Silencing of Glycogen Synthase Kinase 3 Significantly Inhibits Chitin and Fatty Acid Metabolism in Asian Citrus Psyllid, Diaphorina citri

Glycogen is a predominant carbohydrate reserve in various organisms, which provides energy for different life activities. Glycogen synthase kinase 3 (GSK3) is a central player that catalyzes glucose and converts it into glycogen. In this study, a GSK3 gene was identified from the D. citri genome database and named DcGSK3. A reverse transcription quantitative PCR (RT-qPCR) analysis showed that DcGSK3 was expressed at a high level in the head and egg. The silencing of DcGSK3 by RNA interference (RNAi) led to a loss-of-function phenotype. In addition, DcGSK3 knockdown decreased trehalase activity, glycogen, trehalose, glucose and free fatty acid content. Moreover, the expression levels of the genes associated with chitin and fatty acid synthesis were significantly downregulated after the silencing of DcGSK3. According to a comparative transcriptomics analysis, 991 differentially expressed genes (DEGs) were identified in dsDcGSK3 groups compared with dsGFP groups. A KEGG enrichment analysis suggested that these DEGs were primarily involved in carbon and fatty acid metabolism. The clustering analysis of DEGs further confirmed that chitin and fatty acid metabolism-related DEGs were upregulated at 24 h and were downregulated at 48 h. Our results suggest that DcGSK3 plays an important role in regulating the chitin and fatty acid metabolism of D. citri.


Introduction
Citrus Huanglongbing (HLB) is one of the most destructive citrus diseases worldwide, caused by the bacterium Candidatus Liberibacter asiaticus (CLas) and spread by Diaphorina citri Kuwayama [1,2]. The management of HLB can be achieved based on preventive measures, such as planting healthy nursery trees, eliminating CLas-infected trees and controlling the transmission vector [3]. In all the prevention and control measures of HLB, the most effective method is the vector control of D. citri [4]. To date, the control of D. citri mainly relies on the use of chemical insecticides, including neonicotinoids, pyrethroid and organophosphate [5]. However, the excessive use of chemical insecticides has led to high levels of insecticide resistance in D. citri [6]. Therefore, it is crucial to screen and investigate novel targets for controlling D. citri.
In most organisms, glucose metabolism plays a crucial role in energy metabolism, which provides the energy required for life processes, such as growth and development [7,8]. Glycogen and triglyceride are two major energy reserves in animal cells; however, their storage forms are different [9]. Glycogen is stored in a bulky hydrated form, whereas triglyceride is stored in an anhydrous form [10]. Glycogen is a branched glucose polymer that stores energy and carbon in organisms, which plays an essential role in maintaining glucose and energy homeostasis [11]. The glycogen in animals is synthesized under the catalysis of glycogen synthase, and glycogen degradation requires the action of glycogen phosphorylase, debranching enzymes and acid α-glucosidase [12,13]. However, in insects, glycogen is primarily synthesized from UDP-glucose and is used as spare sugar along with trehalose in order to maintain glucose availability [14]. Therefore, glycogen metabolism in insects is closely related to the formation and utilization of trehalose. Glycogen synthase kinase 3 (GSK3) is a serine-threonine kinase that acts as an important regulator of metabolism and development in diverse organisms [15]. Numerous studies have confirmed that GSK3 plays an essential role in regulating glucose metabolism [16][17][18]. GSK3 can inhibit glycogen synthase in insects by phosphorylating and blocking the transmission of the insulin signaling pathway [19]. Ding et al. revealed that GSK3 is a key enzyme that regulates energy and chitin metabolism in N. lugens [20]. However, the molecular mechanisms by which GSK3 regulates chitin and fatty acid metabolism in D. citri have not yet been reported.
RNA interference (RNAi) is a conserved mechanism in most organisms causing a sequence-specific gene silencing phenomenon induced by double-stranded RNA (dsRNA) [21]. RNAi has recently been developed as a useful tool for functional gene research in animals and plants [22]. The silencing of some target genes in insects by ingesting dsRNA suggests that RNAi technology may have application in pest control. Mao et al. revealed that silencing the Helicoverpa armigera P450 monooxygenase gene by plant-mediated RNAi impaired the larval tolerance of gossypol [23]. Therefore, RNAi has shown tremendous potential in pest management [24]. There are three approaches to deliver dsRNA to insects, including oral feeding, microinjection and topical applications, but the microinjection of dsRNA can only be used for fundamental research in the laboratory [25]. Furthermore, the transgenic plants expressing the dsRNA of specific genes are also receiving increasing attention [26]. However, RNAi efficiency is influenced by several external factors, including incomplete dsRNA internalization, instability of dsRNA, impaired systemic spreading of the RNAi signal and refractory target genes [27]. Hence, an efficient dsRNA delivery method and the selection of target genes are the two prerequisites for RNAi-mediated pest management. RNAi by the oral delivery of dsRNA in insects has great potential as a tool for integrated pest management [28]. Cao et al. demonstrated that silencing Mythimna separate chitinase genes by an oral delivery of dsRNA caused lethal phenotypes [29]. The silencing of D. citri NADPHcytochrome P450 reductase (DcCPR) by oral delivery significantly decreased the expression level of DcCPR and increased the susceptibility of D. citri to chemical insecticides [30]. Moreover, the silencing of the D. citri trehalase (DcTre) gene by RNAi significantly reduced the expression levels of chitin-metabolism-related genes, led to a malformed phenotype and dsRNA delivery was conducted by an artificial diet [31]. In hemipterans, dsRNA is easily degraded by either the high pH or dsRNase in the gut [32]. Consequently, we conclude that the microinjection of dsRNA is more than that of oral administration.
In the present study, a glycogen synthase kinase 3 (GSK3) gene, named DcGSK3, was identified from the D. citri genome database, and the sequences and spatial and temporal expression profiles of DcGSK3 were analyzed. The functions of DcGSK3 were investigated using RNAi combined with transcriptome sequencing. Furthermore, the levels of glucose, trehalose, glycogen, free fatty acids and trehalase activity were measured. The results showed that the silencing of the DcGSK3 gene significantly inhibited chitin and fatty acid metabolism, indicating that DcGSK3 is involved in regulating chitin and fatty acid synthesis. These findings provide a theoretical basis for D. citri control through chitin and fatty acid metabolism.

