High- or Low-Yielding F2 Progeny of Wheat Is Result of Specific TaCKX Gene Coexpression Patterns in Association with Grain Yield in Paternal Parent

Members of the TaCKX gene family (GFM) encode oxidase/dehydrogenase cytokinin degrading enzymes (CKX), which play an important role in the homeostasis of phytohormones, affecting wheat development and productivity. Therefore, the objective of this investigation was to test how the expression patterns of the yield-related TaCKX genes and TaNAC2-5A (NAC2) measured in 7 days after pollination (DAP) spikes and the seedling roots of parents are inherited to apply this knowledge in the breeding process. The expression patterns of these genes were compared between parents and their F2 progeny in crosses of one mother with different paterns of awnless cultivars and reciprocal crosses of awned and awnless lines. We showed that most of the genes tested in the 7 DAP spikes and seedling roots of the F2 progeny showed paternal expression patterns in crosses of awnless cultivars as well as reciprocal crosses of awned and awnless lines. Consequently, the values of grain yield in the F2 progeny were similar to the pater; however, the values of seedling root mass were similar to the mother or both parents. The correlation analysis of TaCKX GFMs and NAC2 in spikes and spikes per seedling roots reveals that the genes correlate with each other specifically with the pater and the F2 progeny or the mother and the F2 progeny, which shape phenotypic traits. The numbers of spikes and semi-empty spikes are mainly correlated with the specific coexpression of the TaCKX and NAC2 genes expressed in spikes or spikes per roots of the pater and F2 progeny. Variable regression analysis of grain yield and root mass with TaCKX GFMs and NAC2 expressed in the tested tissues of five crosses revealed a significant dependency of these parameters on the mother and F2 and/or the pater and F2 progeny. We showed that the inheritance of yield-related traits depends on the specific cooperative expression of some TaCKX GFMs, in some crosses coupled with NAC2, and is strongly dependent on the genotypes used for the crosses. Indications for parental selection in the breeding of high-yielding lines are discussed.

The cytokinin content in plant organs is regulated mainly by metabolic processes, as well as their transport [11][12][13].The predominant role in metabolic processes is played by the irreversible degradation of cytokinins by CKX enzymes, which can regulate rice grain yield [14][15][16][17].In wheat, CKX enzymes are encoded by the TaCKX GFM represented by 13 basic genes; however, 11 of them have their homoeologs in each of the three subgenomes of wheat, A, B, and D [18].A significant decrease in expression by silencing the HvCKX1 and TaCKX1 genes using RNAi technology in barley and wheat showed that a low level of expression of these genes determines high yield [19,20].Similarly, the silencing of the TaCKX2.1,TaCKX2.2.1, and TaCKX2.2.2 genes regulates yield-related traits in different ways, which are dependent on awned or awnless wheat spikes [21,22].The change in the expression level of one of the TaCKX genes resulted in a decrease or increase in the expression levels of other genes.Therefore, the silencing of one of the TaCKX genes resulted in a specific expression pattern of other TaCKX genes, which regulates the content of cytokinins and other phytohormones, as well as yield-related traits [20][21][22].There is a wide range of natural variability between the expression levels of the TaCKX genes and their patterns of coexpression in different cultivars and breeding lines [23].Since the TaCKX gene expression pattern is related to wheat yield parameters, it was interesting to check how this pattern is inherited.In our earlier research with TaCKX GFMs, we also included TaNAC2-5A, which encodes the NAC-type transcription factor [22][23][24][25].The TaNAC2-5A was found to increase wheat yield by controlling the nitrate response [26,27], and in our research, the gene was coexpressed with selected TaCKX GFMs and was correlated with several yield-related traits.
The most common way of inheritance of expression patterns, observed in diploid species, referred to as 'parental expression additivity' is the average of expression of the parental genes typically observed in diploid species [28].Exceptions from this scheme were mainly observed in the progeny of polyploid species, in which the expression level was similar to that of one of the parents or was lower or higher than in both parents, or unequal.Such nonadditive gene expression levels associated with phenotypic heterosis in F 1 plants have already been reported and reviewed [29][30][31][32][33].These deviations from the general role are the result of different factors, such as epigenetic regulation by transcription factors [34][35][36][37][38], the balance of gene dosage [39,40], small interfering RNAs (sRNA) [36,41], histone modifications [42], R-loop formation [43], or distally acting factors [31].Furthermore, noncoding RNAs have been described as regulators of the development of shoots and grains in barley [44] and dominant epigenetic regulators of early meiotic stages in wheat, ensuring reproductive success [45].Small noncoding RNAs were involved in photosynthesis, glycolysis, hormone biosynthesis, and cellular homeostasis; however, long noncoding RNAs increased the expression of nearby genes.
The first exception from the common way of inheritance of expression pattern is when the expression level is similar to that of one parent, as described by Yoo et al. [28]; expression level dominance is more commonly called genomic imprinting [42,[46][47][48][49][50].It takes place when genes adopt the parent-of-origin expression pattern.One of the functions of this phenomenon is the regulation of seed dormancy through epigenetic mechanisms in gametes [50].Genes with triple repressive marks H3K27me3/H3K9me2/CHGm remained stably imprinted.Some genes in cereals showed conserved imprinting associated with positive selection pressure [46].Such parent-of-origin gene expression could affect a single gene or a group of genes.The transcriptome-wide identification of allele-specific imprinting genes in the maize embryo and endosperm of three reciprocal crosses revealed their involvement in nutrient transport, signaling pathways, and the transcriptional regulation of kernel development [51].
Most imprinted genes were described as maternally expressed and inherited [52][53][54], suggesting their predominant role in early cereal grain development [55,56].There is rare evidence for paternally expressed imprinted genes.In Capsella, paternally imprinted genes allow for the overcoming of hybridization barriers [57].In maize, the imprinted dosage-effect defective1 (ded1) locus has been identified as a paternal regulator of seed size [58].Ded1 encodes one of the MYB transcription factors and is expressed specifically during early endosperm development, resulting in the repression of late grain-filling genes.
Most of the described epigenetic changes are regulated during plant development and are called developmental epigenetics.However, some of these changes in DNA methylation are stably inherited between generations, which is called transgenerational epigenetics [59], and most of them occur during seed formation [34].
All reports cited above concentrated on the molecular aspects of developmental and transgenerational epigenetics, usually up to the generation of F 1 .Since crossbreeding and selection are the basic steps in breeding, and the expression pattern of TaCKX genes regulates yield-related traits, it was important to check how the expression patterns of these genes together with yield-related traits are inherited.In our primary report on the transgenerational inheritance of agronomically important genes, we tested the expression patterns of TaCKX genes and TaNAC2-5A in segregating the F 2 generation of reciprocal crosses of polyploid wheat [24].We documented that some of them were paternally imprinted, together with the yield parameter.The research was conducted based on reciprocal crosses of selected, awnless cultivars.In this article, we continue research on the inheritance of patterns of expression of TaCKX genes and TaNAC2-5A between parents and the F 2 generation of crosses of one mother with different paterns of awnless cultivars and reciprocal crosses of awned and awnless lines.For the first time, variable regression analysis is used to find a significant dependence of TaCKX GFMs and NAC2 expressed in tested tissues on grain yield and root mass in the mother and F 2 and/or the pater and F 2 progeny.The relative values (related to mother = 1.0) of the expression profiles of the TaCKX family genes in 7 DAP spikes, seedling roots, and phenotypic characteristics of the mother S12B crossed with the pater S6C (C1) in the F 2 progeny are different from the profiles of the same traits in crosses of the same mother S12B crossed with another pater, S5C (C2) (Figure 1; the measured values of phenotypic traits are in Table S1).Similarly, crosses of another mother, S6C, with three different paterns, S3C (C3), S12B (C4), and S5C (C5), lead to different patterns of TaCKX expression and yield-related trait profiles (Figure S1).These differences are mainly influenced by the pater.For example, in S12B crossed with S6C, the high relative expression of TaCKX1 and NAC2 in the 7 DAP spikes of the pater is also observed in the F 2 progeny.Furthermore, the high expression of TaCKX5 and NAC2 and the low expression of TaCKX11 in the seedling roots of the S6C pater are also inherited in F 2 .Similarly, in S12B crossed with S5C, the high relative expression of TaCKX1 and 11 in spikes and TaCKX5 and NAC2 in pater seedling roots was observed in F 2 progeny.In both crosses, the activity of CKX in the roots of the pater and F 2 progeny was increased; however, the yield that included the grain number of the pater and F 2 progeny was decreased compared to the activity of CKX and the mother's yield.

