Activation of Gossypium hirsutum ACS6 Facilitates Fiber Development by Improving Sucrose Metabolism and Transport

Cotton fiber yield depends on the density of fiber cell initials that form on the ovule epidermis. Fiber initiation is triggered by MYB-MIXTA-like transcription factors (GhMMLs) and requires a sucrose supply. Ethylene or its precursor ACC (1-aminocyclopropane-1-carboxylic acid) is suggested to affect fiber yield. The Gossypium hirsutum (L.) genome contains 35 ACS genes (GhACS) encoding ACC synthases. Here, we explored the role of a GhACS family member in the regulation of fiber initiation. Expression analyses showed that the GhACS6.3 gene pair was specifically expressed in the ovules during fiber initiation (3 days before anthesis to 5 days post anthesis, −3 to 5 DPA), especially at −3 DPA, whereas other GhACS genes were expressed at very low or undetectable levels. The expression profile of GhACS6.3 during fiber initial development was confirmed by qRT-PCR analysis. Transgenic lines overexpressing GhACS6.3 (GhACS6.3-OE) showed increased ACC accumulation in ovules, which promoted the formation of fiber initials and fiber yield components. This was accompanied by increased transcript levels of GhMML3 and increased transcript levels of genes encoding sucrose transporters and sucrose synthase. These findings imply that GhACS6.3 activation is required for fiber initial development. Our results lay the foundation for further research on increasing cotton fiber production.


Introduction
Cotton is the most important textile fiber crop in the world.One of the determining factors for cotton fiber yield is the frequency of fiber initials that form on the ovule epidermis.However, there is still much to learn about the mechanism of cotton fiber initiation and development.Each cotton fiber is a unicellular seed trichome differentiated from the ovule epidermis [1,2].There are four distinct and overlapping stages of fiber differentiation and development: differentiation and initiation (3 days before anthesis to 5 days post anthesis, −3 to 5 DPA), rapid elongation or primary cell wall synthesis (5 to 16 DPA), transition from primary wall synthesis to secondary wall synthesis (16 to 20 DPA), secondary cell wall biosynthesis (20 to 40 DPA), and maturation (40 to 50 DPA) [3].Generally, fiber initial development determines fiber yield by controlling the number of fibers on each ovule.The mechanism underlying this process involves a subclass of MYB transcription factors, among which MYB-MIXTA-like transcription factor 3 (MML3)/GhMML3 regulates fiber initiation [2,[4][5][6][7].A GhMML3-deficient mutant (N1) shows a fuzz fiber-less phenotype [8].Interestingly, GhMML3 also functions as an upstream regulator of other GhMML genes, such as GhMML4 and GhMML7 [8].
It is well known that cotton ovules use sugars, especially sucrose, to fuel fiber development; that is, the ovule is a sink organ that consumes sucrose.Sucrose produced by the leaves (the source, and the main photosynthetic organ) is transported into ovules for fiber initiation.Studies have shown that the direct product of photosynthesis, glucose, is converted into sucrose in source tissues [9].It is then loaded into the phloem and transported to sink tissues by two transporters: SWEETs (Sugar Will Eventually be Exported Transporters) and SUTs (Sucrose Transporters) [10,11].In sink tissues, SuSy (sucrose synthase) converts sucrose into glucose (which then binds uridine diphosphate to form UDPG) and fructose, which provide the building blocks and energy needed for fiber development.Therefore, SuSy expression and SuSy activity are directly related to hexose levels in ovules and the fiber initiation rate [12][13][14][15].
Ethylene and its precursor 1-aminocyclopropane-1-carboxylic acid (ACC) have been shown to promote cotton fiber development [15][16][17].For example, ethylene treatment induces SuSy expression, thus promoting fiber initiation [1].However, the mechanism by which ethylene or ACC regulates SuSy-dependent fiber development is largely unclear.In plants, ethylene is synthesized by two enzymes: ACS (ACC synthase), which converts S-adenosylmethionine (SAM) to ACC [18][19][20], and ACO (ACC oxidase), which oxidizes ACC to generate ethylene [21,22].There is an increasing body of evidence showing that the activity of cotton ACS or ACO is necessary for fiber development [15].For example, G. hirsutum ACS6 (GhACS6) is highly expressed in developing ovules [1], and GhACS6.3 is expressed during fiber initial development [23].In contrast, a lack of GhACS activity abolishes fiber initial development, as demonstrated by the fact that application of aminoethoxyvinylglycine (AVG), a specific inhibitor of ACS, results in a fiber-free phenotype in cultured ovules [1].The results of those studies suggest that the activity of cotton ACS family members plays a very important role in fiber initial development.Plant ACS proteins are encoded by multiple genes [24].For example, the G. hirsutum genome contains 35 ACS genes [23].Studies have found that the activity of plant ACS members can be unique, overlapping, and specific [23][24][25][26].However, it is still unknown which GhACS members are involved in the different stages of fiber initial development, and what their mechanism of action is.
In this study, we analyzed the spatiotemporal expression patterns of ACS genes and found that GhACS6.3genes are specifically expressed in ovules during fiber initiation (−3 to 5 DPA).Further experiments showed that GhACS6.3overexpression increased the ACC content in the ovules, triggered the expression of genes encoding a sucrose metabolism enzyme (GhSuSy) and sugar transporters (GhSWEET and GhSUT), and increased the sucrose and glucose (or UDPG) contents in ovules, which increased the fiber yield components under open field conditions.