Bioinformatics Analysis of DcGSK3
By blasting the glycogen synthase kinase 3 protein sequence from N. lugens (XP_039293499), Bombyx mori (XM_004928285) and Drosophila melanogaster (NP_476714) against the D. citri genome database in NCBI, the DcGSK3 cDNA sequence of 981 bp was identified ( Figure 1A). The DcGSK3 encodes a protein containing 326 amino acids and has a predicted molecular weight of 36.71 kDa and a theoretical isoelectric point of 7.08. The structural domain analysis revealed that DcGSK3 consisted of one Pkinase_Tyr domain at the N-terminal and two low complexity (LC) domains at the C-terminal ( Figure 1B). The phylogenetic tree analysis showed that the DcGSK3 is more closely related to the glycogen synthase kinase from Hemiptera (A. pisum, M. persicae and A. glycines) than its counterparts from Lepidoptera, Diptera, Coleoptera and Hymenoptera ( Figure 1C). A multiple alignment analysis revealed that the DcGSK3 protein sequence shares a high level of identity with A. pisum and M. persicae (Figure 2A). Further, 13 conserved phosphorylation sites were found when comparing the DcGSK3 with the GSK3 sequence from A. pisum and M. persicae (Figure 2A). By using the SWISS-MODEL software, the tertiary structure of DcGSK3 was predicted. The results showed that the DcGSK3 protein is primarily composed of alpha helices ( Figure 2B).

Bioinformatics Analysis of DcGSK3
By blasting the glycogen synthase kinase 3 protein sequence from N. lugens (XP_039293499), Bombyx mori (XM_004928285) and Drosophila melanogaster (NP_476714) against the D. citri genome database in NCBI, the DcGSK3 cDNA sequence of 981 bp was identified ( Figure 1A). The DcGSK3 encodes a protein containing 326 amino acids and has a predicted molecular weight of 36.71 kDa and a theoretical isoelectric point of 7.08. The structural domain analysis revealed that DcGSK3 consisted of one Pkinase_Tyr domain at the N-terminal and two low complexity (LC) domains at the C-terminal ( Figure  1B). The phylogenetic tree analysis showed that the DcGSK3 is more closely related to the glycogen synthase kinase from Hemiptera (A. pisum, M. persicae and A. glycines) than its counterparts from Lepidoptera, Diptera, Coleoptera and Hymenoptera ( Figure 1C). A multiple alignment analysis revealed that the DcGSK3 protein sequence shares a high level of identity with A. pisum and M. persicae (Figure 2A). Further, 13 conserved phosphorylation sites were found when comparing the DcGSK3 with the GSK3 sequence from A. pisum and M. persicae (Figure 2A). By using the SWISS-MODEL software, the tertiary structure of DcGSK3 was predicted. The results showed that the DcGSK3 protein is primarily composed of alpha helices ( Figure 2B).

Expression Patterns of DcGSK3 at Different Tissues and Developmental Stages
The expression patterns of DcGSK3 in different tissues and at different developmental stages were analyzed by RT-qPCR. The results showed that DcGSK3 expression was detected in all tissues, including the midgut, leg, head, wing, integument and fat body ( Figure 3). However, the highest expression level of DcGSK3 was found in the head, whereas it was low in the midgut. The expression level of DcGSK3 in the head was 6.3 times higher than in the midgut. In addition, the expression level of DcGSK3 was significantly downregulated from the egg to the nymph stages. However, the expression level of DcGSK3 was constant without significant variations from the first-instar nymph to the adult stage ( Figure 3).

Expression Patterns of DcGSK3 at Different Tissues and Developmental Stages
The expression patterns of DcGSK3 in different tissues and at different developmental stages were analyzed by RT-qPCR. The results showed that DcGSK3 expression was detected in all tissues, including the midgut, leg, head, wing, integument and fat body ( Figure 3). However, the highest expression level of DcGSK3 was found in the head, whereas it was low in the midgut. The expression level of DcGSK3 in the head was 6.3 times higher than in the midgut. In addition, the expression level of DcGSK3 was significantly downregulated from the egg to the nymph stages. However, the expression level of DcGSK3 was constant without significant variations from the first-instar nymph to the adult stage ( Figure 3).

Expression Patterns of DcGSK3 at Different Tissues and Developmental Stages
The expression patterns of DcGSK3 in different tissues and at different developmental stages were analyzed by RT-qPCR. The results showed that DcGSK3 expression was detected in all tissues, including the midgut, leg, head, wing, integument and fat body ( Figure 3). However, the highest expression level of DcGSK3 was found in the head, whereas it was low in the midgut. The expression level of DcGSK3 in the head was 6.3 times higher than in the midgut. In addition, the expression level of DcGSK3 was significantly downregulated from the egg to the nymph stages. However, the expression level of DcGSK3 was constant without significant variations from the first-instar nymph to the adult stage ( Figure 3).  . Data were normalized using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and are represented as the means ± standard errors of the means from three independent experiments. Different lowercase letters indicate the significant differences, for example, a, b and c (p < 0.05).