Crosses of One
The expression profiles of the TaCKX genes in 7 DAP spikes, seedling roots, and the phenotypic traits of S6C crossed with three paterns, S3C (C3), S12B (C4*), and S5C (C5), are presented in Figure S1.Similarly to the expression results above, the patterns of most TaCKX genes and NAC2 in the spikes and roots of the F 2 progeny are more comparable to the pater.Unfortunately, the grain yield in the F 2 of each cross of the S6C as a mother with one of the three paterns was lower than in the mother and comparable to the pater, and this result was opposite to a much higher root mass.The high or low expression of the TaCKX GFM in 7 DAP spikes and seedling roots and the yield-related traits in the mother, pater, and F2 progeny of C1 to C5 crosses are presented in Table 1.The data of the F2 progeny show a similar tendency to increase or decrease as colored paterns are in red, and those similar to those of the mother are colored green.Most of the gene expression levels in both the 7 DAP spikes and the seedling roots, as well as the yield and CKX activity data, are comparable to the pater (red).For example, in the C3 cross, S6C (mother) showed a low level of expression of TaCKX5, 11, 9, and 10 and a higher level of expression of TaCKX2.1 and NAC2 in 7 DAP spikes, just opposite to S3C, which is the paternal component of this cross.In their F2 generation, TaCKX11 and 10 were highly expressed as in the pater.In the seedling roots of S6C, the expression of TaCKX5 and the expression of NAC2 were very low, and the expression of TaCKX3 was high, opposite to the pater, S3C; however, in F2, the expression level of TaCKX11 and 8 was very high and high, respectively, and the expression level of TaCKX3 was low, as in the pater.Similarly, in other crosses, the expression patterns of the TaCKX GFM and NAC2 in the spikes and roots of F2 were inherited from the pater.The F2 progeny in crosses of C1 and C2, where S12B was a better-yielding mother component, and C5, where S6C was the mother with a better yield, showed a decrease in yield comparable to the pater.The decrease in yield was in opposition to the increase in root mass in crosses of S6C as the mother with S3C, S12B, and S5C.However, in the first two crosses, the mass of the root in F2 was higher than in both parents, and in the third cross, it was comparable to the high The high or low expression of the TaCKX GFM in 7 DAP spikes and seedling roots and the yield-related traits in the mother, pater, and F 2 progeny of C1 to C5 crosses are presented in Table 1.The data of the F 2 progeny show a similar tendency to increase or decrease as colored paterns are in red, and those similar to those of the mother are colored green.Most of the gene expression levels in both the 7 DAP spikes and the seedling roots, as well as the yield and CKX activity data, are comparable to the pater (red).For example, in the C3 cross, S6C (mother) showed a low level of expression of TaCKX5, 11, 9, and 10 and a higher level of expression of TaCKX2.1 and NAC2 in 7 DAP spikes, just opposite to S3C, which is the paternal component of this cross.In their F 2 generation, TaCKX11 and 10 were highly expressed as in the pater.In the seedling roots of S6C, the expression of TaCKX5 and the expression of NAC2 were very low, and the expression of TaCKX3 was high, opposite to the pater, S3C; however, in F 2 , the expression level of TaCKX11 and 8 was very high and high, respectively, and the expression level of TaCKX3 was low, as in the pater.Similarly, in other crosses, the expression patterns of the TaCKX GFM and NAC2 in the spikes and roots of F 2 were inherited from the pater.The F 2 progeny in crosses of C1 and C2, where S12B was a better-yielding mother component, and C5, where S6C was the mother with a better yield, showed a decrease in yield comparable to the pater.The decrease in yield was in opposition to the increase in root mass in crosses of S6C as the mother with S3C, S12B, and S5C.However, in the first two crosses, the mass of the root in F 2 was higher than in both parents, and in the third cross, it was comparable to the high mass of roots in the pater.On the contrary, only in one C1 cross, the expression of TaCKX5 (together with TaCKX9) in the mother spike was higher than in the pater spike, and in the F 2 progeny, it was similar to that of the mother.Interestingly, the high expression of TaCKX5 in F 2 was accompanied by the high expression of TaCKX11 (and paternal TaCKX1 and NAC2).
Table 1.TaCKX GFMs and NAC2 expression levels in spikes and roots, and yield-related traits of mother (M), pater (P), and their F 2 siblings from crosses of S12B × S6C, S12B × S5C, S6C × S3C, S6C × S12B, and S6C × S5C.Colors of characters indicate similar expression patterns and yield-related traits in F 2 and pater (red) or in F 2 and mother (green).The positive or negative correlations between TaCKX GFMs and NAC2 expression in 7 DAP spikes and in seedling roots, CKX activity, and the yield-related traits of the mother and F 2 or the pater and F 2 in different crosses are illustrated in Table 2.The correlation coefficients are presented in Table S2.
The expression of TaCKX1 was positively correlated with the expression of TaCKX5 and 1 in the 7 DAP spikes of the mother and F 2 (Table 2, blue) but not in the pater and F 2 progeny.However, in the 7 DAP spikes of the pater and F 2 progeny, TaCKX1 was positively correlated with the expression of TaCKX2.1 and 2.2.2 (Table 2, green).Furthermore, the levels of expression of these genes (TaCKX2.1 and 2.2.2) were positively correlated with NAC2 only in P + F 2 .Another specific positive correlation of P + F 2 was between TaCKX2.2.2 and 10, and a negative correlation was between TaCKX2.1 and 9.In contrast, in M + F 2 , TaCKX2.2.2 was positively correlated with 9 or 11, TaCKX2.1 with 10, and TaCKX10 with 11 and NAC2.Others, such as TaCKX2.2.2 alone, TaCKX5 and 9 or TaCKX5 and 10 or TaCKX5 and 11, TaCKX10 and 11, and TaCKX2.1 with NAC2, are positively correlated in both M + F 2 and P + F 2 .
The expression data of TaCKX GFMs and NAC2 in spikes correlated with those tested in roots (TaCKX1, 3, 5, 8, 10, 11, and NAC2) are also different in both groups, M + F 2 and P + F 2 .Expressed in spikes, TaCKX1 was positively correlated, depending on the cross, with TaCKX1, and with TaCKX5 and 10 or negatively with TaCKX9 or TaCKX11 only in M + F 2 .Specific to P + F 2 , positive or negative correlations were between TaCKX5 and 5, TaCKX5 and 1, TaCKX9 and 1, TaCKX2.1 and 11, NAC2 and TaCKX3, and TaCKX11 and NAC2.The expression of TaCKX9, 10, and 11 in spikes was correlated with the expression of multiple TaCKX and NAC2 genes in roots.All three and some others are strongly correlated with TaCKX8 and 11 in M + F 2 or P + F 2 .For example, TaCKX9 was negatively correlated with TaCKX8 and 11 but positively correlated with TaCKX10 in the roots of both groups.
Among yield-related traits, PH and SN were positively correlated with CKX activity in the spikes of P + F 2 and M + F 2 (Table S2), and PH data were correlated with the expression of TaCKX1 and 11 or with TaCKX11 alone in the spikes of M + F 2 .Furthermore, PH was also positively correlated with spike TaCKX1 per root TaCKX5 and 10 and with spike TaCKX11 per root TaCKX8 or negatively with spike TaCKX11 per root TaCKX1 and 3 of M + F 2 .The ES trait was correlated with the expression of TACKX1 in M + F 2 exclusively.In another cross (M2 + P 2 ), ES together with SN was also correlated with TaCKX2.1 and NAC2 or SN with TaCKX2.2.2 in both M + F 2 and P + F 2 .The expressions in the spikes of M + F 2 , TaCKX2.1, and TaCKX10 with 11 and NAC2 were positively correlated with TGW.GY was negatively correlated with TaCKX11 and NAC2, and TaCKX5 and 9 in the M + F 2 of two crosses.Both PH and ES, as well as RM, GN, and GY, were correlated with the expression of selected TaCKX and NAC2 genes in group M + F 2 exclusively and with TGW predominantly.SN and SES were observed more frequently but not exclusively in the group of P + F 2 from different crosses.SN was positively correlated with TaCKX2.1 in P + F 2 , and both SN and SES were strongly negatively correlated with spike TaCKX9 per root TaCKX8 and 11 in both M + F 2 and P + F 2 or strongly negatively correlated with spike NAC2 per root TaCKX3 in P + F 2 .RM was positively correlated with the expression of TaCKX5, 11 in spikes (M + F 2 and P + F 2 ), negatively correlated with spike NAC2 per root NAC2 in P + F 2 , and positively correlated with NAC2 in M + F 2 .
The correlation coefficients between TaCKX and NAC2 expression, CKX activity, and yield-related traits are related to both parents or the mother or pater separately.In the first cross, the awned spike line representing the mother was crossed with the awnless line (C6).In the reciprocal cross, the awnless line was the mother component, and the awned line was the pater component (C7).The expression of the TaCKX GFM and NAC2 in 7 DAP spikes and seedling roots, as well as yield-related traits in the mother, pater, and their six F 2 progeny in C6 and C7 crosses, is presented in Figure 2. The awned mother line showed higher expression of almost all tested TaCKX genes, as well as lower yield-related parameters compared to the pater.The gene with the lowest expression level in pater spikes, related to the mother (=1.00), was TaCKX5 and then TaCKX10 and 1. TaCKX2.1 was at the same level in both parents.The expression of TaCKX5 in F 2 remained at the same very low level as in the pater; for TaCKX1, the expression level was similar or slightly higher than in the pater, and in the case of TaCKX10, it was several times higher compared to the pater.In seedling roots, the awnless pater showed several times higher expression of TaCKX1 and 8 and much lower expression of TaCKX5 and NAC2 than in the awned mother.This pattern of expression of TaCKX1, 8, 3 and TaCKX5 and NAC2 (much higher or much lower expression levels than in the mother, respectively) was observed in the F 2 progeny.Yield-related traits such as grain number and grain yield in four of the six F 2 progeny were similar to the pater and much higher than in the mother.The expression of the TaCKX GFM and NAC2 in 7 DAP spikes and seedling roots, as well as yield-related traits, in the awnless mother crossed with the awned pater (C7) were opposite to those of the C6 cross.The pattern of higher expression of TaCKX1, 5, 10, 11 in the 7 DAP spikes of the pater was transmitted to the F2 progeny.The very low expression of TaCKX1, 3 in the pater roots was also low in the F2 progeny.However, the very high expression of TaCKX5 and NAC2 was much lower in F2, and in the case of TaCKX5, it was even lower than in the mother.Very low parameters of grain number, grain yield, and TGW in the pater were in the F2 progeny slightly higher than in the pater or similar to the mother.

