Integrated Multi-Omics Analysis to Reveal the Molecular Mechanisms of Inflorescence Elongation in Medicago sativa

The morphological architecture of inflorescence influences seed production. The regulatory mechanisms underlying alfalfa (Medicago sativa) inflorescence elongation remain unclear. Therefore, in this study, we conducted a comparative analysis of the transcriptome, proteome, and metabolome of two extreme materials at three developmental stages to explore the mechanisms underlying inflorescence elongation in alfalfa. We observed the developmental processes of long and short inflorescences and found that the elongation capacity of alfalfa with long inflorescence was stronger than that of alfalfa with short inflorescences. Furthermore, integrative analysis of the transcriptome and proteome indicated that the phenylpropanoid biosynthesis pathway was closely correlated with the structural formation of the inflorescence. Additionally, we identified key genes and proteins associated with lignin biosynthesis based on the differential expressed genes and proteins (DEGs and DEPs) involved in phenylpropanoid biosynthesis. Moreover, targeted hormone metabolome analysis revealed that IAA, GA, and CK play an important role in the peduncle elongation of alfalfa inflorescences. Based on omics analysis, we detected key genes and proteins related to plant hormone biosynthesis and signal transduction. From the WGCNA and WPCNA results, we furthermore screened 28 candidate genes and six key proteins that were correlated with lignin biosynthesis, plant hormone biosynthesis, and signaling pathways. In addition, 19 crucial transcription factors were discovered using correlation analysis that might play a role in regulating candidate genes. This study provides insight into the molecular mechanism of inflorescence elongation in alfalfa and establishes a theoretical foundation for improving alfalfa seed production.


Introduction
The inflorescence architecture of flowering plants displays great diversity in nature and includes racemes, spikes, and capitates, which determine flower growth and seed production [1,2].This is an effective way to investigate the potential mechanisms of inflorescence development to improve the seed yield of plants and crops [3][4][5].Many small flowers emerge on the inflorescence peduncle, and the elongation capacity of the inflorescence peduncle affects raceme length and eventually influences flower and seed development [6].Alfalfa (Medicago sativa) is an important forage, food, and Chinese herbal medicine planted worldwide [7].The inflorescence architecture of alfalfa is a typical raceme, and its inflorescence length is a crucial factor affecting seed production [8].The length of the inflorescence is the structural basis for the growth of more florets, and the number of florets is positively correlated with the seed yield of alfalfa [9].The insufficient production of alfalfa seeds has limited the long-term development of the alfalfa industry.Breeders select a long-spike alfalfa cultivar based on the length of the inflorescence to increase the seed output of alfalfa, which provides an important germplasm resource for alfalfa.However, the molecular mechanisms underlying alfalfa inflorescence elongation have rarely been reported.
Lignin is an important component of plant secondary cell walls and is distributed in the cell walls of both supportive and conductive tissues [10].Lignin hardens the cell wall by forming interwoven webs, allowing for the xylem to remain highly rigid; this allows for lignin to carry the weight of aerial structures [11].Lignin is an essential metabolite for the development of vascular tissues; many studies have demonstrated that the accumulation and distribution of lignin affect the elongation and growth of plant organs [12,13].Additionally, lignin precursors are produced by the phenylpropanoid biosynthesis pathway; some crucial enzymes are involved in the synthesis of lignin, including PAL (L-phenylalanine ammonia-lyase), C4H (cinnamate 4-hydroxylase), 4CL (4-coumarate: CoA ligase) HCT (hydroxycinnamoyl CoA: shikimate hydroxycinnamoyl transferase), CCoAOMT (caffeoyl-CoA O-methyltransferase), CCR (cinnamoyl-CoA reductase), CAD (cinnamyl alcohol dehydrogenase), F5H (ferulate 5-hydroxylase), COMT (caffeic acid/5-hydroxyconiferaldehyde 3-O-methyltransferase), and POX/LAC (peroxidase/laccase) [14,15].In alfalfa, the inflorescent peduncle is the carrier for floret development that continuously synthesizes lignin during peduncle growth.Therefore, investigating lignin biosynthesis in the peduncle may reveal the molecular mechanisms of inflorescence elongation in alfalfa.
Plant hormones play an important role in regulating lignin biosynthesis, organic matter accumulation, and other physiological activities and eventually influence the growth and development of plant tissues and organs [16][17][18].Auxins (IAA) are frequently associated with stem elongation and root development and can promote or inhibit apical dominance [19,20].Gibberellin (GA), one of the plant hormones necessary for plant growth, plays a role in stem development, plant flowering, seed germination, and other metabolisms [21,22].CK can induce cell division and elongation and promote tissue differentiation, and the interaction between cytokinins and auxins plays a critical role in regulating xylem development [23,24].Moreover, other phytohormones, such as abscisic acid (ABA), jasmonate (JA), salicylic acid (SA), and ethylene (ETH), are closely connected with participating in plant growth and development [25][26][27].
Multi-omics techniques have been widely applied to understand the mechanisms underlying inflorescence growth and development [28,29].For example, Weng et al. investigated the transcriptional profiling between the two major inflorescence ecotypes in Panicum hallii based on the global transcriptome analysis; the result suggested the underlying effect of cytokinin signaling in heterochronic changes and provided some novel insights into the transcriptome of inflorescence divergence in Panicum hallii [30].Moreover, previous study has profiled differentially expressed genes (DEGs) at three developmental stages of flower growth and compared them with those from vegetative seedling tissue using RNA-sequencing analysis, which confirmed that key genetic regulators, plant hormones, and cell-cycle genes were involved in barley inflorescence development [31].In this study, we conducted transcriptomic, proteomic, and targeted phytohormone metabolomic analyses at three stages of inflorescence development, contrasting the two extreme materials to elucidate the regulatory mechanisms of inflorescence elongation in alfalfa.