Analysis of D. citri Phenotype and DcGSK3 Expression Level after RNAi
In order to confirm the functions of DcGSK3 in the development of D. citri, the DcGSK3 expression level and effects on the D. citri phenotype were investigated following a microinjection of dsDcGSK3. The results showed that the DcGSK3 expression level was significantly downregulated at 24 h and 48 h after the injection of dsRNA ( Figure 4B). The transition from a fifth-instar nymph to an adult was disrupted in the dsDcGSK3 treatment group, and the emerged adult showed two abnormal phenotypes. The first phenotype was that fifth-instar nymphs could not undergo a normal molt. They also had abnormal dorsal tergites and malformed wings. In the second phenotype, the fifth-instar nymphs were capable of molting, but their wings were irregular and curled at the distal end or were small in size. However, the nymphs in the control group (dsGFP treatment) were able to molt normally ( Figure 4C). 4B). The transition from a fifth-instar nymph to an adult was disrupted in the dsDcGSK3 treatment group, and the emerged adult showed two abnormal phenotypes. The first phenotype was that fifth-instar nymphs could not undergo a normal molt. They also had abnormal dorsal tergites and malformed wings. In the second phenotype, the fifth-instar nymphs were capable of molting, but their wings were irregular and curled at the distal end or were small in size. However, the nymphs in the control group (dsGFP treatment) were able to molt normally ( Figure 4C). The mortality and malformation rates were significantly upregulated from 12 h to 48 h after the silencing of DcGSK3. The mortality rate in the treatment group (dsDcGSK3) was 18.6% while it was 4% in the control group (dsGFP) at 12 h, and the mortality reached 71.5% at 48 h ( Figure 5A). However, the mortality rate was consistently below 20% from 12 h to 48 h in the control group. D. citri had no obvious abnormal phenomenon at 12 h after treatment with dsGFP or dsDcGSK3. Nevertheless, the malformation rate increased rapidly from 8% at 24 h to 32% at 48 h in the dsDcGSK3 groups (Figure The mortality and malformation rates were significantly upregulated from 12 h to 48 h after the silencing of DcGSK3. The mortality rate in the treatment group (dsDcGSK3) was 18.6% while it was 4% in the control group (dsGFP) at 12 h, and the mortality reached 71.5% at 48 h ( Figure 5A). However, the mortality rate was consistently below 20% from 12 h to 48 h in the control group. D. citri had no obvious abnormal phenomenon at 12 h after treatment with dsGFP or dsDcGSK3. Nevertheless, the malformation rate increased rapidly from 8% at 24 h to 32% at 48 h in the dsDcGSK3 groups ( Figure 5B). However, the malformation rate was increased slightly in the dsGFP groups, but the difference was not significant. In contrast, the rate of the cumulative molting from 12 h to 48 h was increased after the injection of dsGFP and dsDcGSK3 ( Figure 5C). In the dsDcGSK3 groups, the cumulative molting rate at 48 h reached 34%, while it was approximately 15% in the dsGFP groups. These results indicated that the silencing of DcGSK3 impaired the molting process of fifth-instar D. citri nymphs.

Analysis of Trehalase Activity and Trehalose, Glycogen, Glucose and Free Fatty Acid Contents after Silencing of DcGSK3
In order to determine whether the DcGSK3 gene can regulate sugar metabolism and fatty acid synthesis, the trehalase activity, trehalose, glucose, glycogen and free fatty acid contents were measured at 24 h and 48 h after the silencing of DcGSK3. The results showed that the membrane-bound trehalase activity decreased at 24 h, while it had no significant change between the treatment group (dsDcGSK3 treatment) and control group (dsGFP treatment) ( Figure 6A). On the contrary, the soluble trehalase activity increased at 24 h; however, it had no obvious change at 48 h after silencing DcGSK3 ( Figure 6B). Following the inhibition of DcGSK3 expression, glucose, trehalose and glycogen content were reduced at 24 h and 48 h ( Figure 6C-E). Furthermore, the free fatty acid content was also downregulated at 24 h in the treatment group (dsDcGSK3 treatment) compared with the control group (dsGFP treatment) ( Figure 6F). These results suggest that DcGSK3 regulates the metabolism of sugar and fatty acids in D. citri.
Int. J. Mol. Sci. 2022, 23, x FOR PEER REVIEW 6 of 20 5B). However, the malformation rate was increased slightly in the dsGFP groups, but the difference was not significant. In contrast, the rate of the cumulative molting from 12 h to 48 h was increased after the injection of dsGFP and dsDcGSK3 ( Figure 5C). In the dsDcGSK3 groups, the cumulative molting rate at 48 h reached 34%, while it was approximately 15% in the dsGFP groups. These results indicated that the silencing of DcGSK3 impaired the molting process of fifth-instar D. citri nymphs.

Analysis of Trehalase Activity and Trehalose, Glycogen, Glucose and Free Fatty Acid Contents after Silencing of DcGSK3
In order to determine whether the DcGSK3 gene can regulate sugar metabolism and fatty acid synthesis, the trehalase activity, trehalose, glucose, glycogen and free fatty acid contents were measured at 24 h and 48 h after the silencing of DcGSK3. The results showed that the membrane-bound trehalase activity decreased at 24 h, while it had no significant change between the treatment group (dsDcGSK3 treatment) and control group (dsGFP treatment) ( Figure 6A). On the contrary, the soluble trehalase activity increased at 24 h; however, it had no obvious change at 48 h after silencing DcGSK3 (Figure 6B). Following the inhibition of DcGSK3 expression, glucose, trehalose and glycogen content were reduced at 24 h and 48 h ( Figure 6C-E). Furthermore, the free fatty acid content was also downregulated at 24 h in the treatment group (dsDcGSK3 treatment) compared with the control group (dsGFP treatment) ( Figure 6F). These results suggest that DcGSK3 regulates the metabolism of sugar and fatty acids in D. citri.