Reciprocal Crosses of Awned and Awnless Lines Showed That the Expression Patterns of Most TaCKX GFMs and Yield Are Mainly Inherited from Pater
TaCKX GFMs with high or low relative expression in 7 DAP spikes, seedling roots, and the parameters of yield-related traits in the mother, pater, and F2 progeny of reciprocal C6 and C7 crosses are presented in Table 3.
Very low expression of TaCKX5 in the 7 DAP spikes of the awnless, high-yielding pater (C6) was transmitted to the F2 progeny (red).Similar expression levels between the pater and F2 progeny in seedling roots were shown by highly expressed TaCKX1 and 8 and lowly expressed TaCKX5 and NAC2.The yielding parameters were very high in both the pater and F2 progeny.The expression of the TaCKX GFM and NAC2 in 7 DAP spikes and seedling roots, as well as yield-related traits, in the awnless mother crossed with the awned pater (C7) were opposite to those of the C6 cross.The pattern of higher expression of TaCKX1, 5, 10, 11 in the 7 DAP spikes of the pater was transmitted to the F 2 progeny.The very low expression of TaCKX1, 3 in the pater roots was also low in the F 2 progeny.However, the very high expression of TaCKX5 and NAC2 was much lower in F 2 , and in the case of TaCKX5, it was even lower than in the mother.Very low parameters of grain number, grain yield, and TGW in the pater were in the F 2 progeny slightly higher than in the pater or similar to the mother.

