Genome-Wide Analysis and Hormone Regulation of Chitin Deacetylases in Silkworm

Chitin deacetylases (CDAs) are a group of enzymes involved in chitin metabolism in insects; they play a critical role in molting, pupation, and the modification of chitin. In this study, we identified several CDAs in the silkworm, Bombyx mori (BmCDA), and investigated the effect of various hormones on their expression in B. mori larvae and embryo cell lines (BmE). Eight genes encoding BmCDAs were identified in the silkworm genome. They showed different expression patterns in different tissues, and were classified into three types based on where they were expressed: the exoskeleton, digestive organs, and genital organs. Moreover, we found that some BmCDAs showed upregulated expression during the molting period, especially during the fourth molting period in larvae. We also verified that the expression of BmCDA1–6 was upregulated by treatment with 20-hydroxyecdysone not only in larvae, but also in BmE cells. Interestingly, juvenile hormone analog treatment also upregulated the expression of some BmCDAs. The overexpression of several transcription factors revealed that the POU transcription factor POUM2 may play a major role in the regulation of BmCDA expression. Finally, the silencing of BmCDA1 and BmCDA2 did not lead to abnormal phenotypes or death, but may have led to delays in silkworm pupation. These results provide important information about lepidopteran insects in terms of chitin deacetylases and the regulation of their expression.


Introduction
Chitin is a polysaccharide that is formed by the polymerization of N-acetylglucosamine. It is widely found in the shells of crustaceans, skin of insects, and cytoderm of fungi [1], as well as in many green algae [2]. In animals, chitin mainly functions to support the exoskeleton of the body and protect the organism from the environment outside. Chitosan is obtained by the deacetylation of chitin, and has wide applications in medicine, food, chemicals, cosmetics, biochemistry, and biomedical engineering [3]. In insects, growth and development is majorly associated with the biosynthesis and modification of chitin [4]. The initial substrate in the insect chitin biosynthesis pathway is trehalose-this pathway leads to the synthesis of chitin precursors by chitin synthase [5]. Chitin precursors are partially deacetylated to form chitin, and the deacetylation of chitin is necessary for the structure, permeability, and mechanical properties of the cuticle that makes it more soft and soluble in most insects [6][7][8][9][10].
Chitin deacetylases (CDAs) are a kind of metalloenzyme that belong to the carbohydrate esterase family 4 (CE-4), and chitin deacetylation is catalyzed by CDAs. CDAs were first discovered in the BmCDA3a and BmCDA3b, respectively (Table 1). The identified genes were different in terms of their sequences, lengths, and number of exons ( Figure 1A).

Sequence Analysis of Eight Silkworm CDAs
Based on a previous study of domain architecture, the eight silkworm CDAs were divided into five groups (I--V) [14]. BmCDA1, BmCDA2, and BmCDA3, which were classified as group I and II CDAs, have a CDA catalytic domain, as well as a chitin-binding domain type 2 (ChtBD2) and a lowdensity lipoprotein receptor (LDLa) binding domain ( Figure 1B). BmCDA4 and BmCDA5, which were

Sequence Analysis of Eight Silkworm CDAs
Based on a previous study of domain architecture, the eight silkworm CDAs were divided into five groups (I-V) [14]. BmCDA1, BmCDA2, and BmCDA3, which were classified as group I and II CDAs, have a CDA catalytic domain, as well as a chitin-binding domain type 2 (ChtBD2) and a low-density lipoprotein receptor (LDLa) binding domain ( Figure 1B). BmCDA4 and BmCDA5, which were classified as group III and IV CDAs, respectively, had CDA and ChtBD2 domains, but lacked the LDLa binding domain. Finally, BmCDA6, BmCDA7, and BmCDA8, which were classified as group V CDAs, contained only the CDA domain ( Figure 1B).

