Identification, Phylogenetic and Expression Analyses of the AAAP Gene Family in Liriodendron chinense Reveal Their Putative Functions in Response to Organ and Multiple Abiotic Stresses

In this study, 52 AAAP genes were identified in the L. chinense genome and divided into eight subgroups based on phylogenetic relationships, gene structure, and conserved motif. A total of 48 LcAAAP genes were located on the 14 chromosomes, and the remaining four genes were mapped in the contigs. Multispecies phylogenetic tree and codon usage bias analysis show that the LcAAAP gene family is closer to the AAAP of Amborella trichopoda, indicating that the LcAAAP gene family is relatively primitive in angiosperms. Gene duplication events revealed six pairs of segmental duplications and one pair of tandem duplications, in which many paralogous genes diverged in function before monocotyledonous and dicotyledonous plants differentiation and were strongly purification selected. Gene expression pattern analysis showed that the LcAAAP gene plays a certain role in the development of Liriodendron nectary and somatic embryogenesis. Low temperature, drought, and heat stresses may activate some WRKY/MYB transcription factors to positively regulate the expression of LcAAAP genes to achieve long-distance transport of amino acids in plants to resist the unfavorable external environment. In addition, the GAT and PorT subgroups could involve gamma-aminobutyric acid (GABA) transport under aluminum poisoning. These findings could lay a solid foundation for further study of the biological role of LcAAAP and improvement of the stress resistance of Liriodendron.


Introduction
As a result of natural selection, there are currently only two species of Liriodendron in nature, namely, L. chinense and L. tulipifera, which are distributed in Asia and North America, respectively. Both are widely grown in China, mainly for landscaping and timber supply [1]. However, in the process of planting Liriodendron, adverse external environmental conditions such as temperature extremes and the presence of heavy metals ions will still affect the growth of Liriodendron, which hinders its popularization. Therefore, understanding the growth, development, and resistance mechanism of Liriodendron is conducive to its promotion and application.
Amino acids, as the main circulation form of organic nitrogen in the process of plant growth and development, are involved in various life processes, including protein synthesis, hormone regulation, and energy storage, as well as nucleotides, chlorophyll, some plant hormones, and most secondary metabolites [2,3]. It is a necessary source of nutrition and an important regulatory mechanism. The absorption of amino acids in the substrate Although many functional AAAP genes have been revealed, proving their importance in the process of plant growth and development and in resistance to abiotic stress [36,37], the AAAP gene in L. chinense has not been characterized. In order to explore the composition of AAAP gene family members in L. chinense and the possible biological processes involved, we identified and analyzed them based on the whole genome data of L. chinense and systematically analyzed the AAAP gene family from the aspects of phylogeny, gene function differentiation, expression pattern, and stress response, which has provided the foundation for further gene function research and genetic improvement.

Identification of LcAAAP Gene Family
Through the local BLASTP and HMMER to search the AAAP protein in the L. chinense genomic database, 73 putative protein sequences were obtained. Next, the Pfam and Conserved Domain Databases were applied to verify the conserved domains. Finally, the 52 LcAAAP protein sequences were retained; these genes were designated based on the AtAAAP gene family (Table 1, Figure 1 and Figure S1). Gene characteristics, including chromosomal location, open reading frame length, amino acid length, molecular weight, isoelectric point, and subcellular location were analyzed (Table 1). Through analysis of the chromosome distribution based on the L. chinense genome, 48 LcAAAP genes were located on 14 chromosomes. Chromosome 4 had the most LcAAAP genes (14), followed by chromosome 2 (seven genes), chromosomes 1 and 17 (five genes), chromosome 14 (four genes), chromosomes 3 and 18 (three genes), chromosomes 11, 13, and 16 (two genes), and chromosome 8, 10 and 12 (one gene). The genes LcAPP6c, LcATL7d, LcATL7b, and LcATL8a were located on the contigs. The LcAAAP proteins ranged from 288 (LcAPP17) to 1109 (LcAPP7a) amino acids, with molecular weights between 31.31 (LcAPP17) and 123.62 (LcAPP7a) kDa. The predicted pI values ranged from 5.56 (LcATL8b) to 9.9 (LcPorT1). The subcellular location predicted result showed the 48 genes can be located in the cell membrane (Table 1). LcATL1b only was located on the chloroplast, and LcAPP6a, LcATL8a, and LcAPP6b only were located on the Golgi apparatus. Additionally, the 17 genes could be located in other parts of the cell without the cell membrane, such as the chloroplast, cytoplasm, mitochondrion, and nucleus. There was a certain relationship between multiple positioning sites and gene function.  (Zm), and V. vinifera (Vv) AAAP proteins. Multiple sequence alignment of full-length proteins was performed by muscle and the phylogenetic tree using the maximum-likelihood estimation with IQtree2. 13 and 1000 bootstrap replications: the optimal model was JTT+F+R7. The tree was divided into eight subgroups, marked by different-colored backgrounds.  alignment of full-length proteins was performed by muscle and the phylogenetic tree using the maximum-likelihood estimation with IQtree2. 13 and 1000 bootstrap replications: the optimal model was JTT+F+R7. The tree was divided into eight subgroups, marked by different-colored backgrounds.

