↓ *CKX3*,*8*↑ *CKX8*↑↑

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

CKX expression 7 DAP *CKX2.2.2*↑ *CKX2.2.2*↓ *CKX2.2.2*↓

CKX expression root *CKX5*, *8*, *NAC2*↓↓ *CKX5*, *8*, *NAC2*↑↑ *CKX1*, *NAC2*↑↑

CKX expression root *CKX 5*, *8*, *NAC2*↑ *CKX5*, *8*, *NAC2*↓↓ *CKX3*, *5*, *8*, *11*↓

yield-related traits yield↓ yield↑↑ yield↑

yield-related traits yield↑ yield↓ yield↓↓

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

CKX expression root *CKX5*, *8*, *NAC2*↓↓ *CKX5*, *8*, *NAC2*↑↑ *CKX1*, *NAC2*↑↑

yield-related traits yield↓ yield↑↑ yield↑

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

**C7 = D19 × KOH7**

**C5 = D19 × D16**

**C6 = D16 × D19** CKX expression 7 DAP *CKX2.2.2*, *5*, *9*↓ *CKX2.2.2*, *5*, *9*↑

**C7 = D19 × KOH7**

**C6 = D16 × D19** CKX expression 7 DAP *CKX2.2.2*, *5*, *9*↓ *CKX2.2.2*, *5*, *9*↑

**C7 = D19 × KOH7**

*CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

CKX act. spike = ↓ CKX act. spike=↑ CKX act. spike= root↓ root↑ root=↓

CKX act. spike=↓ CKX act. spike=↑ CKX act. spike=↑ root= root= root↓ semi-empty spikes↓ semi-empty spikes↑ semi-empty spikes↑↑↑

CKX act. spike= CKX act. spike= CKX act. spike↓↓ root = ↓ root=↑ root = ↓

*CKX10*↓ *CKX10*↑ *CKX9*↑

*CKX11*, *10*↑ *CKX11*, *10*↓ *CKX11*↓

*CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓

*CKX10*↓ *CKX10*↑ *CKX9*↑

*CKX1*, *3*, *10*, *11*↑ *CKX1*, *3*, *10*, *11*↓ (*CKX10*, *NAC2*↑)

*CKX1*, *3*, *10*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX8*, *10*↑

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

CKX act. spike=↓ CKX act. spike=↑ CKX act. spike=↑ root= root= root↓ semi-empty spikes↓ semi-empty spikes↑ semi-empty spikes↑↑↑

*CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓

*CKX1*, *3*, *10*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX8*, *10*↑

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

*CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓

*CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓

*CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓

*NAC2*↓

*NAC2*↓

*NAC2*↓


*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 6 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 5 of 26

CKX expression 7 DAP *CKX5*, *9*↓ *CKX5*, *9*↑ *CKX9*, *10*, *11*, *5*↑

CKX expression root *CKX10*, *11*↓↓ *CKX10*, *11*↑↑ *CKX5, 8*, *10*, *11*↑↑

*Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 6 of 26

yield-related traits yield↓ yield↑ yield=↓

CKX expression 7 DAP *CKX11*↓↓ *CKX11*↑↑ *CKX5,9*↑↑

CKX expression 7 DAP *CKX5*, *9*↓ *CKX5*, *9*↑ *CKX9*, *10*, *11*, *5*↑

ANOVA test followed by the LSD post hoc test (STATISTICA 10, StatSoft).

CKX act. root↓↓ CKX act. root↑↑ CKX act. root↑↑

*CKX1*, *2.1*, *NAC2*↑ *CKX1*, *2.1*, *NAC2*↓ *CKX1*, *NAC2*↓

*CKX1, 3*, *5*, *NAC2*↑ *CKX1, 3*, *5*, *NAC2*↓ *CKX1*, *3*, *NAC2*↓

CKX act. spike= CKX act. spike= CKX act. Spike= root↑ root=↓ root↑↑ CKX act. root↑ CKX act. root↓ CKX act. Root=

CKX act. root↓↓ CKX act. root↑↑ CKX act. root↑↑

*CKX1*, *2.1*, *NAC2*↑ *CKX1*, *2.1*, *NAC2*↓ *CKX1*, *NAC2*↓

**Figure 1.** Example of *TaCKX* GFM and *NAC2* expression profiles in 7 DAP spikes, seedling roots, and phenotypic traits in the maternal parent, paternal parent, and their six F<sup>2</sup> progeny, from reciprocal cresses of S12B × S6C, C1 (**A**) and S6C × S12B, C2 (**B**). The data represent mean values with standard deviation. Black and red asterisks indicate statistical significance compared to the maternal parent or paternal parent, respectively (\* 0.05 > *p* ≥0.01, \*\* 0.01 > *p* ≥ 0.001, \*\*\* *p* < 0.001) using the