Phylogenetic Analysis of CDAs
The predicted CDA catalytic domain of putative CDA proteins, which belong to the CE-4 family, in insects, fungi, bacteria, and other invertebrates, were analyzed by ClustalX [30]. Multiple sequence alignment analysis showed that all silkworm CDAs contained five conserved motifs ( Figure 2A); this was found in most other organisms [31,32]. In addition, three motifs, namely motif 1 (T[F/Y]DD), motif 2 (H[S/T]xxH) and motif 3 (RxPY), were conserved across various species, whereas motif 4 and motif 5 showed major differences depending on the species. Interestingly, we identified a highly conserved area, with a predicted amino acid sequence of threonine-phenylalanine-phenylalanine-valine (TFFV), between motif 1 and motif 2.
To further analyze the homology of the enzymes in these different species, a maximum likelihood (ML) evolutionary tree was constructed ( Figure 2B). The ML evolutionary tree was divided into three branches comprising arthropods, fungi, and bacteria ( Figure 2B). All eight B. mori CDA proteins were within the arthropod clade. BmCDA1 and BmCDA2 were classified into one subclade, and were homologous with the serpentine and vermiform homologues of Drosophila CDAs (Dmserp and Dmverm), respectively ( Figure 2B).BmCDA3, BmCDA4, and BmCDA5 were placed in two separate subclades, distinct from BmCDA1 and BmCDA2 ( Figure 2B). BmCDA6, BmCDA7, and BmCDA8 were classified into another subclade that was clearly distinct from the other silkworm CDA proteins ( Figure 2B). Furthermore, fungi and bacteria were classified into a separate clade; insects were found to be more relevant to the evolution of fungi than to that of bacteria ( Figure 2B).

Tissue and Developmental Expression Patterns of BmCDAs
The tissue expression patterns and developmental expression patterns of eight putative BmCDAs were determined by RT-qPCR. The results of tissue expression pattern analysis showed that BmCDA1 and BmCDA2, both of which are group I CDAs, were mainly expressed in the epidermis and head, suggesting that they have functions in the integument ( Figure 3A,B). BmCDA3, BmCDA4, and BmCDA5, which are group II, III, and IV CDAs, respectively, were irregularly expressed in the genital organs or Malpighian tube ( Figure 3C-E). BmCDA6, BmCDA7, and BmCDA8, which are group V CDAs, were highly expressed in the midgut and seemed to be gut-specific proteins ( Figure 3F-H

Tissue and Developmental Expression Patterns of BmCDAs
The tissue expression patterns and developmental expression patterns of eight putative BmCDAs were determined by RT-qPCR. The results of tissue expression pattern analysis showed that BmCDA1 and BmCDA2, both of which are group I CDAs, were mainly expressed in the epidermis and head, suggesting that they have functions in the integument ( Figure 3A,B). BmCDA3, BmCDA4, and BmCDA5, which are group II, III, and IV CDAs, respectively, were irregularly expressed in the genital organs or Malpighian tube ( Figure 3C-E). BmCDA6, BmCDA7, and BmCDA8, which are group V CDAs, were highly expressed in the midgut and seemed to be gut-specific proteins ( Figure 3F-H). Total RNA was extracted from different tissues including the head, epidermis, midgut, fat body, ovary, testis, silk gland, and Malpighian tube, which were dissected from larvae at day 3 of the fifth instar.
The analysis of developmental expression patterns showed that the expression levels of BmCDA1-BmCDA6 were significantly increased in the molting stage, and then declined rapidly after molting ( Figure 4A-F). Interestingly, the expression of most of them was substantially increased during the fourth larval molting stage, rather than during pupation; this was observed for BmCDA2-BmCDA6 ( Figure 4B-F). Only BmCDA1 was highly expressed during pupation, rather than during larval molting ( Figure 4A). BmCDA7 and BmCDA8, the gut-specific group V CDAs, were only expressed during the feeding period ( Figure 4G,H). Total RNA was extracted from different tissues including the head, epidermis, midgut, fat body, ovary, testis, silk gland, and Malpighian tube, which were dissected from larvae at day 3 of the fifth instar.
The analysis of developmental expression patterns showed that the expression levels of BmCDA1-BmCDA6 were significantly increased in the molting stage, and then declined rapidly after molting ( Figure 4A-F). Interestingly, the expression of most of them was substantially increased during the fourth larval molting stage, rather than during pupation; this was observed for BmCDA2-BmCDA6 ( Figure 4B-F). Only BmCDA1 was highly expressed during pupation, rather than during larval molting ( Figure 4A). BmCDA7 and BmCDA8, the gut-specific group V CDAs, were only expressed during the feeding period ( Figure 4G,H).