Evolutionary Analysis of LcAAAP Genes
To explore the LcAAAP gene family's evolutionary relationship with other species, a phylogenetic tree was constructed based on the AAAP protein sequences of L. chinense (Lc), Oryza sativa (Os), Arabidopsis thaliana (At), Amborella trichopoda (Atr), Sorghum bicolor (Sb), Populus trichocarpa (Pt), Zea mays (Zm), and Vitis vinifera (Vv) ( Figure S1). The basic information on this protein is listed in Table S1. The AAAP gene family could be divided into eight subgroups: Lysine and Histidine Transporters (LHT), Amino acid permease (AAP), proline transporter (ProT), gamma-aminobutyric acid transporter (GAT), auxin transporter (AUX), Amino acid-like transporter (ATLa and ATLb), and Aromatic and neutral amino acid transporters (ANT). The LcAAAP gene family also was divided into eight subgroups, including 6 LHT members, 15 AAP members, 2 ProT members, 5 GAT members, 5 AUX members, and 2 ANT members. A total of 10 members were classed into ATLa, and seven members belong to the ATLb ( Figure 2). The phylogenetic tree result showed that many members of LcAAAP were clustered into a clade with VvAAAP and AtrAAAP; it was indicated that the AAAP gene family of L. chinense, A. trichopoda and V. vinifera have a closer kinship (Figure 1 and Figure S1).
Understanding gene structure can provide information about evolution and gene function. Therefore, the gene structures, protein conserved motif., and transmembrane topology were analyzed. The LcAAAP contains different exon numbers from 2 to 15 ( Figure 3). Different subfamilies show different patterns of gene structure. Most AAAP gene family members within each subgroup share the same or similar gene structure and gene length. However, there were also certain differences in the structure of genes between members of the same subfamily: for example, LHT4 only has two exons, but 15 exons were detected in LHT2. Gene structure analysis showed that the gene structure of the LcAAAP gene family was relatively diverse, which may indicate the evolutionary trend of diversity and diverse gene functions. At the same time, if the positions of exons and introns in the same subfamily are relatively conserved, this may indicate that members of the same subfamily share a close evolutionary relationship. In addition, we found that multiple LcAAAP gene family members possess ultra-long introns (>10 kb); this result may negatively affect gene transcription. Motif analysis indicated that each subfamily has its own unique motif distribution pattern ( Figure 4 and Table S2). Most of the genes that were revealed to be closely related by the phylogenetic analysis had a conserve motif type and distribution. Most of the genes that were revealed to be closely related by the phylogenetic analysis had a conserved motif type and distribution. Moreover, some conserved motifs were widely distributed among AAAP members, such as motif1, 4, and 6. In contrast, most conserved motifs are specific, appearing only in specific subfamilies. The motifs 14,18,19, and 20 belong only to the AUX subgroup, and motif 5 only exists in the APP subgroup; this result may herald their association with functions unique to subfamilies. In addition, some motifs such as 9,11,12,13, and 17 were mainly distributed in one subgroup, but they were sporadically distributed among individual members of other subfamilies; this may affect changes in the function of certain genes.
Most AAAP gene family members are responsible for the transport of amino acids and mediate their transport across the cell's membrane structure. This indicates that the transmembrane structure of the AAAP protein has a specific connection with the function of the gene. We predicted the LcAAAP gene family transmembrane domain ( Figure 5 and Figure S3 and Table S3). The result showed that the LcAAAP gene family typically has 8-12 transmembrane domains, and the location and number of transmembrane domains were relatively conserved, indicating that members of the same subfamily may have similar functions. However, there were also some members whose transmembrane domains were quite different, such as LcATL1a (14), LcLHT2 (17), LcAPP4b (17), LcATL11 (4), LcAAP17 (5), LcATL9 (6), and LcPorT1 (6). Reduced or expanded numbers of transmembrane domains indicate that their functions may change.
ci. 2022, 23, x FOR PEER REVIEW 6 of 29 into eight subgroups: Lysine and Histidine Transporters (LHT), Amino acid permease (AAP), proline transporter (ProT), gamma-aminobutyric acid transporter (GAT), auxin transporter (AUX), Amino acid-like transporter (ATLa and ATLb), and Aromatic and neutral amino acid transporters (ANT). The LcAAAP gene family also was divided into eight subgroups, including 6 LHT members, 15 AAP members, 2 ProT members, 5 GAT members, 5 AUX members, and 2 ANT members. A total of 10 members were classed into ATLa, and seven members belong to the ATLb ( Figure 2). The phylogenetic tree result showed that many members of LcAAAP were clustered into a clade with VvAAAP and AtrAAAP; it was indicated that the AAAP gene family of L. chinense, A. trichopoda and V. vinifera have a closer kinship (Figures 1 and S1). Understanding gene structure can provide information about evolution and gene function. Therefore, the gene structures, protein conserved motif., and transmembrane topology were analyzed. The LcAAAP contains different exon numbers from 2 to 15 (Figure 3). Different subfamilies show different patterns of gene structure. Most AAAP gene analysis had a conserved motif type and distribution. Moreover, some conserved motifs were widely distributed among AAAP members, such as motif1, 4, and 6. In contrast, most conserved motifs are specific, appearing only in specific subfamilies. The motifs 14,18,19, and 20 belong only to the AUX subgroup, and motif 5 only exists in the APP subgroup; this result may herald their association with functions unique to subfamilies. In addition, some motifs such as 9,11,12,13, and 17 were mainly distributed in one subgroup, but they were sporadically distributed among individual members of other subfamilies; this may affect changes in the function of certain genes.   Most AAAP gene family members are responsible for the transport of amino acids and mediate their transport across the cell's membrane structure. This indicates that the transmembrane structure of the AAAP protein has a specific connection with the function of the gene. We predicted the LcAAAP gene family transmembrane domain ( Figures 5 and  S3 and Table S3). The result showed that the LcAAAP gene family typically has 8-12 transmembrane domains, and the location and number of transmembrane domains were rela-  Table S3). The result showed that the LcAAAP gene family typically has 8-12 transmembrane domains, and the location and number of transmembrane domains were relatively conserved, indicating that members of the same subfamily may have similar functions. However, there were also some members whose transmembrane domains were quite different, such as LcATL1a (14), LcLHT2 (17), LcAPP4b (17), LcATL11 (4), LcAAP17 (5), LcATL9 (6), and LcPorT1 (6). Reduced or expanded numbers of transmembrane domains indicate that their functions may change.

Synteny Analysis of LcAAAP Genes in L. chinense, Grape, Arabidopsis, and Rice
The expansion and contraction of gene families are affected mainly by whole genomic duplication and by tandem and segmental duplication. L. chinense has a single lineage-specific WGD event that occurred approximately 116 million years ago. This may