Distinct Phenotypes of Long and Short Inflorescences between Two Extreme Materials in Alfalfa
In the present study, inflorescence development between long and short inflorescences was investigated in alfalfa.We classified the developmental process of the inflorescence into six stages, namely, inflorescence growth, including early budding stage 'a', full budding stage 'b', early flowering stage 'c', flowering stage 1 'd', flowering stage 2 'e', and full flowering stage 'f' (Figure 1A).As shown in Figure 1B, long inflorescence continuously elongated throughout the growth period; the length extended from 0.7 cm of 'a' to 6.2 cm of 'f'.It displayed that the length difference was not prominent in the short inflorescence; the length extended from 0.4 cm of 'a' to 1.6 cm of 'f' (Figure 1B).The results revealed that long inflorescences had a stronger ability for peduncle elongation than short inflorescences in alfalfa, which provided adequate space for more florets to grow and perform photo-synthesis.Notably, the inflorescence underwent pollination and fruiting when it reached full flowering, and the inflorescence peduncle almost stopped elongating and reached its maximum length.Based on these results, we speculated that the lack of ability for elongation during inflorescence development and most of the small flower buds blooming in a short amount of time were the major reasons for the structural formation of the short inflorescence.Conversely, the length of the long inflorescence continuously increased during inflorescence development; the florets gradually bloomed from the base, and the inflorescence peduncle elongated sustainably.The growth strategy of short inflorescences even led to the death of flower buds and florets owing to insufficient space, time, and nutrition for development, whereas the growth strategy of long inflorescences was able to support the growth and development of all florets and produce more seeds.
continuously elongated throughout the growth period; the length extended from 0.7 cm of 'a' to 6.2 cm of 'f'.It displayed that the length difference was not prominent in the short inflorescence; the length extended from 0.4 cm of 'a' to 1.6 cm of 'f' (Figure 1B).The results revealed that long inflorescences had a stronger ability for peduncle elongation than short inflorescences in alfalfa, which provided adequate space for more florets to grow and perform photosynthesis.Notably, the inflorescence underwent pollination and fruiting when it reached full flowering, and the inflorescence peduncle almost stopped elongating and reached its maximum length.Based on these results, we speculated that the lack of ability for elongation during inflorescence development and most of the small flower buds blooming in a short amount of time were the major reasons for the structural formation of the short inflorescence.Conversely, the length of the long inflorescence continuously increased during inflorescence development; the florets gradually bloomed from the base, and the inflorescence peduncle elongated sustainably.The growth strategy of short inflorescences even led to the death of flower buds and florets owing to insufficient space, time, and nutrition for development, whereas the growth strategy of long inflorescences was able to support the growth and development of all florets and produce more seeds.

Genetic and Proteomic Analysis of Two Extreme Materials at the Three Developmental Stages
To precisely elucidate the potential molecular mechanism of inflorescence elongation in alfalfa, we conducted a comparative transcriptome in two extreme materials at three developmental stages: the full budding stage of long and short inflorescences (L1 and S1), early flowering stage of long and short inflorescences (L2 and S2), and full flowering stage of long and short inflorescences (L3 and S3); the inflorescences were removed from the florets and the remaining inflorescence peduncles were used as experimental material (Figure 2).We detected a total of 127.56 Gb of clean data after removing low-quality reads, with more than 6 Gb of clean reads per library.The percentage of Q30 bases was, on average, 92%.No less than 73% of the clean reads from 18 samples could be mapped to

Genetic and Proteomic Analysis of Two Extreme Materials at the Three Developmental Stages
To precisely elucidate the potential molecular mechanism of inflorescence elongation in alfalfa, we conducted a comparative transcriptome in two extreme materials at three developmental stages: the full budding stage of long and short inflorescences (L1 and S1), early flowering stage of long and short inflorescences (L2 and S2), and full flowering stage of long and short inflorescences (L3 and S3); the inflorescences were removed from the florets and the remaining inflorescence peduncles were used as experimental material (Figure 2).We detected a total of 127.56 Gb of clean data after removing low-quality reads, with more than 6 Gb of clean reads per library.The percentage of Q30 bases was, on average, 92%.No less than 73% of the clean reads from 18 samples could be mapped to the reference genome; this was indicative of the high quality of the transcriptome dataset (Table S2).Principal component analysis (PCA) showed an obvious separation among the six samples, and the contribution rates of PC1 and PC2 were 24.52% and 15.99%, respectively (Figure 3A), suggesting that significant differences between the six samples and the experiment could be reliable.For simplicity, we hereafter refer to the comparison of L1 vs. S1 as L1_S1, L2 vs. S2 as L2_S2, L3 vs. S3 as L3_S3, L2 vs. L1 as L2_L1, L3 vs. L2 as L3_L2, L3 vs. L1 as L3_L1, S2 vs. S1 as S2_S1, S3 vs. S2 as S3_S2, and S3 vs. S1 as S3_S1.
Among groups, 1666 DEGs in L1_S1, 3993 in L2_S2, and 4232 in L3_S3 were identified at an FDR of < 0.05 and |log 2 Fold Change| ≥ 1 (Figure 3B).Additionally, we selected 11 key genes to perform qRT-PCR to determine the accuracy and reliability of the transcriptome datasets.The relative expression levels of these genes were similar to the RNA-Seq results, indicating that the RNA-Seq data could be trusted in this study (Figure S1).
Among groups, 1666 DEGs in L1_S1, 3993 in L2_S2, and 4232 in L3_S3 were identified at an FDR of < 0.05 and |log2Fold Change| ≥ 1 (Figure 3B).Additionally, we selected 11 key genes to perform qRT-PCR to determine the accuracy and reliability of the transcriptome datasets.The relative expression levels of these genes were similar to the RNA-Seq results, indicating that the RNA-Seq data could be trusted in this study (Figure S1).
Proteomic analysis was used to identify protein differences between the two extreme materials at three developmental stages.Similarly, PCA revealed that the proteomes were significantly separated among the six samples (Figure 3C), indicating that the proteome datasets could be trusted.We obtained and quantified 11570 proteins from six samples.Among these proteins, we detected 638, 1235, and 1298 DEPs in L1_S1, L2_S2, and L3_S3, respectively (Figure 3D).Proteomic analysis was used to identify protein differences between the two extreme materials at three developmental stages.Similarly, PCA revealed that the proteomes were significantly separated among the six samples (Figure 3C), indicating that the proteome datasets could be trusted.We obtained and quantified 11,570 proteins from six samples.Among these proteins, we detected 638, 1235, and 1298 DEPs in L1_S1, L2_S2, and L3_S3, respectively (Figure 3D).

