Genome-Wide Identification of the MAPK and MAPKK Gene Families in Response to Cold Stress in Prunus mume

Protein kinases of the MAPK cascade family (MAPKKK–MAPKK–MAPK) play an essential role in plant stress response and hormone signal transduction. However, their role in the cold hardiness of Prunus mume (Mei), a class of ornamental woody plant, remains unclear. In this study, we use bioinformatic approaches to assess and analyze two related protein kinase families, namely, MAP kinases (MPKs) and MAPK kinases (MKKs), in wild P. mume and its variety P. mume var. tortuosa. We identify 11 PmMPK and 7 PmMKK genes in the former species and 12 PmvMPK and 7 PmvMKK genes in the latter species, and we investigate whether and how these gene families contribute to cold stress responses. Members of the MPK and MKK gene families located on seven and four chromosomes of both species are free of tandem duplication. Four, three, and one segment duplication events are exhibited in PmMPK, PmvMPK, and PmMKK, respectively, suggesting that segment duplications play an essential role in the expansion and evolution of P. mume and its gene variety. Moreover, synteny analysis suggests that most MPK and MKK genes have similar origins and involved similar evolutionary processes in P. mume and its variety. A cis-acting regulatory element analysis shows that MPK and MKK genes may function in P. mume and its variety’s development, modulating processes such as light response, anaerobic induction, and abscisic acid response as well as responses to a variety of stresses, such as low temperature and drought. Most PmMPKs and PmMKKs exhibited tissue-specifific expression patterns, as well as time-specific expression patterns that protect them through cold. In a low-temperature treatment experiment with the cold-tolerant cultivar P. mume ‘Songchun’ and the cold-sensitive cultivar ‘Lve’, we find that almost all PmMPK and PmMKK genes, especially PmMPK3/5/6/20 and PmMKK2/3/6, dramatically respond to cold stress as treatment duration increases. This study introduces the possibility that these family members contribute to P. mume’s cold stress response. Further investigation is warranted to understand the mechanistic functions of MAPK and MAPKK proteins in P. mume development and response to cold stress.


Introduction
During growth and development, plants are continuously stimulated by mixed signals from internal and external environments. To adapt to changes in the external environment, they have gradually formed complex and delicate signal transduction mechanisms [1][2][3][4][5][6][7][8][9]. The mitogen-activated protein kinase (MAPK or MPKs) cascade pathway is a highly conserved signaling module in eukaryotes that links different extracellular stimuli to a broad range of intracellular responses through a cascade of phosphorylation reactions, which are thought to play vital roles in a variety of biological processes [10][11][12]. The MAPK cascade contains

Identification of MPK and MKK Gene Family Members
A total of 11, 12, and 69 MPKs were identified in P. mume, P. mume var. tortuosa genome, and the other five Rosaceae species (including 12 in P. armeniaca, 10 in P. persica, 13 in P. salicina, 22 in M. domestica, and 12 in R. chinensis), respectively. Furthermore, 7 MKKs were detected in P. mume, 7 were detected in P. mume var. tortuosa genome, and 36 were detected in the other five Rosaceae species (including 7 in P. armeniaca, 9 in P. persica, 5 in

Identification of MPK and MKK Gene Family Members
A total of 11, 12, and 69 MPKs were identified in P. mume, P. mume var. tortuosa genome, and the other five Rosaceae species (including 12 in P. armeniaca, 10 in P. persica, 13 in P. salicina, 22 in M. domestica, and 12 in R. chinensis), respectively. Furthermore, 7 MKKs were detected in P. mume, 7 were detected in P. mume var. tortuosa genome, and 36 were detected in the other five Rosaceae species (including 7 in P. armeniaca, 9 in P. persica, 5 in P. salicina, 8 in M. domestica, and 7 in R. chinensis) (sequence details are shown in File S1). Each newly detected MPK and MKK gene was given a name based on its similarity to Arabidopsis MPK and MKK proteins (Tables 1, S1 and S2). The number of amino acids, molecular weight (MW), and isoelectric point (pI) were computed based on the identified MPK and MKK proteins' sequences. As shown in Table 1, the predicted PmMPK proteins in P. mume ranged in length from 368 (PmMPK7) to 818 (PmMPK8) amino acids with relative molecular weights of 42.39 kDa (PmMPK7) to 92.92 kDa (PmMPK8) and theoretical pIs of 5.18 (PmMPK13) to 9.31 (PmMPK19). Moreover, 7 PmMKKs were predicted to encode 324-518 aa, with MW ranging from 36.17-57.79 kDa and pIs from 5. 36-8.04. In P. mume var. tortuosa, 12 PmvMPKs were predicted to encode 370-1011 aa, with MW ranging from 42.58-115.17 kDa and pIs from 5. 18-9.29. Furthermore, 7 PmvMKKs were predicted to encode 321-518 aa, with MW ranging from 35.95-57.77 kDa and pIs from 5. 51-7.58. It can be seen that the number of amino acids, molecular weight, and isoelectric point of P. mume var. tortuosa are relatively similar to those of wild P. mume, and only one gene, PmvMPK7, had several indices higher than the others. Subcellular localization was predicted, showing that MPKs and MKKs in P. mume and P. mume var. tortuosa are located in the nucleus with the exception of PmvMPK7, which may be present in the cell membrane.