Expression Profiles of GhACS Genes in Ovules and Leaves
To explore the roles of GhACS genes during the different stages of fiber initial development, the expression patterns of GhACS family members were determined.There are 35 GhACS genes in the genome of G. hirsutum [23], and transcriptome data are available at the CottonFGD (Cotton Function Genomics Database).Data on the expression of 35 GhACS genes in ovules from −3 to 30 DPA and in leaves were downloaded from the CottonFGD, and then the expression profiles of the GhACS gene family in various tissues, including the ovule, leaf, root, stem, torus, stamen, pistil, and calycle, were determined and displayed as a heat map.
As shown in the heat map, two GhACS6.3genes (Gh_AACS6.3 in the A genome and Gh_DACS6.3 in the D genome) were predominantly expressed in the flower (including ovules, stamen, pistil, petal, and calycle) and root (Figure 1A and Supplementary Figure S1).In ovules, the relative expression levels of this gene pair showed two peaks at −3 DPA and 10 DPA (Figure 1A).In addition to the GhACS6.3genes, two GhACS10.1 genes (Gh_AACS10.1 in the A genome and Gh_DACS10.1 in the D genome) were expressed in ovules from −3 to 10 DPA, but their relative expression levels were lower than those of GhACS6.3.In leaves, GhACS10.1 genes were expressed at higher levels than were other GhACS genes.
pressed at significantly higher levels than were GhACS10.1 genes in ovules from −3 to 0 DPA (Figure 1B).The expression levels of GhACS6.3 in ovules were 21.06-fold and 38.23fold higher than that in leaves at −3 DPA and 10 DPA, respectively (Figure 1B,C).GhACS6.3 and GhACS10.1 were continuously expressed in the leaves, but the expression level of GhACS10.1 was higher than that of GhACS6.3 (Figure 1C).The fact that GhACS6.3genes was preferentially expressed in ovules implied that GhACS6 plays an important role for the formation of fiber yield components.

Effects of GhACS6.3 Overexpression on ACC Production in Ovules and Leaves
To explore the relationship between GhACS6.3 and fiber yield components, we examined the fiber yield components in GhACS6.3-overexpressing(OE) lines, namely GhACS6.3-OE(#1),GhACS6.3-OE(#4), and GhACS6.3-OE(#5).First, qRT-PCR analyses confirmed the significantly increased mRNA levels of GhACS6.3 in these lines and showed that GhACS6.3 was expressed to different degrees in the subtending leaves and the ovules but always at higher levels than that in the control.Among all the OE lines, GhACS6.3-OE(#4)had the highest expression levels of GhACS6.3 (Figure 2A).
Next, we determined the ACC content in the control and the GhACS6.3-OElines.Compared with the control, the GhACS6.3-OElines showed significantly increased ACC production to different degrees in the subtending leaves and the ovules.For example, in GhACS6.3-OE(#4) at −3 DPA, the ACC contents in the subtending leaves (117.24 ± 2.95 ng/g) and the ovules (380.97 ± 10.41 ng/g) were significantly higher than those in the control (32.68 ± 1.54 and 211.02 ± 5.28 ng/g, respectively) (Figure 2B).At 0 DPA, the ACC content was higher in the subtending leaves (102.34 ± 9.56 ng/g) and the ovules (251.22 ± 7.50 ng/g) of GhACS6.than in the control (30.68 ± 1.70 and 104.37 ± 5.69 ng/g, respectively).The transcript level of GhACS6.3 was directly proportional to the ACC content, indicating that GhACS6.3increased ACC production in the ovules during fiber initiation.Next, qRT-PCR analyses were conducted to check the expression levels of GhACS6.3 and GhACS10.1 in ovules from −3 to 20 DPA.The expression level of GhACS6.3 reached the maximum at −3 DPA (relative value 0.26), then decreased to a relative value of 0.0036 at 3 DPA (fiber initiation stage), then increased to reach another maximum (relative value 0.19) at 10 DPA (fiber elongation stage), and then rapidly decreased to the minimum (relative value 0) after 10 DPA (Figure 1B).Among the 35 GhACS genes, GhACS6.3 was expressed at significantly higher levels than were GhACS10.1 genes in ovules from −3 to 0 DPA (Figure 1B).The expression levels of GhACS6.3 in ovules were 21.06-fold and 38.23-fold higher than that in leaves at −3 DPA and 10 DPA, respectively (Figure 1B,C).GhACS6.3 and GhACS10.1 were continuously expressed in the leaves, but the expression level of GhACS10.1 was higher than that of GhACS6.3 (Figure 1C).The fact that GhACS6.3genes was preferentially expressed in ovules implied that GhACS6 plays an important role for the formation of fiber yield components.