Enrichment Analysis of Integrative Genes and Proteins in Two Extreme Materials at the Three Developmental Stages
We conducted comprehensive transcriptome and proteome analyses to elucidate the underlying mechanisms of inflorescence elongation.GO term analysis indicated that genes and proteins commonly participated in three predominant categories: biological processes; cellular components; and molecular functions.Among them, some essential GO terms were discovered to be correlated with inflorescence growth and development, including phenylpropanoid biosynthetic process, lignin biosynthetic process, plant-type cell wall, lignin catabolic process, plant-type secondary cell wall biogenesis, phloem development, flavonoid biosynthetic process, and other terms (Figure S2).Additionally, KEGG enrichment revealed that a large number of differentially regulated genes and proteins were enriched in the top 20 pathways associated with plant organ elongation and development in L1_S1, L2_S2, and L3_S3, such as phenylpropanoid biosynthesis, flavonoid biosynthesis, starch and sucrose metabolism, nitrogen metabolism, plant hormone signal transduction, and hormone biosynthesis-related pathways (Figure 4A-C), which might be closely related to inflorescence elongation.

Enrichment Analysis of Integrative Genes and Proteins in Two Extreme Materials at the Three Developmental Stages
We conducted comprehensive transcriptome and proteome analyses to elucidate the underlying mechanisms of inflorescence elongation.GO term analysis indicated that genes and proteins commonly participated in three predominant categories: biological processes; cellular components; and molecular functions.Among them, some essential GO terms were discovered to be correlated with inflorescence growth and development, including phenylpropanoid biosynthetic process, lignin biosynthetic process, plant-type cell wall, lignin catabolic process, plant-type secondary cell wall biogenesis, phloem KEGG enrichment revealed that a large number of differentially regulated genes and proteins were enriched in the top 20 pathways associated with plant organ elongation and development in L1_S1, L2_S2, and L3_S3, such as phenylpropanoid biosynthesis, flavonoid biosynthesis, starch and sucrose metabolism, nitrogen metabolism, plant hormone signal transduction, and hormone biosynthesis-related pathways (Figure 4A-C), which might be closely related to inflorescence elongation.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Integrated Analysis of Genes and Proteins Associated with Lignin Biosynthesis
The phenylpropanoid biosynthesis pathway produces lignin as an end product in vascular plants and is closely correlated with the growth of plant stems, roots, inflorescence peduncles, and other prop tissues.From the integrative profiles of the transcriptome and proteome, we found that phenylpropanoid biosynthesis was a crucial pathway related to the structural formation of the inflorescence.Furthermore, 24 DEGs and 24 DEPs, 49 DEGs and 29 DEPs, and 81 DEGs and 53 DEPs were detected in L1_S1, L2_S2, and L3_S3, respectively.To screen key genes and proteins related to the structural formation of long inflorescences, we established a lignin biosynthesis pathway (Figure 5).Furthermore, we detected that many genes and proteins were commonly upregulated in the lignin biosynthesis of key enzymes in the three comparison groups, including POX/LAC in L1_S1; 4CL, COMT, CAD, and POX/LAC in L2_S2; and PAL, 4CL, CCoAOMT, CCR, CAD, F5H, COMT, and POX/LAC in L3_S3 (Table S3), which might be crucial genes and proteins involved in lignin accumulation in long inflorescences.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Integrated Analysis of Genes and Proteins Associated with Lignin Biosynthesis
The phenylpropanoid biosynthesis pathway produces lignin as an end product in vascular plants and is closely correlated with the growth of plant stems, roots, inflorescence peduncles, and other prop tissues.From the integrative profiles of the transcriptome and proteome, we found that phenylpropanoid biosynthesis was a crucial pathway related to the structural formation of the inflorescence.Furthermore, 24 DEGs and 24 DEPs, 49 DEGs and 29 DEPs, and 81 DEGs and 53 DEPs were detected in L1_S1, L2_S2, and L3_S3, respectively.To screen key genes and proteins related to the structural formation of long inflorescences, we established a lignin biosynthesis pathway (Figure 5).Furthermore, we detected that many genes and proteins were commonly upregulated in the lignin biosynthesis of key enzymes in the three comparison groups, including POX/LAC in L1_S1; 4CL, COMT, CAD, and POX/LAC in L2_S2; and PAL, 4CL, CCoAOMT, CCR, CAD, F5H, COMT, and POX/LAC in L3_S3 (Table S3), which might be crucial genes and proteins involved in lignin accumulation in long inflorescences.
Compared with three developmental stages of long and short inflorescence, we identified one (indole-3-acetic acid, Log2 fold-change value = 1.18) and two (indole-3-acetic acid, Log2 fold-change value = Inf; indole-3-acetyl-L-tryptophan, Log2 fold-change value = 2.44) auxins that were significantly up-accumulated in L1_S1 and L2_S2, respectively.Quantitative analysis showed that a high content of indole-3-acetic acid (IAA) was maintained in both L1 (473.66 ng/mL) and L2 (328.06 ng/mL).These results suggest that IAA may play an essential role in inflorescence elongation.Moreover, we found that two GAs (GA9, Log2 fold-change value = 1.11;GA53, Log2 fold-change value = 2.10) and one GA (GA9, Log2 fold-change value = Inf) were prominently up-accumulated in L2_S2 and L3_S3, respectively.Based on the morphological characteristics of inflorescences and the function of GAs, we speculated that GA9 and GA53 might be closely related to inflorescence elongation, expansion of the peduncle diameter, and floret blooming.Additionally, some up-accumulated CKs were detected in L1_S1 (BAPR, Log2 fold-change value = Inf; cZRMP, Log2 fold-change value = 2.01) and L2_S2 (tZ9G, Log2 fold-change value = Inf; iPRMP, Log2 fold-change value = 1.54; iP7G, Log2 fold-change value = Inf); these CKs might play an essential role in inflorescence growth at the full budding and early flowering stages.In addition, differential accumulations of Jas and SA were detected Compared with three developmental stages of long and short inflorescence, we identified one (indole-3-acetic acid, Log 2 fold-change value = 1.18) and two (indole-3-acetic acid, Log 2 fold-change value = Inf; indole-3-acetyl-L-tryptophan, Log 2 fold-change value = 2.44) auxins that were significantly up-accumulated in L1_S1 and L2_S2, respectively.Quantitative analysis showed that a high content of indole-3-acetic acid (IAA) was maintained in both L1 (473.66 ng/mL) and L2 (328.06 ng/mL).These results suggest that IAA may play an essential role in inflorescence elongation.Moreover, we found that two GAs (GA9, Log 2 fold-change value = 1.11;GA53, Log 2 fold-change value = 2.10) and one GA (GA9, Log 2 fold-change value = Inf) were prominently up-accumulated in L2_S2 and L3_S3, respectively.Based on the morphological characteristics of inflorescences and the function of GAs, we speculated that GA9 and GA53 might be closely related to inflorescence elongation, expansion of the peduncle diameter, and floret blooming.Additionally, some up-accumulated CKs were detected in L1_S1 (BAPR, Log 2 fold-change value = Inf; cZRMP, Log 2 fold-change value = 2.01) and L2_S2 (tZ9G, Log 2 fold-change value = Inf; iPRMP, Log 2 fold-change value = 1.54; iP7G, Log 2 fold-change value = Inf); these CKs might play an essential role in inflorescence growth at the full budding and early flowering stages.In addition, differential accumulations of Jas and SA were detected in L1_S1, L2_S2, and L3_S3, which may be involved in the growth and development of inflorescences.
Notably, most phytohormones were significantly down-accumulated in L3_L2, L3_L1, S3_S2, and S3_S1 (Table S4).The results suggested that the metabolic activity of the inflorescence peduncle at the full flowering stage was significantly weaker than that at the full budding and early flowering stages, which was consistent with the developmental regularity in which the inflorescence peduncle stopped elongating and mainly played a role in structural support and nutrient transport.