Reciprocal Crosses of Awned and Awnless Lines Showed That the Expression Patterns of Most TaCKX GFMs and Yield Are Mainly Inherited from Pater
TaCKX GFMs with high or low relative expression in 7 DAP spikes, seedling roots, and the parameters of yield-related traits in the mother, pater, and F 2 progeny of reciprocal C6 and C7 crosses are presented in Table 3. High (↑), very high (↑↑), extremally high (↑↑↑), the same (=), slightly low (=↓), the same, lower or higher (=↓↑), low (↓), very low (↓↓), or extremally low (↓↓↓) expression levels, CKX activity or yield-related traits.Relative expression levels: high-1.5-2higher than in mother; very high-above twice as high as in mother; extremally high-several times higher as in mother; low-below 0.4 than in mother.
Very low expression of TaCKX5 in the 7 DAP spikes of the awnless, high-yielding pater (C6) was transmitted to the F 2 progeny (red).Similar expression levels between the pater and F 2 progeny in seedling roots were shown by highly expressed TaCKX1 and 8 and lowly expressed TaCKX5 and NAC2.The yielding parameters were very high in both the pater and F 2 progeny.
In the C7 cross, where the pater was an awned component, the very high expression of TaCKX5 in the 7 DAP spikes of the pater was lower in the F 2 progeny but still higher than in the mother.Furthermore, the very high level of expression of TaCKX10 and the high level of TaCKX1 and 11 in the pater were similar in the F 2 progeny.The low expression levels of TaCKX1 and 8 in the roots of the awned pater were similar in the F 2 progeny; however, the low expression level of TaCKX5 was similar in the F 2 progeny to the mother.In this cross, the yield-related traits of the F 2 progeny of the very high-yielding awnless mother crossed with the very low-yielding awned pater were higher compared to the pater or on a level similar to that of the mother.The correlation between the TaCKX GFM and NAC2, as well as yield-related traits, in the groups of M + F 2 and P + F 2 of reciprocal crosses of awned and awnless parents is presented in Table 4.The correlation coefficients are presented in Table S3.

The Correlation between the Expression of TaCKX
The expression levels of TaCKX GFM and NAC2 in the spikes of M + F 2 correlate positively with the expression levels of each gene in the spikes of P + F 2 of the C6 and C7 crosses (Table 4, first row, and Table S3, yellow).The largest differences in the correlations of expression are between TaCKX GFM and NAC2 in the spikes and roots of M + F 2 and P + F 2 .Most genes expressed in the spikes of awned M + F 2 correlate with the TaCKX GFM and NAC2 in roots, starting from the positive correlations of spike TaCKX1 with root TaCKX5 and NAC2 through the negative correlations of spike NAC2 with root TaCKX1 and 8.The only negative correlation of spike TaCKX1 with root TaCKX10 was observed in both M + F 2 and P + F 2 (bold).Similarly, correlations between the TaCKX GFM and NAC2 in roots were observed mainly in awned M + F 2 ; however, the correlation between TaCKX1 and 5 in M + F 2 is negative, but in P + F 2 , it is positive.
Numerous correlations between TaCKX GFM and NAC2 expression were observed in the spikes and roots of M + F 2 and P + F 2 of the reciprocal C7 cross, where M was awnless, and P was awned (Table 4).Many of them, such as the positive correlations between TaCKX1 in spikes and TaCKX5 in roots and TaCKX5, 9, 10, 11, and NAC2 in spikes and TaCKX5 in roots, represented both awnless M + F 2 and awned P + F 2 .However, many correlations between TaCKX2.1, 5, 9, 10, 11, and NAC2 in spikes and TaCKX8 in roots were positive in M + F 2 but negative in P + F 2 , and the correlation between TaCKX2.2.2 and TaCKX11 was negative in both groups.However, the correlations between other TaCKX genes in the spike and the root differed in these two groups and were more frequent in P + F 2 , where, similar to C6, P was the awned parent.
The correlations between yield-related traits and the expression of the TaCKX GFM and NAC2 in spikes and roots in the two groups of the M + F 2 and P + F 2 of C6 and C7 crosses are presented in Table 5. Numerous correlations between yield-related traits and TaCKX GFM and NAC2 expression were observed in both spikes and roots in the awned M + F 2 or awned P + F 2 of the C6 or C7 cross, respectively.On the contrary, in the C6 group of awnless P + F 2 and the C7 group of awnless M + F 2 , there were only a few correlations between yield-related traits and the TaCKX GFM and NAC2 in spikes: negative between root weight and TaCKX1, 2.1, 2.2.2 in C6 P + F 2 , which was the same in M + F 2 ; negative between the semi-empty spike number and TaCKX2.1,2.2.2, 10, and NAC2 and positive between spike length and TaCKX1, 9, 11, and NAC2 (C6) or positive between spike length and TaCKX1, 2.1, 5, 9, 10, 11, and NAC2 (C7).Furthermore, in the awnless P + F 2 of the C6 group, the spike number was correlated with the NAC2 expressed in roots, and in the awnless M + F 2 , spike length was correlated with the root TaCKX5, 8.