Effects of GhACS6.3 Overexpression on ACC Production in Ovules and Leaves
To explore the relationship between GhACS6.3 and fiber yield components, we examined the fiber yield components in GhACS6.3-overexpressing(OE) lines, namely GhACS6.3-OE(#1),GhACS6.3-OE(#4), and GhACS6.3-OE(#5).First, qRT-PCR analyses confirmed the significantly increased mRNA levels of GhACS6.3 in these lines and showed that GhACS6.3 was expressed to different degrees in the subtending leaves and the ovules but always at higher levels than that in the control.Among all the OE lines, GhACS6.3-OE(#4)had the highest expression levels of GhACS6.3 (Figure 2A).
Next, we determined the ACC content in the control and the GhACS6.3-OElines.Compared with the control, the GhACS6.3-OElines showed significantly increased ACC production to different degrees in the subtending leaves and the ovules.For example, in GhACS6.3-OE(#4) at −3 DPA, the ACC contents in the subtending leaves (117.24 ± 2.95 ng/g) and the ovules (380.97 ± 10.41 ng/g) were significantly higher than those in the control (32.68 ± 1.54 and 211.02 ± 5.28 ng/g, respectively) (Figure 2B).At 0 DPA, the ACC content was higher in the subtending leaves (102.34 ± 9.56 ng/g) and the ovules (251.22 ± 7.50 ng/g) of GhACS6.3-OE(#4)than in the control (30.68 ± 1.70 and 104.37 ± 5.69 ng/g, respectively).The transcript level of GhACS6.3 was directly proportional to the ACC content, indicating that GhACS6.3increased ACC production in the ovules during fiber initiation.The ACC content was also increased in the leaves of GhACS6.3-OE(#4) at −3 and 10 DPA, but the increase was lower than that in the ovules (Figure 2B), implying that GhACS6-dependent ACC production in the ovules may be involved in fiber initial development.The ACC content was also increased in the leaves of GhACS6.3-OE(#4) at −3 and 10 DPA, but the increase was lower than that in the ovules (Figure 2B), implying that GhACS6dependent ACC production in the ovules may be involved in fiber initial development.

Effects of GhACS6.3-Dependent ACC Production on Fiber Initiation and Yield Components
To explore the relationship between ACC production and fiber yield components, we analyzed the cotton fiber yield components of field-grown plants from 2017.The data from 2021 and 2022 showed that the fiber yield components were clearly increased in the GhACS6.3-OElines compared with the control.For example, in GhACS6.3-OE(#4), the number of bolls per plant (43.8 ± 2.6) and the single boll weight (5.6 ± 0.3 g) of GhACS6.3-OE(#4) were 18.06% and 24.44% higher, respectively, that their respective values in the control (37.1 ± 2.3 and 4.5 ± 0.3 g, respectively), suggesting that overexpression of GhACS6.3 gene was beneficial for the development of cotton bolls.
In addition to the above fiber yield components, the LI (lint index), SI (seed index), LP (lint percentage), and fruit branches of GhACS6.3-OE(#4) were higher than those of the control (Table 1).