MPK and MKK Genes' Phylogenetic Analysis and Classification
To comprehend how homologous MPK and MKK genes have evolved, we established phylogenetic trees of all MPK and MKK sequences of A. thaliana (model dicots), O. sativa (model monocots), P. mume, and P. mume var. tortuosa using the ML method. According to AtMPK and AtMKK reported in earlier studies [14], MPK and MKK in P. mume and P. mume var. tortuosa are classified into four clades (i.e., Clade A, B, C, and D) (Figures S1 and S2). In order to examine the evolutionary relationship between MPK and MKK in P. mume and P. mume var. tortuosa and Rosaceae species, an ML phylogeny was constructed using nine species, including five other Rosaceae species. The MPK and MKK gene family members from the nine species were divided into four clades ( Figure 2A). Clade D had the highest number of the MPK gene, comprising 4 PmMPKs, 5 PmvMPKs, 8 AtMPKs, 11 OsMPKs, and 27 Rosaceae MPKs. The number of MPK genes found in the evolutionary clades of A, B, and C did not differ significantly. Clades A, B, and C had two, three, and two PmMPKs and PmvMPKs, respectively ( Figure 2A and Table S3). Clade D also had the highest numbers of the MKK gene, comprising 4 PmMKKs, 4 PmvMKKs, 4 AtMKKs, 3 OsMKKs, and 19 Rosaceae MKKs. Notably, Clade C had only four members, i.e., two AtMKKs and two OsMKKs. None of the selected species in Rosaceae were found in Clade C. Each of the nine species had one member in Clade B. Clade A had a similar number of members, ranging from two to four; for example, there were two PmMKKs and two PmvMKKs ( Figure 2B; the exact number is shown in Table S3). It can be seen that in the MPK and MKK families, the number of genes in each taxonomic clade is equal for both wild P. mume and P. mume var. tortuosa, except for the gene PmvMPK7, which was more abundant in P. mume var. tortuosa than wild P. mume. To investigate the sequence features of MPK and MKK proteins in P. mume and P. mume var. tortuosa, the MEME program and TBtools were utilized to predict and map conserved domains. A total of 13 unique motifs were investigated in MPK and MKK proteins, which are illustrated schematically in Figure 3. The number of motifs in the MPKs was similar, ranging from 9 to 10. Motifs 1, 2, 3, 4, 5, 6, and 12 were highly conserved and existed in all 11 PmMPK and 12 PmvMPK proteins. Motifs 7 and 8 were present only in Clade D. Motifs 9 and 10 were present only in Clades A, B, and C. Motif 11 was present only in Clade A and B; motif 13 existed only in four members (PmMPK19, PmvMPK19, PmMPK20, PmvMPK20) of Clade D ( Figure 3). The number of MKK motifs in P. mume and P. mume var. tortuosa ranged from 7 to 10. Motifs 1, 4, and 5 were highly conserved and were present in all 7 PmMKK and 7 PmvMKK proteins. Motif 3 was found in 13 MKK proteins in P. mume and P. mume var. tortuosa with the exception of PmMKK2. Motif 6 was present in 12 MKK proteins in P. mume and P. mume var. tortuosa with the exception of PmMKK3 and PmvMKK3. Motifs 7, 9, and 10 were present only in Clades A and B. Motifs 11, 12, and 13 were present only in Clade B ( Figure 4A). Detailed information on MPK and MKK motifs is shown with logos acquired from the MEME Suite website in Figures S3  and S4. It can be seen that the conserved motif composition and gene structure of the homologous genes of wild P. mume and P. mume var. tortuosa were identical.  To investigate the sequence features of MPK and MKK proteins in P. mume and P. mume var. tortuosa, the MEME program and TBtools were utilized to predict and map conserved domains. A total of 13 unique motifs were investigated in MPK and MKK proteins, which are illustrated schematically in Figure 3. The number of motifs in the MPKs was similar, ranging from 9 to 10. Motifs 1, 2, 3, 4, 5, 6, and 12 were highly conserved and existed in all 11 PmMPK and 12 PmvMPK proteins. Motifs 7 and 8 were present only in Clade D. Motifs 9 and 10 were present only in Clades A, B, and C. Motif 11 was present only in Clade A and B; motif 13 existed only in four members (PmMPK19, PmvMPK19, PmMPK20, PmvMPK20) of Clade D (Figure 3). The number of MKK motifs in P. mume and P. mume var. tortuosa ranged from 7 to 10. Motifs 1, 4, and 5 were highly conserved and were present in all 7 PmMKK and 7 PmvMKK proteins. Motif 3 was found in 13 MKK proteins in P. mume and P. mume var. tortuosa with the exception of PmMKK2. Motif 6 was present in 12 MKK proteins in P. mume and P. mume var. tortuosa with the exception of PmMKK3 and PmvMKK3. Motifs 7, 9, and 10 were present only in Clades A and B. Motifs 11, 12, and 13 were present only in Clade B ( Figure 4A). Detailed information on MPK and MKK motifs is shown with logos acquired from the MEME Suite website in Figures S3 and S4. It can be seen that the conserved motif composition and gene structure of the homologous genes of wild P. mume and P. mume var. tortuosa were identical.   The multiple sequence alignment of MPKs in P. mume and P. mume var. tortuosa showed that they all contained the characterized multiple domains I-V, VIa, VIb, VII, VIII, IX, X, XI, and CD-domain ( Figure 5) [18]; Clades A, B, and C contained TEY activation motifs, while Clade D contained TDY activation motifs ( Figure S5). Multiple alignments of the MKK sequences in P. mume and P. mume var. tortuosa were generated, showing that MKK proteins are highly similar and that the eight specific domains, as well as ATP-binding, MAPK-binding, and activation domains, are highly conserved. By contrast, the NTF2binding domains only existed in Clade B (PmMKK3 and PmvMKK3) (Figure 6 and S6). In addition, we conducted a comparative analysis of the homologous genes of the MPK and MKK gene families in P. mume and P. mume var. tortuosa, respectively, and found that sequence similarity for homologous gene pairs was above 93% in all cases except PmMPK7/PmvMPK7, as shown in Figures S7 and S8. Sequence alignment shows that PmvMPK7 had 643 bp more sequence than PmMPK7, which was not a structural domain of the MPK gene family. The multiple sequence alignment of MPKs in P. mume and P. mume var. tortuosa showed that they all contained the characterized multiple domains I-V, VIa, VIb, VII, VIII, IX, X, XI, and CD-domain ( Figure 5) [18]; Clades A, B, and C contained TEY activation motifs, while Clade D contained TDY activation motifs ( Figure S5). Multiple alignments of the MKK sequences in P. mume and P. mume var. tortuosa were generated, showing that MKK proteins are highly similar and that the eight specific domains, as well as ATP-binding, MAPK-binding, and activation domains, are highly conserved. By contrast, the NTF2binding domains only existed in Clade B (PmMKK3 and PmvMKK3) (Figure 6 and S6). In addition, we conducted a comparative analysis of the homologous genes of the MPK and MKK gene families in P. mume and P. mume var. tortuosa, respectively, and found that sequence similarity for homologous gene pairs was above 93% in all cases except PmMPK7/PmvMPK7, as shown in Figures S7 and S8. Sequence alignment shows that PmvMPK7 had 643 bp more sequence than PmMPK7, which was not a structural domain of the MPK gene family.  In order to clarify the structural characteristics of the MPKs and MKKs in P. mume and P. mume var. tortuosa, the exon-intron structure was examined further. As shown in Figure 3B, the number of introns in the MPK gene family members varied widely from 1 to 15; those in Clade A and Clade B each had 4-5 introns. Clade C family members had one intron with the exception of PmvMPK7, which contained 15 introns, which should be related to the fact that it had an additional part of the sequence than its homolog PmMPK7. There were 8-12 introns in Clade D family members. Similarly, P. mume and P. mume var. tortuosa's MKK gene family members demonstrated a wide range of intron counts from 0 to 8, with those in Clade D having no introns. Those in Clade B contained eight introns; those in Clade A contained seven, excluding PmMKK6, which contained eight ( Figure 4B). In order to clarify the structural characteristics of the MPKs and MKKs in P. mume and P. mume var. tortuosa, the exon-intron structure was examined further. As shown in Figure 3B, the number of introns in the MPK gene family members varied widely from 1 to 15; those in Clade A and Clade B each had 4-5 introns. Clade C family members had one intron with the exception of PmvMPK7, which contained 15 introns, which should be related to the fact that it had an additional part of the sequence than its homolog PmMPK7. There were 8-12 introns in Clade D family members. Similarly, P. mume and P. mume var. tortuosa's MKK gene family members demonstrated a wide range of intron counts from 0 to 8, with those in Clade D having no introns. Those in Clade B contained eight introns; those in Clade A contained seven, excluding PmMKK6, which contained eight ( Figure 4B).