Stepwise Regression Analysis of Grain Yield and Root Mass with TaCKX GFMs and NAC2 Expression
The map of significant dependent variable regression of grain yield and root mass with TaCKX GFMs and NAC2 in the mother and F 2 and the pater and F 2 in different crosses is presented in Figure 3.

Stepwise Regression Analysis of Grain Yield and Root Mass with TaCKX GFMs and NAC2 Expression
The map of significant dependent variable regression of grain yield and root mass with TaCKX GFMs and NAC2 in the mother and F2 and the pater and F2 in different crosses is presented in Figure 3.The regression of grain yield and root mass was generally significant for a pair or three of genes in the mother and F 2 and the pater and F 2 in crosses of awnless cultivars (C1-C5).The most frequent for grain yield in 7 DAP spikes was the positive regression of TaCKX2.1 (0.92) and negative TaCKX1 (−0.55); the positive regression of TaCKX2.1 (0.82) and negative TaCKX9 (−0.56); the positive regression of TaCKX2.1 (0.79) coupled with the positive regression of TaCKX3 in roots (0.51) and negative with NAC2 (−0.29).In the roots of the same cross, the negative regression of NAC2 (−0.89) was coupled with the negative regression of TaCKX3 (−0.42) and with the negative regression of TaCKX2.1 (−0.28) in spikes.Furthermore, the negative regression of NAC2 (−0.82) in the roots of M + F 2 was coupled with the negative regression of root TaCKX1 (−0.40) and spike TaCKX2.2.2 (−0.56).However, the highest negative regression was for NAC2 (−1.46) in the spikes of the pater and F 2 coupled with root TaCKX10 (−0.83).In the C2 cross, the negative regression of the spike NAC2 was coupled with TaCKX8 in the roots of the mother and F 2 ; however, in the pater and F 2 , negative regression of the root NAC2 (−0.55) was coupled with the negative regression of the root TaCKX5 (−0.73).In the same group of the pater and F 2 of the C2 cross, the negative high regression of the root TaCKX8 (−1.03) was combined with the positive regression of the spike TaCKX11 (0.46) and TaCKX1 (0.21) and the negative regression of the spike TaCKX2.2.2 (−0.37).In the C4 cross, the positive regression of the spike TaCKX2.1 was coupled with the positive regression of the root TaCKX8 but only in the pater and F 2 .The only regression coefficients for the grain yield of the C3 and C5 crosses were for the root TaCKX5 (−0.23) coupled with the spike TaCKX2.2.2 (−0.15) and for the root TaCKX10 (+0.34) coupled with the spike TaCKX10 (−0.09) in the mother and pater and F 2 .
The regression coefficients of grain yield with TaCKX GFMs and NAC2 in the reciprocal crosses of awned and awnless cultivars were significant only in the pater and F 2 when the pater was an awned component (C7).In this cross, grain yield showed positive regression with root TaCKX11 (0.99) coupled with the negative regression of root TaCKX3 (−0.28) and spike TaCKX9 (−0.50).The mass of the root showed strong, negative regression coefficients with spike TaCKX2.2.2 (−1.15) positively coupled with root TaCKX9 (0.60) in the awned mother and F 2 of the C6 cross.In our previous research [24], it was documented for the first time that the expression levels of most of the yield-related TaCKX genes, TaNAC2-5A, and grain yield were inherited in F 2 from the paternal parent.The experiment was carried out using reciprocal crosses of awnless cultivars.This unexpected way of inheritance was proved in the present research based on other crosses: crosses of one mother with different parents of awnless cultivars and reciprocal crosses of awned and awnless lines.Interestingly, the yield values in F 2 were similar to those of the pater as well.Therefore, presented by us in our previous research, the paternal pattern of inheritance of yield-related TaCKX gene expression and the yield in F 2 generation was the first example of transgenerational paternal inheritance of these traits, and this research is a valuable confirmation of these data, including awned and awnless genotypes.
These different crosses of one mother with different paterns or reciprocal crosses of awned and awnless lines led to different patterns of the coexpression of TaCKX family genes, which together regulate yield-related traits in F 2 .The genotype dependency of specific, cooperative expression on yield-related traits was also very distinct in the silencing experiments of TaCKX1 and TaCKX2 genes in two cultivars, awnless Kontesa and awned Ostka [20][21][22].The decreased silencing expression of one TaCKX gene mediates the decreased or increased expression of other TaCKX genes, regulating phytohormone content and yield-related traits in different ways.For example, in TaCKX1 lines of the awnless wheat cultivar silenced by RNAi, the expression of TaCKX11 was significantly decreased, but the expression of TaCKX2 genes was significantly increased, resulting in a change in the content of cytokinins and other phytohormones and the obtaining of the wheat phenotype with a significantly increased spike and grain number but a decrease in TGW [20].Interestingly, a similar pattern of expression of TaCKX genes in the silenced TaCKX1 lines of awned cultivars regulated the content of cytokinins and other phytohormones differently, resulting in a significant increase in TGW and seedling root mass [22].Similarly, the expression patterns in different crosses are regulated by the cooperation of the TaCKX genes in the same positive way and, for some, in the opposite way, and the obtained phenotype is a result of this cooperation.However, in all crosses of this research and a previous one [24], high-yielding F 2 progeny was obtained, when a low-yielding mother was crossed with a high-yielding pater, suggesting that groups of paternally inherited TaCKX genes determinate, similar to the pater, high or low yield in F 2 .