Synteny Analysis of LcAAAP Genes in L. chinense, Grape, Arabidopsis, and Rice
The expansion and contraction of gene families are affected mainly by whole genomic duplication and by tandem and segmental duplication. L. chinense has a single lineagespecific WGD event that occurred approximately 116 million years ago. This may cause the expansion of the LcAAAP gene family members. In addition, the whole-genome collinearity analysis result showed that one pair of tandemly duplicated even members (LcGAT1a-LcGAT1c) and six pairs of segmentally duplicated members (LcAPP4a-LcAPP4b, LcAUX1a-LcAUX1b, LcLHT1-LcLHT2, LcLAX5a-LcLAX5b, LcATL5a-LcATL5c, LcLAX2-LcLAX5b) were detected in the LcAAAP gene family ( Figure 6, Table S4). The GAT, APP, AUX, ATL, and LHT subgroups have member expansion through tandem and segmental duplication, especially the AUX subgroups. Moreover, the number of collinear genes shared by different species may reflect the gene family phylogenetic relationships. We performed the synteny analysis of the AAAP gene family between L. chinense, Arabidopsis thaliana, Oryza sativa, and Vitis vinifera (Figure 7). A total of 38 orthologous pairs were investigated between L. chinense and the other three species' AAAP genes. Among them, 10, 4, and 24 orthologous pairs were presented with genes in A. thaliana, rice, and grape, respectively. The LcAAAP and VvAAAP gene families have a large number collinear gene pairs, which shows that LcAAAP and VvAAAP are closer in evolutionary position. thaliana, Oryza sativa, and Vitis vinifera (Figure 7). A total of 38 orthologous pairs were investigated between L. chinense and the other three species' AAAP genes. Among them, 10, 4, and 24 orthologous pairs were presented with genes in A. thaliana, rice, and grape, respectively. The LcAAAP and VvAAAP gene families have a large number collinear gene pairs, which shows that LcAAAP and VvAAAP are closer in evolutionary position.
To verify whether the duplicated homologous genes of AAAP were selected during the evolution to adapt to external changes, we measured the Ka/Ks nucleotide substitution ratios of collinear genes to study the exerted selective pressure (Table S4). In the LcAAAP, we found that their Ka/Ks << 1, indicating that the duplicated LcAAAP genes have undergone a purifying selection during their evolutionary history. Compared with the Arabidopsis thaliana, Oryza sativa, and Vitis vinifera, most of the AAAP gene pairs Ka/Ks << 1, it was mean that the orthologous LcAAAP genes of the three species were subjected to purifying selection during evolution. However, for two gene pairs between LcAAAP and VvAAAP gene families Ka/Ks>1, it was indicated that the two pairs have positive selection during their evolution.

Codon Usage Bias Analysis
The codon usage pattern reflects the evolution and mutation of species or genes. The relative synonymous codon usage and the relative frequency of synonymous codons of LcAAAP genes were calculated with CodonW (Table S5). Here, 30, 33, and 1 codon had a positive bias, negative bias (RSCU < 1), and no bias (RSCU =1), respectively. In the more frequently used codons, some amino acids (RSCU > 1) such as Leucine (Leu), Isoleucine (Ile), serine (Ser), and threonine (Thr) have two or more codons with positive bias. Moreover, if an RFSC value exceeds 60% or is 0.5 times greater than the average frequency of synonymous codons, the codons are considered high frequency. The RFSC analysis result showed that UUC, AAG, and AGC showed high frequency. Compared with the AtAAAP, OsAAAP, and AtrAAAP codon usage frequency ratios, 1, 29, and 1 codon had ratios greater than 2.00 or lower than 0.50, respectively. Among them, 29 codons had significant differences with the AAAP genes of O. sativa. This result showed that the LcAAAP codons usage frequency is like that of AtAAAP and AtrAAAP and that Arabidopsis can be used as one of the choices for LcAAAP gene receptors.

Analysis of Cis-Acting Elements in LcAAAP Promoters
To further investigate which process could be regulated through the LcAAAP, the cisacting element promoter was analyzed by the online tool PlantCARE. We detected 111 type cis-acting elements and multiple elements that could be involved in different processes. Among them, 43 type elements were related to light responsiveness within all the To verify whether the duplicated homologous genes of AAAP were selected during the evolution to adapt to external changes, we measured the Ka/Ks nucleotide substitution ratios of collinear genes to study the exerted selective pressure (Table S4). In the LcAAAP, we found that their Ka/Ks << 1, indicating that the duplicated LcAAAP genes have undergone a purifying selection during their evolutionary history. Compared with the Arabidopsis thaliana, Oryza sativa, and Vitis vinifera, most of the AAAP gene pairs Ka/Ks << 1, it was mean that the orthologous LcAAAP genes of the three species were subjected to purifying selection during evolution. However, for two gene pairs between LcAAAP and VvAAAP gene families Ka/Ks>1, it was indicated that the two pairs have positive selection during their evolution.

Codon Usage Bias Analysis
The codon usage pattern reflects the evolution and mutation of species or genes. The relative synonymous codon usage and the relative frequency of synonymous codons of LcAAAP genes were calculated with CodonW (Table S5). Here, 30, 33, and 1 codon had a positive bias, negative bias (RSCU < 1), and no bias (RSCU =1), respectively. In the more frequently used codons, some amino acids (RSCU > 1) such as Leucine (Leu), Isoleucine (Ile), serine (Ser), and threonine (Thr) have two or more codons with positive bias. Moreover, if an RFSC value exceeds 60% or is 0.5 times greater than the average frequency of synonymous codons, the codons are considered high frequency. The RFSC analysis result showed that UUC, AAG, and AGC showed high frequency. Compared with the AtAAAP, OsAAAP, and AtrAAAP codon usage frequency ratios, 1, 29, and 1 codon had ratios greater than 2.00 or lower than 0.50, respectively. Among them, 29 codons had significant differences with the AAAP genes of O. sativa. This result showed that the LcAAAP codons usage frequency is like that of AtAAAP and AtrAAAP and that Arabidopsis can be used as one of the choices for LcAAAP gene receptors.

Analysis of Cis-Acting Elements in LcAAAP Promoters
To further investigate which process could be regulated through the LcAAAP, the cis-acting element promoter was analyzed by the online tool PlantCARE. We detected 111 type cis-acting elements and multiple elements that could be involved in different processes. Among them, 43 type elements were related to light responsiveness within all the LcAAAP genes (Table S6). In addition, we mainly analyzed and classified cis-acting elements related to hormones and stress. (Figure 8). Here, 11 and 3 type elements related to hormones and stress were identified, respectively. Most obviously, the ABRE (cis-acting element involved in the abscisic acid responsiveness) elements were abundantly enriched in certain genes, such as LcATL7d/8b/9 and LcAVT6 et al. Furthermore, the elements related to drought induction (MBS), low temperature responsiveness (LTR), auxin response (AAuxRR-core and TGA-element), and MeJA response (CGTCA-motif and TGACG-motif) were also recognized in many LcAAAP promoters. It was suggested that these genes may be involved in hormone signaling and stress response. In general, different gene promoters have different regulatory elements, suggesting the complexity of gene regulatory networks. In Figure 6, we found that 51 LcAAAP genes have one or more stress response elements. Among them, 32 and 37 genes could respond to low temperature signals and drought signals by MYB gene binding sites, and 18 genes could be involved in defense and stress responsiveness. Auxin, abscisic acid, salicylic acid, gibberellin, and methyljasmone response elements were distributed on all LcAAAP gene promoters. The most widely distributed was the ABA element (50), followed by the MeJA element (44 genes), GA response element (40), auxin response element (33), and SA response element (1).