inflorescences.
Notably, most phytohormones were significantly down-accumulated in L3_L2, L3_L1, S3_S2, and S3_S1 (Table S4).The results suggested that the metabolic activity of the inflorescence peduncle at the full flowering stage was significantly weaker than that at the full budding and early flowering stages, which was consistent with the developmental regularity in which the inflorescence peduncle stopped elongating and mainly played a role in structural support and nutrient transport.

Integrated Analysis of Genes, Proteins, and Metabolites Related to Plant Hormone Biosynthesis
Based on the results above, we conclude that IAA, GA, and CK are closely correlated with inflorescence elongation at the full budding and early flowering stages.To further investigate the underlying mechanism of plant hormone biosynthesis in L1_S1 and L2_S2,

Integrated Analysis of Genes, Proteins, and Metabolites Related to Plant Hormone Biosynthesis
Based on the results above, we conclude that IAA, GA, and CK are closely correlated with inflorescence elongation at the full budding and early flowering stages.To further investigate the underlying mechanism of plant hormone biosynthesis in L1_S1 and L2_S2, we conducted an integrative analysis of the transcriptome, proteome, and metabolome of long and short inflorescences.KEGG enrichment analysis showed that some differentially regulated genes, expressed proteins, and accumulated metabolites were enriched in pathways related to plant hormone biosynthesis in L1_S1 and L2_S2, including tryptophan metabolism (ko00380), diterpenoid biosynthesis (ko00904), and zeatin biosynthesis (ko00908) (Figure 7A,B).In tryptophan metabolism (ko00380), the integrative analysis showed that one amiE (amidase) protein (A0A072TU58, Log 2 fold-change value = 1.32) and two amiE genes

Integrated Analysis of Genes, Proteins, and Metabolites Related to Plant Hormone Signal Transduction
Moreover, we identified many DEGs and DEPs involved in plant hormone signal transduction between long and short inflorescences, which were mainly enriched in IAA, CK, GA, and BR signaling pathways.As shown in Figure 9, we examined the IAA, CK, GA, and BR signaling pathways.In plant hormone signal transduction among three comparisons, indole-3-acetic acid was a unique differential metabolite detected in L1_S1 and L2_S2.Previous studies have revealed that the auxin signaling pathway was closely related to plant growth and organ elongations [19].In general, we discovered that many DEGs involved in plant hormone signal transduction were enriched in the auxin signaling pathway, and most DEGs involved in the auxin signaling pathway were upregulated in long inflorescences rather than in short inflorescences, according to transcriptome In diterpenoid biosynthesis (ko00904), integrative analysis showed that one KAO (ent-kaurenoic acid monooxygenase) protein (A0A072V5K8, Log 2 fold-change value = 0.72) and two KAO genes (MsG0580024673.01,Log 2 fold-change value = 1.22;MsG0580024666.01,Log 2 fold-change value = 1.61) were commonly upregulated in L2 compared to S2 (Figure 8B), suggesting that they participated in the synthesis of GA53 and GA9.In addition, two GA2ox (gibberellin 2beta-dioxygenase) proteins (G7IJL7, Log 2 fold-change value = 0.65; G7K254, Log 2 fold-change value = 1.13) and one GA2ox gene (MsG0580029209.01,Log 2 fold-change value = 1.01) were upregulated in L2_S2 (Figure 8B), which might be closely correlated with the accumulation of other GAs in L2.
In addition, we detected some differentially expressed genes and proteins related to IAA, GA, and CK biosynthesis in the transcriptome and proteome, respectively, such as YUCCA, TAA1, GA20ox, IPT, and other enzymes that might also participate in regulating inflorescence growth.