Effects of GhACS6.3-Dependent ACC Production on Fiber Initiation and Yield Components
To explore the relationship between ACC production and fiber yield components, we analyzed the cotton fiber yield components of field-grown plants from 2017.The data from 2021 and 2022 showed that the fiber yield components were clearly increased in the GhACS6.3-OElines compared with the control.For example, in GhACS6.3-OE(#4), the number of bolls per plant (43.8 ± 2.6) and the single boll weight (5.6 ± 0.3 g) of GhACS6.3-OE(#4) were 18.06% and 24.44% higher, respectively, that their respective values in the control (37.1 ± 2.3 and 4.5 ± 0.3 g, respectively), suggesting that overexpression of GhACS6.3 gene was beneficial for the development of cotton bolls.
In addition to the above fiber yield components, the LI (lint index), SI (seed index), LP (lint percentage), and fruit branches of GhACS6.3-OE(#4) were higher than those of the control (Table 1).Then the initial density of fibers on the ovule epidermis was investigated.The fiber cell density (6649 ± 78/mm 2 ) was higher in GhACS6.3-OE(#4)than in the control (6117 ± 65 /mm 2 ) at 0 DPA (Figure 3A,B).The GhACS6.3-dependentACC production (Figure 2) was positively correlated with the formation of fiber yield components (Table 1).Then the initial density of fibers on the ovule epidermis was investigated.The fiber cell density (6649 ± 78/mm 2 ) was higher in GhACS6.3-OE(#4)than in the control (6117 ± 65 /mm 2 ) at 0 DPA (Figure 3A,B).The GhACS6.3-dependentACC production (Figure 2) was positively correlated with the formation of fiber yield components (Table 1).To prove that ACC facilitates fiber initial development, the effects of ACC treatment on fiber cell initiation from the cultured ovules were tested by adding ACC at different concentrations (0, 0.5, 1.0, 1.5, 2.0, 5.0 µM) into the culture medium.After 48 h, the density of fiber cells on the ovule epidermis was significantly higher in the 0.5, 1.0, and 1.5 µM ACC treatments than in the untreated control (0 µM ACC) (Figure 4).However, high concentrations of ACC (2.0 and 5.0 µM) reduced the density of fiber cells (Figure 4).The size of the cultured ovules did not differ significantly between the ACC treatments and the control (Supplementary Figure S2).These data further demonstrate that GhACS6.3-dependentACC production may be a positive regulator of fiber initial development.To prove that ACC facilitates fiber initial development, the effects of ACC treatment on fiber cell initiation from the cultured ovules were tested by adding ACC at different concentrations (0, 0.5, 1.0, 1.5, 2.0, 5.0 µM) into the culture medium.After 48 h, the density of fiber cells on the ovule epidermis was significantly higher in the 0.5, 1.0, and 1.5 µM ACC treatments than in the untreated control (0 µM ACC) (Figure 4).However, high concentrations of ACC (2.0 and 5.0 µM) reduced the density of fiber cells (Figure 4).The size of the cultured ovules did not differ significantly between the ACC treatments and the control (Supplementary Figure S2).These data further demonstrate that GhACS6.3dependentACC production may be a positive regulator of fiber initial development.
Next, the transcript profiles of genes involved in the initiation and development of cotton fibers were determined.The mRNA levels of GhMML3 in ovules were significantly higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA.For example, at −1 DPA, the transcript level of GhMML3 was 2-fold higher in GhACS6.3-OE(#4)than in the control (Figure 5A).Accordingly, the transcript levels of GhMML4 and GhMML7 in ovules from −1 to 2 DPA were higher in GhACS6.3-OE(#4)than in the control (Figure 5B).These findings indicate that ACC produced by GhACS6.3 promoted the expression of GhMML3, GhMML4, and GhMML7, consistent with the increased density of fiber cells (Figure 3) and formation of fiber yield components (Table 1).concentrations (0, 0.5, 1.0, 1.5, 2.0, 5.0 µM) into the culture medium.After 48 h, the density of fiber cells on the ovule epidermis was significantly higher in the 0.5, 1.0, and 1.5 µM ACC treatments than in the untreated control (0 µM ACC) (Figure 4).However, high concentrations of ACC (2.0 and 5.0 µM) reduced the density of fiber cells (Figure 4).The size of the cultured ovules did not differ significantly between the ACC treatments and the control (Supplementary Figure S2).These data further demonstrate that GhACS6.3-dependentACC production may be a positive regulator of fiber initial development.Next, the transcript profiles of genes involved in the initiation and development of cotton fibers were determined.The mRNA levels of GhMML3 in ovules were significantly higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA.For example, at −1 DPA, the transcript level of GhMML3 was 2-fold higher in GhACS6.3-OE(#4)than in the control (Figure 5A).Accordingly, the transcript levels of GhMML4 and GhMML7 in ovules from −1 to 2 DPA were higher in GhACS6.3-OE(#4)than in the control (Figure 5B).These findings indicate that ACC produced by GhACS6.3 promoted the expression of GhMML3, GhMML4, and GhMML7, consistent with the increased density of fiber cells (Figure 3) and formation of fiber yield components (Table 1).

Sucrose Content and Transcript Levels of Sucrose Transporter Genes in Leaves and Ovules
Adequate sucrose is necessary for fiber development.The sucrose content in ovules was higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA (Figure 6).For example, at 0 DPA, the sucrose content in GhACS6.3-OE(#4) was 18.47 ± 0.25 mg/g, which was 2.15-fold that in the control (8.60 ± 0.17 mg/g).Because sucrose is transported from the leaves (source) to the ovules (sink), the effects of GhACS6.3 overexpression on the transcript levels of genes encoding the sucrose transporters GhSUT and GhSWEET were investigated.Based on RNA-seq data [27], the transcript profiles of 18 GhSUT [11] and 21 GhSWEET [28] genes in ovules and leaves were determined.As shown in the heat maps, GhSUT2 was specifically expressed in ovules, GhSUT1 was predominantly expressed in leaves (Supplementary Figure S3A), GhSWEET2 specifically expressed in leaves, and GhSWEET5 was predominantly expressed in ovules (Supplementary Figure S3B).
The results of qRT-PCR analyses showed that, from −1 to 2 DPA, the transcript levels of GhSUT1 (Figure 7A) and GhSWEET2 (Figure 7B) in the subtending leaves were higher

Sucrose Content and Transcript Levels of Sucrose Transporter Genes in Leaves and Ovules
Adequate sucrose is necessary for fiber development.The sucrose content in ovules was higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA (Figure 6).For example, at 0 DPA, the sucrose content in GhACS6.3-OE(#4) was 18.47 ± 0.25 mg/g, which was 2.15-fold that in the control (8.60 ± 0.17 mg/g).Next, the transcript profiles of genes involved in the initiation and development of cotton fibers were determined.The mRNA levels of GhMML3 in ovules were significantly higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA.For example, at −1 DPA, the transcript level of GhMML3 was 2-fold higher in GhACS6.3-OE(#4)than in the control (Figure 5A).Accordingly, the transcript levels of GhMML4 and GhMML7 in ovules from −1 to 2 DPA were higher in GhACS6.3-OE(#4)than in the control (Figure 5B).These findings indicate that ACC produced by GhACS6.3 promoted the expression of GhMML3, GhMML4, and GhMML7, consistent with the increased density of fiber cells (Figure 3) and formation of fiber yield components (Table 1).