Expression Patterns of LcAAAP Genes in Various Organs, Somatic Embryogenesis, and Response to Different Stress
We combined the published transcriptome data to determine the expression patterns of individual LcAAAP genes in seven organs (bract, leaf, shoot apex, stamen, petal, pistil, and sepal) (Figure 9). A total of 52 LcAAAP genes were divided into four groups based on their expression profiles in a hierarchical clustering heat map. In group 1, the 13 LcAAAP genes (LcAUX1a/1b, LcLHT1/4/8/11, LcATL5a/7b/9, LcAPP2/7b, LcLAX5a, and LcAVT6) had a high transcripts per million (TPM) value in various organs, except for the LcLHT8 expression level in the leaf. It was suggested this group gene could have an important role in the flower. In group 2, the genes expression level of the six LcAAAP genes (LcATL1b/5c, LcAPP7a/7c, LcANT2, and LcPorT2) maintain a steady state in different organs. The remaining 7 genes (LcAPP1a/4a and LcATL1a/7b/8b/12a/12b) were only specifically expressed in certain organs. In groups three and four, 12 genes (LcAPP1b/4b/4c/6b/6c/6d/17, LcGAT1b/1c/4, LcLHT3, and LcPorT1) lacked expression or had low expression levels. The LcATL15 and LcLHT2 were specifically highly expressed in the stamen; it was suggested that they could be involved in stamen development. The LcLAX5b only was found in the shoot apex. In addition, some genes only were detected at a relatively high expression level in the leaf, shoot apex, stamen, and pistil, such as LcLAX5b and LcLHT6, which had higher expression levels only in the shoot, apex, and pistil.
networks. In Figure 6, we found that 51 LcAAAP genes have one or more stress response elements. Among them, 32 and 37 genes could respond to low temperature signals and drought signals by MYB gene binding sites, and 18 genes could be involved in defense and stress responsiveness. Auxin, abscisic acid, salicylic acid, gibberellin, and methyljasmone response elements were distributed on all LcAAAP gene promoters. The most widely distributed was the ABA element (50), followed by the MeJA element (44 genes), GA response element (40), auxin response element (33), and SA response element (1).

Expression Patterns of LcAAAP Genes in Various Organs, Somatic Embryogenesis, and Response to Different Stress
We combined the published transcriptome data to determine the expression patterns of individual LcAAAP genes in seven organs (bract, leaf, shoot apex, stamen, petal, pistil, and sepal) (Figure 9). A total of 52 LcAAAP genes were divided into four groups based on their expression profiles in a hierarchical clustering heat map. In group 1, the 13 LcAAAP genes (LcAUX1a/1b, LcLHT1/4/8/11, LcATL5a/7b/9, LcAPP2/7b, LcLAX5a, and LcAVT6) had a high transcripts per million (TPM) value in various organs, except for the LcLHT8 expression level in the leaf. It was suggested this group gene could have an important role in the flower. In group 2, the genes expression level of the six LcAAAP genes (LcATL1b/5c, LcAPP7a/7c, LcANT2, and LcPorT2) maintain a steady state in different organs. The re- We compared the expression levels of the LcAAAP gene family in mid-petal development (1-4) of L. chinense and L. tulipifera (Figure 9 and Table S7). The expressions of some members of the second and fourth groups caught our attention. In L. tulipifera, the expressions of LcAPP1a/1b/4a and LcANT8a were upregulated to a very high level (58-fold, 67-fold, 79-fold, and 23-fold, respectively) in the second or third stage of petal development, whereas in L. chinense petals, the expression levels of these genes were low or significantly reduced. In contrast, the expression level of LcLHT8, LcATL9, and LcAUX1a in the petals of L. tulipifera was higher than that of L. chinense through the development of petals. In addition, we also explored the expression pattern of LcAAAP during somatic embryogenesis of hybrid Liriodendron (Figure 9 and Table S7). Based on the laboratory-established hybrid Liriodendron somatic embryogenesis system, we divided the process into 11 stages (PEM: embryogenic callus; ES1: 10 days after liquid culture; ES2: 2 days after screening; ES3: ABA 1 day of treatment; ES4: ABA treatment for 3 days; ES5: globular embryo; ES6: heart-shaped embryo; ES7: torpedo embryo; ES8: immature cotyledon embryo; ES9: mature cotyledon embryo; PL: plantlet). The expression levels of LcLHT1 and LcATL5a were continually upregulated 23.9-fold and 22.5-fold in the embryogenic callus stage (ES1-ES4), respectively. In LcLHT1, LcATL5a, and LcAUX1a/1b, there were certain genes (LcLHT1/8, LcATL1a/5a/8, LcAUX1a/1b, LcAPP4/9, and LcLAX2/5a) that were consistently highly expressed in the somatic embryogenesis stage, and the LcAUX1a/1b were stably expressed in all 11 stages of hybrid Liriodendron somatic embryogenesis. These results indicate that these genes may be involved in some important regulatory pathways in the process of somatic embryogenesis of hybrid Liriodendron. genes (LcLHT1/8, LcATL1a/5a/8, LcAUX1a/1b, LcAPP4/9, and LcLAX2/5a) that were consistently highly expressed in the somatic embryogenesis stage, and the LcAUX1a/1b were stably expressed in all 11 stages of hybrid Liriodendron somatic embryogenesis. These results indicate that these genes may be involved in some important regulatory pathways in the process of somatic embryogenesis of hybrid Liriodendron. We determined the expression pattern of the LcAAAP gene family under drought, low temperature (4 °C), and heat stress in the leaf (Figure 10). The 52 members were clustered into multiple groups following the different stresses. In response to the heat (40 °C) treatment, four groups were found by gene expression level. The group 3 gene (LcAPP4a, LcLHT1, and LcATL12b) expression levels were downregulated quickly in a short period of time (12 h) and then upregulated. In group 2, the LcATL5a and LcATL9 gene expression levels showed a trend of downregulation and upregulation within 12 h after heat stress treatment, respectively, and then recovered. Downregulated expression levels were de- We determined the expression pattern of the LcAAAP gene family under drought, low temperature (4 • C), and heat stress in the leaf (Figure 10). The 52 members were clustered into multiple groups following the different stresses. In response to the heat (40 • C) treatment, four groups were found by gene expression level. The group 3 gene (LcAPP4a, LcLHT1, and LcATL12b) expression levels were downregulated quickly in a short period of time (12 h) and then upregulated. In group 2, the LcATL5a and LcATL9 gene expression levels showed a trend of downregulation and upregulation within 12 h after heat stress treatment, respectively, and then recovered. Downregulated expression levels were detected after 1h of treatment for six genes (AUX1a/1b, LcAPP1a/2/7b, LcLAX5b). In the other genes of groups 1 and 4, the expression levels of most genes in leaves were very low or not expressed. Under drought stress, we did not observe significant up-or downregulation of LcAAAP gene expression except for LcATL5a, which was downregulated as processing time grew. In addition, the expression levels of some members only changed at a certain point, and eventually all returned to the original pattern, such as the LcAPP4a and the LcAAP7b, which upregulated only at 24 h and 12 h, respectively. Regarding the effect of low temperature on gene expression, LcATL5a, LcAPP4a, and LcLHT1 upregulated after 12 h exposure to 4 • C. The LcAPP7b and LcAUX1a/1b showed a trend of downregulated expression. In addition, we found that the LcATL12b expression level increased rapidly in 12 h. In this result, the expression abundances of most genes changed in a short period of time during different stress treatments, but the changes were small. It was worth noting that LcATL5a, LcAPP4a, and LcLHT1 showed different expression patterns under multiple stress treatments. These results suggest that diverse mechanisms control LcAAAP gene responses to various stresses.