Integrated Analysis of Genes, Proteins, and Metabolites Related to Plant Hormone Signal Transduction
Moreover, we identified many DEGs and DEPs involved in plant hormone signal transduction between long and short inflorescences, which were mainly enriched in IAA, CK, GA, and BR signaling pathways.As shown in Figure 9, we examined the IAA, CK, GA, and BR signaling pathways.In plant hormone signal transduction among three comparisons, indole-3-acetic acid was a unique differential metabolite detected in L1_S1 and L2_S2.Previous studies have revealed that the auxin signaling pathway was closely related to plant growth and organ elongations [19].In general, we discovered that many DEGs involved in plant hormone signal transduction were enriched in the auxin signaling pathway, and most DEGs involved in the auxin signaling pathway were upregulated in long inflorescences rather than in short inflorescences, according to transcriptome datasets.Therefore, we identified that auxin signaling played a more essential role than other signal transduction pathways in regulating the elongation of inflorescences.By further investigating the pathways of auxin signaling pathway based on the transcriptome, proteome, and metabolome, we identified a total of 18, 23, and 17 upregulated DEGs, 2, 3, and 3 upregulated DEP, and 1, 1, and 0 upregulated DAMs in L1_S1, L2_S2, and L3_S3, respectively.Commonly upregulated genes and proteins were identified in the three comparison groups: AUX1; AUX/IAA; and SAUR (Figure 9).These results suggest that IAA induces the expression of genes involved in the auxin signaling pathway and further induces the expression of downstream genes associated with inflorescence peduncle elongation.
In addition, we found that some DEGs and DEPs were enriched in the CK, GA, and BR signaling pathways between long and short inflorescences (Figure 9).For example, encoding DELLA genes and proteins were commonly downregulated in L2_S2 (Figure 9), which might also affect inflorescence elongation.

Weighted Gene and Protein Co-Expression Network Analysis
To further search for candidate genes highly correlated with inflorescence elongation, we performed weighted gene co-expression network analysis (WGCNA).As a result, a dynamic hierarchical tree cut showed that 19 modules with similar gene expression patterns were detected by WGCNA, which are defined using different colors (Figure 10A).Furthermore, we screened several modules that were highly correlated with each sample from the WGCNA results (R > 0.3) (Figure 10B).Concerning L1, a blue module was selected.For L2, a black module was selected.For L3, purple and green modules were selected.For S1, a turquoise module was selected.For S2, no module was selected.For S3, a brown module was selected.
Based on morphological studies of the alfalfa raceme, we confirmed that the gene expression of the inflorescence peduncle at the full budding and early flowering stages played a more critical role in affecting the growth and development of the inflorescence peduncle than that at the full flowering stage.Eigengene expression analysis of the blue and black modules is shown in Figure 10C, indicating that patterns were closely correlated with the full budding and early flowering stages, respectively.Therefore, genes in the blue and black modules can be speculated to be associated with inflorescence elongation.Among the genes belonging to the blue and black groups, we identified 28 genes that were related to lignin biosynthesis, IAA and GA biosynthesis, and IAA signaling pathway (Table S5), which might be candidate genes involved in inflorescence development.
Similarly, WPCNA (weighted protein co-expression network analysis) results indicated that 11 modules had similar protein expression patterns, and yellow and purple modules (1308 and 76 proteins, respectively) were recognized as crucial modules associated with the elongation of the inflorescence peduncle (Figure S3).From these two modules, we screened a total of six proteins involved in lignin and GA biosynthesis, including POX/LAC (G7IC23, G7JJ71, and G7LDV0), CAD (G7JFC2), KAO (G7IJL7), and GA2ox (G7K254) (Table S6) that might be key proteins associated with inflorescence peduncle elongation.
datasets.Therefore, we identified that auxin signaling played a more essential role than other signal transduction pathways in regulating the elongation of inflorescences.By further investigating the pathways of auxin signaling pathway based on the transcriptome, proteome, and metabolome, we identified a total of 18, 23, and 17 upregulated DEGs, 2, 3, and 3 upregulated DEP, and 1, 1, and 0 upregulated DAMs in L1_S1, L2_S2, and L3_S3, respectively.Commonly upregulated genes and proteins were identified in the three comparison groups: AUX1; AUX/IAA; and SAUR (Figure 9).These results suggest that IAA induces the expression of genes involved in the auxin signaling pathway and further induces the expression of downstream genes associated with inflorescence peduncle elongation.
In addition, we found that some DEGs and DEPs were enriched in the CK, GA, and BR signaling pathways between long and short inflorescences (Figure 9).For example, encoding DELLA genes and proteins were commonly downregulated in L2_S2 (Figure 9), which might also affect inflorescence elongation.

