*2.4. Correlations between TaCKX GFM and NAC2 Expression and Yield-Related Traits in the Group of Maternal Plants, and F<sup>2</sup> and Paternal Plants, and F<sup>2</sup> of Reciprocal Crosses*

Seed number and spike number were positively correlated (Tables 3 and S1); however, each of these yield-related traits was correlated with different *TaCKX* GFMs.

*Int. J. Mol. Sci.* **2023**, *24*, 8196

**Table 3.** Correlations between *TaCKX GFM* and *NAC2* expression in 7 DAP spikes or seedling roots, and yield-related traits in the group of maternal plants and F2 , and paternal plants and F2 of reciprocal crosses.



**Table 3.** *Cont.*





**Table 3.** *Cont.*




−—negative correlation; =−—low negative correlation; −−—strong negative correlation.

#### 2.4.1. Seed Number

The decrease in seed number in maternal plants and their F<sup>2</sup> (M and F2) of the C1 cross (S12B × S6C) was strongly negatively correlated with upregulated *TaCKX1* and *TaCKX5* in spikes and positively correlated with the downregulated *TaCKX11* in the seedling roots of F2. There was no significant correlation between the expression of *TaCKX* GFM and the yield-related traits in the groups of paternal plants (P) and F<sup>2</sup> in the same cross, and M and F2, and P and F<sup>2</sup> in the reverse, C2 cross. The F<sup>2</sup> progeny in this reverse cross showed a similar yield and greater root mass compared to the parents.

In both M and F2, and P and F<sup>2</sup> of C3, the decrease in seed number was strongly positively correlated with *TaCKX2.1* and *TaCKX2.2.2* in spikes and positively correlated with *TaCKX3* and *TaCKX8* but negatively correlated with *TaCKX11* in seedling roots. These correlations were not significant in the reverse, C4 cross, in which the M (KOH7) and F<sup>2</sup> plants showed higher yields.

The decrease in seed number in M × F<sup>2</sup> of C5 was negatively correlated with *TaCKX1* in spikes and positively correlated with downregulated *TaCKX3* and *5,* and upregulated *NAC2* in seedling roots. There were also positive correlations of *TaCKX3* in roots between P and F<sup>2</sup> of the same cross. These correlations were not significant in the reverse C6 cross; however, F<sup>2</sup> of this cross was characterized by higher yield and similar root mass than in the parents.

The increase in seed number was strongly positively correlated with downregulated *TaCKX2.1* in spikes and negatively correlated with downregulated *TaCKX1* in the seedling roots only in F<sup>2</sup> progeny of a C8 cross, 14K (in the case of P × F<sup>2</sup> only for *TaCKX2.1*).

#### 2.4.2. Spike Number

The decrease in the number of spikes in M × F<sup>2</sup> and P × F<sup>2</sup> of the C1 cross was strongly positively correlated with *TaCKX2.1* and *TaCKX2.2.2* in spikes and positively correlated with *TaCKX11* in seedling roots (only in M × F2). There was also a significant and positive correlation between the expression of *TaCKX2*.*2.2* and the number of spikes in the reverse C2 cross of M and F2. Furthermore, in the same cross, the spike number was negatively correlated with downregulated *NAC2*, but only in the P × F<sup>2</sup> group. There were no significant correlations between *TaCKX* GFM expression and spike number in M × F<sup>2</sup> and P × F<sup>2</sup> spikes of C3 and C4. However, there was a strong and positive correlation of the spike number with *TaCKX8* in the roots of C3 and *TaCKX5* in the roots of C4.

The decreased spike number in M × F<sup>2</sup> and P × F<sup>2</sup> of C5 was not correlated with any *TaCKX* expressed in the spikes but was negatively correlated with *TaCKX1*, positively correlated with *TaCKX5,* and positively correlated with *NAC2* expressed in the seedling roots. Conversely, in reverse C6 cross, there was a positive correlation of the spike number with *TaCKX2.1* in a P × F2, which resulted in a higher yield phenotype in the F2.