Prediction and Correlation Analysis of LcAAAP Interacting Proteins
The construction of protein interaction networks is of great significance for the study of gene interactions and regulatory relationships. We demonstrate the protein interaction

Prediction and Correlation Analysis of LcAAAP Interacting Proteins
The construction of protein interaction networks is of great significance for the study of gene interactions and regulatory relationships. We demonstrate the protein interaction between the LcAAAP gene family and the protein interaction between LcAAAP protein and LcMYB/WRKY transcription factor ( Figure 11 and Table S8). The protein interaction network within the LcAAAP gene family was divided into three groups. The first group, LcANT2, can interact with LcAPP4c/6c/9 and LcGAT2/4, and LcPorT1 can interact with LcATL1b/8a as the second group. In addition, a complex network of interactions involving some of the other LcAAAP members was constructed in the third group. LcLAX2, LcAPP2, and LcATL9 can interact with each other. LcATL7d can interact with the LcLAX2, LcAPP2, LcATL12b, and LcANT1. Some transcriptional regulators that may be involved in the regulation of LcAAAP protein were also identified ( Figure 11). The result showed that LcMYB118 and LcWRKY22 could interact with LcAPP2, and LcMYB40/108 could interact with LcLAX2, respectively. LcAUX1b can be independently regulated by multiple transcription factors (LcMYB77/89/108/186), and LcWRKY18, LcMYB91, and LcMYB25 can be regulated with LcAPP7b, LcAVT6, and LcLHT4, respectively.
To further explore whether LcAAAP was involved in heat, drought, and low temperature stresses ( Figure 12 and Table S9), we analyzed the correlation between LcAAAPs and the LcMYB/WRKY transcription factors involved in abiotic stress. The result showed a positive correlation (r > 0.7) between the expression levels of LcAPP4a and LcWRKY28, LcMYB83, LcMYB141, LcWRKY5, LcMYB71, and LcMYB38. LcLHT1 showed a positive correlation with LcWRKY14, LcWRKY15, LcWRKY16, LcWRKY17, LcWRKY27, LcMYB9, LcMYB37, LcMYB38, LcMYB83, and LcMYB105. In the same way, there was a significant positive correlation among the expression levels of LcATL12b, LcMYB28, LcMYB38, LcMYB56, and LcWRKY17. These results further showed that LcAPP4, LcLHT1, and LcATL12b could be involved in the abiotic stress response process. In addition, some genes may be specifically regulated only in certain abiotic stresses: for instance, under heat stress, the expression level of LcAPP2 has a positive correlation with LcMYB32, LcMYB43, LcMYB84, LcMYB89, LcMYB110, LcMYB113, LcMYB138, and LcMYB144. A negative correlation (r < −0.7) was identified in LcMYB12 and LcMYB84. In summary, the correlation analysis shows that a few LcAPPPs participate in the stress response.  To further explore whether LcAAAP was involved in heat, drought, and low temperature stresses (Figure 12 and Table S9), we analyzed the correlation between LcAAAPs and the LcMYB/WRKY transcription factors involved in abiotic stress. The result showed

Response of GAT and PorT Subgroups to Al Stress
Wang et al. showed that gamma-aminobutyric acid (GABA) signaling could enhance tolerance of hybrid Liriodendron to Al stress by promoting organic acid transport and maintaining cellular redox and osmotic balance. The current study shows that GAT is specific for GABA transport, whereas PorT mainly transports proline, but PorT of some species can also respond to GABA. Therefore, we explored the responses of GAT and PorT family members under the three treatment methods: Al stress, GABA treatment ( Figures  13-15), and Al stress followed by GABA treatment. In this process, the six genes of the GAT and PorT subgroups have varying degrees of expression except for LcGAT1b, which were not detected.