Chromosomal Distribution and Gene Duplication Analysis
All the MPKs and MKKs in P. mume and P. mume var. tortuosa were mapped based on gene location information, which showed that they were all located on the chromosome. The distribution of MPK genes on the chromosome was similar in P. mume and P. mume var. tortuosa. Chromosomes 1, 2, 7, and 8 each contained 2 MPKs, whereas chromosomes 3, 4, and 5 each contained 1 ( Figure 7). Likewise, the distribution of MKK genes on chromosomes was also similar in P. mume and P. mume var. tortuosa, with chromosomes 2, 7, and 8 each containing two MKKs and chromosome 4 containing one ( Figure 7). The chromosome distributions of the other five species of Rosaceae are given in Tables S1 and S2.
Tandem and segmental duplication are the two main engines for generating new copies of genes in the evolution of gene families. Tandem duplication generates nearby copies of genes in genomic clusters, and segmental duplication events have a different effect, as they may widely disperse copies of genes throughout the genome [45]. We use Multiple Collinearity Scan Toolkit (MCScanX) and Advanced Circos in TBtools to analyze gene tandem and segment replication events. The results showed that no tandem duplication events were detected in P. mume, P. mume var. tortuosa, or any of the other five Rosaceae species. gene location information, which showed that they were all located on the chromosome. The distribution of MPK genes on the chromosome was similar in P. mume and P. mume var. tortuosa. Chromosomes 1, 2, 7, and 8 each contained 2 MPKs, whereas chromosomes 3, 4, and 5 each contained 1 ( Figure 7). Likewise, the distribution of MKK genes on chromosomes was also similar in P. mume and P. mume var. tortuosa, with chromosomes 2, 7, and 8 each containing two MKKs and chromosome 4 containing one ( Figure 7). The chromosome distributions of the other five species of Rosaceae are given in Tables S1 and S2. Tandem and segmental duplication are the two main engines for generating new copies of genes in the evolution of gene families. Tandem duplication generates nearby copies of genes in genomic clusters, and segmental duplication events have a different effect, as they may widely disperse copies of genes throughout the genome [45]. We use Multiple Collinearity Scan Toolkit (MCScanX) and Advanced Circos in TBtools to analyze gene tandem and segment replication events. The results showed that no tandem A synteny analysis of MPKs and MKKs in P. mume and P. mume var. tortuosa was conducted using the Advanced Circos procedure of TBtools. Four segmental duplication events, including PmMPK13/PmMPK20, PmMPK19/PmMPK20, PmMPK3/PmMPK5, and PmMPK1/PmMPK7, were identified in the MPK gene family of P. mume, and three segmental duplication events, including PmvMPK3/PmvMPK12, PmvMPK4/PmvMPK12, and PmvMPK1/PmvMPK7, were detected in P. mume var. tortuosa. In addition, only one segmental duplication event was identified in the MKK gene family of P. mume, namely PmMKK9-3/PmMKK3. Each collinear gene pair was situated on a different chromosome, as shown by the red, blue, and olive-drab lines in Figure 8A. No MKK gene segment duplication events were detected in P. mume var. tortuosa. In addition, we analyzed segmental duplication events in 5 other Rosaceae species, of which 2 were detected in the MPK gene family of P. armeniaca, 9 in P. persica, 5 in P. salicina, 24 in M. domestica, and 2 in R. chinensis. It is worth noting that there were two MKK genes in the family of P. persica, one in P. salicina, and two in M. domestica. No MKK gene segment duplication events were detected in P. armeniaca and R. chinensis, the details of which are given in Table S4. plication events were detected in P. mume var. tortuosa. In addition, we analyzed segmental duplication events in 5 other Rosaceae species, of which 2 were detected in the MPK gene family of P. armeniaca, 9 in P. persica, 5 in P. salicina, 24 in M. domestica, and 2 in R. chinensis. It is worth noting that there were two MKK genes in the family of P. persica, one in P. salicina, and two in M. domestica. No MKK gene segment duplication events were detected in P. armeniaca and R. chinensis, the details of which are given in Table S4. In order to detect the selection pressure during gene divergence after duplication, the Ka (nonsynonymous)/Ks (synonymous) substitution ratio and divergence time of the duplicated pairs were further calculated. In the evolutionary process, the Ka/Ks ratio > 1 indicates positive selection (adaptive evolution), a ratio = 1 indicates neutral evolution (drift), and a ratio < 1 indicates negative selection (conservation). Result showed that the Ka/Ks ratios for all the duplicated orthologous gene pairs were all <1, indicating that MPK and MKK genes in P. mume, P. mume var. tortuosa, and the five Rosaceae species primarily undergo purifying selection following their duplication. Table S4 illustrates the In order to detect the selection pressure during gene divergence after duplication, the Ka (nonsynonymous)/Ks (synonymous) substitution ratio and divergence time of the duplicated pairs were further calculated. In the evolutionary process, the Ka/Ks ratio > 1 indicates positive selection (adaptive evolution), a ratio = 1 indicates neutral evolution (drift), and a ratio < 1 indicates negative selection (conservation). Result showed that the Ka/Ks ratios for all the duplicated orthologous gene pairs were all <1, indicating that MPK and MKK genes in P. mume, P. mume var. tortuosa, and the five Rosaceae species primarily undergo purifying selection following their duplication. Table S4 illustrates the divergence times of the duplicated gene pairs. The Ka/Ks ratio could not be calculated for some of the duplicated gene pairs, which was possibly because these gene pairs exhibited synonymous mutation at sites where synonymous mutations could occur; that is, the sequence divergence was large, resulting in a large evolutionary distance.

Interspecies Collinearity Analysis of the MPK and MKK Gene Family
In order to further investigate the particular retention of MPKs and MKKs in P. mume and P. mume var. tortuosa, the collinear relationships of these species with A. thaliana and five Rosaceae species were analyzed using the MCScanX algorithm of TBtools. A total of 15 homologous MPK gene pairs were found in P. mume and P. mume var. tortuosa ( Figure 8B, red lines). Furthermore, it was found that P. mume and A. thaliana share 16 pairs of homologous genes, as do P. mume and P. armeniaca. In addition, 14 pairs are shared between P. mume and P. persica; 17 between P. mume and P. salicina; 29 between P. mume and M. domestica; and 14 between P. mume and R. chinensis (Figure 9A-C and Table S5). Similarly, 20 homologous gene pairs were detected between P. mume var. tortuosa and A. thaliana; 18 between P. mume var. tortuosa and P. armeniaca; 14 between P. mume var. tortuosa and P. persica; 17 between P. mume var. tortuosa and P. salicina; 29 between P. mume var. tortuosa and M. domestica; and 17 between P. mume var. tortuosa and R. chinensis ( Figure 9D-F and Table S5). tween P. mume and P. persica; 17 between P. mume and P. salicina; 29 between P. mume and M. domestica; and 14 between P. mume and R. chinensis ( Figure 9A-C and Table S5). Similarly, 20 homologous gene pairs were detected between P. mume var. tortuosa and A. thaliana; 18 between P. mume var. tortuosa and P. armeniaca; 14 between P. mume var. tortuosa and P. persica; 17 between P. mume var. tortuosa and P. salicina; 29 between P. mume var. tortuosa and M. domestica; and 17 between P. mume var. tortuosa and R. chinensis ( Figure  9D-F and Table S5). A total of seven pairs of homologous MKK genes were found to be shared between P. mume and P. mume var. tortuosa ( Figure 7B, blue lines). Similarly, it was found that P. mume and A. thaliana share five pairs of the homologous genes; P. mume and P. armeniaca share seven; P. mume and P. persica share seven; P. mume and P. salicina share five; P. mume and M. domestica share seven; and P. mume and R. chinensis share five (Figure 10A-C and  Table S6). Furthermore, six pairs of homologous genes were found to be shared by P. mume var. tortuosa and A. thaliana; seven are shared by P. mume var. tortuosa and P. armeniaca; seven are shared by P. mume var. tortuosa and P. persica; five are shared by P. mume A total of seven pairs of homologous MKK genes were found to be shared between P. mume and P. mume var. tortuosa ( Figure 7B, blue lines). Similarly, it was found that P. mume and A. thaliana share five pairs of the homologous genes; P. mume and P. armeniaca share seven; P. mume and P. persica share seven; P. mume and P. salicina share five; P. mume and M. domestica share seven; and P. mume and R. chinensis share five (Figure 10A-C and  Table S6). Furthermore, six pairs of homologous genes were found to be shared by P. mume var. tortuosa and A. thaliana; seven are shared by P. mume var. tortuosa and P. armeniaca; seven are shared by P. mume var. tortuosa and P. persica; five are shared by P. mume var. tortuosa and P. salicina; six are shared by P. mume var. tortuosa and M. domestica; and five are shared by P. mume var. tortuosa and R. chinensis ( Figure 10D-F and Table S6).
var. tortuosa and P. salicina; six are shared by P. mume var. tortuosa and M. domestica; and five are shared by P. mume var. tortuosa and R. chinensis ( Figure 10D-F and Table S6).