The spike number in C7 and C8 crosses was not correlated with the level of expression of any gene tested in the spikes; however, it was negatively correlated with the expression of *TaCKX1*, 5, and *8* in seedling roots.

#### 2.4.3. TGW

TGW was positively correlated with *TaCKX2.1*, *10,* and *NAC2* in spikes of M and F<sup>2</sup> of C1 and with *TaCKX2.2.2* of P and F<sup>2</sup> of the same cross. There was no correlation in F<sup>2</sup> between the expression of the genes tested and TGW in spikes of the C2 and roots of the C1 and C2 crosses. There were no correlations between the TGW and *TaCKX* genes in the spikes and roots of C3. However, there was a negative correlation of this trait with *TaCKX2.2.2* in spikes and positive correlations with *TaCKX10* and *NAC2* in roots of the reciprocal C4 cross. Positive correlations of *NAC2* with TGW were also observed in the roots of C5 but not in those of C6. Negative correlations of *TaCKX2.1* and *2.2.2* in spikes with TGW were observed in both reciprocal crosses, C7 and C8. Additionally, *TaCKX9* was negatively correlated with the trait in M + F2, *TaCKX5* was positively correlated with the trait in P + F<sup>2</sup> of C7, and *TaCKX10* was negatively correlated with TGW in P + F<sup>2</sup> of C8. There was no correlation between TGW and any gene expression in roots.

#### 2.4.4. Root Mass

The mass is positively correlated with the expression of *TaCKX5*, *11,* and *NAC2* in spikes of M + F<sup>2</sup> of C1 and negatively correlated with *NAC2* in M + F<sup>2</sup> of C2. A positive correlation between root mass and *TaCKX11,* and *NAC2* was also visible in P + F<sup>2</sup> of C3, and a negative correlation between trait and *NAC2* expression was also observed in spikes of M + F<sup>2</sup> of C5 and P + F<sup>2</sup> of C7. Furthermore, in the C1 cross, this trait was negatively correlated with the expression of *TaCKX1*, *3*, *10,* and *NAC2* in roots of M + F<sup>2</sup> and with *TaCKX11* in roots of P + F2. There were also positive correlations between root mass and *TaCKX1* (M + P of C3), root mass and *TaCKX2.1* (P + F<sup>2</sup> of C4; P + F<sup>2</sup> of C5; M + F<sup>2</sup> of C6), and root mass and *TaCKX2.2.2* (P + F<sup>2</sup> of C4; M + F<sup>2</sup> and P + F<sup>2</sup> of C6). The expression of another gene, *TaCKX10,* in spikes, was positively correlated with root mass in P + F<sup>2</sup> of C4 but negatively correlated in M + F<sup>2</sup> of C8. Correlations between root mass and gene expression tested in roots were dependent on the parent and cross. There were negative correlations with *TaCKX1* in 5 out of 16 combinations tested, negative correlations with *TaCKX3* in 3 combinations, but positive correlations in two combinations, positive correlations with *TaCKX5* in two combinations, and single positive or negative correlations with *TaCKX8*, *11,* and *NAC2*. The root mass in single combinations was positively correlated with the yield (twice), height of the plant (once), and length of the spike (once), and negatively correlated with seed number of seeds (twice).

#### 2.4.5. Semi-Empty Spikes

The number of semi-empty spikes was positively correlated with the expression of *TaCKX9* in the P + F<sup>2</sup> C1, C2, and M + F<sup>2</sup> C3 crosses, positively correlated with *TaCKX5* in the M + F2, and P + F<sup>2</sup> C1 and C6 crosses, and positively correlated with *TaCKX10* in the M + F<sup>2</sup> C7 and P + F<sup>2</sup> C8 crosses, all expressed in 7 DAP spikes. The negative correlation between the number of semi-empty spikes and *TaCKX2.1* was in P + F<sup>2</sup> of C5, and between the same trait and *TaCKX11* was in P + F<sup>2</sup> of C3. In seedling roots of various crosses, this trait was mainly negatively correlated with *TaCKX5*, *8*, *10,* and *NAC2*.