Response of GAT and PorT Subgroups to Al Stress
Wang et al. showed that gamma-aminobutyric acid (GABA) signaling could enhance tolerance of hybrid Liriodendron to Al stress by promoting organic acid transport and maintaining cellular redox and osmotic balance. The current study shows that GAT is specific for GABA transport, whereas PorT mainly transports proline, but PorT of some species can also respond to GABA. Therefore, we explored the responses of GAT and PorT family members under the three treatment methods: Al stress, GABA treatment (Figures 13-15), and Al stress followed by GABA treatment. In this process, the six genes of the GAT and PorT subgroups have varying degrees of expression except for LcGAT1b, which were not detected.   Mock represents the ultrapure water treatment; Al represents the aluminum stress treatment; GABA represents exogenously applied GABA treatment; Al+GABA represents Al stress treatment and then exogenously applied GABA treatment.
In the Al stress, the LcGAT1a, LcGAT1c, LcGAT2, and LcGAT4 increase the expression abundance in root and stem after 48 h under Al stress. LcGAT2 was upregulated in leaves after 48 h treatment, while LcGAT4 expression in leaves and shoots began to increase after Mock represents the ultrapure water treatment; Al represents the aluminum stress treatment; GABA represents exogenously applied GABA treatment; Al+GABA represents Al stress treatment and then exogenously applied GABA treatment.
In the Al stress, the LcGAT1a, LcGAT1c, LcGAT2, and LcGAT4 increase the expression abundance in root and stem after 48 h under Al stress. LcGAT2 was upregulated in leaves after 48 h treatment, while LcGAT4 expression in leaves and shoots began to increase after 24 h treatment. LcPorT1 was consistently upregulated in different organs. LcPorT2 showed two expression patterns in different organs, downregulated first and then upregulated in stem and bud, in contrast to expression in leaves.
Under GABA stress, the expression level of LcGAT1a in roots and shoots was downregulated after 24 h of treatment, and the expression of LcGAT1c in leaves and shoots was downregulated after 12 h of treatment. Compared with other LcGAT genes, the expression level of LcGAT2 in the stem continued to increase after 12 h of treatment. In the PorT subgroup, LcPorT1 could be upregulated in four organs after 12 h of stress. Interestingly, the expression trend of LcPorT2 in stems, leaves, and shoots was the inverse of that of LcPorT1, suggesting that their responses to GABA may have mutually antagonistic effects.
LcGAT1a and LcPorT1 were upregulated in stem and leaf at 12 h after the GABA was applied after the Al stress treatment. A similar situation exists in other GAT genes. For example, expression levels of LcGAT1a, LcGAT1c, and LcGAT2 were upregulated in leaves at 12 h or 48 h; LcGAT2 also was upregulated in stem and bud at 12 h and 24 h, respectively. In addition, the downregulated abundance genes compared with the Al stress in different organs also were detected. LcGAT1a, LcGAT1c, LcGAT2, and LcPorT1 were downregulated in root, stem, and bud after 48 h of treatment. LcPorT2 was downregulated in stem and bud at 48 h.
The above analysis results show that the hybrid Liriodendron may respond to changes in the external environment by changing the content of specific substrates in cells of different organs, thereby affecting the expression of the LcGAT and LcPorT subgroups' genes.