Influence of 20E on Expression of BmCDAs
In order to investigate whether the expression of BmCDAs was affected by 20E exposure, silkworm larvae treated with 20E were subjected to RT-qPCR. The mRNA expression levels of BmCDA1-BmCDA6 were significantly increased after 20E treatment compared to the control group ( Figure 5A-F). However, the expression of BmCDA7 and BmCDA8 was decreased after 20E treatment ( Figure 5G-H). Furthermore, BmCDA8 expression was undetectable after treatment with 20E ( Figure 5H). Next, BmE cells were treated with 20E. Consistent with our results in silkworm larvae, the mRNA expression levels of BmCDA1-BmCDA6 were increased after 20E treatment compared to the control group ( Figure S1). The expression of BmCDA7 was undetectable, and the expression of BmCDA8 was significantly decreased after 20E treatment in BmE cells ( Figure S1). Total RNA was extracted from whole insects using day 1 of the fourth instar larvae (L4D1), day 3 of the fourth instar larvae (L4D3), molting of fourth instar larvae (4M), newly molted larvae (L5D0), day 1 of fifth instar larvae (L5D1), day 3 of fifth instar larvae (L5D3), day 5 of fifth instar larvae (L5D5), day 7 of fifth instar larvae (L5D7), day 1 of wandering (W1), day 2 of wandering (W2) prepupa (PP), day 1 of pupa (P1), day 3 of pupa (P3), day 5 of pupa (P5), day 7 of pupa (P1), and day 1 of adult (A1).

Influence of 20E on Expression of BmCDAs
In order to investigate whether the expression of BmCDAs was affected by 20E exposure, silkworm larvae treated with 20E were subjected to RT-qPCR. The mRNA expression levels of BmCDA1-BmCDA6 were significantly increased after 20E treatment compared to the control group ( Figure 5A-F). However, the expression of BmCDA7 and BmCDA8 was decreased after 20E treatment ( Figure 5G-H). Furthermore, BmCDA8 expression was undetectable after treatment with 20E ( Figure  5H). Next, BmE cells were treated with 20E. Consistent with our results in silkworm larvae, the mRNA expression levels of BmCDA1-BmCDA6 were increased after 20E treatment compared to the control group ( Figure S1). The expression of BmCDA7 was undetectable, and the expression of BmCDA8 was significantly decreased after 20E treatment in BmE cells ( Figure S1).

Influence of JHA on Expression of BmCDAs
In addition to 20E, JH was also used to treat silkworms to analyze its effect on the expression of BmCDAs. Most BmCDAs, namely BmCDA1-BmCDA5, BmCDA7, and BmCDA8 were significantly upregulated after JHA treatment compared to the control group (Figure 6A-E,G,H). BmE cell lines

Influence of JHA on Expression of BmCDAs
In addition to 20E, JH was also used to treat silkworms to analyze its effect on the expression of BmCDAs. Most BmCDAs, namely BmCDA1-BmCDA5, BmCDA7, and BmCDA8 were significantly upregulated after JHA treatment compared to the control group (Figure 6A-E,G,H). BmE cell lines were also treated with JHA. All BmCDAs under investigation were upregulated in cells after JHA treatment compared to the control group ( Figure S2). Of them, the upregulation of BmCDA7 and BmCDA8 was the most significant, with an increase of approximately fourfold. Therefore, it can be concluded that JH plays a crucial role in the regulation of the expression of BmCDAs. Figure 5. Effect of 20-hydroxyecdysone (20E) on the expression of eight Bombyx mori chitin deacetylases (BmCDAs). The mRNA expression levels of BmCDAs were detected using RT-qPCR after injection of 20E for 24 h. Dimethyl sulfoxide (DMSO) was used in the control group. *p < 0.05, **p < 0.01, ***p < 0.001 vs. control.