Cis-Acting Elements in MPK and MKK Gene Promoter Prediction Analysis
To further examine the potential regulatory mechanisms of MPKs and MKKs during P. mume and P. mume var. tortuosa growth or defense mechanisms, especially in response to abiotic stresses such as low temperature, we uploaded the 2.0 kb sequence upstream of each MPK and MKK gene translation start point to the PlantCARE database to search for specific cis-acting elements. Fourteen and thirteen conserved regulatory elements related to plant hormones and environmental stress responses were analyzed in the MPK and MKK promoters, respectively (Figure 11 and S9, and Table S7 and S8). The promoter regions of P. mume and P. mume var. tortuosa toward the MPK/MKK gene family were found to contain several elements related to light response, anaerobic induction, MeJA response, and ABA response. Based on the regulatory elements in their promoters, six MPK gene family members in P. mume and P. mume var. tortuosa are sensitive to low temperature. Similarly, five MKK gene family members are sensitive to low temperatures in P. mume and P. mume var. tortuosa, as shown in Figures 11 and S9 and Table S7 and S8. Figure 12 illustrates the number of each cis-element of the PmMPK, PmMKK, PmvMPK, and PmvMKK genes. The results showed that they contain comparable numbers of low-temperature cis-elements.

Cis-Acting Elements in MPK and MKK Gene Promoter Prediction Analysis
To further examine the potential regulatory mechanisms of MPKs and MKKs during P. mume and P. mume var. tortuosa growth or defense mechanisms, especially in response to abiotic stresses such as low temperature, we uploaded the 2.0 kb sequence upstream of each MPK and MKK gene translation start point to the PlantCARE database to search for specific cis-acting elements. Fourteen and thirteen conserved regulatory elements related to plant hormones and environmental stress responses were analyzed in the MPK and MKK promoters, respectively (Figures 11 and S9, and Tables S7 and S8). The promoter regions of P. mume and P. mume var. tortuosa toward the MPK/MKK gene family were found to contain several elements related to light response, anaerobic induction, MeJA response, and ABA response. Based on the regulatory elements in their promoters, six MPK gene family members in P. mume and P. mume var. tortuosa are sensitive to low temperature. Similarly, five MKK gene family members are sensitive to low temperatures in P. mume and P. mume var. tortuosa, as shown in Figures 11 and S9 and Tables S7 and S8. Figure 12 illustrates the number of each cis-element of the PmMPK, PmMKK, PmvMPK, and PmvMKK genes. The results showed that they contain comparable numbers of low-temperature cis-elements.  Based on the results of the sequence comparison, we found that the sequence similarity between the MPK and MKK genes in wild P. mume and P. mume var. tortuosa is very high. We therefore examined the expression of the PmMPK and PmMKK genes in wild P. mume. Our RNA-seq dataset was used to analyze the expression patterns of MPK and Figure 12. The number of cis-elements in PmMPK, PmvMPK, PmMKK, and PmvMKK promoter.