Analysis of Chitin-Metabolism-Related Genes after Silencing of DcGSK3
We performed an RT-qPCR to detect the expression levels of chitin-metabolism-related genes, including TPS1, Tre1-1, Tre1-2, Tre2, HK, G6PI, GFAT, GNPNA, PAGM, UAP, CHT and NAG, to determine whether DcGSK3 could regulate D. citri chitin metabolism. The results showed that the relative expression levels of D. citri Tre1-1, Tre1-2, Tre2, HK and G6PI were downregulated at 24 h and 48 h after silencing DcGSK3 (Figure 7). Further, the expression levels of the three genes decreased only at 48 h after treatment with DcGSK3, whereas no significant difference was detected between the dsDcGSK3 groups and dsGFP groups at 24 h, including TPS1, GNPNA and PAGM (Figure 7). The relative expression levels of GFAT and NAG were upregulated in the dsDcGSK3 compared with the dsGFP group at 24 h, with no significant difference at 48 h ( Figure 7). Additionally, the expression level of UAP was downregulated at 24 h and exhibited upregulation at 48 h after the silencing of DcGSK3, and the CHT expression level had no significant difference at 24 h and 48 h between the dsDcGSK3 groups and dsGFP groups (Figure 7). These results indicated that DcGSK3 might play an important role in regulating chitin metabolism in D. citri. and free fatty acid content (F) at 24 h and 48 h. Each experiment contained three biological replicates. The significant differences are indicated by * (p < 0.05) or ** (p < 0.01). The "ns" indicated no significant differences.

Analysis of Chitin-Metabolism-Related Genes after Silencing of DcGSK3
We performed an RT-qPCR to detect the expression levels of chitin-metabolism-related genes, including TPS1, Tre1-1, Tre1-2, Tre2, HK, G6PI, GFAT, GNPNA, PAGM, UAP, CHT and NAG, to determine whether DcGSK3 could regulate D. citri chitin metabolism. The results showed that the relative expression levels of D. citri Tre1-1, Tre1-2, Tre2, HK and G6PI were downregulated at 24 h and 48 h after silencing DcGSK3 ( Figure 7). Further, the expression levels of the three genes decreased only at 48 h after treatment with DcGSK3, whereas no significant difference was detected between the dsDcGSK3 groups and dsGFP groups at 24 h, including TPS1, GNPNA and PAGM (Figure 7). The relative expression levels of GFAT and NAG were upregulated in the dsDcGSK3 compared with the dsGFP group at 24 h, with no significant difference at 48 h ( Figure 7). Additionally, the expression level of UAP was downregulated at 24 h and exhibited upregulation at 48 h after the silencing of DcGSK3, and the CHT expression level had no significant difference at 24 h and 48 h between the dsDcGSK3 groups and dsGFP groups (Figure 7). These results indicated that DcGSK3 might play an important role in regulating chitin metabolism in D. citri. The significant differences are indicated by * (p < 0.05) or ** (p < 0.01). The "ns" indicated no significant differences.

Analysis of Fatty Acid Metabolism-Related Genes after Silencing of DcGSK3
The acetyl-CoA produced from glycolysis can be utilized to form lipids. In order to analyze the effect of DcGSK3 on D. citri fatty acid metabolism, a total of six genes involved in the synthesis and degradation of fatty acids were analyzed at 24 h and 48 h after the silencing of DcGSK3. The results suggested that two genes associated with fatty acid synthesis were significantly downregulated at 24 h after the knockdown of DcGSK3. In contrast, there was no significant difference between the dsDcGSK3 group and dsGFP group at 48 h, including acetyl-CoA carboxylase-like (ACC) and fatty acid synthase-like (FAS) products ( Figure 8). In addition, three genes involved in the oxidative degradation of fatty acids exhibited similar expression patterns. They were upregulated at 24 h or 48 h after silencing DcGSK3, including medium-chain-specific acyl-CoA dehydrogenase (MCAD), glutaryl-CoA dehydrogenase (GCD) and Lipase. The expression level of acetyl-CoA acetyltransferase (ACAT) remained unchanged at 24 h and 48 h after silencing DcGSK3 (Figure 8). We speculated that DcGSK3 might be involved in regulating fatty acid metabolism in D. citri.

Transcriptome Sequencing and Reads Assembly
After removing the redundant reads, a total of 40 Figure 7. Effects of silencing DcGSK3 on expression levels of twelve genes involved in chitin metabolism in D. citri. The dsGFP treatment group was used as a control. The mean expression level represents three biological replicates. The asterisks indicate significant differences by * (p < 0.05), ** (p < 0.01) and *** (p < 0.001).

Analysis of Fatty Acid Metabolism-Related Genes after Silencing of DcGSK3
The acetyl-CoA produced from glycolysis can be utilized to form lipids. In order to analyze the effect of DcGSK3 on D. citri fatty acid metabolism, a total of six genes involved in the synthesis and degradation of fatty acids were analyzed at 24 h and 48 h after the silencing of DcGSK3. The results suggested that two genes associated with fatty acid synthesis were significantly downregulated at 24 h after the knockdown of DcGSK3. In contrast, there was no significant difference between the dsDcGSK3 group and dsGFP group at 48 h, including acetyl-CoA carboxylase-like (ACC) and fatty acid synthase-like (FAS) products (Figure 8). In addition, three genes involved in the oxidative degradation of fatty acids exhibited similar expression patterns. They were upregulated at 24 h or 48 h  Figure  8). We speculated that DcGSK3 might be involved in regulating fatty acid metabolism in D. citri. Figure 8. Effects of silencing DcGSK3 on expression levels of six genes related to fatty acid metabolism in D. citri. The dsGFP treatment group was used as a control. The mean expression level represents three biological replicates. The asterisks indicate significance differences by * (p < 0.05) and ** (p < 0.01).  (Table S3).

Identification of DEGs and Enrichment Analysis
The DEGs were identified in comparable groups by using the DESeq method. A total of 991 DEGs were identified in the dsDcGSK3 groups compared with the dsGFP groups at 24 h, of which 444 DEGs were upregulated and 547 DEGs were downregulated ( Figure 9A; Table S4). A total of 1692 DEGs were identified in the dsDcGSK3 groups compared with the dsGFP groups at 48 h, among which 1020 DEGs were upregulated and 672 DEGs were downregulated ( Figure 9B; Table S4).
A GO enrichment analysis suggested that most DEGs were mainly involved in the carbohydrate metabolic process at 24 h after the knockdown of DcGSK3 ( Figure 10A). The upregulated DEGs were associated with a cellular amide metabolic process and amide biosynthetic process, and the downregulated DEGs were related to transporter Figure 8. Effects of silencing DcGSK3 on expression levels of six genes related to fatty acid metabolism in D. citri. The dsGFP treatment group was used as a control. The mean expression level represents three biological replicates. The asterisks indicate significance differences by * (p < 0.05) and ** (p < 0.01).