Are Transcription Factors the Main Epigenetic Regulators in Wheat?
As reviewed in the introduction, any deviations from the non-Mendelian inheritance of a parent-of-origin expression pattern might be the effect of epigenetic regulation.The main factors of this epigenetic regulation are the balance of gene dosage [28], small interfering RNAs [36,41], noncoding RNAs [44,45], or cisand/or trans-regulatory elements [34][35][36][37][38].All of them might be involved in the regulation of these epigenetic types of expression patterns, especially in polyploid species.Noncoding RNAs have been described as the dominant epigenetic regulators of early meiotic stages in wheat, involved in photosynthesis, glycolysis, hormone biosynthesis, and cellular homeostasis [45], and in barley, they were found to regulate the development of shoots and grains [44].Wheat TaNAC transcription factors are involved in the cis-regulation of selected TaCKX family genes [25].Interestingly, one TaNAC can be involved in the regulation of the transcription of two to three TaCKX genes.For example, TaNAC J-1 and TaNAC94 are expected to regulate TaCKX1, 2.1, and 5; NAC13a was found to regulate TaCKX2.2.1 and 10; TaNAC Br-1 was identified as the regulator of TaCKX2.2.1, 9 and 11; and TaNAC6D binds to the cis-regulatory region of TaCKX10.We need to conduct more research to find which of these or other factors can be dominantly involved in the paternal inheritance of the expression pattern of TaCKX GFMs combined with yield.

TaCKX5 Plays a Major Role in Coexpression with Other TaCKX GFMs and TaNAC2-5A
As presented in this research, grain yield was inherited in F 2 progeny as in the pater, regardless of awned or awnless spike cultivars.The same was documented in our previous research with awnless wheat cultivars [24].It is difficult to indicate the common pattern of coexpression of TaCKX genes, which should be represented in the high-yielding pater to transmit this pattern of expression together with the yield to F 2 generation.In the C6 cross of the awned × awnless line, the strong down-regulation of TaCKX5 and 10 and the down-regulation of TaCKX1 and 11 in the 7 DAP spikes of a high-yielding pater resulted in high-yielding F 2 .The coexpression of these genes is in agreement with silencing experiments in awned and awnless cultivars [21,22].The down-regulation of TaCKX1 expression was coordinated with the down-regulation of TaCKX11 expression in both cultivars, as well as the decreased expression of TaCKX5 in the awned, which resulted in an increased TGW, root mass, and grain yield in the awned [22].However, in the case of the high level of coexpression of TaCKX5 with TaCKX9 and the decreased coexpression of TaCKX1, 2.1, and NAC2 of the high-yielding pater with the low-yielding mother in C4, the yield of the progeny in F 2 was between both parents, and the mass of seedling roots was similar to the mother.These results demonstrate the importance of TaCKX5 coexpression with others.This is reasonable since we found the highest expression of the TaCKX5 gene among other TaCKX GFMs in inflorescences and seedling roots and very high in 0 DAP spikes and leaves [60].As mentioned above, this gene might also be regulated by at least two TaNAC transcription factors [25].

Simultaneous Paternal Inheritance of the Expression Pattern of the TaCKX Genes and
TaNAC2-5A with Grain Yield Is Not the Rule for Seedling Root Mass and Other Yield-Related Traits Similarly to 7 DAP spikes, the expression pattern of the TaCKX genes and NAC2 was inherited in the seedling roots of F 2 as in the pater.However, unlike paternally inherited grain yield, seedling root mass was inherited in the F 2 progeny similar to both parents or with values ranging between the two parents, or in two crosses (C3, C4) from the same S6C mother.In both crosses, the same pattern of high expression of TaCKX8, 10, and 11 and decreased TaCKX3 appeared in the roots of the pater and in the F 2 progeny.Therefore, paternally inherited expression patterns in spikes and seedling roots influence grain yield; however, the mass of seedling roots is inherited from the mother or both parents.All these genes (TaCKX3, 8, 10, 11), highly expressed in seedling roots, are also highly expressed in 0 DAP spikes and TaCKX11 in inflorescences [60]; therefore, their coexpression in seedling roots influences grain yield.
The inheritance of coexpression patterns of TaCKX GFMs and NAC2, grain yield, and the mass of seedling roots could be explained based on the correlation coefficients of their expression in the pater and F 2 compared to the mother and F 2 .TaCKX1, 2.1, and 2.2.2 and NAC2, as well as TaCKX 2.2.2 with 10, and CKX2.1 with 9 correlated with each other in 7 DAP spikes only in the pater and F 2 progeny (not in the mother and F 2 progeny) in tested crosses.However, some of them, like TaCKX1 and 5, TaCKX10, 11, and NAC2, TaCKX2.1 and 10, and TaCKX2.2.2 and 9 or 11 correlated with each other only in the mother and F 2 .Similar specific correlations of TaCKX GFMs and NAC2 expressed in spike per these expressed in roots were documented for both groups.There are also a few examples of the same genes or gene pairs correlating with each other in both groups, the mother and F 2 and the pater and F 2 .Generally, most yield-related traits, such as grain number, spike length, plant height, grain yield, and root mass and the most frequent TGW, were correlated with TaCKX GFMs and NAC2 in spikes or spike per root in the mother and F 2 group; however, some very important traits for total grain yield like spike number and semi-empty spike number were mainly correlated with selected spike or spike per root TaCKX GFMs and NAC2 in the pater and F 2 population.Therefore, depending on the crossed genotypes, the correlations between yield-related traits in the mother and F 2 and the pater and F 2 depended on the specific expression or more frequently the coexpression of several TaCKX and NAC2 genes in 7 DAP spikes, as well as the seedling roots of the mother and F 2 and the pater and F 2 .The map of dependent variable regression analysis of grain yield or root mass and the TaCKX GFM and NAC2 expressed in the spikes and seedling roots of different crosses revealed a significant dependency of these parameters in the mother and F 2 or in the pater and F 2 .Both traits were strongly dependent positively or negatively on groups of two to four tested genes expressed in spikes or seedling roots, specifically to both groups (mother and F 2 or pater and F 2 ).For example, grain yield in C1 is highly positively dependent on the spike TaCKX2.1 and negatively on the spike TaCKX1, or the spike TaCKX9, or positively on the root TaCKX3, or negatively on the root NAC2, specific to the mother and F 2 progeny; however, in the pater and F 2 of the same cross, this trait was highly negatively dependent on the spike's NAC2 and the root's TaCKX10.These parameters of the dependent variable regression were specific to the mother and F 2 and/or the pater and F 2 of each cross.Among the most frequent are the positive regression of grain yield and the spike TaCKX2.1 and the negative regression of the grain yield and the spike or root NAC2.TaCKX2.1 was isolated and characterized as a gene related to the grain number per spike by Zhang et al. [61].In recombinant inbred lines, TaCKX6a [62], then renamed TaCKX2.1-3D[18], showed significant correlations with grain size, weight, and grain filling rate.However, in our investigation, the importance of positive or negative coregulation of TaCKX2.1 with other TaCKX GFMs on yield-related traits was highlighted.In the silencing experiment, the down-regulation of wheat TaCKX1 resulted in a strong down-regulation of TaCKX1 and 11 and up-regulation of TaCKX2.1 and others in both awnless and awned cultivars; however, it affected different yield-related traits, and only in the awned one, it resulted in a highyielding phenotype [20,22].Similarly, the spike TaCKX2.1, which is positively correlated with grain number, grain yield, spike number, spike length, and root mass, was coupled with other TaCKX GFMs, and in the case of grain yield, it was negatively correlated with the spike and root TaCKX1 in the high-yielding F 2 progeny [24].In addition to this specific coregulation, TaCKX1, 2.1, and 5 could be regulated by JUNGBRUNNEN 1-like TF, renamed TaNACJ-1 and TaNAC94 TF [25].Based on gene ontology analysis, the TaNACJ-1 takes part in the negative regulation of leaf senescence.This is also in agreement with the silencing experiments of the TaCKX2 genes, which significantly increased chlorophyll content in the flag leaves of awned and awnless cultivars [21,22].The second one, TaNAC94, can be involved in response to auxins, the positive regulation of asymmetric cell division, somatic stem cell division, root cap development, etc. [25]-traits that influence grain yield.And again, the silencing of TaCKX2 genes in both awned and awnless cultivars affected not only the contents of cytokinins but also auxins, however, in different ways.The IAA content along with the active cytokinin content in the awnless cultivar was increased [21], but in the awned cultivar, the pattern of cytokinin content was different from the awnless one, and the IAA content was decreased [22].Therefore, selected TaNACs could regulate the transcription of TaCKX2 genes that influence phytohormone content and yield-related traits.