Expression Pattern Analysis of MPKs and MKKs in P. mume
Based on the results of the sequence comparison, we found that the sequence similarity between the MPK and MKK genes in wild P. mume and P. mume var. tortuosa is very high. We therefore examined the expression of the PmMPK and PmMKK genes in wild P. mume. Our RNA-seq dataset was used to analyze the expression patterns of MPK and MKK family members in the roots, stems, leaves, flower buds, fruits of wild P. mume [46], and flower buds of cultivar P. mume 'Lve' at different stages of dormancy [47] to learn more about the function of PmMPKs and PmMKKs in development and response to low temperature. Tables S9 and S10 present their RPKM values. As illustrated in Figure 13A, 11 PmMPKs were expressed in all the test tissues, and all showed a high level of transcript accumulation (RPKM > 5), of which two PmMPK genes in flower buds (PmMPK13 and PmMPK7), PmMPK4 in leaves, three PmMPK genes in roots (PmMPK3, PmMPK5 and PmMPK16), three PmMPK genes in fruits (PmMPK7, PmMPK8, and PmMPK19), and at least four PmMPK genes in stems (PmMPK1, PmMPK4, PmMPK6, and PmMPK20) have relatively high expression levels. In addition, three (PmMKK9-2, PmMKK9-3, and PmMKK10) out of seven PmMKKs had very low or no detectable expression in the tissues tested ( Figure 13A). The remaining four PmMKKs were relatively highly expressed in different tissues. For instance, PmMKK2 and PmMKK6 levels were high in leaves, PmMKK9-1 and PmMKK2 levels were high in fruits, and PmMKK3 levels were high in stems. All the PmMPKs were expressed during the flower bud dormancy period and at particular development stages ( Figure 13B). Four PmMPKs (PmMPK1, PmMPK3, PmMPK7, and PmMPK13) were preferentially expressed during the endo-dormancy I stage (EDI) in November, six PmMPKs (PmMPK3, PmMPK5, PmMPK7, PmMPK16, PmMPK19, and PmMPK20) showed the highest expression levels during the endo-dormancy II stage Endo-dormancy II, December; EDIII: Endo-dormancy III, January; NF: Natural flush, February. The expression amount is converted to a 2-based log function and then normalized by the row using the normalized method. The relative expression level is indicated by the color scale to the right of the heat map, and an elevated expression level is indicated by the color gradient from dodger blue to red.
All the PmMPKs were expressed during the flower bud dormancy period and at particular development stages ( Figure 13B). Four PmMPKs (PmMPK1, PmMPK3, PmMPK7, and PmMPK13) were preferentially expressed during the endo-dormancy I stage (EDI) in November, six PmMPKs (PmMPK3, PmMPK5, PmMPK7, PmMPK16, PmMPK19, and PmMPK20) showed the highest expression levels during the endo-dormancy II stage (EDII) in December, and four PmMPKs (PmMPK7, PmMPK16, PmMPK19, and PmMPK20) showed up-regulated expression during the endo-dormancy III stage (EDIII) in January. Three PmMPKs (PmMPK4, PmMPK6, PmMPK8) in particular showed greater expressions during the natural flush stage (NF) in February. Six up-regulated genes (PmMPK4, PmMPK6, PmMPK7, PmMPK8, PmMPK16, and PmMPK20) (Table S7) exhibited low-temperature response elements in their putative promoter regions. In contrast to three PmMKKs (PmMKK9-2, PmMKK9-3, and PmMKK10) that had very low or no detectable expression in the tissues tested, the other four PmMKKs were highly expressed at one or several stages ( Figure 13B). For example, PmMKK2/6/9-1 was preferentially expressed in November in EDI, PmMKK2/6 showed the highest level of expression in EDII, and PmMKK3/6 showed up-regulated expression in EDIII. Among them, PmMKK2 and PmMKK6 exhibited low-temperature response elements in their putative promoter regions.
To further evaluate the expression patterns of PmMPKs and PmMKKs under exposure to cold, we examined the stems of the cold-tolerant cultivar P. mume 'Songchun' in three different geographic locations, whose FPKM values are provided in Table S11. As shown in Figure 14A, five PmMPKs (PmMPK3, PmMPK7, PmMPK16, PmMPK19, and PmMPK20) showed higher expression in winter, of which three PmMPKs (PmMPK3, PmMPK16, and PmMPK20) were also highly expressed in autumn in Beijing and Chifeng. Additionally, two PmMPKs (PmMPK1 and PmMPK6) showed higher expression levels in winter in Beijing (−5.4 • C). Five PmMPKs (PmMPK1, PmMPK4, PmMPK7, PmMPK13, and PmMPK19) showed higher expression in autumn (6.1~7.9 • C). The expression of PmMPK8 was higher in the spring (3.2~5.3 • C), and PmMPK5 showed higher expression levels in the spring of Chifeng and Gongzhuling (3.2, 5.3 • C). Among these genes with up-regulated expression, six PmMPKs (PmMPK4, PmMPK6, PmMPK7, PmMPK8, PmMPK16, and PmMPK20) (Table  S7) were found to contain low-temperature response elements in their putative promoter regions. Another heatmap was created to compare the expression patterns of PmMPKs for the year ( Figure 14B). PmMPKs in the samples obtained from the three locations displayed comparable expression patterns at the same times throughout the year, as seen in Figure 14B. Most PmMPKs were expressed at higher levels in autumn and winter and downregulated in the spring ( Figure 14B). Similarly, three PmMKKs (PmMKK9-2, PmMKK9-3, and PmMKK10) also had very low or no detectable expression ( Figure 14A,B). Of the other four genes, PmMKK3 was relatively highly expressed in spring, PmMKK6 was relatively highly expressed in fall in Gongzhuling, PmMKK2 was relatively highly expressed in winter in Gongzhuling, and PmMKK9-1 was irregularly expressed, with high expression in spring, autumn, and winter in different locations. Seasonal expression patterns for the four genes at the three locations were not obvious.
(PmMKK9-2, PmMKK9-3, and PmMKK10) also had very low or no detectable expression ( Figure 14A,B). Of the other four genes, PmMKK3 was relatively highly expressed in spring, PmMKK6 was relatively highly expressed in fall in Gongzhuling, PmMKK2 was relatively highly expressed in winter in Gongzhuling, and PmMKK9-1 was irregularly expressed, with high expression in spring, autumn, and winter in different locations. Seasonal expression patterns for the four genes at the three locations were not obvious.

PmMPK and PmMKK Expression Patterns in Response to Cold Treatment
To validate the results of transcriptome analysis of P. mume MPK and MKK genes during naturally low-temperature, and evaluate the involvement of PmMPKs and PmMKKs in cold stress, expression profiles under 4 °C stress treatments for 0, 1, 4, 6, 12,

PmMPK and PmMKK Expression Patterns in Response to Cold Treatment
To validate the results of transcriptome analysis of P. mume MPK and MKK genes during naturally low-temperature, and evaluate the involvement of PmMPKs and PmMKKs in cold stress, expression profiles under 4 • C stress treatments for 0, 1, 4, 6, 12, 24, 48, and 72 h were studied using qRT-PCR with the cold-tolerant cultivar P. mume 'Songchun' and the cold-sensitive cultivar P. mume 'Lve'. We used a qRT-PCR assay on the 11 identified PmMPKs and 4 PmMKKs, and relative expression was calculated with the PP2A and Actin genes of P. mume as the reference genes, respectively (Figures 15 and S10). All the tested MPK and MKK genes showed potentially induced expression after 1-72 h of cold stress in the stems (Figures 15 and S10). The expression levels of 11 PmMPK and 4 PmMKK genes changed at different rates in the two cultivars under the imposed cold stress treatment. Particularly noteworthy was the cold-induced expression of PmMPK3, which after 1 h in 'Lve' and 'Songchun' reached its greatest expression levels (13-and 9-fold increases, respectively), which consistent with the transcriptome results, PmMPK3 expression was upregulated in 'Lve' when the temperature decreases in November and December, and in 'Songchun' during the winter months. In addition, the responses of the cold-induced genes could sometimes be remarkably quick, and distinct expression changes were discovered at earlier time points. For example, PmMPK16 genes were remarkably up-regulated after 1 h of cold stress, while PmMPK1, PmMPK5, PmMPK6, PmMPK7, PmMPK8, PmMPK19, PmMPK20, PmMKK3, and PmMKK9-1 were significantly down-regulated in 'Songchun'. In 'Lve', PmMPK1 and PmMPK3 were remarkably up-regulated after 1 h of cold stress, while PmMPK6, PmMPK13, PmMPK20, PmMKK3, PmMKK6, and PmMKK9-1 were remarkably down-regulated. The expression trends of these genes in the face of sudden cooling were almost identical to the transcriptome results obtained in the autumn of 'Songchun' in Beijing, when the temperature starts to drop, and in the buds of 'Lve' in November, when the temperature drops ( Figures 13B and 14B). The expression of the four genes PmMPK5, PmMPK6, PmMPK20, and PmMKK3 decreased in the two cultivars as the length of the treatment increased. The expression patterns of PmMPK3, PmMPK19, and PmMKK9-1 were similar in both varieties, with expressions first up-regulated and then down-regulated or first down-regulated and then up-regulated after cold treatment. The expression patterns of five genes (PmMPK4, PmMPK7, PmMPK8, PmMKK2, and PmMKK6) varied between the two cultivars. Of these five genes, PmMPK4/7/8 showed a decreasing trend in 'Songchun', while in 'Lve' it showed a rising trend as the length of the treatment increased. PmMKK2 was initially up-regulated, followed by down-regulation as the length of the treatment increasedin 'Songchun', while in 'Lve' it exhibited no significant changes overall. PmMKK6 was up-regulated with treatment increase in 'Songchun', while no significant changes with prolonged treatment were demonstrated in 'Lve'.