#### **Table 1.** *Cont. CKX2.1*, *2.2.2*↓ *CKX2.1*, *2.2.2*↑ **Table 1.** *TaCKX* GFMs and *NAC2* with high (↑), very high (↑↑), low (↓) or very low (↓↓) expression CKX expression root *CKX10*, *11*↓↓ *CKX10*, *11*↑↑ *CKX5, 8*, *10*, *11*↑↑ *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 6 of 26

**C2 = S6C × S12B**

**C3 = D16 × KOH7** 

**C2 = S6C × S12B**

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓ root↓ root↑ root=↓ root = ↓ root=↑ root = ↓ **C4 = KOH7 × D16** Bold—cross number and parents.

**C7 = D19 × KOH7**

**C5 = D19 × D16**

**C7 = D19 × KOH7**

**C6 = D16 × D19** CKX expression 7 DAP *CKX2.2.2*, *5*, *9*↓ *CKX2.2.2*, *5*, *9*↑

**C7 = D19 × KOH7**

**C6 = D16 × D19** CKX expression 7 DAP *CKX2.2.2*, *5*, *9*↓ *CKX2.2.2*, *5*, *9*↑

root= root= root↓ **C7 = D19 × KOH7** CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑ *CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓ *NAC2*↓ semi-empty spikes↓ semi-empty spikes↑ semi-empty spikes↑↑↑ **C6 = D16 × D19** CKX expression 7 DAP *CKX2.2.2*, *5*, *9*↓ *CKX2.2.2*, *5*, *9*↑ *CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓ CKX expression root *CKX5*, *8*, *NAC2*↓↓ *CKX5*, *8*, *NAC2*↑↑ *CKX1*, *NAC2*↑↑ *CKX1*, *3*, *10*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX8*, *10*↑ yield-related traits yield↓ yield↑↑ yield↑ **C5 = D19 × D16** CKX expression 7 DAP *CKX2.2.2*↑ *CKX2.2.2*↓ *CKX2.2.2*↓ *CKX10*↓ *CKX10*↑ *CKX9*↑ CKX expression root *CKX 5*, *8*, *NAC2*↑ *CKX5*, *8*, *NAC2*↓↓ *CKX3*, *5*, *8*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX1*, *3*, *10*, *11*↓ (*CKX10*, *NAC2*↑) yield-related traits yield↑ yield↓ yield↓↓ CKX act. spike=↓ CKX act. spike=↑ CKX act. spike=↑ CKX expression 7 DAP *CKX9*, *10*↓ *CKX9*, *10*↑ *CKX5*,*9*↑↑ *CKX11*↑↑ *CKX11*↓↓ *CKX11*, *NAC2*↓↓ *CKX2.1*, *2.2.2*↑ *CKX2.1*, *2.2.2*↓ *CKX2.1*, *2.2.2*↓ CKX expression root *CKX3*, *8*↓ *CKX3*,*8*↑ *CKX8*↑↑ *CKX1*, *5*, *10*, *11*↑ *CKX1*, *5*, *10*, *11*↓ *CKX3*, *10*↓ *NAC2*↑↑ *NAC2*↓↓ yield-related traits yield↓ yield↑ yield↑ Yield-related traits are represented by total grain yield and root mass (Table 1). Interestingly, grain yield in 7 out of 8 crosses is inherited from the paternal parent. The exceptions are the F<sup>2</sup> progeny of S6C × S12B (C2), which show very large differences in yield, exceeding parental data. The root mass in F<sup>2</sup> was lower than that in both parents or higher than that in both parents. In the first case, the root mass in the paternal parent was higher than that in the maternal parent, and in the second, the root mass in the paternal parent was lower than that in the maternal parent.

> CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

> CKX act. spike = ↓ CKX act. spike=↑ CKX act. spike= root↓ root↑ root=↓

*CKX1*, *3*, *10*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX8*, *10*↑

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

CKX act. spike=↓ CKX act. spike=↑ CKX act. spike=↑ root= root= root↓ semi-empty spikes↓ semi-empty spikes↑ semi-empty spikes↑↑↑

*CKX1*, *3*, *10*, *11*↑ *CKX1*, *3*, *10*, *11*↓ (*CKX10*, *NAC2*↑)

*CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓

*CKX1*, *3*, *10*, *11*↓ *CKX1*, *3*, *10*, *11*↑ *CKX8*, *10*↑

CKX act. spike=↑ CKX act. spike=↓ CKX act. spike= root= root= root↓

*CKX9*↑ *CKX9*↓ *CKX2.2.2*, *9*↓

*CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓

root= root= root↓ semi-empty spikes↓ semi-empty spikes↑ semi-empty spikes↑↑↑

*CKX10*↑ *CKX10*↓ *CKX1*, *10*, *11*, *NAC2*↓

*NAC2*↓

*NAC2*↓

*NAC2*↓

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

CKX expression 7 DAP *CKX2.2.2*↑ *CKX2.2.2*↓ *CKX2.2.2*↓

yield-related traits yield↓ yield↑↑ yield↑

yield-related traits yield↑ yield↓ yield↓↓

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

CKX expression root *CKX5*, *8*, *NAC2*↓↓ *CKX5*, *8*, *NAC2*↑↑ *CKX1*, *NAC2*↑↑

yield-related traits yield↓ yield↑↑ yield↑

CKX expression 7 DAP *CKX2.1*, *11*, *NAC2*↓ *CKX2.1*, *11*, *NAC2*↑ *CKX10*↑

The results of crossing the low-yielding maternal parent with the higher-yielding paternal parent and their accompanying up- or down-regulated *TaCKX* GFMs and *NAC2* in F<sup>2</sup> generations are presented in Figure 2. *Int. J. Mol. Sci.* **2023**, *24*, x FOR PEER REVIEW 8 of 26

**Figure 2.** Models of up- (↑) or down-regulation (↓) of *TaCKX* GFMs and *NAC2* in low-yielding maternal parent crossed with higher-yielding paternal parent and their F<sup>2</sup> progeny. **Figure 2.** Models of up- (↑) or down-regulation (↓) of *TaCKX* GFMs and *NAC2* in low-yielding maternal parent crossed with higher-yielding paternal parent and their F<sup>2</sup> progeny.

*2.2. Cooperating and Opposite-Functioning Genes TaCKX5* with *TaCKX9* (yellow) and *TaCKX2.1* with *TaCKX11* (green) showed coordinated up- or downregulation in 7 DAP spikes of the paternal parent of C1, C2, C6, and C8 crosses; and C3, C4, C7, and C8 crosses, respectively (Table 2). Higher expression of *TaCKX5* and *9* in this parent was associated with a higher yield in F2. However, a higher coordinated expression of *TaCKX2.1* with *TaCKX11* in the paternal parent determined a lower yield in F2, and, in contrast, a lower expression of these two genes in the paternal parent was associated with a higher yield. In the 7 DAP spikes, the paternal parent of the C3, C4, C5, and C6 crosses, *TaCKX2.2.2,* showed opposite expression to *TaCKX10* (blue), and upregulation of the first and downregulation of the second were associated with lower yield (but not in C6). In the paternal parent of the C1, C2, C7, and C8 crosses, *NAC2* was oppositely expressed to *TaCKX9*; in these crosses, a high yield was observed when *TaCKX9* was upregulated and *NAC2* was downregulated, and vice versa (only in C7 and C8). Furthermore, upregulated *TaCKX5* and downregulated *TaCKX1* were associated with high root mass in C2 and conversely in Depending on the crosses, downregulation of *TaCKX5* with *TaCKX9* and upregulation of *NAC2* in spikes of the low-yielding maternal parent and the opposite regulation of these genes in spikes of the higher-yielding paternal parent resulted in high-yielding F2, characterized, as in the paternal component, by a higher expression level of *TaCKX9* and a lower expression level of *NAC2*. Upregulation of *TaCKX2.1* and *11* in spikes of the maternal parent and downregulation of these genes in the paternal parent were associated with downregulation of *TaCKX11* in high-yielding F2. Similarly, the upregulation of *TaCKX2.2.2* and the downregulation of *TaCKX10* in the spikes of the low-yielding maternal parent and opposite regulation of these genes, and the yield in the paternal parent, resulted in the downregulation of *TaCKX2.2.2* and the upregulation of *TaCKX10* in the spikes of the high-yielding F2. The expression of *TaCKX3* in seedling roots of the high-yielding paternal parent and F<sup>2</sup> was upregulated. However, *TaCKX8* expression was upregulated, and *NAC2* was downregulated in the same organ of the paternal parent, but these genes were up- or down-regulated in F2, depending on the cross.

#### Among the *TaCKX* genes coordinately expressed in the paternal seedling roots were *2.2. Cooperating and Opposite-Functioning Genes*

the reverse cross (C1).