Influence of JHA on Expression of BmCDAs
In addition to 20E, JH was also used to treat silkworms to analyze its effect on the expression of BmCDAs. Most BmCDAs, namely BmCDA1-BmCDA5, BmCDA7, and BmCDA8 were significantly upregulated after JHA treatment compared to the control group (Figure 6A-E,G,H). BmE cell lines were also treated with JHA. All BmCDAs under investigation were upregulated in cells after JHA treatment compared to the control group ( Figure S2). Of them, the upregulation of BmCDA7 and BmCDA8 was the most significant, with an increase of approximately fourfold. Therefore, it can be concluded that JH plays a crucial role in the regulation of the expression of BmCDAs.

Regulation of Expression of BmCDAs by Overexpression of Transcription Factors
To study the mechanisms of the regulation of BmCDAs, six sequences, approximately 2000 bp upstream of each BmCDA gene, were analyzed to identify binding sites for several important transcription factors. Many potential transcription factor binding sites were predicted upstream of these genes, including binding sites for broad-complex zinc finger 2 (BRC-Z2), POU transcription factor (POUM2), and Ets transcription factor (E74) ( Figure 7A). These three factors were overexpressed by transfecting the relevant vectors into BmE cells. The transcription factors were

Regulation of Expression of BmCDAs by Overexpression of Transcription Factors
To study the mechanisms of the regulation of BmCDAs, six sequences, approximately 2000 bp upstream of each BmCDA gene, were analyzed to identify binding sites for several important transcription factors. Many potential transcription factor binding sites were predicted upstream of these genes, including binding sites for broad-complex zinc finger 2 (BRC-Z2), POU transcription factor (POUM2), and Ets transcription factor (E74) ( Figure 7A). These three factors were overexpressed by transfecting the relevant vectors into BmE cells. The transcription factors were successfully overexpressed in BmE cells, as shown using RT-qPCR ( Figure S3). The expression levels of BmCDA1, BmCDA3, BmCDA5, and BmCDA6 were found to be upregulated after overexpressing BRC-Z2 in cells ( Figure 7B). Furthermore, the expression levels of BmCDA2 and BmCDA5 were increased by the overexpressing of E74 in BmE cells ( Figure 7C). In addition, the expression levels of BmCDA1-BmCDA6, which were upregulated by 20E treatment (Figure 5A-F), were also upregulated by the overexpressing of POUM2 in BmE cells ( Figure 7D). These results show that the three transcription factors investigated, especially POUM2, potentially played a key role in regulating the expression of BmCDAs mediated by 20E. Figure 7B). Furthermore, the expression levels of BmCDA2 and BmCDA5 were increased by the overexpressing of E74 in BmE cells ( Figure 7C). In addition, the expression levels of BmCDA1-BmCDA6, which were upregulated by 20E treatment (Figure 5A-F), were also upregulated by the overexpressing of POUM2 in BmE cells ( Figure 7D). These results show that the three transcription factors investigated, especially POUM2, potentially played a key role in regulating the expression of BmCDAs mediated by 20E.

RNAi of BmCDA1 and BmCDA2
BmCDA1 and BmCDA2 were the most highly expressed CDAs in the exoskeletal system. To demonstrate their roles in the molting process of silkworms, dsRNA sequences of BmCDA1 and BmCDA2 were injected into larvae. The expression levels of BmCDA1 and BmCDA2 were confirmed to be decreased after RNAi ( Figure 8D). We then observed the process of metamorphosis from larvae to pupae. The results did not reveal any abnormal phenotypes during pupation, and the structure of the new epidermal tissues was complete and regular. However, the pupation time was delayed after BmCDA1 and BmCDA2 RNAi treatment compared to that in the control group ( Figure 8B,C).