Weighted Gene and Protein Co-Expression Network Analysis
To further search for candidate genes highly correlated with inflorescence elongation, we performed weighted gene co-expression network analysis (WGCNA).As a result, a dynamic hierarchical tree cut showed that 19 modules with similar gene expression patterns were detected by WGCNA, which are defined using different colors indicated that 11 modules had similar protein expression patterns, and yellow and purple modules (1308 and 76 proteins, respectively) were recognized as crucial modules associated with the elongation of the inflorescence peduncle (Figure S3).From these two modules, we screened a total of six proteins involved in lignin and GA biosynthesis, including POX/LAC (G7IC23, G7JJ71, and G7LDV0), CAD (G7JFC2), KAO (G7IJL7), and GA2ox (G7K254) (Table S6) that might be key proteins associated with inflorescence peduncle elongation.11A).Among the TFs families, some TFs might play a crucial role in the growth and elongation of the inflorescence peduncle.Subsequently, we screened the differentially expressed TF genes in the blue and black modules to identify the critical TFs that regulated candidate genes.In L1_S1 and L2_S2, 24 differentially expressed TFs were identified in two modules (FPKM > 2).Furthermore, we performed a correlation analysis between these TFs and 28 candidate genes (|r| > 0.7, p < 0.05).Nineteen TFs were positively correlated with 16 structural genes ((Figure 11B and Table 1).In the lignin biosynthesis pathway, we identified 13 TFs that were positively related to five structural genes, including MYB (four), bHLH (two), and NAC (one).In the plant hormone biosynthesis and signal transduction, we identified 17 TFs that were positively associated with 11 candidate genes, including MYB (seven), bHLH (four), NAC (two), GRAS (one), and HB (one).These TFs may play an important role in regulating structural and regulatory genes and eventually prompt the elongation of the inflorescence peduncle.11A).Among the TFs families, some TFs might play a crucial role in the growth and elongation of the inflorescence peduncle.Subsequently, we screened the differentially expressed TF genes in the blue and black modules to identify the critical TFs that regulated candidate genes.In L1_S1 and L2_S2, 24 differentially expressed TFs were identified in two modules (FPKM > 2).Furthermore, we performed a correlation analysis between these TFs and 28 candidate genes (|r| > 0.7, p < 0.05).Nineteen TFs were positively correlated with 16 structural genes ((Figure 11B and Table 1).In the lignin biosynthesis pathway, we identified 13 TFs that were positively related to five structural genes, including MYB (four), bHLH (two), and NAC (one).In the plant hormone biosynthesis and signal transduction, we identified 17 TFs that were positively associated with 11 candidate genes, including MYB (seven), bHLH (four), NAC (two), GRAS (one), and HB (one).These TFs may play an important role in regulating structural and regulatory genes and eventually prompt the elongation of the inflorescence peduncle.

Discussion
The architecture of plant inflorescence determines flower and seed growth and development, and length is an essential character influencing the morphological structure of alfalfa inflorescences, which is also closely correlated with seed production of alfalfa [8].To clarify the potential mechanism of alfalfa inflorescence elongation, phenylpropanoid biosynthesis was selected as a critical pathway between the two extreme materials by analyzing the integrative transcriptomic and proteomic results.Combined with the targeted phytohormone metabolome, we identified key genes, proteins, and metabolites associated with plant hormone biosynthesis and signal transduction based on omics analysis.
The phenylpropanoid pathway is important for lignin synthesis and is closely associated with the elongation of plant organs [32].In this study, we found that many differentially expressed genes and proteins were enriched in the phenylpropanoid pathway between long and short inflorescences, which suggested that lignin biosynthesis played an essential role in inflorescence structure formation.PAL, C4H, and 4CL are upstream enzymes in the phenylpropanoid biosynthesis pathway, and their expression levels in plants significantly affect lignin accumulation [33,34].HCT, F5H, COMT, and CCoAOMT affect the synthesis of lignin monomers, and the inhibition of these enzymes significantly reduces the accumulation of lignin [35,36].Downstream lignin biosynthesis determines the synthesis of different lignins, which are associated with four enzymes, including CCR, CAD, and POD/LAC [37][38][39].In the present study, we screened many commonly upregulated genes and proteins related to enzymes involved in lignin biosynthesis in L1_S1, L2_S2, and L3_S3, including PAL, 4CL, CCR, CAD, CCoAOMT, F5H, COMT, and POD/LAC.High expression levels of these genes or proteins may contribute to the synthesis of related enzymes, thereby increasing lignin accumulation in long inflorescences.
Phant hormones are pivotal for the growth of the inflorescence peduncle.The auxin indole-3-acetic acid (IAA) is a crucial promoter of plant tissue growth.Previous studies

Discussion
The architecture of plant inflorescence determines flower and seed growth and development, and length is an essential character influencing the morphological structure of alfalfa inflorescences, which is also closely correlated with seed production of alfalfa [8].To clarify the potential mechanism of alfalfa inflorescence elongation, phenylpropanoid biosynthesis was selected as a critical pathway between the two extreme materials by analyzing the integrative transcriptomic and proteomic results.Combined with the targeted phytohormone metabolome, we identified key genes, proteins, and metabolites associated with plant hormone biosynthesis and signal transduction based on omics analysis.
The phenylpropanoid pathway is important for lignin synthesis and is closely associated with the elongation of plant organs [32].In this study, we found that many differentially expressed genes and proteins were enriched in the phenylpropanoid pathway between long and short inflorescences, which suggested that lignin biosynthesis played an essential role in inflorescence structure formation.PAL, C4H, and 4CL are upstream enzymes in the phenylpropanoid biosynthesis pathway, and their expression levels in plants significantly affect lignin accumulation [33,34].HCT, F5H, COMT, and CCoAOMT affect the synthesis of lignin monomers, and the inhibition of these enzymes significantly reduces the accumulation of lignin [35,36].Downstream lignin biosynthesis determines the synthesis of different lignins, which are associated with four enzymes, including CCR, CAD, and POD/LAC [37 -39].In the present study, we screened many commonly upregulated genes and proteins related to enzymes involved in lignin biosynthesis in L1_S1, L2_S2, and L3_S3, including PAL, 4CL, CCR, CAD, CCoAOMT, F5H, COMT, and POD/LAC.High expression levels of these genes or proteins may contribute to the synthesis of related enzymes, thereby increasing lignin accumulation in long inflorescences.
Phant hormones are pivotal for the growth of the inflorescence peduncle.The auxin indole-3-acetic acid (IAA) is a crucial promoter of plant tissue growth.Previous studies have shown that IAA played a crucial role in promoting stem and root elongation.Consistent with our findings, we found that the IAA content of the full budding and flowering stages was significantly upregulated in long inflorescences compared to short inflorescences, suggesting that IAA might be a key phytohormone contributing to the elongation of alfalfa inflorescences.Pollmann discovered that amidase plays a role in catalyzing the conversion of IAM to IAA in Arabidopsis thaliana [40].In this study, we identified that one gene and one protein, amid, were commonly upregulated in L1_S1, suggesting that amidases might play an important role in IAA accumulation in long inflorescences.Subsequently, IAA acts on related receptors and prompts the expression of auxin response genes, eventually resulting in its biological effects.AUX1 plays a critical role in auxin transport.A previous report demonstrated that AtAUX1 mutations led to tAUX1 mutations, attenuating auxin trafficking in Arabidopsis seedlings and altering IAA distribution in young leaf and root tissues [41].Aux/IAA, an important functional gene, participates in plant growth and development by regulating the downstream genes of the auxin signaling pathway [42].SAUR gene family can respond early in auxin induction; overexpression of AtSAUR63 elongated the hypocotyls and stamen filament of the transgenic plants in Arabidopsis [43].In this study, we discovered that most DEGs and DEPs maintained higher expression levels in long inflorescences than in short inflorescences.In addition, we further identified that many genes and proteins encoding AUX1, Aux/IAA, and SAUR were commonly up-or down-regulated in L1_S1, L2_S2, and L3_S3, revealing that AUX1, Aux/IAA, and SAUR might play an essential role in regulating the elongation of long inflorescences.
Gibberellins facilitate cell wall extension, stem development, and tissue growth [44].The targeted phytohormone metabolome showed that the concentrations of GA9 and GA53 were significantly higher in L2 than in S2, suggesting that GA9 and GA53 might be involved in the elongation and growth of inflorescences at the early flowering stage.KAO is an upstream enzyme of the GA biosynthesis pathway.In the present study, we identified that two KAO genes and one KAO protein maintained higher expression in L2 than in S2, which might be involved in the accumulation of GA and contribute to the growth and elongation of inflorescences at the early flowering stage.Most studies have demonstrated that DELLA proteins inhibited plant growth by binding to transcription factors related to the regulation of plant growth [45][46][47].In this study, we discovered that DELLA proteins were downregulated in L1_S1 and L2_S2, and DELLA genes maintained low expression levels in L2 compared to S2.The results revealed that DELLA proteins might play a negative role in the elongation and development of alfalfa inflorescences.
Additionally, some pathways were connected to plant growth and development based on an integrated analysis of the transcriptome and proteome, such as starch and sucrose metabolism, nitrogen metabolism, and flavonoid biosynthesis [48][49][50][51].These results show that these pathways might also be involved in the structural formation of inflorescences.Previous studies revealed that MYB, NAC, and bHLH TFs participated in the regulation of lignin biosynthesis [52,53].In this study, we identified that four MYB, two bHLH, and one NAC TFs were positively correlated with candidate genes associated with lignin biosynthesis based on a correlation analysis, revealing that these TFs might play an essential role in regulating structural genes related to lignin accumulation.