Discussion
As one of the largest gene superfamilies of AAT, the AAAP gene family can be divided into eight subfamilies according to existing research, which plays an important role in plant physiology and stress resistance research and has great potential value. Relevant research work has been carried out in a variety of model plants. such as Arabidopsis, rice, potato, and polar et al. [10,11,39,40]. Although the functions and roles of AAAP genes have been explored in multiple species, systematic analysis of the AAAP gene family has not been completed after the publication of the genome of L. chinense. In this research, 52 LcAAAP genes were identified based on the genome of L. chinense, and eight subgroups were divided by the AAAP gene family phylogenetic tree of multiple species. Although the cluster clade was like other AAAP gene families, the numbers of members in different subgroups differs significantly, suggesting that different degrees of expansion occurred in different species.
In addition to LcAPP6c, LcATL7d, LcATL7b, and LcATL8a, the remaining 48 genes were unevenly mapped to 14 chromosomes, and most genes were distributed in chromosomes 4, 2, and 17. Subcellular localization results showed that 48 LcAAAP genes were located on the cell membrane; it was suggested that these genes were mainly used for membrane localization to experiment with the transport of substances inside and outside the cell. Among the LcAAAP genes, many genes have multiple locations (17) or are located only in the chloroplast or Golgi (4). It was indicated that these genes could also function in different organelles. To better understand LcAAAP gene function and evolutionary history, we explored different perspectives. Exon-intron structure analysis indicated that most LcAAAP genes in the same subgroup had the same or similar numbers. However, there are still structural inconsistencies in some genes, such as differences in the number of exons and the existence of ultra-long introns. Most AAAP gene family members are responsible for the transport of amino acids with different specificities and characteristics and mediate their transport across the cell's membrane structure. It was suggested that the existence of the transmembrane (TM) domain plays an extremely important role in the AAAP genes. 7-11 TM were identified in most of the LcAAAP genes of the same subgroups. However, there are still some genes (LcLHT2/4, LcATL11, LcPorT1, and LcAPP4/17) whose numbers of TMs have expanded or shrunk. Compared with the results of the motif analysis, the conserved motif types of different subfamilies were found to be consistent with the transmembrane domains. This result indicates that the composition of the transmembrane domain is closely related to gene function and further suggests a direct connection between gene motifs and gene function.
WGD and gene duplication events were thought to be the main causes of gene family expansion and functional diversity [41,42]. In the Magnolia plants, two and one wholegenome duplication events occurred in C. kanehirae and L. chinense [1,43], which could lead to the members of the gene family expanding. Currently, 25% (13/52) of AAAP genes are duplicated genes in L. chinense, which exist as one pair of tandemly duplicated and six pairs of segmental duplicated events, especially in the LcAUX subgroup, including three pairs of gene duplication events. These results showed that the segmentally duplicated event was the main cause of expression of the LcAAAP gene family. In addition, we analyzed the collinear relationship of AAAP genes between L. chinense, Arabidopsis thaliana, Oryza sativa, and Vitis vinifera. The 24 orthologous gene pairs were identified between LcAAAPs and VvAAAPs, greater than the 10 and 4 orthologous gene pairs found in LcAAAP-AtAAAP and LcAAAP-OsAAAP. It was indicated that there may be a closer kinship between LcAAAPs and VvAAAPs. At the same time, there were collinear relationships between the LcAAAP genes and VvAAAP/AtAAAP/OsAAAP, which also proves that the LcAAAP genes may have been formed before monocotyledonous and dicotyledonous plants. These opinions can be verified from a genome-wide perspective: magnoliids arose before the divergence of eudicots and monocots, represented by Liriodendron [1]. The Ka/Ks ratio of orthologous genes' duplicated events was used to evaluate the AAAP genes contribute to organism fitness: Ka/Ks < 1 for most paralog gene pairs, both within and between species, which indicated that these paralog gene pairs encountered various degree purifying selective pressures, demonstrating that complex selective pressures drove the evolution of the AAAP gene family. Gene duplication events provide raw materials for regulating physiological and morphological changes in plants and provide a basis for species to adapt to changes in the external environment [44]. LcAUX1a and LcAUX1b have similar expression patterns in various organs and stresses, which indicated that they may have redundant functions. However, the other paralog gene pairs all showed different expression patterns, which, because of the duplicated genes, always tend to be sub-functionalized, neo-functionalized, or both [45]. This result suggested that the strong purifying selection may have inflected the gene functional divergence to adapt to diversity in the environment.
There was a very important relationship between the biological function of genes and their expression level, and a comprehensive analysis of the expression level of the genes can analyze the putative gene function in the process of plant growth and development [46]. In this study, the expression profiles of the LcAAAP gene family were analyzed in different organs and stages of Liriodendron. Approximately 28 and 34 LcAAAP genes were expressed at relatively high levels (TPM > 1) in the leaves and shoot apex of L. chinense, respectively. Here, 40 genes showed relatively high expression levels in the flowers of L. chinense. We found that LcLHT1, LcLHT2, and LcATL15 have high expression in leaves, bract, and stamens, respectively. This result is like that of paralogous genes in Arabidopsis [19,28,47]. Therefore, LcLHT2 could play an important role in amino acid transport in pollen development and maturation, and LcALT15 also might be involved in the long-distance transport of amino acids in stamens. In addition, the expression levels of LcLHT8 were significantly higher in floral organs than in leaves; it was indicated that LHT8 could also have an important role in flowering. Nectar generally contains substances such as carbohydrates, amino acids, inorganic ions, proteins, and lipids [48] and the central area of the petals was thought to be the location of the Liriodendron nectary [49,50]. The LcAAP1a gradually upregulated during the development of nectar in L. chinense; and the expression levels of LcAAP1a, LcAAP1b, and LcAAP4 in the nectar of L. tulipifera were significantly upregulated in the second stage relative to L. chinense, which indicated that the increased levels of amino acid transport in the nectar implies that the amount of nectar secreted by L. tulipifera would be greater than that of L. chinense, which is consistent with reality. This result showed that a more complex amino acid transport network exists in the nectar secretion of L. tulipifera. As protein synthesis substrates, amino acids play an indispensable role in the process of plant somatic embryogenesis [51,52]. As the embryogenic cells develop toward the globular embryo stage in hybrid Liriodendron (PEM-ES4), the LcLHT1 and LcATL5a were specifically upregulated in this progress. During subsequent embryonic development (ES5-ES8), the LcLHT1/8, LcATL1a/5a, and LcAAP4a maintained stable levels, and the expression levels of LcAUX1a, and LcLAX2 were significantly upregulated relative to the previous stage. In addition, the LcATL5a and LcAAP9 were upregulated, and the LcLAX2 and LcLAX5a were downregulated during plant morphogenesis (ES9-PL). These results indicate that protein and auxin transport plays an important role during somatic embryogenesis. In the first stage, a large amount of protein accumulates as the basis for subsequent embryonic development. In the second stage, the content of intracellular auxin decreases, resulting in assists in embryonic development, similar to the expression pattern of LcPIN genes [53]. In the third stage, the specific regulation of auxins and proteins in cells provides the basis for plant morphogenesis.
According to previous studies, AAAPs were involved in the low temperature, heat temperature, drought stress treatments in many plants [35,54,55]. In this study, we analyze the LcAAAP gene expression pattern based on the stress transcriptome data of hybrid Liriodendron [56,57]. A total of 13 genes in hybrid Liriodendron leaves were able to respond to the three abiotic stress treatments. Most genes' expression levels were downregulated in a short period of time. Interestingly, the significant changes occurred in the expression patterns of LcATL5a, LcAPP4a, and LcLHT1 in these data, and they showed completely opposite trends in heat and low temperature stress treatments. Cis-acting element analysis results showed that the promoters of LcAAAP genes' response to the abiotic have at least one of the LTR (response to the low temperature), MBS (MYB binding site involved in drought-inducibility), and TC-rich repeat elements (cis-acting element involved in defense and stress responsiveness), which showed that promoters could control structural and morphological changes in plant-environment interactions to adapt to unfavorable external environments [58]. How plants adapt to external abiotic stresses involves a very complex gene regulatory network, and a variety of transcription factors are usually involved in this process [58][59][60]. In previous studies, we found that WRKY and MYB genes were involved in regulating the response of hybrid Liriodendron to stress [56,57]. Therefore, we constructed the co-expression network based on the LcAAAP, LcMYB, and LcWRKY genes to explore the interactions between them. Multiple MYB and WRKY genes showed a significantly positive correlation, and LcWRKY5/27 and LcMYB9/28/83 could rise to a relatively high level in a short time under abiotic stress [56,57]. These results indicated that the LcLHT1, LcAPP4a, and LcATL5a could transport amino acids under abiotic stress, provide support for the internal homeostasis of plants, and then adapt to changes in the external environment, and the specific expression pattern of these genes was directly or indirectly regulated by upstream transcription factors.
Gamma-aminobutyric acid (GABA) enhances aluminum tolerance in hybrid Liriodendron [61]. Therefore, we explored the gene expression pattern of LcGAT and LcPorT, where GAT can specifically transport GABA and the PorT can barely transport GABA in addition to its ability to transport proline [34,36]. The LcGAT1a/1c/2/4 were significantly upregulated in different organs after Al stress treatment; and the treatment of exogenous GABA causes GATs to respond at different rates in different organs. For example, the expression level of LcGAT1c/2 in roots was gradually increased from 12 to 48 h, whereas it was gradually downregulated in buds. When exogenous GABA treatment was applied to Al stressed plants, the expression levels of LcGAT1/2 in stems and leaves were rapidly upregulated within 12 h, indicating that this treatment may accelerate the transport of GABA in plants, thereby revealing that the treatment might accelerate the transport of GABA in the plant, more quickly alleviating the effect of Al poisoning on the growth and development of hybrid Liriodendron. For example, the accumulation of GABA in leaves facilitates the closure of plant stomata, thereby increasing its tolerance [62]. In addition, proline accumulated in hybrid Liriodendron leaves under the effect of Al toxicity based on previous studies; therefore, the expression level of LcPorTs showed different degrees of upregulation under aluminum stress treatment. LcPorT1/2 showed opposite expression patterns in stems, leaves, and buds under GABA treatment; this indicates that there is also a relationship between the transport of LcPorTs and GABA, but the specific mode and pattern need further research.