Generally, negative correlations between the expression of *TaCKX2.1*, *2.2.2,* and *10* in spikes and TGW, seed number, seed yield, and spike number were correlated with higher yield, and positive correlations were correlated with lower yield. On the other hand, positive correlations between the expression of these genes and root mass determine a higher yield in F2. Higher yield in F<sup>2</sup> is also associated with balanced CKX enzyme activity in spikes and seedling roots.

A summary of the regulation of yield-related traits by *TaCKX* GFMs and *NAC2* in the high-yielding progeny of F<sup>2</sup> is presented in Figure 3.


**Figure 3.** Regulation of yield-related traits by *TaCKX* GFMs and *NAC2* in the high-yielding progeny of F<sup>2</sup> based on correlation coefficients. **Figure 3.** Regulation of yield-related traits by *TaCKX* GFMs and *NAC2* in the high-yielding progeny of F<sup>2</sup> based on correlation coefficients.

### **3. Discussion**

**3. Discussion** Common wheat is a very important cereal crop for feeding the world's population; therefore, continued improvement of the yield of this species is significant. *CKX* GFMs have already been documented to perform a pivotal role in determining yield-related traits in many plant species, including wheat [4,12]. The genes are tissue-specific; they encode cytokinin oxidase/dehydrogenase, the enzyme that irreversibly degrades cytokinins. We have already characterized the role of *TaCKX1* and *TaCKX2* in the regulation of yield traits in awnless and owned-spike cultivars [5–7]. The range of natural variation in the expression levels of most *TaCKX* genes among breeding lines and cultivars was very high, indicating the possibility of selecting beneficial genotypes for breeding purposes [8]. Common wheat is a very important cereal crop for feeding the world's population; therefore, continued improvement of the yield of this species is significant. *CKX* GFMs have already been documented to perform a pivotal role in determining yield-related traits in many plant species, including wheat [4,12]. The genes are tissue-specific; they encode cytokinin oxidase/dehydrogenase, the enzyme that irreversibly degrades cytokinins. We have already characterized the role of *TaCKX1* and *TaCKX2* in the regulation of yield traits in awnless and owned-spike cultivars [5–7]. The range of natural variation in the expression levels of most *TaCKX* genes among breeding lines and cultivars was very high, indicating the possibility of selecting beneficial genotypes for breeding purposes [8]. Therefore, we were interested in how the expression of these genes is inherited.

#### Therefore, we were interested in how the expression of these genes is inherited. *3.1. The Expression Patterns of Most TaCKX GFMs and TaNAC2-5A Are Mainly Inherited from 3.1. The Expression Patterns of Most TaCKX GFMs and TaNAC2-5A Are Mainly Inherited from the Paternal Parent*

*the Paternal Parent* Comparison of the expression patterns of most of the *TaCKX* GFMs and yield-related traits between parents and F<sup>2</sup> progeny in all reciprocal crosses tested indicated their inheritance from the paternal parent. This rule includes expression patterns in both tissues tested, 7 DAP spikes, and seedling roots, and all *TaCKX* GFMs and *TaNAC2-5A* tested were represented in different crosses. The exception was *TaCKX5* expressed in 7 DAP spikes, and *TaCKX3* expressed in seedling roots, for which the expression level in single crosses was inherited from the maternal parent. Furthermore, high or low yield was predominantly inherited from the paternal parent, and root mass was inherited from both parents or in one reciprocal cross from the maternal parent. We have not found such examples of inheritance in the literature; however, some deviations from parental additivity Comparison of the expression patterns of most of the *TaCKX* GFMs and yield-related traits between parents and F<sup>2</sup> progeny in all reciprocal crosses tested indicated their inheritance from the paternal parent. This rule includes expression patterns in both tissues tested, 7 DAP spikes, and seedling roots, and all *TaCKX* GFMs and *TaNAC2-5A* tested were represented in different crosses. The exception was *TaCKX5* expressed in 7 DAP spikes, and *TaCKX3* expressed in seedling roots, for which the expression level in single crosses was inherited from the maternal parent. Furthermore, high or low yield was predominantly inherited from the paternal parent, and root mass was inherited from both parents or in one reciprocal cross from the maternal parent. We have not found such examples of inheritance in the literature; however, some deviations from parental additivity of expression in polyploid plants were described [2]. An example of such non-additive gene expression takes place when the gene expression level in progeny is higher than that of one parent. The