Identification of DEGs and Enrichment Analysis
The DEGs were identified in comparable groups by using the DESeq method. A total of 991 DEGs were identified in the dsDcGSK3 groups compared with the dsGFP groups at 24 h, of which 444 DEGs were upregulated and 547 DEGs were downregulated ( Figure 9A; Table S4). A total of 1692 DEGs were identified in the dsDcGSK3 groups compared with the dsGFP groups at 48 h, among which 1020 DEGs were upregulated and 672 DEGs were downregulated ( Figure 9B; Table S4). activity and the carbohydrate metabolic process. At 48 h between the dsDcGSK3 treatment groups and dsGFP treatment groups, most DEGs were mainly associated with transition metal ion binding and transporter activity. The upregulated DEGs were related to transporter activity and iron ion binding, while the downregulated DEGs were mainly associated with the ribosome's structural molecular activity and structural constituent ( Figure 10B). A KEGG enrichment revealed that most DEGs were significantly enriched for starch, sucrose metabolism and fatty acid metabolism at 24 h after the knockdown of DcGSK3 ( Figure 11A). At 48 h between the dsDcGSK3 treatment groups and dsGFP treatment groups, most DEGs were significantly enriched for lysosome, ribosome and fatty acid metabolism ( Figure 11B).  A GO enrichment analysis suggested that most DEGs were mainly involved in the carbohydrate metabolic process at 24 h after the knockdown of DcGSK3 ( Figure 10A). The upregulated DEGs were associated with a cellular amide metabolic process and amide biosynthetic process, and the downregulated DEGs were related to transporter activity and the carbohydrate metabolic process. At 48 h between the dsDcGSK3 treatment groups and dsGFP treatment groups, most DEGs were mainly associated with transition metal ion binding and transporter activity. The upregulated DEGs were related to transporter activity and iron ion binding, while the downregulated DEGs were mainly associated with the ribosome's structural molecular activity and structural constituent ( Figure 10B). A KEGG enrichment revealed that most DEGs were significantly enriched for starch, sucrose metabolism and fatty acid metabolism at 24 h after the knockdown of DcGSK3 (Figure 11A). At 48 h between the dsDcGSK3 treatment groups and dsGFP treatment groups, most DEGs were significantly enriched for lysosome, ribosome and fatty acid metabolism ( Figure 11B).

Identification of DEGs Involved in Chitin and Fatty Acid Metabolism
Based on the transcriptome analysis, numerous DEGs involved in chitin and fatty acid metabolism were altered in different comparable groups. In the chitin metabolism, 16 DEGs related to chitin metabolism were identified based on transcriptome data, of which 13 (81.2%) DEGs were downregulated at 24 h in the dsDcGSK3 groups compared with the dsGFP groups ( Figure 12; Table 1). Furthermore, three genes were upregulated at 24 h after the silencing of DcGSK3, including IDGF2, PAGM and HK2. It was found that 11 DEGs were upregulated and 5 DEGs showed downregulation at 48 h after the silencing of DcGSK3. These results indicated that the inhibition of DcGSK3 at 24 h significantly disrupted D. citri chitin metabolism and that the effect could be recovered at 48 h.

Identification of DEGs Involved in Chitin and Fatty Acid Metabolism
Based on the transcriptome analysis, numerous DEGs involved in chitin and fatty acid metabolism were altered in different comparable groups. In the chitin metabolism, 16 DEGs related to chitin metabolism were identified based on transcriptome data, of which 13 (81.2%) DEGs were downregulated at 24 h in the dsDcGSK3 groups compared with the dsGFP groups ( Figure 12; Table 1). Furthermore, three genes were upregulated at 24 h after the silencing of DcGSK3, including IDGF2, PAGM and HK2. It was found that 11 DEGs were upregulated and 5 DEGs showed downregulation at 48 h after the silencing of DcGSK3. These results indicated that the inhibition of DcGSK3 at 24 h significantly disrupted D. citri chitin metabolism and that the effect could be recovered at 48 h. For fatty acid metabolism, a total of 16 DEGs involved in fatty acid metabolism were identified, among which 11 DEGs were downregulated at 24 h in the dsDcGSK3 groups compared with the dsGFP groups ( Figure 13A; Table 2). Moreover, four DEGs were upregulated at 24 h, including Lipase 3, FA2H, ELOVL1 and Lipase 1 ( Figure 13A). At 48 h after silencing DcGSK3, nine DEGs were upregulated and four DEGs were downregulated. The relative expression levels of 16 DEGs associated with fatty acid were further validated by RT-qPCR ( Figure 13B).   For fatty acid metabolism, a total of 16 DEGs involved in fatty acid metabolism were identified, among which 11 DEGs were downregulated at 24 h in the dsDcGSK3 groups compared with the dsGFP groups ( Figure 13A; Table 2). Moreover, four DEGs were upregulated at 24 h, including Lipase 3, FA2H, ELOVL1 and Lipase 1 ( Figure 13A). At 48 h after silencing DcGSK3, nine DEGs were upregulated and four DEGs were downregulated. The relative expression levels of 16 DEGs associated with fatty acid were further validated by RT-qPCR ( Figure 13B).     The asterisks indicate significance differences by * (p < 0.05) and ** (p < 0.01).