Collection of Query Sequences and Identification of the AAAP Gene Family Based on L. chinense Genomic Data
AtAAAP gene family protein sequences were acquired from the UniProT database (https://www.uniprot.org/ (accessed on 12 February 2022)) [63], and all the AAAP gene family sequences of other species were obtained from the JGI (https://phytozome.jgi. doe.gov/pz/portal.html (accessed on 13 February 2022)) [64]. The index sequences consisted of 46 AtAAAP protein sequences based on the BLASTP. The 69 putative sequences were found from the L. chinense protein database through a local BLASTP search with an E-value threshold of e −10 . The Hidden Markov mode (HMM) profile for the AAAP domain (PF01490) downloaded from the Pfam database (http://pfam.xfam.org (accessed on 15 February 2022)) [65], was used to identify LcAAAP genes from the L. chinense genome with HMMER 3.3.2 (http://hmmer.janelia.org/ (accessed on 15 February 2022)), with an E-value of e −5 . A total of 73 putative sequences were identified from the genomic database. After all putative LcAAAP protein sequences were merged, the candidate sequences were further analyzed with the online tools Conserved Domain Database (http://www.ncbi.nlm.nih.gov/cdd/ (accessed on 16 February 2022)) [66] and pfam [67]. Finally, the 52 target sequences were identified and used for further analysis. Molecular weight and isoelectric point of each protein were determined using the online tool ExPASy (https://web.expasy.org/protparam/ (accessed on 17 January 2022)) [68]. The subcellular localization of all LcAAAP proteins was predicted based on the online tool Cell-PLoc 2.0 (http://www.csbio.sjtu.edu.cn/bioinf/Cell-PLoc-2/ (accessed on 17 January 2022)) [69].

Phylogenetic Tree, Gene Structure, and Conserved Motif and Transmembrane Region Analyses of LcAAAP Genes
Multiple sequence alignments analyses of AAAP amino acid sequences of Arabidopsis thaliana, Oryza sativa, Vitis vinifera, Sorghum bicolor, Populus trichocarpa, Zea mays, Amborella trichopoda, and L. chinense were performed with Muscle 3.8.31 [70]. We built the phylogenetic tree using the Maximum-Likelihood estimation with IQtree2.13 [71] and 1000 bootstrap replications; the optimal model was JTT+F+R7. The method for constructing the phylogenetic tree from the LcAAP protein sequences was consistent with the above; the optimal model was VT+F+G4. Evolutionary tree beautification was carried out using online tools iTOL (https://itol.embl.de/tree/2182103287321642042781 (accessed on 16 April 2022)) [72] and Adobe Illustrator CS3 software (version 13.0.0).

Chromosomal Location, Syntenic Analyses and Calculation of the Ka/Ks Value
The positions of the LcAAAP genes were acquired from the annotation file. Among them, 48 LcAAAP genes were mapped on 14 chromosomes, and the remaining four genes were located on the three contigs, with results listed in Table 1. We used MCScanX [76] with the default settings to identify LcAAAP gene pairs of segmental/tandem duplications in the L. chinense genome and analyzed the LcAAAP genes' syntenic relation with the AtAAAP, OsAAAP, and VvAAAP gene pairs. The KaKs_Calculator2.0 [77] software was used to calculate the Ka/Ks values of homologous genes.

Analysis of the Codon Usage Pattern and The Cis-Acting Element
We obtained the coding sequences of L. chinense, Arabidopsis thaliana, Oryza sativa, and Amborella trichopoda based on the above AAAP genes. The CodonW1.4.4 software [78] was used to calculate the relative synonymous codon usage (RSCU) and the relative frequency of synonymous codons (RFSC). As promoter sequences, 2500 bp upstream from the translation start sites for LcAAAP genes were considered, and the cis-acting elements were predicted and analyzed using PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html/ (accessed on 16 April 2022)) [79].

Plant Materials and Stress Treatment
According to the previous research, the GAT and ProT could transport the gammaaminobutyric acid (GABA), and GABA signaling could enhance the tolerance of hybrid Liriodendron to Al stress by promoting organic acid transport and maintaining cellular redox and osmotic balance [61]. To compare the GAT and ProT subgroup expression levels in organs of different stages in the Al stress, we chose the one-month hybrid Liriodendron with basically the same growth for stress treatment (Nanjing, China). The cultivation environment is the greenhouse of the Key Laboratory of Forest Genetics and Biotechnology of the Ministry of Education, Nanjing Forestry University, under white light (light for 16 h, darkness for 8 h). The concentrations of AlCl 3 and GABA were 30 uM and 10 mM, respectively; the culture medium was 3/4 MS. Hybrid Liriodendron seedlings were soaked in Al solution for 3 h and then inoculated on the corresponding medium; we obtained materials from three time periods, 12 h, 24 h, and 48 h, respectively. We quickly froze the material in liquid nitrogen and stored it at −80 • C.

RNA Extraction and Quantitative Real-Time PCR Analysis
Total RNAs were extracted using the KK Fast Plant Total RNA Kit (ZOMANBIO, Beijing, China). The Evo M-MLV RT Kit with gDNA Clean for qPCRII AG11711 (Accurate Biotechnology (Hunan) Co., Ltd., Changsha, China) was applied to synthesize the firststrand cDNA from 1.0 mg RNA. Equalbit 1× dsDNA HS Assay Kit (EQ121-01, vazyme) completed quantification of all reversed cDNAs. PCR amplifications were carried out with SYBR ® Green Premix Pro Taq HS qPCR Kit (Accurate Biotechnology (Hunan) Co., Ltd., Changsha, China) in 20 µL volumes using Roche LightCycler ® 480 Real-Time PCR System. Three replicates were performed for each selected gene. The 18S was taken as a reference gene. The relative expression levels were calculated by the ∆∆ CT method [80]. All qRT-PCR primers were designed by Primer5.0 [81] and are listed in Table S5.

Conclusions
In this study, we identified 52 AAAP genes from the L. chinense genome. Multiple analyses, including the construct phylogenetic tree, gene exon-intron structure, motif, and transmembrane domain prediction, indicated that the LcAAAP gene family was divided into eight subgroups. Most of the identified LcAAAP proteins have a close evolutionary relationship with A. trichopoda proteins, implying that they were relatively primitive among angiosperms. WGD and gene duplication events suggested that the expansion of some subgroups during the evolutionary history led to divergent gene functions. The expression patterns in different organs showed that the LcAAAPs were specifically expressed in the nectary development and somatic embryogenesis of Liriodendron, indicating that the LcAAAP gene family has coordinated to regulate growth and development. In addition, the LcAAAP gene family can also respond to low temperature, heat, and drought stresses through the regulation of WRKY/MYB genes; and the members of the GAT and PorT subgroups members can participate in the process of hybrid Liriodendron to resist Al poisoning.  Data Availability Statement: The datasets supporting the conclusions and description of a complete protocol can be found within the manuscript and its additional files. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.