Sucrose Content and Transcript Levels of Sucrose Transporter Genes in Leaves and Ovules
Adequate sucrose is necessary for fiber development.The sucrose content in ovules was higher in GhACS6.3-OE(#4)than in the control from −1 to 2 DPA (Figure 6).For example, at 0 DPA, the sucrose content in GhACS6.3-OE(#4) was 18.47 ± 0.25 mg/g, which was 2.15-fold that in the control (8.60 ± 0.17 mg/g).Because sucrose is transported from the leaves (source) to the ovules (sink), the effects of GhACS6.3 overexpression on the transcript levels of genes encoding the sucrose transporters GhSUT and GhSWEET were investigated.Based on RNA-seq data [27], the transcript profiles of 18 GhSUT [11] and 21 GhSWEET [28] genes in ovules and leaves were determined.As shown in the heat maps, GhSUT2 was specifically expressed in ovules, GhSUT1 was predominantly expressed in leaves (Supplementary Figure S3A), GhSWEET2 specifically expressed in leaves, and GhSWEET5 was predominantly expressed in ovules (Supplementary Figure S3B).
The results of qRT-PCR analyses showed that, from −1 to 2 DPA, the transcript levels Because sucrose is transported from the leaves (source) to the ovules (sink), the effects of GhACS6.3 overexpression on the transcript levels of genes encoding the sucrose transporters GhSUT and GhSWEET were investigated.Based on RNA-seq data [27], the transcript profiles of 18 GhSUT [11] and 21 GhSWEET [28] genes in ovules and leaves were determined.As shown in the heat maps, GhSUT2 was specifically expressed in ovules, GhSUT1 was predominantly expressed in leaves (Supplementary Figure S3A), GhSWEET2 specifically expressed in leaves, and GhSWEET5 was predominantly expressed in ovules (Supplementary Figure S3B).
The results of qRT-PCR analyses showed that, from −1 to 2 DPA, the transcript levels of GhSUT1 (Figure 7A) and GhSWEET2 (Figure 7B) in the subtending leaves were higher in GhACS6.3-OE(#4)than in the control; and the transcript levels of GhSUT2 (Figure 7C) and GhSWEET5 (Figure 7D) in ovules were higher in GhACS6.3-OE(#4)than in the control.These results implied that the increased abundance of GhACS6.3 promoted sucrose transport from the leaves to the ovules.and GhSWEET5 (Figure 7D) in ovules were higher in GhACS6.3-OE(#4)than in the control.These results implied that the increased abundance of GhACS6.3 promoted sucrose transport from the leaves to the ovules.

Involvement of GhACS6.3 in Sucrose Metabolism in Ovules
Ovules are sink organs in which SuSy converts sucrose into UDPG and fructose to support fiber initiation and development.We analyzed the transcript profiles of GhSuSy genes that were specifically expressed in the ovule.As shown in the heat map, we determined the transcript profiles of 15 GhSuSy [29] in ovules and leaves from −3 DPA to 3 DPA.Of them, GhSuSy1 and GhSuSy3 were predominantly expressed in the ovules (Supplementary Figure S4).
The results of qRT-PCR analyses confirmed that the mRNA levels of GhSuSy1 and GhSuSy3 in ovules were higher in GhACS6.than in the control.For example, at 0 DPA, the relative transcript GhSuSy1 and GhSuSy3 in ovules of GhACS6.3-OE(#4) were 3.5-and 3.1-times higher than their respective levels in ovules of the control (Figure 8A).Meanwhile, the UDPG (Figure 8B) or fructose (Supplementary Figure S5) contents showed similar increases in ovules of GhACS6.3-OE(#4)from −1 to 2 DPA.These findings suggested that increased activities of GhSuSy2 and GhSuSy3 in GhACS6.3-OEplants increased the UDPG and fructose contents to facilitate fiber initial development.

Involvement of GhACS6.3 in Sucrose Metabolism in Ovules
Ovules are sink organs in which SuSy converts sucrose into UDPG and fructose to support fiber initiation and development.We analyzed the transcript profiles of GhSuSy genes that were specifically expressed in the ovule.As shown in the heat map, we determined the transcript profiles of 15 GhSuSy [29] in ovules and leaves from −3 DPA to 3 DPA.Of them, GhSuSy1 and GhSuSy3 were predominantly expressed in the ovules (Supplementary Figure S4).
The results of qRT-PCR analyses confirmed that the mRNA levels of GhSuSy1 and GhSuSy3 in ovules were higher in GhACS6.3-OE(#4)than in the control.For example, at 0 DPA, the relative transcript GhSuSy1 and GhSuSy3 in ovules of GhACS6.3-OE(#4) were 3.5-and 3.1-times higher than their respective levels in ovules of the control (Figure 8A).Meanwhile, the UDPG (Figure 8B) or fructose (Supplementary Figure S5) contents showed similar increases in ovules of GhACS6.3-OE(#4)from −1 to 2 DPA.These findings suggested that increased activities of GhSuSy2 and GhSuSy3 in GhACS6.3-OEplants increased the UDPG and fructose contents to facilitate fiber initial development.
GhSuSy3 in ovules were higher in GhACS6.than in the control.For example, at 0 DPA, the relative transcript GhSuSy1 and GhSuSy3 in ovules of GhACS6.3-OE(#4) were 3.5-and 3.1-times higher than their respective levels in ovules of the control (Figure 8A).Meanwhile, the UDPG (Figure 8B) or fructose (Supplementary Figure S5) contents showed similar increases in ovules of GhACS6.3-OE(#4)from −1 to 2 DPA.These findings suggested that increased activities of GhSuSy2 and GhSuSy3 in GhACS6.3-OEplants increased the UDPG and fructose contents to facilitate fiber initial development.