RNAi of BmCDA1 and BmCDA2
BmCDA1 and BmCDA2 were the most highly expressed CDAs in the exoskeletal system. To demonstrate their roles in the molting process of silkworms, dsRNA sequences of BmCDA1 and BmCDA2 were injected into larvae. The expression levels of BmCDA1 and BmCDA2 were confirmed to be decreased after RNAi ( Figure 8D). We then observed the process of metamorphosis from larvae to pupae. The results did not reveal any abnormal phenotypes during pupation, and the structure of the new epidermal tissues was complete and regular. However, the pupation time was delayed after BmCDA1 and BmCDA2 RNAi treatment compared to that in the control group ( Figure 8B,C).

Discussion
CDAs have been widely studied in insects by genome-wide analysis, including six CDAs from D. melanogaster; five each from Aedes aegypti, Anopheles gambiae, and Culex quinquefasciatus; nine from T. castaneum; CDAs from Apis mellifera; and more from a variety of other insects [18]. Alternatively spliced transcripts have also been found for the CDAs of insects [8,17,20]. We discovered two alternatively spliced transcripts for both BmCDA2 and BmCDA3, and other alternative splicing forms can potentially be found in other developmental stages or other tissues. In a previous study, different phenotypes were observed after downregulating the two alternatively spliced transcripts of TcCDA2 [20]. These alternative splice forms in silkworms may also play various roles in different processes and pathways. In addition to those studied in insects, several microbial CDAs and chitin oligosaccharide deacetylases have been widely studied in fungi and bacteria since 1993 [34,35]. Chitin deacetylation is required during spore wall and cytoderm formation in fungi [36,37]. In the marine bacterium Vibrio parahaemolyticus, chitin oligosaccharide deacetylase interacts with chitinase to promote the formation of beta-D-N-acetylglucosaminyl-(1,4)-D-glucosamine (GlcNAc-GlcN) [38]. Fungi, bacteria, and arthropods share a polysaccharide deacetylase catalytic domain, even though they belong to three distinct branches of the evolutionary tree. A previous study has speculated that chitinase was originally obtained from bacteria or baculoviruses [39]. However, we did not find evidence of genetic relatedness or horizontal gene transfer for CDAs between bacteria, fungi, and insects.
Based on our sequence analysis, CDAs from fungi, bacteria, and insects share three conserved motifs: motif 1 (TFDD), motif 2 (H[S/T]xxH), and motif 3 (RxPY). In contrast, motif 4 and motif 5 showed major differences depending on the species. In terms of the structure of fungal chitin deacetylase, two aspartic acid residues in motif 1 are considered to interact with zinc, cobalt, and acetate; two histidine residues in motif 2 are also thought to bind to the metal ions. Motif 3 forms one side of the active site groove and has multiple functions; motif 4 forms the other side of the active site groove and tryptophan is thought to be the most critical residue for acetate binding. Motif 5 forms a hydrophobic pocket with a leucine and a histidine residue [31,32]. Interestingly, we found a highly conserved region between motif 1 and motif 2, containing the amino acid sequence TFFV. Based on our observations of the structure of CDA, we found that TFFV was located at the center of the protein. Therefore, we concluded that the conserved TFFV motif may stabilize the protein structure of CDA.
CDA is widely distributed across various insect tissues, including the epidermis, head, trachea, midgut, and Malpighian tube. In silkworms, BmCDAs can be divided into three categories according to their tissue expression patterns. The first group, which includes BmCDA1 and BmCDA2, is expressed in the exoskeleton, while the second group is highly expressed in the digestive organ, and includes BmCDA6, BmCDA7, and BmCDA8. The other group is expressed in other internal organs such as the gonads or Malpighian tube. Interestingly, the classification of BmCDAs based on their tissue expression patterns was consistent with their distribution on the evolutionary tree. We also found similar expression patterns in other insects such as T. castaneum and N. lugens [18]. It was suggested that gene differentiation between groups occurs at an earlier stage, while differentiation of genes within a group occurs at a later stage.