Plant Materials
Alfalfa (Medicago sativa) samples with long and short inflorescences were cultivated at the Grassland Institute of the Chinese Academy of Agricultural Sciences in Hohhot (40 • 58 ′ N, 111 • 78 ′ E).All samples were collected when the alfalfa reached the blooming stage, as we could obtain all samples during this period.We selected three alfalfas with long inflorescence and nine alfalfas with short inflorescence as candidate materials; inflorescences were collected based on the developmental stages and inflorescence lengths (1.5-2.0 cm for the full budding stage of long inflorescence 'L1', 3.5-4.0cm for the early flowering stage of long inflorescence 'L2', 6.0-6.5 cm for the full flowering stage of long inflorescence 'L3', 0.8 to 1.2 cm for the full budding stage of short inflorescence 'S1', 1.2-1.5 cm for the early flowering stage of short inflorescence 'S2', and 1.5-2 cm for the full flowering stage of short inflorescence 'S3').We removed all flowers and retained the inflorescence peduncle from the first floret, from the base to the top.Three biological replicates were obtained for the six samples, and the inflorescence peduncle for each replicate was heavier than 3.0 g.

Transcriptome Sequencing and Data Analysis
The total RNA of the inflorescence peduncle (L1, L2, L3, S1, S2, and S3) was extracted by ethanol precipitation and CTAB-PBIOZOL.RNA purity and integrity were analyzed using an Agilent 2100 Bioanalyzer and a Qubit 2.0 Fluorometer.Eighteen cDNA libraries were sequenced using an Illumina Sequencing 6000 platform.After the low-quality sequences were removed, clean reads were assembled using fastp software (fastp v0.19.4).All non-redundant transcripts were mapped using the Medicago sativa reference genome (https://figshare.com/articles/dataset/Medicago_sativa_genome_and_annotation_files/12623960 (accessed on 10 August 2023)).DESeq2 software (DESeq2 v3.19) was used to determine the differential expression profiles among the samples.Subsequently, we obtained notably differential genes with a false discovery rate (FDR < 0.01) and fold change (FC ≥ 2).Enrichment analysis was conducted based on a hypergeometric test, with pathway-based hypergeometric distribution checking for Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) term-based profiles.Finally, a weighted gene co-expression network analysis (WGCNA) was performed using the varFilter function in the R genefilter package.The correlation network diagram was conducted by using R version 3.5.1.

Proteomic Analysis
Proteins were extracted from the samples using acetone precipitation.Protein samples extracted from the inflorescence peduncle were incubated in L3 buffer (0.15 M, pH 8.0) containing 1% SDS, 100 mMTris-HCl, 7 M urea, 2 M thiourea, 1 mM PMSF, and 2 mM EDTA and ultrasonically cracked on ice for 10 min.After the protein solution was obtained by centrifuging the supernatant, we added 4× volume of frozen acetone into the protein solution, precipitated at −20 • C overnight, and centrifuged at 4 • C to obtain the precipitate.After obtaining the precipitate, we washed it with cold acetone, dissolved it in 8 M urea, and measured protein concentration.Equal amounts of protein from each sample were subjected to tryptic digestion.Then, we used tryptic to digest equal content of proteins from each sample, added 8 M urea to 200 µL to the supernatants, reduced with 10 mM DTT for 45 min at 37 • C, and alkylated with 50 mM iodoacetamide (IAM) for 15 min in a dark room at 20 • C. The protein precipitate was air-dried and resuspended in 200 µL of 25 mM ammonium bicarbonate solution and 3 µL of trypsin (Promega) and digested overnight at 37 • C after adding 4× volume of chilled acetone, precipitated at −20 • C for 2 h and centrifuged.Subsequently, we desalted for peptides, dried and concentrated, using a C18 cartridge, vacuum concentration meter, and vacuum centrifugation, respectively, and eventually re-dissolved in 0.1% (v/v) formic acid.