Discussion
This study identified a GSK3 gene with the complete ORF sequence from the D. citri genome database. A bioinformatics analysis revealed that DcGSK3 contains a Pkinase_Tyr domain and two LC domains. Moreover, DcGSK3 comprises 13 phosphorylation sites, liking serine, threonine and tyrosine modification sites. The GSK3 enzyme plays a pivotal role in the β-catenin signaling pathway, which is responsible for the phosphorylation and degradation of β-catenin through ubiquitin proteasome [33]. The results indicated that DcGSK3 is a kinase that is involved in regulating the downstream signaling pathways. In Drosophila, a homolog of GSK3 named Shaggy regulates development by controlling Wnt/β-catenin signaling [34]. In Tribolium castaneum, the Wnt/β-catenin signaling pathway integrates the patterning and metabolism of the insect growth zone [35]. Chen et al. revealed that Wnt/β-catenin signaling regulates pupal development by upregulating c-Myc and AP-4 in Helicoverpa armigera [36]. Consequently, we speculated that DcGSK3 might regulate Wnt/β-catenin signaling to influence the growth and development of D. citri.
The relative expression level of DcGSK3 was analyzed in different tissues and developmental stages. The results indicated that DcGSK3 expression was highest in the head of D. citri, followed by the wing. In insects, the head tissue is central to the brain and is responsible for hormone secretion and chemosensory response. GSK3 is involved in the regulation of insulin. The binding of insulin to the corresponding receptor can activate the tyrosine kinase activity of the receptor, and further regulate phosphoinositol-3 kinase (PI3K), 3-phosphoinositide-dependent protein kinase (PDK1) and protein kinase B (PKB) to induce the phosphorylation of the GSK3 substrates [37]. In Drosophila, a total of eight insulin-like peptides (ILPs) have been identified, among which ILPs 2, 3 and 5 are produced by neurosecretory cells (NCSs) in the brain [38]. We also found that DcGSK3 exhibited a high expression in the wing. Intriguingly, Xu et al. reported that two insulin receptors regulate wing morphs in N. lugens [37]. Accordingly, we considered that the insulin-related peptides secreted from the D. citri brain regulate the downstream crosstalk between the insulin signal pathways and sugar metabolism pathways modulated by DcGSK3. DcGSK3 was expressed at a higher level in the egg at different development stages. By acting as a downstream regulatory switch, GSK3 controls the output of numerous signaling pathways induced by various stimuli [39]. In Drosophila, a membrane-tethered form of GSKS activates Wnt signaling, thereby regulating embryonic development [40]. The expression of GSK3 was found to be restricted to the embryonic tissue of T. castaneum, which indicates that embryonic and extra-embryonic cells display different metabolic activities [41]. In the hemipteran Rhodnius prolixus, the knockdown of GSK3 induced an impaired oogenesis and early embryogenesis, suggesting that GSK3 is essential for both processes in R. prolixus [42]. Hence, we speculated that DcGSK3 might play an important role in D. citri embryonic development.
Trehalose is the major blood sugar in insects and plays a crucial role in providing an instant energy source and responding to abiotic stresses [43]. Trehalose is synthesized by the catalysis of trehalose-6-phosphate synthase (TPS) and is degraded by the catalysis of trehalase [44]. The synthetic substrates of trehalose, uridine diphosphate glucose and phosphate-6-glucose are derived from the decomposition of glycogen [45]. Therefore, trehalose metabolism and glycogen utilization are closely related. The present study aimed to determine whether DcGSK3 plays a role in the growth and development of D. citri through an RNAi bioassay. Even though RNAi is one of the most powerful tools for functional gene research and the control of agricultural pests, there are a few potential challenges associated with the lack of effective dsRNA delivery methods [23]. Particularly for D. citri, dsRNA delivery mainly relied on topical feeding, soaking and artificial feeding [46][47][48]. However, dsRNA delivery by microinjection has not been reported in D. citri. In this study, DcGSK3 was effectively silenced at 24 h and 48 h after the injection of dsRNA. By silencing DcGSK3, the D. citri molting from a fifth-instar nymph to an adult was significantly disrupted. According to Ding et al., the knockdown of GSK3 significantly affected the molting and wing formation of N. lugens [20].
Chitin is a major structural component of the insect exoskeleton and plays an important role in molting [49]. Therefore, we considered that the abnormal molt observed after the inhibition of DcGSK3 may have resulted from a chitin metabolism block in D. citri. The pathway of insect chitin biosynthesis begins with trehalose and is catalyzed by hexokinase (HK), glucose-6-phosphate isomerase (G6PI), fructose 6-phosphate transaminase (GFAT), glucosamine-6-phosphate N-acetyltranferase (GNPNA), phosphor acetylglucosamine mutase (PAGM), UDP-N-acetylglucosamine pyrophosphorylase (UAP) and chitin synthase (CHS) [8]. We found that the silencing of DcGSK3 significantly reduced the expression levels of chitin-metabolism-related genes at 24 h and 48 h. Furthermore, membrane-bound trehalase activity was significantly reduced at 24 h after the knockdown of DcGSK3, while soluble trehalase activity was increased significantly. In our previous research, three trehalase genes were identified based on the D. citri genome database, including Tre1-1, Tre1-2 and Tre2, of which the Tre1s were related to soluble trehalase and Tre2 was associated with membrane-bound trehalase [31]. Here, we also found that the expression levels of Tre1-1, Tre1-2 and Tre2 were downregulated at 24 h and 48 h after silencing DcGSK3. These results suggest that DcGSK3 may regulate chitin metabolism by influencing D. citri Tres expression levels. Intriguingly, a transcriptome analysis revealed that the genes related to chitin metabolism were downregulated at 24 after the knockdown of DcGSK3, whereas they were upregulated at 48 h. The fifth-instar nymph of D. citri is an important developmental stage in the transition from nymph to adult [48]. In the early stage of D. citri molting, chitin synthesis and degradation are essential for forming a new cuticle. D. citri molt was significantly influenced by genes that inhibit chitin metabolism. Nonetheless, D. citri needs to increase the expression level of the genes related to chitin metabolism in the late stage of molt to produce more chitin by a regulatory feedback mechanism. Additionally, two transport genes were identified as being involved in glucose and trehalose transport, including facilitated glucose transporter membrane 6 (GLUT6) and facilitated trehalose transporter 1 (TRET1). Sugar transporters play an important role in controlling carbohydrate transport in organisms and are responsible for mediating the movement of sugars into cells [50]. The results showed that GLUT6 and TRET1 expression levels significantly decreased at 24 h following DcGSK3 silencing, indicating that DcGSK3 might indirectly regulate GLUT6 and TRET1 expression levels to influence the transport of glucose and trehalose into cells.
Free fatty acids (FFAs) are a diverse group of lipids that play an important role in maintaining homeostasis and managing cellular processes. In insects, FFAs are essential components of the cell and cell membrane, as well as sources of energy [51]. The energy reserves of insects are stored in the form of glycogen and triglycerides. Fatty acids stored as triglyceride can also be used for energy production through β-oxidation [14]. This study found that free fatty acid content was downregulated at 48 h after silencing DcGSK3. Furthermore, the expression levels of fatty-acid-synthesis-related genes were downregulated, while the genes associated with fatty acid degradation were upregulated after silencing DcGSK3. We speculated that the inhibition of DcGSK3 reduced energy production and promoted fatty acid degradation to compensate energy shortages. Based on transcriptome sequencing results, 16 DEGs involved in fatty acid metabolism were identified, of which 12 genes were downregulated at 24 h in the treatment groups (dsDcGSK3 treatment) compared with the control groups (dsGFP treatment). In addition, four genes were upregulated at 24 h after silencing DcGSK3, including lipase 3, FA2H, ELOVL1 and lipase 1. Lipases are ubiquitous enzymes in nature that play a crucial role in fat metabolism by catalyzing the hydrolysis of triacylglycerol to free fatty acids and glycerol [52]. The results showed that the expression levels of lipase1 and lipase 3 were significantly reduced after the knockdown of DcGSK3. We speculated that the inhibition of DcGSK3 significantly reduced the glucose and trehalose contents in D. citri and then lipase-related genes were activated to promote the hydrolysis of triacylglycerol. Further research is needed to determine the specific molecular mechanisms involved.