Regression Analysis
Another, the most dependent on grain yield in two crosses of one mother with two paterns, is TaNAC2-5A.This gene is specifically or not specifically correlated with others expressed in the spikes or spikes per roots TaCKX GFMs of all crosses and showed negative regression with grain yield and root mass, coupling positively or negatively with other TaCKX GFMs.The TaNAC2-5A belongs to the large family of NAC genes, which encode NAC-type transcription factors that are involved in the regulation of important agronomic traits [26,38,63,64].The overexpression of the TaNAC2-5A enhanced root growth and increased the ability of the root to acquire nitrogen and, under field conditions, increased nitrate uptake and grain yield [26].However, in a controlled environment, this gene is positively correlated with the activity of the CKX enzyme in seedling roots and negatively with tiller number [23], was expressed in roots together with TaCKX3 and 8, and was negatively correlated with root mass [24].Here, the spike TaNAC2-5A together with the spike TaCKX11 in the mother and F 2 was negatively correlated with root mass, spike number, and grain yield, however, positively with TGW; but in the case of a negative correlation of spike TaNAC2-5A with root TaCKX3 or root TaNAC2-5A in the pater and F 2 , negative correlations with the spike number and semi-empty spikes were observed.In summary, the expression of TaNAC2-5A in spikes and/or roots is coregulated by other genes from TaCKX GFMs.This coregulation is not direct, since we did not find TaNAC2-5A TF binding sites in the cis-regulatory sequences of TaCKX GFMs; however, the TF binding sites of the other five TaNACs were identified [25].As reported by Li et al. [27], the NAC TF of TaNAC2-5A binds directly to the promoter of the nitrate transporter gene, TaNRT2.5-3B,playing a key role in seed vigor.Another NAC2, OsNAC2 in rice, which can regulate the expression of auxin-and cytokinin-responsive genes, was shown to be an integrator of auxin and cytokinin pathways, playing a role in modulating root development [65], and through the ABA pathway delayed the germination of seeds [66].

Growing Conditions and Crossbreeding
The parent plants and the F 2 plants were grown in the same growth chamber at the same time, to provide the same, controlled environment.Temperatures were maintained at 20 • C during the day and 18 • C at night, with a day/night cycle of 16 h of light followed by 8 h of darkness.The intensity of light was 350 µmol•s −1 •m 2 .The plants were watered three times a week and fertilized once a week with Florovit, following the manufacturer's guidelines.
The experimental tissue samples were collected from parental lines (3 plants per parent) and their six F 2 progeny using the same methods as described in Szala et al. [24].There were roots from 5-day-old seedlings, cut 0.5 cm from the base before replanting, and spikes from the same plants 7 days after pollination.All samples were taken at 9:00 a.m. and kept in freezer in liquid nitrogen at −80 • C until needed.
The crossbreeding was performed like in Szala et al. [24].

RNA Extraction, cDNA Synthesis, and RT-qPCR
Total RNA was extracted from the collected samples using TRI reagent according to the manufacturer's instructions.RNA concentration and quality were determined according to Szala et al. [24].High-quality RNA was used for cDNA synthesis using the RevertAid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Vilnius, Lithuania).RT-qPCR assays were carried out for 10 genes, TaCKX1, TaCKX2.1,TaCKX2.2.2, TaCKX3, TaCKX5, TaCKX8, TaCKX9, TaCKX10, TaCKX11, and TaNAC2-5A, and all reactions were carried out in triplicate on a Rotor Gene Q thermal cycler (QIAGEN, Hilden, Germany) using HOT FIREPol EvaGreen qPCR Mix Plus (Solis BioDyne, Tartu, Estonia).Reaction conditions and pairs of primers for the genes studied were the same as in the previous publication and in Table S3.The expression of TaCKX genes was calculated using ADP-ribosylation factor as a normalizer.