Discussion
In this study, we detected specific expression of GhACS6.3 in the ovules during fiber initial development.Our results show that GhACS6.3-dependentACC production is positively correlated with the fiber yield components and reveal some of the mechanisms underlying increased GhMML3 activity and sucrose transport and utilization.

Specificity of GhACS6.3 Activation during Fiber Initiation
Using data from CottonFGD, we analyzed the transcript profiles of the GhACS family in cotton.Among 35 GhACS genes [23], a pair of GhACS6.3 genes was expressed in the flower (including ovules, stamen, pistil, petal, and calycle) and in the root.Notably, from −3 DPA to 0 DPA, the transcript levels of GhACS6.3 in ovules were higher than those of other GhACS genes (Figure 1A,B).Because the period of −3 to 5 DPA corresponds to the fiber initial development stage [7], the transcription of GhACS6.3 genes during this time may play an important role in fiber initial development.GhACS6.3-OElines showed significantly increased ACC content in ovules from −3 to 10 DPA, and the transcript levels of GhACS6.3 (Figure 2A) were positively correlated with ACC content (Figure 2B).These results imply that GhACS6.3catalyzes ACC production in ovules to facilitate fiber initial development.

Overexpression of GhACS6.3 Increases Fiber Yield Components
Compared with the control, the transgenic GhACS6.3-OElines showed significantly increased values of fiber yield components, such as the number of bolls per plant, boll weight, LI, SI, LP, and fruit branches under open field conditions (Table 1).These changes were related to the increased density of fiber cell initiation (Figure 3) and ultimately resulted in increased fiber yield components in the GhACS6.3-OElines (Table 1).The LI and SI are the important fiber yield components [31].
The results of this study provide evidence that ACC promotes fiber initiation in cotton.First, both ACC (≤1.5 µM) treatment (Figure 4A,B) and GhACS6.3-producedACC (Figure 2A,B) increased the density of fiber cell initiation (Figure 3).We explored the molecular and genetic mechanisms by which the density of fiber cell initials was increased in the GhACS6.3-OElines and found that these lines showed increased transcript levels Plants 2023, 12, 3530 9 of 13 of GhMML3/4/7 (Figure 5A).GhMML3 is known to control fiber cell initiation [4,8] by regulating the expression of GhMML4 and GhMML7 [2,5,32].

Increased Transcript Levels of GhACS6.3 and Increased Sucrose Transport from Leaves to Ovules
During fiber cells' initiation, cotton ovules need sugar to provide nutrients and material basis for the initiation and development of fiber cells.In addition to an increase of fiber initial development (Figure 3) and yield components (Table 1), the GhACS6.3-OElines showed significantly increased sucrose contents in ovules (Figure 6).These results indicate that GhACS6-produced ACC facilitates sucrose transport from leaves to ovules.
As the most ubiquitous phloem-transported sugar, sucrose is synthesized in source organs (mainly in photosynthetic leaves) and consumed or stored in sink organs [33].Sucrose is transported between the source and sink organs by transporters such as SUTs and SWEETs [11,34,35].The cotton ovule is a sink that consumes sucrose, and thus, it requires sucrose to be transported from the leaves.In this study, transcript profiling analyses revealed specific expression of GhSUT2 in ovules and the predominant expression of GhSUT1 in the leaves (Supplementary Figure S3A).Consistent with this, a previous study showed that cotton uses a single SUT for phloem loading in mature leaves but uses multiple SUTs for loading into fibers [11].At the same time, the transcript profiling analyses revealed specific expression of GhSWEET2 in the leaves and the predominant expression of GhSWEET5 in ovules (Supplementary Figure S3B), in accordance with the previous finding that the activities of GhSWEET2/5 promote sucrose accumulation in fibers [28].Our results show that in the GhACS6.3-OElines, the transcript levels of GhSUT1 and GhSWEET2 in leaves and those of GhSUT2 and GhSWEET5 in ovules were increased from −1 to 2 DPA (Figure 7).These findings imply that in the GhACS6.3-OElines, the cooperation between GhSUT1/2 and GhSWEET2/5 ensures fiber development.

GhACS6.3 Overexpression Affects Sucrose Utilization in Ovules
The activity of SuSy is regarded as a biochemical marker for sink strength, especially in crop species [14].Sucrose is degraded by SuSy into UPDG and fructose to provide the building blocks and energy necessary for fiber development [12], and this scenario has been proven in upland cotton [29].In this work, transcript profiling analyses (Supplementary Figure S4) revealed specific and predominant expression of GhSuSy1/3 in ovules.Interestingly, compared with the control, the GhACS6.3-OElines showed significantly increased transcript levels of GhSuSy1/3 (Figure 8A) as well as increased contents of UDPG (Figure 8B) and fructose (Supplementary Figure S5) in ovules from −1 to 2 DPA.Consistent with this, the transgenic cotton plants with suppressed GhSuSy showed reduced fiber initiation [12].Our results show that an increase in the abundance of GhACS6.3 is beneficial for sucrose metabolism into UDPG and fructose in the young ovules, thus promoting fiber initial development.