D. citri Rearing and Sample Collection
The D. citri adults were collected from Ganzhou China and were continuously reared on Murraya exotica inside steel cages (60 × 60 × 90 cm 3 ) at Gannan Normal University, China. D. citri rearing was conducted at a constant temperature (27 ± 1 • C), relative humidity (75 ± 5%) and with a 16:8 (light: dark) photoperiod. To maintain the consistent growth and development of D. citri, the females, after mating, were placed into the flourishing M. exotica. After 24 h, all the D. citri adults were removed using a portable aspirator and the eggs continued to hatch and develop into nymphs. Using a stereomicroscope, different stages of D. citri were obtained based on morphological characteristics. All the collected samples were kept at −80 • C. Each group of samples contained three biological replicates.

Identification of DcGSK3 and in Silico Analysis
The protein sequences of GSK3 from N. lugens (XP_039293499), Bombyx mori (XM_004928285) and Drosophila melanogaster (NP_476714) were downloaded from the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/, accessed on 12 January 2022) and these GSK3 protein sequences were used as queries for blasting against the D. citri genome database (https://citrusgreening.org/ftp/genomes/Diaphorina_ citri/assembly/DIACI_v2.0/, accessed on 12 January 2022). The candidate GSK3 sequences were analyzed to obtain the complete DcGSK3 cDNA sequence. A complete open reading frame of DcGSK3 was amplified by PCR. The purified product was linked to the pMD19-T vector and sequenced by a biotechnology company (Sangon Biotech, Shanghai, China).
The amino acid sequence of DcGSK3 was analyzed using DNASTAR software. The theoretical molecular weights (MWs) and isoelectric points (pIs) of DcGSK3 were calculated with ExPASy online software (http://web.expasy.org/compute_pi, accessed on 1 May 2021). The conserved domain was identified by using SMART online software (http://smart.embl-heidelberg.de/, accessed on 20 May 2021) and Illustrate for Biological Sequences (IBS) software (http://ibs.biocuckoo.org/, accessed on 1 May 2021). The multi-ple sequence alignment was performed using DNAMAN software. The SWISS-MODEL software (https://swissmodel.expasy.org/interactive, accessed on 1 January 2022) was used to predict the sense structure of the protein. MEGA 7.0 software was used for constructing the phylogenetic tree with the neighbor-joining method with 1000 replicates [53].

RNA Isolation, cDNA Synthesis and RT-qPCR Analysis
The total RNA was isolated from D. citri in different tissues (midgut, leg, head, wing, integument and fat body) and at different developmental stages (egg, 1st-, 2nd-, 3rd-, 4th-and 5th-instar nymphs and adults) using the animal tissue total RNA kit (Simgen). RNA quality and quantity were measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, New York, NY, USA). The total RNA from different samples was reverse transcribed in a 20 µL reaction system using a cDNA synthesis master mix kit according to a previous protocol [54]. All the cDNA samples were diluted to the same concentration as a template for RT-qPCR analysis. RT-qPCR reactions were conducted using a LightCycler ® 96 PCR Detection System (Roche, Basel, Switzerland). The specific sample addition procedures were performed according to a previous report [44]. The relative expression levels were analyzed using the 2 −∆∆Ct method. The constitutively expressed glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene was used as a reference gene. All primers were presented in Table S1. There were three biological and technical replicates for each sample.