Analysis of CKX Activity
CKX enzyme activity was measured on the same samples used for analysis of TaCKX gene expression according to the procedure developed by Frebort et al. [67] and was optimized for wheat tissues according to Szala et al. [24].The procedure involved the extraction of plant material, incubation in a reaction mixture, and measurement of the concentration of the product.Total protein concentration was approximated by referring to the standard graph created using bovine serum albumin (BSA), following the Bradford method, as outlined by Bradford and Williams [68].

Measurement of Yield-Related Traits
The following yield-related traits were measured: the height of the plant, number of spikes, number of partially empty spikes, number of tillers, length of the spike, yield of grains, number of grains, weight of 1000 grains (TGW), and weight of 5-day-old seedling roots.

Statistical Analysis
For statistical analysis, Statistica version 13 software (TIBCO Software Inc., Palo Alto, Santa Clara, CA, USA) was utilized.Changes in significance were assessed through ANOVA variance analysis followed by the least significant difference (LSD) post hoc test.The correlation coefficients were calculated using parametric correlation matrices (Pearson test) or nonparametric correlation analysis (Spearman test).Progressive stepwise regression was calculated.

Conclusions
Expression patterns in spikes and seedling roots, as well as grain yield, in the F 2 progeny are inherited like in the pater, while the mass of seedling roots is inherited from the mother or both parents.However, particular yield-related traits are regulated by specific, cooperative expression of a few TaCKX GFMs and in some crosses with NAC2.Both spike number and semi-empty spike number are mainly correlated with the specific coexpression of TaCKX and NAC2 genes expressed in spikes or spikes per roots of the pater and F 2 progeny, suggesting that these traits from the parent site are the main factors influencing grain yield.Regression analysis showed a strong dependence of grain yield or root mass on the coexpression of TaCKX genes and NAC2 in the mother or pater, depending on the cross.Therefore, this specific cooperative expression is also very strongly dependent on the genotype.Parents and the F 2 progeny of each cross used to have their own expression and gene cooperation pattern that influenced the traits in F 2 .Interestingly, in reciprocal crosses of awned and awnless lines and their F 2 progeny, the predominant role was played by the awned component, regardless of whether it was the mother or the pater.Also, in these crosses, the grain yield was inherited after the pater.All of these data indicate that the pater component, which is selected for breeding, should be characterized by a specific TaCKX expression pattern and higher yield compared to the mother component.
Mother with Different Paterns 2.1.1.Crosses of One Mother with Different Paterns Lead to Different Patterns of TaCKX Expression and Data of Yield-Related Traits in F 2

Figure 1 .
Figure 1.TaCKX GFM and NAC2 expression patterns in 7 DAP spikes, seedling roots, and phenotypic traits in mother, pater, and their six F2 progeny, from crosses of S12B × S6C (C1*) and S12B × S5C (C2).Data represent mean values with standard deviation and are related to mother set as 1.00.Black and red asterisks indicate statistical significance compared to maternal parent or paternal parent, respectively (* 0.05 > p ≥ 0.01, ** 0.01 > p ≥ 0.001, *** p < 0.001).Data for C1 cross were already presented in Szala et al. [24], where S12B was component of reciprocal cross.2.1.2.Crosses of One Mother with Different Paterns Show That the Expression Patterns of Most of the TaCKX GFM and the Yield Are Mainly Inherited from the Pater

Figure 1 .
Figure 1.TaCKX GFM and NAC2 expression patterns in 7 DAP spikes, seedling roots, and phenotypic traits in mother, pater, and their six F 2 progeny, from crosses of S12B × S6C (C1*) and S12B × S5C (C2).Data represent mean values with standard deviation and are related to mother set as 1.00.Black and red asterisks indicate statistical significance compared to maternal parent or paternal parent, respectively (* 0.05 > p ≥ 0.01, ** 0.01 > p ≥ 0.001, *** p < 0.001).Data for C1 cross were already presented in Szala et al. [24], where S12B was component of reciprocal cross.2.1.2.Crosses of One Mother with Different Paterns Show That the Expression Patterns of Most of the TaCKX GFM and the Yield Are Mainly Inherited from the Pater
GFM and NAC2, as Well as the Yield-Related Traits, in Reciprocal Crosses of Awned and Awnless Parents and Their F 2 Progeny Indicates the Predominant Role of the Awned Component

Figure 3 .
Figure 3. Map of dependent variable regression of grain yield and root mass with TaCKX GFMs and NAC2 in mother and F 2 and pater and F 2 of different crosses.

1 .
The Expression Patterns of the TaCKX Genes, TaNAC2-5A, and Grain Yield Are Inherited from the Paternal Parent and Are Genotype-Dependent Proved a Significant Dependence of the Specific Expression Pattern on the Parameters of Yield-Related Traits in Mother and F 2 or in Pater and F 2
root ↑↑ CKX act.root ↑ CKX act.root ↓ CKX act.root = 2.1.3.How Expression Levels of TaCKX GFMs and NAC2, CKX Activity in 7 DAP Spikes and Roots, and Yield-Related Traits of the Mother and the Pater Were Correlated with These Traits in F 2

Table 2 .
Positive (+) or negative (−) correlations between TaCKX GFMs and NAC2 expression in spikes, in spikes and roots, and yield-related traits of mother (M) + F 2 and pater (P) + F 2 from different crosses.

Table 3 .
TaCKX GFM and NAC2 expression in spikes and roots, and yield-related traits of mother (M), pater (P), and their F 2 progeny from crosses of awned × awnless parent lines (C6) and awnless × awned parent lines (C7).Character colors indicate similar expression patterns and yield-related traits in pater and F 2 progeny (red) or in mother and F 2 progeny (green).

Table 4 .
Positive (+) or negative (−) correlations between the expression of TaCKX GFM and NAC2, as well as yield-related traits, in the groups of M + F 2 and P + F 2 of reciprocal crosses of parents: awned × awnless (C6) and awnless × awned (C7).The same correlations between genes in M + F 2 and P + F 2 are in bold.The opposing correlations are colored red.

Table 5 .
Positive (+) and negative (−) correlations between yield-related traits and TaCKX GFM and NAC2 expression in spikes and roots in two groups of M + F 2 and P + F 2 of reciprocal crosses of awned and awnless parents (C6, C7).Divergent correlations are colored red; same correlations are highlighted by gray; lc-lack of correlation with TaCKX and NAC2 genes.