*TaCKX3*, *5,* and 8 (green) in C5, C6, C7, and C8 crosses; *TaCKX3* and *8* (green) in C1 and C2 crosses; *TaCKX10*, *11*, and *1* (yellow), and *NAC2* in C3 to C7 crosses; and *TaCKX10* and *11* (yellow) in C1 and C2 crosses. However, in one reciprocal cross, C3 and C4, the expression of *TaCKX3* and *TaCKX5* was opposite, and in the case of upregulation of *TaCKX3* and downregulation of *TaCKX5*, the grain yield in F<sup>2</sup> was higher. *TaCKX5* with *TaCKX9* (yellow) and *TaCKX2.1* with *TaCKX11* (green) showed coordinated up- or downregulation in 7 DAP spikes of the paternal parent of C1, C2, C6, and C8 crosses; and C3, C4, C7, and C8 crosses, respectively (Table 2). Higher expression of *TaCKX5* and *9* in this parent was associated with a higher yield in F2. However, a higher coordinated expression of *TaCKX2.1* with *TaCKX11* in the paternal parent determined a lower yield in F2, and, in contrast, a lower expression of these two genes in the paternal parent was associated with a higher yield.

In the 7 DAP spikes, the paternal parent of the C3, C4, C5, and C6 crosses, *TaCKX2.2.2,* showed opposite expression to *TaCKX10* (blue), and upregulation of the first and downregulation of the second were associated with lower yield (but not in C6). In the paternal parent of the C1, C2, C7, and C8 crosses, *NAC2* was oppositely expressed to *TaCKX9*; in these crosses, a high yield was observed when *TaCKX9* was upregulated and *NAC2* was

downregulated, and vice versa (only in C7 and C8). Furthermore, upregulated *TaCKX5* and downregulated *TaCKX1* were associated with high root mass in C2 and conversely in the reverse cross (C1).

**Table 2.** Coordinated expression of *TaCKX* GFMs and *NAC2* genes in 7 DAP spikes and seedling roots of the paternal parent (P) of four reciprocal crosses.


P—paternal parent; expr.+—upregulated; expr.—downregulated; 1, 3, 5, 9 . . . -*TaCKX* GFMs, As—CKX activity spike; s-e+++—high number of semi-empty spikes, y—yield, r—root mass.

Among the *TaCKX* genes coordinately expressed in the paternal seedling roots were *TaCKX3*, *5,* and 8 (green) in C5, C6, C7, and C8 crosses; *TaCKX3* and *8* (green) in C1 and C2 crosses; *TaCKX10*, *11*, and *1* (yellow), and *NAC2* in C3 to C7 crosses; and *TaCKX10* and *11* (yellow) in C1 and C2 crosses. However, in one reciprocal cross, C3 and C4, the expression of *TaCKX3* and *TaCKX5* was opposite, and in the case of upregulation of *TaCKX3* and downregulation of *TaCKX5*, the grain yield in F<sup>2</sup> was higher.

In three reciprocal crosses, *NAC2* was downregulated in paternal roots (C2, C4, and C8), and in two of them (C2 and C8), *NAC2* was downregulated in paternal spikes as well. This negative regulation of *NAC2* occurred in the F<sup>2</sup> progeny, which was accompanied by a higher yield and a higher or similar to the parents' mass of the seedling roots. In contrast, in another way crosses, when expression of *NAC2* was increased in paternal roots (C1, C3, and C7) and was upregulated in paternal spikes, the same was observed in F<sup>2</sup> progeny, characterized by lower yield and lower or similar to the parent mass of the roots.

The higher yield in F<sup>2</sup> has been associated with the same or higher CKX activity, as in the paternal parent, in 7 DAP spikes. A higher number of semi-empty spikes, which occurred in low-yielding F<sup>2</sup> of the C3, C5, and C7 crosses, was accompanied by downregulated *TaCKX10* and/or upregulated *TaCKX11* in 7 DAP spikes and upregulated *TaCKX10*, *11,* and *NAC2* or downregulated *TaCKX10* and *NAC2* in seedling roots.

### *2.3. The Correlation Coefficients between TaCKX GFMs and NAC2 Expression, CKX Activity, and Yield-Related Traits Were Significant for Both Parents or the Maternal or Paternal Parent Separately*

The correlation coefficients between *TaCKX* GFMs and *NAC2* expression, CKX activity, and yield-related traits in reciprocal crosses were analyzed separately for the maternal parent and F2, paternal parent, and F<sup>2</sup> for each cross (Table S1).