Cotton Materials
G. hirsutum cv.YZ1 was used as the control.The transgenic GhACS2/6-OE lines were created as previously reported [36].Briefly, the open reading frame of GhACS6.3 (GH_A12G2512) was inserted into the vector pK7WG2 with CaMV 35S promoter and introduced into Agrobacterium tumefaciens strain EHA105 and then infected hypocotyl of YZ1.The positive plants were screened on 1/2 MS medium containing 50 µg/mL kanamycin, until obtaining T 3 lines for research analysis.During cotton plant culturing, seeds were soaked in H 2 O at 25 • C for 24 h and then were germinated on moist cotton wool at 25 • C for 2 days.Cotton young seedings grew in the greenhouse (25 ± 2 • C, 80% relative humidity, 120 µmol m −2 s −1 light intensity and a 16-h light/8-h dark photoperiod) for two weeks.
According to experimental needs, some seedlings were transplanted into the experimental field in Henan University (Kaifeng, China) with a within-row plant-to-plant distance of 25 to 30 cm.This field experiment has been repeated for many years, and the results are consistent.This manuscript presents data for the two years 2021 and 2022 were averaged over three replications every year with 30 plants per replicate.

Transcriptome Data Analysis
Genome-wide transcriptome data were downloaded from the G. hirsutum database (SRA:PRJNA248163) as previously reported [27].Log 2 (TPM+1) normalization was performed, and the R-4.0.2 language was used to compile the standardized data.

Statistics of Cotton Fiber Yield Components
The methods of statistically determining cotton fiber yield components were used as described in reported literature [31].The boll weight is the average dry weight of bolls per plant.The lint index (LI) is the lint weight from 100 seeds, and the seed index (SI) is the weight of 100 seeds with fiber.These data of cotton fiber yield components were calculated in 2021 and 2022 employing a complete randomized block design with three replications every year and 30 individuals per replication.

In Vitro Ovule Culturing
The buds were harvested from the −1 DPA ovules (at this point, there was no fibers protruding from the ovule epidermis).Referring to the methods provided in the literature [37], the whole tissue was immediately immersed in 75% ethanol for 5 min with continuous shaking, then rinsed in distilled and deionized water, and soaked again in 0.1% HgCl 2 solution containing 0.05% Tween-80 for 20 min to sterilize it.Then the shuck of the ovary was removed carefully to avoid damage to the ovules.The ovules were then placed gently on the surface of the medium within each culture plate having 5 mL of nutrient medium.During in vitro culturing, the floating ovules were incubated at 30 ± 1 • C in darkness without shaking.
The ovules were cultured on BT medium with some modifications.The BT medium was prepared at pH 5 according to the literature [37].According to experimental requirements, the BT media was supplemented with various concentrations of ACC (0, 0.5, 1.0, 1.5, 2.0, or 5.0 µM).The original ACC reagent was purchased from Sigma-Aldrich (Steinheim, Germany).After 48 h of in vitro culturing of the ovules, the fiber initial development on the ovule epidermis was analyzed.

Analysis of the Density of Fiber Cells on the Ovule Epidermis
The number of fiber cells was observed and photographed by using a scanning electron microscope (SEM, FEI Quanta 250, Rock Hill, SC, USA).The number of fiber initial cells was counted from the same middle region of the ovule.The fiber cells in each image were counted by using ImageJ software, and then the density of the fiber initial cells was calculated per 0.01 mm 2 on the ovules.The experiments were repeated five times with at least 30 ovules per replicate.

RNA Analysis
The ovules and subtending leaves were harvested and immediately immersed in liquid nitrogen and stored at −80 • C. The frozen samples were ground to a fine powder in liquid nitrogen with a mortar and pestle.Total RNA extractions were performed from 100 mg of each macerate of plant tissue using an RNAprep Pure Plant kit (TIANGEN Biotech, Beijing, China).Complementary DNA (cDNA) was synthesized using a Reverse Transcription system (Vazyme, Nanjing, China) and was used as the template for qRT-PCR analyses along with 2× SYBR Green I master mix (Vazyme, Nanjing, China) on a Roche 480 real-time PCR system (Roche, Basle, Sweden).The relative expression levels were calculated based on the comparative ∆C T method [36,38].The reference gene was cotton GhUBQ7 (DQ116441).The primers used for PCR amplification are listed in Table S1.

Measurements of ACC Contents
The ovules and subtending leaves were harvested for ACC extraction.ACC was extracted as described in a previous study [36].Fresh samples (approximately 100 mg) were harvested and ground to a fine powder in liquid nitrogen, and 1 mL of cold extraction buffer (800 µL of methanol, 190 µL of water, and 10 µL of acetic acid) was added to it.The mixture was homogenized thoroughly by shaking for 16 h at 4 • C in the dark, and it was then centrifuged at 10,000 rpm for 15 min at 4 • C. The supernatant was filtered through the 0.2 µm microfilter and collected.The filtrates were dried using nitrogen gas at room temperature and were then dissolved in 200 µL of methanol.The sample was diluted 100 times and analyzed by using an Applied Biosystems MDS SCIEX 4000 QTRAP liquid chromatography-tandem mass spectrometry system (AB Sciex, Foster City, CA, USA).The ACC standard was used for the quantitative analyses.