Evolutionarily, BmCDA1 and BmCDA2, which belong to group I CDAs, were highly expressed in the exoskeletal system. In Coleoptera, Hemiptera, and Lepidoptera, interference of CDAs homologous with BmCDA1 or BmCDA2 led to shedding failure in old cuticles and eventually to high mortality [17,18,20]. In contrast, in B. mori, interference of BmCDA1 and BmCDA2 led to a delayed pupation time. We speculated that other BmCDA proteins may have compensated for the reduction in BmCDA1 and BmCDA2 activity.
Silkworms require large quantities of mulberry leaves during the fifth larval stage; the leaves are digested in the midgut. BmCDA6, BmCDA7, and BmCDA8 were highly expressed in the midgut, and their expression was only detected during the feeding period. Previous studies have shown that BmCDA7 should protect silkworms from microbial infections from mulberry leaves by modifying and changing the permeability of the PM [23]. We speculated that other BmCDAs expressed in the midgut potentially modified and maintained midgut properties to stabilize the metabolism during digestion in the fifth larval stage.
As a holometabolous insect, the silkworm has two different physiological stages: molting and metamorphosis. In this study, most CDAs were upregulated during molting and downregulated after ecdysis, particularly BmCDA1-BmCDA6. Interestingly, most BmCDAs showed the highest expression in the fourth molting stage of the larval-pupal metamorphosis period. The distinction between the larval molting stage and pupal metamorphosis was mainly induced by JH. Our analysis of the effects of JHA showed that almost all BmCDAs were upregulated in both tissues and cells, indicating that increased BmCDA expression was induced by both 20E and JH in the fourth larvae molting period.
It has been reported that E74 and BRC are the primary response genes induced by 20E treatment, due to the EcR/USP complex [40]. BRC-Z4 is an upstream regulatory factor of POUM2, that controls the pupal-specific expression of BmWCP4 [41]. POUM2 may act together with BRC-Z2 to regulate vitellogenin gene expression in the silkworm [42]. In this study, we found that some BmCDAs were upregulated by BRC-Z2 or E74, but all BmCDAs that were activated by 20E showed increased expression after POUM2 was overexpressed. Therefore, POUM2 appears to play a crucial role in the regulation of BmCDAs induced by 20E. Moreover, we found that the expression of POUM2 was significantly increased after the overexpression of E74 and BRC-Z2 (data not shown). However, the extent to which POUM2 regulated the expression of BmCDAs as well as its relationship with other important regulators require future investigation.
In conclusion, we identified eight CDA genes in the silkworm in this study. The regulation of these genes by both 20E and JH was studied extensively in the B. mori CDA family. Moreover, we examined the impact of multiple transcription factors on silkworm CDAs. The results suggested that the transcription factor POUM2 played the most important role in regulating BmCDA expression. These results improve our understanding of insect CDAs, and provide a background for further studies on chitosan for industrial, agricultural, and pharmaceutical applications.

Experimental Insects and in Vitro Culture of B. Mori Embryo (BmE) Cells
B. mori (Dazao) individuals were provided by the College of Biotechnology at Southwest University (Chongqing, China). Silkworm larvae were reared on fresh mulberry leaves at 25 • C. Larvae at day 3 of the fifth instar (5L3D) were dissected to obtain various tissues for expression analysis, including the head, epidermis, midgut, fat body, ovary, testis, and Malpighian tubule. Larvae between day 1 of the fourth instar (4L1D) and day 1 of adulthood (A1) were ground using a mortar and pestle for developmental expression analysis. Samples were stored at -80 • C for subsequent use. BmE cells were cultured in Grace's insect cell culture Medium, Gibco (Waltham, MA, USA) supplemented with 10% (v/v) FBS, Bio Basic Inc. (Toronto, Ontario, Canada).