expression level dominance of one parent, also called genomic imprinting, is epigenetic in origin and was investigated primarily at the molecular level in plants and animals [36,37]. The main regulators of gene imprinting are DNA and histone methylation asymmetries between parental genomes. Most of the imprinted genes in the endosperm of grains of different rice cultivars are imprinted across cultivars, and their functions are associated with the regulation of transcription, development, and signaling [39]. Imprinting might affect a single gene or a group of genes. Genes that showed conserved imprinting in cereals have been shown to reveal positive selection and were suggested to regulate seed development in a dose-dependent manner [38]. The only example of a paternally imprinted locus in maize is *ded1*, which encodes a transcription factor specifically expressed during early embryo development and activates early embryo genes that contribute to grain set and weight [47]. To our knowledge, there are no examples of paternally inherited expression patterns. According to Arabidopsis research, imprinted paternally expressed genes during seed development are mainly related to hypomethylated maternal alleles, repressed by small RNAs or less frequently with transposable elements [48,49]. Contrary to developmental epigenetics, in the case of transgenerational epigenetics, these epigenetic changes do not reset between generations, and this type of inheritance is more related to plants than animals (heritable changes in DNA methylation) [54]. Therefore, we suggest that this paternal inheritance of selected *TaCKX* GFMs is an effect of transgenerational epigenetic changes, not reset between generations. These heritable epigenetic changes might be effects of DNA methylation, repression of maternal alleles by small RNAs, transposable elements, or, most likely, transcription factors. From our in silico analysis and expression analysis (Iqbal et al., not published yet), several NAC transcription factors appear to strongly regulate the expression of *TaCKX* GFMs and *TaIPT* GFMs, influencing yield-related traits [24].

#### *3.2. Cooperation of TaCKX GFMs and TaNAC2-5A in the Determination of Yield-Related Traits*

The coordinated high or low level of expression of a few groups of genes in the paternal parent positively or negatively regulates higher or lower yield. In two reciprocal crosses, where both *TaCKX5* and *TACKX9* showed high expression in 7 DAP spikes of the paternal parent, the yield in the F<sup>2</sup> progeny was high and vice versa. In others, the high yield in the F<sup>2</sup> progeny was determined by a low level of expression of *TaCKX2.1* and *TaCKX11* in spikes of the paternal parent and high levels of their expression in the maternal parent. The level of expression of *TaNAC2-5A* in the paternal parent and/or F<sup>2</sup> was in opposition to *TaCKX5* and *9*; however, it was in agreement with *TaCKX11* and *TaCKX2.1,* suggesting their role in the regulation of transcription of these genes. In fact, it was proven by correlation analysis of its expression with yield-related traits [8]. Opposite cooperation of some of the genes in paternal spikes, which resulted in high or low yield in F2, has also been observed. The high level of expression of *TaCKX2.2.2* and the low level of expression of *TaCKX10* predominantly resulted in the low yield in F<sup>2</sup> progeny and vice versa.

Such common rules of gene expression in the paternal parent associated with yield in the paternal parent and F<sup>2</sup> progeny were also observed in the seedling roots. The high level of expression of *TaCKX3* and *TaCKX8* in the paternal parents of three reciprocal crosses resulted in high yield in the F<sup>2</sup> progeny and vice versa. The expression of *TaNAC2-5A* in spikes and seedling roots of high-yielding paternal parents and F<sup>2</sup> progeny showed a predominantly low expression level and inversely. These principles of paternal inheritance of selected *TaCKX* GFMs and *TaNAC2-5A* expression associated with high yield could be directly involved as molecular markers in high-yielding wheat breeding.