Soluble Carbohydrate Analysis
The ovules at −1 to 2 DPA were harvested.Approximately 50 mg samples were ground in liquid nitrogen.The samples were extracted three times with 1 mL of preheated 75% ethanol for 5 min at 80 • C. The aqueous phases of the samples were collected, filtered, and dried under a vacuum at 37 • C and then redissolved in 200 µL distilled water for further analysis.Sucrose, UDP-glucose, and fructose contents were measured and analyzed with an Applied Biosystems MDS SCIEX 4000 QTRAP liquid chromatographytandem mass spectrometry system (AB Sciex, Foster City, CA, USA).Standard samples (sucrose, UDP-glucose, and fructose) (Sigma-Aldrich, Steinheim, Germany) were used for the quantitative analyses.

Statistical Analysis
At least three biological replicates and three technical replicates were conducted in all of the experiments.Data are presented as means ± SDs.Asterisks indicate statistically significant differences, as determined by Student's t-tests (*, p < 0.05 indicate significant differences, **, p < 0.01 indicate extremely significant differences).

Conclusions
During fiber initial development, GhACS6.3-dependentACC accumulation in the ovule promotes GhMML3-triggered fiber initiation and improves sucrose transport to the ovules and metabolism in the ovules and thus increases fiber yield components.This study provides new ideas for improving cotton fiber yield components.
Bolls per plant = the total number of mature bolls, young bolls, and batting bolls, Boll weight = the average dry weight of all batting bolls per plant, LI (g) = the lint weight of 100 unginned seeds, SI (g) = the weight of 100 seeds, Lint percentage (%) = lint weight (g)/unginned seeds weight (g) × 100.These cotton fiber yield components showed in 2021 and 2022 with a complete randomized block design with three replications every year, and 30 individuals per replication were employed.Asterisks indicate statistically significant differences between the control and GhACS6.3-OE(#4)as determined by Student's t-test (* p < 0.05, ** p < 0.01).Error bars represent ± SD.

Figure 3 .
Figure 3. Fiber initial density of GhACS6.3-OElines grown under open field conditions.(A) Images of ovule epidermis at 0 DPA.Scale bar = 500 µm for whole ovules and 100 µm in higher-magnification images.(B) Statistical analysis of fiber cell density on ovule epidermis of control and GhACS6.3-OE(#4)lines.Experiments were repeated five times with similar results, with at least 30 ovules per replicate.Values are means ± SDs (Student's t-test; ** p < 0.01).

Figure 3 .
Figure 3. Fiber initial density of GhACS6.3-OElines grown under open field conditions.(A) Images of ovule epidermis at 0 DPA.Scale bar = 500 µm for whole ovules and 100 µm in higher-magnification images.(B) Statistical analysis of fiber cell density on ovule epidermis of control and GhACS6.3-OE(#4)lines.Experiments were repeated five times with similar results, with at least 30 ovules per replicate.Values are means ± SDs (Student's t-test; ** p < 0.01).

Figure 7 .
Figure 7. Transcript profiles of GhSUTs and GhSWEETs in ovules and leaves.(A-D) qRT-PCR analysis of transcript levels of GhSUT1, GhSUT2, GhSWEET2, and GhSWEET5 in leaves and ovules at −1 to 2 DPA, normalized against GhUBQ7 as the internal control.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; ** p < 0.01).

Figure 8 .
Figure 8. Transcript levels of GhSuSys in ovules of GhACS6.3-OElines.(A) qRT-PCR analysis of mRNA levels of GhSuSy1 and GhSuSy3 genes in ovules at −1 to 2 DPA, normalized against GhUBQ7 as the internal control.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).(B) HPLC analysis of UPDG in ovules at −1 to 2 DPA.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).

Figure 7 .
Figure 7. Transcript profiles of GhSUTs and GhSWEETs in ovules and leaves.(A-D) qRT-PCR analysis of transcript levels of GhSUT1, GhSUT2, GhSWEET2, and GhSWEET5 in leaves and ovules at −1 to 2 DPA, normalized against GhUBQ7 as the internal control.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; ** p < 0.01).

Figure 8 .
Figure 8. Transcript levels of GhSuSys in ovules of GhACS6.3-OElines.(A) qRT-PCR analysis of mRNA levels of GhSuSy1 and GhSuSy3 genes in ovules at −1 to 2 DPA, normalized against GhUBQ7 as the internal control.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).(B) HPLC analysis of UPDG in ovules at −1 to 2 DPA.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).

Figure 8 .
Figure 8. Transcript levels of GhSuSys in ovules of GhACS6.3-OElines.(A) qRT-PCR analysis of mRNA levels of GhSuSy1 and GhSuSy3 genes in ovules at −1 to 2 DPA, normalized against GhUBQ7 as the internal control.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).(B) HPLC analysis of UPDG in ovules at −1 to 2 DPA.Experiments were repeated three times with similar results.Values are means ± SDs (Student's t-test; * p < 0.05, and ** p < 0.01).