Detection of Trehalase Activity, Sugar and Free Fatty Acid Content
The trehalase activity was assayed based on a previous protocol [31]. After treatment with dsDcGSK3 and dsGFP, a total of 30 5th-instar nymphs were added to a 1.5 mL centrifuge tube and then 1 mL precooled PBS (pH 7.0) was added. The mixture was homogenized on ice using a glass homogenizer and the homogenate was centrifuged at 1000× g for 20 min at 4 • C. Approximately 400 µL of supernatant was collected for trehalase activity analysis. The remaining supernatant was used to determine the trehalase, glucose and glycogen contents. The 400 µL of supernatant was centrifuged at 20,800× g for 60 min at 4 • C. The ultracentrifugal supernatant was obtained to determine the soluble trehalase activity and the pellet was suspended with PBS (pH 7.0) to measure membranebound trehalase activity. Finally, the soluble and membrane-bound trehalase activities were assayed using a glucose assay kit (Simgen, Hangzhou, China).
For trehalose measurement, the suspension was incubated with trehalase (Sigma-Aldrich) at 37 • C overnight. The free glucose calculated from the sample without an amyloglucosidase or trehalase treatment was subtracted from the glycogen or trehalose value, respectively. For the glucose measurement, 30 nymphs were homogenized in 300 µL of PBS before measuring the glucose levels with the GO reagent (Sigma-Aldrich, St. Louis, MO, USA). The protein concentration was determined using a BCA Protein Assay Reagent (Pierce Chemical Company, Rockford, IL, USA) according to the manufacturer's instructions. In order to measure glycogen, the suspension was incubated for 1 h with amyloglucosidase (catalog no. 10115, Sigma-Aldrich) at 50 • C.
Free fatty acids were determined from the dsGFP-and dsDcGSK3-treated samples according to the protocol reported by Yang et al. with minor modifications [55]. Briefly, 40 D. citri were added to a 1.5 mL centrifugation tube and 1 mL of extracting solution was added before homogenizing on ice. All the samples were centrifuged at 8000× g for 10 min at 4 • C and the supernatants were collected. The protein concentrations from different samples were determined using a BCA Protein Assay Reagent (Pierce Chemical Company, Rockford, IL). A mother solution containing 5 µmol/mL of palmitic acid was diluted to different concentrations (0.05, 0.1, 0.2, 0.4, 0.6, 0.8 and 1.0 µmol/mL) as a standard substance. A standard curve was constructed using an FFA content determination kit (Solarbio, Beijing, China) with the above standard substance at an absorbance of 550 nm. Finally, the contents of FFA from different samples were analyzed based on three biological replicates.

dsRNA Synthesis and Microinjection
The RNAi-based silencing of DcGSK3 gene expression was carried out to analyze the functions of DcGSK3 on D. citri. The dsDcGSK3 and dsGFP were synthesized using the T7 RioMAXTM Express RNAi System (Promega, Madison, WI, USA) according to the manufacturer's instructions. The specific dsDcGSK3 and dsGFP primers containing a T7 promoter sequence were designed and are listed in Table S1. The synthetic dsDcGSK3 was diluted to 600 ng/mL of working solution using RNase-free water and a moderate 0.1% red food dye was added as an indicator. For the RNAi experiment, 300 fifth-instar D. citri nymphs were divided into three groups. A sticker was attached on the dorsum to maintain the ventrum in an upright position for injection ( Figure 4A). A total of 20 nL of dsDcGSK3 was injected into the fifth-instar D. citri nymphs and then transferred onto the fresh M. exotica seedlings. An equal amount of dsGFP was injected as a control. All the experiments contained three biological replicates. All the alive D. citri were collected at 24 h and 48 h after dsRNA treatment. The efficiency of silencing was determined using an RT-qPCR as described above.

cDNA Library Preparation and Illumina Sequencing
Transcriptome sequencing was carried out at Novogene Biological Information Technology Co., Ltd. (Tianjin, China). Approximately 100 D. citri were collected from each treatment group (treated with dsDcGSK3) and control group (treated with dsGFP) at 24 h and 48 h following the injection of dsRNA. All samples contained three biological replicates. The concentration and purity of the RNA were measured by a Qubit RNA Assay Kit (Life Technologies, CA, USA). For cDNA library construction, a total of 1 µg of RNA was used for constructing a cDNA library with the TruSeq RNA Sample Preparation Kit v2 (Illumina, San Diego, CA, USA). The prepared cDNA library was sequenced by the Illumina HiSeq platform, generating 150 bp paired-end reads. The clean reads were obtained by removing reads that contained the adapter from raw data. Furthermore, the Q20, the Q30 and the GC content of the clean data were calculated.

Transcriptome Analysis after Silencing of DcGSK3
The transcriptome data were mapped to a D. citri reference genome (https://www. citrusgreening.org/, accessed on 12 January 2022) using a Hisat2 (version 2.0.5; https: //anaconda.org/biobuilds/hisat2, accessed on 12 January 2022) aligner. This generated a database of splice junctions based on the gene model annotation file. The expression levels of the genes were calculated using reads per kilobase per million mapped reads (RPKM). Differential expression analyses of the genes between the dsGFP and dsDcGSK3 groups at 24 h and 48 h were performed using the DESeq2 R package. The p-values were adjusted using the Benjamini-Hochberg method to control for the false-discovery rate. A corrected p-value of 0.05 and an absolute |log2 (fold change)| of 0 was set as the thresholds for significantly different gene expression. The hierarchical cluster analysis of DEGs was conducted using Genesis software. The topGO R package, which implements the Gene ontology (GO) terms, was used for the enrichment analysis of length-corrected DEGs. A Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis for the DEGs was performed using KOBAS. A p-value of <0.01 was set as the threshold.

Statistical Analysis
In this study, all data were analyzed using a one-way analysis of variance and Turkey's test. p-values of less than 0.05 and 0.01 were defined as significant and extremely significant, respectively.

Conclusions
In summary, we identified a GSK3 gene with the complete ORF from the D. citri genome database and named it DcGSK3. DcGSK3 showed a high expression in the head and egg. Furthermore, the inhibition of DcGSK3 by RNAi led to an abnormal phenotype and downregulated the synthesis of glucose, trehalose, glycogen and free fatty acid contents. Moreover, an RT-qPCR and comparative transcriptome sequencing analysis revealed that the silencing of DcGSK3 significantly suppressed the expression levels of the genes involved in chitin and fatty acid metabolism. In addition to providing a foundation for further research on DcGSK3, these results also provide a novel target for controlling D. citri.