Hormone Analysis
The LC-MS/MS analysis was conducted using a Q Exactive HF-X mass spectrometer combined with an Easy-nLC 1000 system (Thermo Fisher Scientific, Waltham, Massachusetts, USA).LC-MS/MS was performed as previously described by Zhu et al. (2022) [5].The protein sequences obtained above were blasted in the specified tax ID of the nr database using the BLASTP algorithm, with the principle that proteins with the same or similar amino acid sequences share similar functions.Annotations from the mapped protein hits, mainly GO terms (including Biological Process, Cellular Component, and Molecular Function) and KEGG pathway information, were transferred to the original submitted proteins.

qRT-PCR Analysis
The RNA-Seq results were confirmed by using quantitative real-time PCR.The primers were designed using Primer3 (https://primer3.ut.ee/ (accessed on 8 January 2024)).Primers used are listed in Supplementary Table S1.The actin gene (MsG0380015289.01)was selected as the reference gene in this study because of its high and steady expression levels in all samples based on transcriptome data.All results were obtained from three repetitions.

Conclusions
In this study, the regulatory mechanisms of inflorescence elongation in alfalfa were investigated using transcriptome, proteome, and targeted phytohormone metabolome analyses.Specifically, we analyzed the developmental processes of long and short inflorescences in alfalfa and used the three developmental stages of two inflorescences as experimental materials.Compared to short inflorescences, we found that IAA, GA, and CK played crucial roles in regulating peduncle elongation according to the metabolome results.Additionally, based on omics analyses, we detected candidate genes and proteins correlated with lignin biosynthesis, GA biosynthesis, auxin biosynthesis, and signaling pathways.Moreover, TFs related to lignin biosynthesis, GA biosynthesis, auxin biosynthesis, and signaling pathways were identified using correlation analysis.The results of this study highlight the developmental processes and the potential mechanisms underlying inflorescence elongation in alfalfa and provide a theoretical foundation for germplasm innovation.

Figure 1 .
Figure 1.(A) Phenotypes of long and short inflorescence at the six developmental stages (a-f).(B) Values of length of long and short inflorescences at the six developmental stages (from the first floret at the base to the top).

Figure 1 .
Figure 1.(A) Phenotypes of long and short inflorescence at the six developmental stages (a-f).(B) Values of length of long and short inflorescences at the six developmental stages (from the first floret at the base to the top).

Figure 4 .
Figure 4. KEGG enrichment analysis of (A) L1_S1, (B) L2_S2, and (C) L3_S3 based on the integrative profiles of transcriptome and proteome.Color changes (red to green) represent low to high p-values.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Figure 4 .
Figure 4. KEGG enrichment analysis of (A) L1_S1, (B) L2_S2, and (C) L3_S3 based on the integrative profiles of transcriptome and proteome.Color changes (red to green) represent low to high p-values.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Figure 7 .
Figure 7. KEGG enrichment analysis of (A) L1_S1 and (B) L2_S2 based on integrative profiles of transcriptome, proteome, and metabolome.Color changes (red to green) represent low to high pvalues.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Figure 7 .
Figure 7. KEGG enrichment analysis of (A) L1_S1 and (B) L2_S2 based on integrative profiles of transcriptome, proteome, and metabolome.Color changes (red to green) represent low to high p-values.The area size of the circle and triangle represents the number of genes and proteins enriched in a pathway, respectively; the larger the area, the more there are.

Figure 8 .
Figure 8. Key enzymes and metabolites participated in (A) auxin and (B) GA biosynthesis pathways; red words represented crucial enzymes or metabolites.The red font in the box represents the high accumulation of metabolites, and the red font next to the arrow represents the commonly high expression of genes and proteins in the long inflorescences compared to the short inflorescences.

Figure 8 .
Figure 8. Key enzymes and metabolites participated in (A) auxin and (B) GA biosynthesis pathways; red words represented crucial enzymes or metabolites.The red font in the box represents the high accumulation of metabolites, and the red font next to the arrow represents the commonly high expression of genes and proteins in the long inflorescences compared to the short inflorescences.

Figure 9 .
Figure 9. IAA, CK, GA, and BR signal transduction diagram.The six rows of small squares represent the three comparison groups, including L1_S1, L2_S2, and L3_S3.The genes and proteins are located on upper and lower sides, respectively.Red, green, blue, and white boxes represent up-regulation, down-regulation, complex, and insignificant, respectively.

Figure 9 .
Figure 9. IAA, CK, GA, and BR signal transduction diagram.The six rows of small squares represent the three comparison groups, including L1_S1, L2_S2, and L3_S3.The genes and proteins are located on upper and lower sides, respectively.Red, green, blue, and white boxes represent up-regulation, down-regulation, complex, and insignificant, respectively.

Figure 10 .
Figure 10.(A) Dendrogram displaying module eigengenes detected by WGCNA and clustering dendrogram of expressed genes.(B) The heatmap of correlation coefficient between samples and modules with positive and negative correlations is shown in red and blue, respectively.(C) Eigengene expression analysis of blue and black module.

Figure 10 .
Figure 10.(A) Dendrogram displaying module eigengenes detected by WGCNA and clustering dendrogram of expressed genes.(B) The heatmap of correlation coefficient between samples and modules with positive and negative correlations is shown in red and blue, respectively.(C) Eigengene expression analysis of blue and black module.

Figure 11 .
Figure 11.(A) TF family classification among genes.(B) Correlation analysis between candidate genes and TFs; purple and green circles represent TFs and candidate genes, respectively.

Figure 11 .
Figure 11.(A) TF family classification among genes.(B) Correlation analysis between candidate genes and TFs; purple and green circles represent TFs and candidate genes, respectively.

Table 1 .
The identified transcription factors in the blue and black modules.

Table 1 .
The identified transcription factors in the blue and black modules.