Discussion
MAP kinase cascades in plants are the oldest conserved signal transduction pathways. They regulate many biological functions, including hormone signaling, growth, and development [12,14,48,49] as well as various types of stress [10][11][12]50]. The complete genome-wide sequencing of a large number of species has led to the identification of numerous MPK cascade family members in numerous plants [19][20][21][22][23][24][25][26][27][28][29][30][31]. However, P. mume and its variety P. mume var. tortuosa MAPK cascade genes' identification and functions remain mostly unknown. P. mume is the collective name for an essential early spring flowering plant in China and Southeast Asia that is often challenged by cold stress in northern China. In order to investigate the tolerance mechanisms of P. mume and its variety P. mume var. tortuosa for cold stresses, we conducted a thorough analysis of the MPK and MKK family genes in wild P. mume, P. mume var. tortuosa, and five associated Rosaceae species along with the expression traits of PmMPK and PmMKK genes under cold stresses. The present study used Arabidopsis protein sequences from 20 AtMPKs and 10 AtMKKs to identify the complete set of MPK and MKK proteins in P. mume, P. mume var. tortuosa, and five Rosaceae species. A total of 11 PmMPKs and 7 PmMKKs from P. mume; 12 PmvMPKs and 7 PmvMKKs from P. mume var. tortuosa; 12 PaMPKs and 7 PaMKKs from P. armeniaca; 10 PpMPKs and 9 PpMKKs from P. persica; 13 PsMPKs and 5 PsMKKs from P. salicina; 22 MdMPKs and 8 MdMKKs from M. domestica; and 12 RcMPKs and 7 RcMKKs from R. chinensis were identified (Table S3). The genome sizes of these plants are~219.9 Mb for P. mume,~237.7 Mb for P. mume var. tortuosa,~221.9 Mb for P. armeniaca,~224.6 Mb for P. persica,~284.2 Mb for P. salicina,~658.9 Mb for M. domestica, and~503 Mb for R. chinensis [46,[51][52][53][54][55][56]. This phenomenon suggests that there is no direct correlation between the number of MPK and MKK genes and plant genome size. The numbers of MPK gene families among the selected Rosaceae species were similar with the exception of the apple, which had 29 MdMPK genes-a number that is far greater than those of other species. Such a high number of homologous genes in apple is consistent with the extensive events of duplication in its genome [57].
According to Arabidopsis [14], these MPKs and MKKs are classified into four clades (A-D). The MPK proteins in Clades A, B, and C all have TEY phosphorylation sites, while the protein in D has TDY phosphorylation sites. A close genetic association was found between P. mume and P. mume var. tortuosa in terms of MPK and MKK genes. Consistent clade division and high sequence similarity were demonstrated ( Figures S7 and S8), indicating that MPK and MKK members in P. mume and P. mume var. tortuosa are relatively conservative in evolution, excepting PmvMPK7, which is~640 bp longer than its homologous gene PmMPK7. This extra sequence belongs to other genes according to the annotation ( Figure  S5) and may have been caused by a gene annotation error. In addition, P. mume var. tortuosa has one additional gene, PmvMPK17, compared with wild P. mume, which may be related to the expansion of the MPK gene family after the divergence of the two species and their independent evolution [51]. The gene numbers of Clades A, B, and C varied little among the different species studied. Clade D was demonstrated to have the largest number of members (Table S3), which is consistent with previous studies [14,58,59]. It is worth noting that no MKK family members from Clade C were found in the selected Rosaceae species, and the numbers from Clade A and B were similar to those of Arabidopsis and rice, indicating that members of Clade C tended to evolve into Clade D during the evolutionary process ( Figure 2B). The NTF2-binding domain only existed in Clade B (Figure 6), which is consistent with previous studies on angiosperms, gymnosperms, pteridophytes, and bryophytes [14,18,59].
The structure of gene family members is closely linked to gene expression and function [60]. Members within the same clade have similar motif compositions. Different clades contain 1-2 unique motifs within them, suggesting that the conserved motifs in MPK and MKK genes support their close evolutionary relationship. Unique motifs may participate in diverse biological processes between different clades [61]. Exon/intron structure is essential to the biological evolution of organisms [62]. In this study, it was found that similar intron-exon structures exist in members belonging to the same clade of MPK and MKK families, with the exception of PmvMPK7, which had 15 introns-far more than members of the same clade and consistent with its long genetic sequence possibly due to a gene annotation error ( Figures S5 and S7). The MKK gene family was known to have a smaller number of gene families and the ability to participate in multiple biological processes. Being intermediate between MAPK and MAPKKK and able to react with multiple MAPKs and MAPKKKs to form multiple distinct signaling pathways, MAPKK may be the intersection of signaling networks in plants consisting of multiple MAPK cascade pathways [14,21].
Segmental and tandem duplications have been proposed to represent two of the primary causes of gene family expansion in plants [45]. Our data showed that no tandem duplication events occur in the selected Rosaceae species, while segment duplication is frequently present in Rosaceae genomes, especially in the MPK gene family (Table S4), suggesting that chromosomal segments may play an essential role in the expansion of these gene families. Previous studies have also suggested that tandem duplication is rare in expanding the MAPK, MAPKK, and MAPKKK gene families [19,[63][64][65][66]. The intergenomic collinearity analysis of MPK/MKK genes from the species investigated in this study found the genes to have good homology, providing further evidence that the MPK and MKK gene families are conserved in angiosperms (Figures 8-10), which may imply functional consistency among these homologous genes.
Gene function and regulation are determined mainly by cis-acting regulatory elements in the promoter region [67]. In the course of long-term evolution, plants have evolved intricate mechanisms of gene regulation in response to adverse environmental impacts. The promoters of both MPKs and MKKs include stress-related cis-acting elements such as abscisic acid response, anaerobic induction, drought inducibility, and low-temperature responsive cis-acting elements in P. mume and P. mume var. tortuosa. Our results showed that a large proportion (≥1/2) of the number of genes contained low-temperature-associated cis-elements in both the MPK and MKK gene families of P. mume and P. mume var. tortuosa. In addition, the promoters of the two gene families from sugarcane [63] and tea [58] were also found to contain the aforementioned cis-acting elements, and expression levels of MPKs and MKKs changed after stress treatment.
MPK and MKK genes play critical roles in a wide range of biological processes [12]. We examined their expression in different tissues of wild P. mume, and the results showed preferential tissue expression among MPK and MKK gene pairs (Figure 13), which supports previous findings [19,59,64]. Duplicated gene pairs of PmMPK (PmMPK20/PmMPK19; PmMPK7/PmMPK1; PmMPK20/PmMPK13; and PmMPK3/PmMPK5) and PmMKK (PmMKK3/ PmMKK9-3) had different expression patterns. For example, PmMPK20 was more highly expressed in the stem, but this was not the case for the similar duplicated gene PmMPK19. Moreover, the expression of PmMPK7 was higher in the flower bud and fruit, but its duplication PmMPK1 was predominantly expressed in the stem and root. This phenomenon has also been observed in J. curcas [19]. Thus, regardless of duplicated gene pairs having similar genetic compositions, they may not share similar functions or participate in the same metabolic pathways [27]. In the evolution of P. mume, some functions might be lost or gained after duplication.
Several studies have investigated the role of the MAPK cascade in response to cold stress in different plants [19,[37][38][39][40][41][42][43]. In this study, we found that the PmMPK and PmMKK genes are specifically expressed during different periods of flower bud dormancy from winter to spring. Thus, we hypothesize that these PmMPKs and PmMKKs are involved in the cold response of protected flower buds at low temperatures during different stages of flower bud dormancy. In addition, we identified several PmMPK genes that are specifically highly expressed in autumn (PmMPK4, PmMPK13, PmMPK16) and winter (−5 •~− 22 • ) (PmMPK1, PmMPK3, PmMPK6, PmMPK7, PmMPK14, PmMPK19) ( Figure 14). This suggests that the expression of these genes starts to increase in autumn from the onset of cold domestication to withstand the coming cold, continues to increase in winter to help the plant survive the harsh winter, and decreases in spring after temperatures start to rise again.
According to promoter analysis, the majority of these genes (5/9) have low-temperature cis-acting elements. By contrast, the other four genes do not have low-temperature response elements (Table S7) but are induced into expression by low temperatures. Hence, the ciselements of the genes are not the only determinants of the stress response, which may also be induced by other stresses. This phenomenon has also been observed in other studies [58]. The expression pattern of the PmMKK gene at three sites and three periods is less pronounced, suggesting that the MKK gene has distinct patterns in stress response across different plant species.
According to the qRT-PCR investigation results, PmMPKs and PmMKKs were activated at low temperatures (4 • C), and their expression either increased or decreased as the treatment duration was prolonged ( Figure 15). The expression trends of many genes in the face of sudden cooling are consistent with the transcriptome results. Some genes (PmMPK4 and PmMPK8) were slightly down-regulated or up-regulated after 1 h of treatment but were significantly up-regulated after 48/72 h of cold stress, indicating the generalized response of these MAPK cascade pathway kinase genes to a variety of adversity stresses [42,68]. Four genes (PmMPK5, PmMPK6, PmMPK20, and PmMKK3) in 'Songchun' and five (PmMPK5, PmMPK6, PmMPK13, PmMPK20, and PmMKK3) in 'Lve' showed a decrease in expression levels with increasing treatment time (Figure 15), suggesting that these genes may be negatively regulated by low temperatures, leading to enhanced cold sensitivity. This phenomenon has also been observed in cucumber and watermelon, where most CsMAPK and ClMAPK genes are down-regulated in expression after chilling treatment [27,41]. The expression levels of PmMKK6 in 'Songchun' and PmMPK4 in 'Lve' gradually increased with continued treatment (Figure 15), suggesting that these two genes may be positively regulated by cold stress responses and enhance the cold tolerance of P. mume. The discrepancies in the expression patterns of PmMPK4, PmMPK7, PmMKK8, PmMKK2, and PmMKK6 between 'Songchun' and 'Lve' may have been due to genetic variation between these two cultivars that has made them differently resistant to cold. In terms of species classification, 'Songchun' belongs to the apricot mume and 'Lve' to the true mume. The branches of Songchun are pale grey bronze violet with pink flowers and pale brown violet sepals, resembling apricots. The branches of 'Lve' are pale yellowgreen, with pale cream to white flowers and pale yellow-green sepals ( Figure 16). Differences in branching morphology between the two cultivars may be one reason why some genes respond differently to cold. Some homologs of the MAPK cascade gene showing different expression patterns under the same stress conditions have also been found in other species. For example, AtMPK7 is remarkably down-regulated after cold stress, while CsMPK7 is remarkably up-regulated; the OsMKK4 gene is up-regulated after cold stress, while CsMKK4 is down-regulated under the same conditions [27]. After cold stress treatment, some MPK and MKK genes have similar expression patterns, suggesting that these gene pairs may had similar functions. Some homologs are remarkably different, suggesting that they may have acquired new functions in evolution and play roles in different signaling pathways [27,69,70].