Identification and Analysis of Silkworm CDA Gene Family
B. mori chitin deacetylases (BmCDAs) were identified using the NCBI BLAST [43], SilkDB [44], and KAIKO databases [45] based on reported chitin deacetylases in GenBank, including NlCDA1, NlCDA2, NlCDA3, and NlCDA4 from the brown planthopper [18]. The molecular mass and isoelectric points of the genes were predicted using ExPASy [46]. Conserved domains in the protein sequences were identified using NCBI's conserved domain database [47]. The chromosome position and exon or intron organization of each BmCDA was predicted based on the genome-wide model using a specialized BLAST tool in the KAIKO database.
Protein sequences containing the chitin deacetylase catalytic domain in fungi, bacteria, shellfish, and other insects were obtained from NCBI. Next, multiple sequence alignment analyses for the different homologues mentioned above were constructed using ClustalX and GENEDOC [30,48,49]. MEGA6 software [50] was used for phylogenetic analyses, and a phylogenetic tree was constructed by the maximum likelihood method with 1000 bootstrap tests [51].

Developmental and Tissue Expression Profiles of BmCDA Genes
Total RNA was extracted using TRIzol reagent, Invitrogen (Carlsbad, CA, USA) using tissues from 5L3D larvae, as well as from silkworms between 4L1D and adulthood. cDNA synthesis was carried out using the two-step reaction synthesis system, comprising denatured total RNA and reverse transcription using M-MLV Reverse Transcriptase, Promega (Madison, WI, USA) with 5 µg total RNA for each reaction. Real-time quantitative PCR (RT-qPCR) was performed to analyze the tissue expression and developmental expression patterns of BmCDAs. The reaction was performed in a 20-µL reaction volume containing 2 µL 200 ng/µL diluted cDNA, 400 nM each primer, and 10 µL NovoStart SYBR qPCR SuperMix Plus, Novoprotein Scientific Inc. (Shanghai, China) in the real-time PCR system, analytikjena qTOWER 2.2, Analytik Jena (Beijing, China) under the following reaction conditions; 95 • C for 30 s, followed by 40 cycles of 95 • C for 3 s, and 60 • C for 30 s. The gene-specific primers are listed in Table S1 and the expression levels of the target genes were normalized against sw22934. 4.4. Influence of 20E and Juvenile Hormone Analog (JHA) on the Expression of BmCDA Genes 20E (Sigma Aldrich, St. Louis, MO, USA) was dissolved in dimethyl sulfoxide (DMSO) at a working concentration of 1 µg/µL. Next, 10 µg of 20E solution was injected into the spiracles of 5L3D larvae; pure DMSO was injected into the control group. For JH, juvenile hormone analog (JHA), methoprene, Sigma-Aldrich (Shanghai, China) was used instead of JH in larvae. A total of 10 µg JHA solution, dissolved in dimethyl ketone (DMK), was injected into silkworms, as described for 20E treatment. The same volume of DMK was injected into the control group. After 24 h, the cuticles and midguts of larvae were collected for RNA isolation. Reverse transcription and RT-qPCR analysis of BmCDA expression were carried out as described in Section 4.3.
For experiments on BmE cells, a 3-µL solution containing 3 µg 20E or JHA was added to BmE cell lines in a six-well Cell Culture Cluster, Corning Inc. (Corning, NY, USA). After 24 h, cells were collected and total RNA was isolated and reverse transcribed, followed by RT-qPCR to detect the expression of BmCDAs at a cellular level as described in Section 4.3.

Overexpression of Transcription Factors Targeting 20E
The promoter sequences (2000 bp) of six BmCDA genes were obtained from the SilkDB database [44]. Potential cis-regulatory elements (CREs) were also identified by MatInspector within the six gene promoters [33]. Finally, three elements involved in regulation by ecdysone were selected