Identification of MPK and MKK Gene Family Members
To identify the MPK and MKK genes in P. mume and P. mume var. tortuosa, we conducted a BLASTP against 2 P. mume and P. mume var. tortuosa genomes and 5 Rosaceae species with MPK and MKK protein sequences of Arabidopsis as queries with an E-value threshold set at 10 −10 . The SMART database (http://smart.embl-heidelberg.de/, accessed on 10 October 2022) and the Conserved Domain Database of NCBI (https://www.ncbi.nlm.nih.gov/cdd, accessed on 10 October 2022) were then used to confirm all putative MPK and MKK proteins.
The MPK and MKK genes were named based on their similarity to Arabidopsis MPK and MKK proteins. Additionally, the number of amino acids, molecular weight (MW), and isoelectric point (pI) were computed using the Python script. The subcellular localization of MPK and MKK gene family members was evaluated using the Cell-PLoc 2.0 online server [71] (http://www.csbio.sjtu.edu.cn/bioinf/Cell-PLoc-2/, accessed on 20 October 2022).

Identification of MPK and MKK Gene Family Members
To identify the MPK and MKK genes in P. mume and P. mume var. tortuosa, we conducted a BLASTP against 2 P. mume and P. mume var. tortuosa genomes and 5 Rosaceae species with MPK and MKK protein sequences of Arabidopsis as queries with an E-value threshold set at 10 −10 . The SMART database (http://smart.embl-heidelberg.de/, accessed on 10 October 2022) and the Conserved Domain Database of NCBI (https://www.ncbi.nlm. nih.gov/cdd, accessed on 10 October 2022) were then used to confirm all putative MPK and MKK proteins.
The MPK and MKK genes were named based on their similarity to Arabidopsis MPK and MKK proteins. Additionally, the number of amino acids, molecular weight (MW), and isoelectric point (pI) were computed using the Python script. The subcellular localization of MPK and MKK gene family members was evaluated using the Cell-PLoc 2.0 online server [71] (http://www.csbio.sjtu.edu.cn/bioinf/Cell-PLoc-2/, accessed on 20 October 2022).

Phylogenetic Analysis
In order to investigate the phylogenetic relationships between MPK and MKK genes in P. mume and P. mume var. tortuosa and other species, the alignment of full-length MPK and MKK protein sequences from P. mume, P. mume var. tortuosa, A. thaliana, and O. sativa along with five other Rosaceae species was carried out using MAFFT software with the auto setting [72]. Subsequently, phylogenetic trees with maximum likelihood (ML) were created using FastTree (version 2.1.11) [73] with default parameters. The phylogenetic tree was then annotated and beautified using EvolView v2 [74] [75] predicted the conserved motifs of each identified PmMPK, PmvMPK, PmMKK, and PmvMKK protein. The number of motifs for the conserved domains was set to 13, the motif width was set to 6-50 amino acids, and the residuals were set to the default parameters. The P. mume and P. mume var. tortuosa genome gff files were used to extract the gene structure data, which were then visualized with TBtools software [76] and edited in AI CS6.

Chromosome Location, Duplication, and Synteny Analysis
Gene location and genome chromosome length information for each PmMPK, PmMKK, PmvMPK, and PmvMKK gene of P. mume and P. mume var. tortuosa were retrieved from the gff file downloaded from the P. mume genome project (http://prunusmumegenome. bjfu.edu.cn/, accessed on 6 October 2022) and the Genome Database for Rosaceae (https: //www. rosaceae.org/, accessed on 6 October 2022), respectively. Using TBtools, a chromosomal location figure was created. The Multiple Collinearity Scan Toolkit (MCScanX) and Advanced Circos in TBtools were used to analyze gene tandem and segment replication events. The criteria used for identifying gene duplication were as follows: (a) all genes are initially classified as 'singletons' and assigned gene ranks according to their order of appearance along chromosomes; (b) in any BLASTP hit, the two genes are re-labeled as 'tandem duplicates' if they have a difference of gene rank = 1, that is, they were continuous repeat; (c) the anchor genes in collinear blocks are re-labeled as 'segment replications' [76,77]. The synteny analysis of the MPK and MKK genes of P. mume and P. mume var. tortuosa, respectively, with A. thaliana and five Rosaceae species was performed using MCScanX in TBtools. The rate of Ka (non-synonymous substitution)/Ks (synonymous substitution) was employed to assess the codon evolutionary rate between the duplicated gene pairs based on the alignment of the coding sequence using the Ka/Ks calculator program embedded in the TBtools. According to 2 ordinary rates (λ) of 1.5 × 10 −8 and 6.1 × 10 −9 substitutions per site per year [78,79], the formula t = Ks/2λ × 10 −6 Mya was used to calculate the divergence time.
4.6. Cis-Acting Element Analysis of P. mume and P. mume var. tortuosa MPK and MKK Gene Promoter Regions The upstream genomic sequences (2.0 kb) of each identified PmMPK, PmMKK, PmvMPK, and PmvMKK gene were extracted from the genomic sequence data using TBtools and then uploaded to the PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/ plantcare/html/, accessed on 22 November 2022) [80] for cis-acting element analysis. After much deliberation, we selected 14 elements for MPK and 13 for MKK, including some that are activated by hormones such as methyl jasmonate (MeJA) and abscisic acid (ABA). Additionally, some elements respond to growth and development (such as light response and meristem expression) as well as stress, such as low temperatures. The map was created by TBtools using these data along with phylogenetic tree data (nwk file) and then edited by AI CS6 software.

PmMPK and PmMKK Gene Expression Analysis
We examined the expression patterns of PmMPKs and PmMKKs in various tissues using data from RNA sequencing of the root, stem, leaf, flower bud, and fruit of wild P. mume collected from Tongmai, Tibet, China (30 • 06 N, 95 • 05 E) (RNAseq data are available in the NCBI Gene Expression Omnibus (GEO) under accessions GSE40162) [46], and we examined the responses of PmMPKs and PmMKKs to naturally low temperatures from November to February using data from RNA sequencing of P. mume 'Lve' flower bud dormancy, the specimens for which were grown in the Jiufeng International Plum Blossom Garden, Beijing, China (40 • 07 N, 116 • 11 E) (the data was acquired from the corresponding author) [47]. Additionally, we investigated the expression of MPK and MKK gene family members of P. mume 'Songchun' for three seasons, including autumn (cold acclimation, October), winter (the final period of endo-dormancy, January), and spring (deacclimation, March) in three geographical locations: Beijing (BJ, 39 • 54 N, 116 • 28 E), Chifeng (CF, 42 • 17 N, 118 • 58 E), and Gongzhuling (ZGL, 43 • 42 N, 124 • 47 E) to better understand the role of PmMPKs and PmMKKs in reacting to natural low temperatures (the data was acquired from the author). Heatmaps were created using TBtools [76].

qRT-PCR Analysis of PmMPK and PmMKK Genes
To validate the results of transcriptome analysis of P. mume MPK and MKK genes during naturally low-temperature, we collected the annual branches of the cold-tolerant P. mume 'Songchun' and the cold-sensitive cultivar P. mume 'Lve' from the Jiufeng International Plum Blossom Garden, Beijing, China (40 • 07 N, 116 • 11 E) to study how PmMPKs and PmMKKs react to low temperatures ( Figure 16). Before chilling treatment, the shoots were incubated at 22 • C for an overnight period; then, they were lowered to 4 • C under long-day circumstances (16 h of light and 8 h of darkness) for 0, 1, 4, 6, 12, 24, 48, and 72 h. The stems were immediately removed and put in liquid nitrogen, then kept at −80 • C for long-term storage in preparation for RNA extraction. Each treatment had three biological replicates.
The RNAprep Pure Plant Plus Kit (Tiangen, Beijing, China) was used to extract the total RNA from each sample. Using ReverTra Ace®qPCR RT master mix with gDNA remover (Toyobo, Osaka, Japan), complementary cDNA was synthesized. Primer 3 (https://bioinfo. ut.ee/primer3--0.4.0/, accessed on 2 December 2022) designed the specific primers based on the cDNA sequences (Table S12). Quantitative real-time polymerase chain reaction (qRT-PCR) was undertaken on a qTOWER2.2 system (Analytik Jena, Jena, German) with a SYBR®Green Premix Pro Taq HS qPCR kit (Accurate Biology, Hunan, China) to examine the expression levels of PmMPKs and PmMKKs at low temperatures. The reactions took place at a 20 µL volume and contained 4.0 µL 10 × each of forward and reverse primers, 2.0 µL 10× of cDNA, and 10.0 µL SYBR®Green Premix Pro Taq HS qPCR master mix. The reactions were carried out using the following protocol: 95 • C for 30 s, followed by 40 cycles of 95 • C for 5 s, 55 • C for 30 s, and 72 • C for 30 s. The annealing temperature was adjusted according to the actual situation; specifically, temperatures were 55 • C for PmMPK4/5/6/7/13; 57 • C for PmMPK1 and PmMKK3; 58 • C for PmMPK3/16/19/20 and PmMKK9-1/2/6; and 60.6 • C for PmMPK8. Using the PHOSPHATASE 2A (PP2A) and Actin gene of P. mume as the reference gene, the relative expression was calculated using the delta-delta CT method [81]. For each qRT-PCR, three biological replicates were used. Using SPSS 22.0, separate statistical analyses of 'Songchun' and 'Lve' were performed. With a significant difference at level p = 0.05, the Student-Newman-Keuls test and least significant difference (LSD) test were used to calculate the one-way ANOVA analysis of variance. The graphs were produced using GraphPad Prism6 software.

Conclusions
In conclusion, our research is the first to carry out the genome-wide identification and characterization of MPKs and MKKs in the species P. mume and its variety P. mume var. tortuosa, including chromosomal location, the identification of duplication genes, analysis of gene structure, phylogenetic relationships, and conserved motifs. Based on the RNA-seq data, the expression profiles of the PmMPK and PmMKK genes in various tissues and geographical locations were also further examined. Furthermore, under cold stress conditions, a qRT-PCR assay was used to examine the expression profiles of the PmMPK and PmMKK genes. Our findings may be very important for further research into the biological roles of PmMPKs and PmMKKs.