Expression Analysis of mRNA Decay of Maternal Genes during Bombyx mori Maternal-to-Zygotic Transition

Maternal genes play an important role in the early embryonic development of the silkworm. Early embryonic development without new transcription depends on maternal components stored in the egg during oocyte maturation. The maternal-to-zygotic transition (MZT) is a tightly regulated process that includes maternal mRNAs elimination and zygotic transcription initiation. This process has been extensively studied within model species. Each model organism has a unique pattern of maternal transcriptional clearance classes in MZT. In this study, we identified 66 maternal genes through bioinformatics analysis and expression analysis in the eggs of silkworm virgin moths (Bombyx mori). All 66 maternal genes were expressed in vitellogenesis in day eight female pupae. During MZT, the degradation of maternal gene mRNAs could be divided into three clusters. We found that eight maternal genes of cluster 1 remained stable from 0 to 3.0 h, 17 maternal genes of cluster 2 were significantly decayed from 0.5 to 1.0 h and 41 maternal genes of cluster 3 were significantly decayed after 1.5 h. Therefore, the initial time-point of degradation of cluster 2 was earlier than that of cluster 3. The maternal gene mRNAs decay of clusters 2 and 3 is first initiated by maternal degradation activity. Our study expands upon the identification of silkworm maternal genes and provides a perspective for further research of the embryo development in Bombyx mori.


Introduction
The transition from the oocyte depending on maternally supplied RNA and protein complements the commencement of zygotic transcription is a key process in the earliest stages of early embryonic development [1][2][3]. Early embryonic development is maternally regulated. Maternal mRNAs and proteins stored in oocytes are activated to initiate and regulate embryonic development. Following the period of maternal transcriptional silence, the embryonic zygote's own genome starts transcription and plays a role in the development of embryos [4]. The transition from the maternal to the zygotic genome is a key process in the final transformation of the zygotic regulation of individual development [2]. Following the maternal-to-zygotic transition (MZT) period, the maternal control of development begins to decline and maternal mRNAs begin to degrade [5,6]. Therefore, later developmental control to identify the maternal genes. In this study, the expression of the 66 successfully identified maternal genes was analyzed in the developing oocytes from day eight female pupae, and during the MZT period in silkworm.

Identification of Potential Maternal Genes
In our previous study, we obtained 76 potential maternal genes in the B. mori genome [48]. In this study, we blasted the NCBI database and the newly assembled silkBase by the sequence of each gene that was obtained from the silkworm database (SilkDB) [49]. BGIBMGA012517 and BGIBMGA012518 are orthologous genes to MAMO in Drosophila melanogaster [48]. The sequences of BGIBMGA012517 and BGIBMGA012518 were found to be part of the KWMTBOMO05086 gene that was annotated in silkBase (Table 1). Similarly, BGIBMGA002518and 002519 were found to be part of KWMTBOMO005319; BGIBMGA000972, 000973, and 000974 were part of KWMTBOMO007913; BGIBMGA004415 and 004416 were part of XM_012695102; and BGIBMGA013473 and 013474 were part of XM_012690736. The BLAST results of BGIBMGA007314 and BGIBMGA001094 were very poor in the NCBI database and silkBase. Therefore, 68 preliminary potential maternal genes were identified in the B. mori genome. The mRNAs of maternal genes are produced by the females and stored in embryos [2,6]. Thus, undetectable expression in the embryo can be considered a non-maternal gene. The results of the transcriptional analysis of the 68 potential maternal genes in eggs of virgin moths by reverse transcription-PCR (RT-PCR) showed that for 66 genes, transcriptional signals were detected, whereas two had no transcriptional signals (Figure 1), BGIBMGA003296 and BGIBMGA002069 had no transcriptional signals ( Figure 1). The specific primers for each gene were used in RT-PCR, as shown in Table S1. We finally identified 66 maternal genes in the silkworm genome, and information, including amino-acid length, chromosomal distribution, signal peptide, and gene name, was collected for each (Table 1).

Tissue Expression Patterns on Day 3 of the Fifth Instar
The silkworm feeds and grows quickly in the fifth larval period. Day 3 of the fifth instar is typical for larval development with more active biological processes [50]. Therefore, studying this time point will enrich the expression patterns and help with further understanding of the functions of maternal genes in different developmental stages. The microarray data of 10 silkworm tissues on day 3 of the fifth instar were downloaded from the SilkMDB [50]. The probes of SPE, BAEE and Pabn2 were not found in SilkMDB from the attached BLAST search ( Table 1). The microarray data of the other 63 maternal genes are provided in Table S2. The expression patterns of these 63 maternal genes are listed as found in various tissues and both sexes of silkworm in Figure 2. The expressed genes are defined as previously described [51]. Most of the maternal genes usually showed very low expression levels overall in the tissues and sex. The expression levels of sw10899 (aub) and sw14777 (me31B) were higher in the ovary and testis than in other tissues overall, sw19434 (Nelf-E) was only higher in the testis. The expression level of sw20327 (proPPAE) was higher in the testis, head, epidermis, and hemocyte, and sw13482 (Th) was higher in the head and epidermis. The expression levels of sw22934 (Eif-4a), sw12663 (eIF4AIII), sw1118 (Bin1), and sw21871 (Sod2) were higher overall. Most maternal genes showed low expression levels in multiple silkworm larval tissues on day 3 of the fifth instar. This is contrary to the abundant expression in the eggs of virgin moth ( Figure 1). "-" indicates that no signal peptide was predicted and no probe number was found.

Figure 2.
Tissue expression profile of maternal genes in larvae on day 3 of the silkworm fifth instar. The columns represent ten different tissues with both sexes: Testis, ovary, head, epidermis, fat body, midgut, hemocyte, Malpighian tubule, anterior/median silk gland (A/MSG), posterior silk gland (PSG), female (F), and male (M). Gene expression levels are represented by red (higher expression) and blue (lower expression) boxes.

Expression Analysis in Developing Oocytes in Day 8 Pupae
The silkworm has a pair of ovaries, each of which is composed of four ovarioles. The developmental stage of each ovariole is opportune, being found in vitellogenesis, choriogenesis, and mature eggs from day 8 pupae [43,44]. A large amount of yolk proteins and no chorion proteins exist in oocytes during vitellogenesis. Chorion proteins appear just after vitellogenesis and continue throughout the whole of choriogenesis and until the formation of the eggshell of mature eggs

Expression Analysis in Developing Oocytes in Day 8 Pupae
The silkworm has a pair of ovaries, each of which is composed of four ovarioles. The developmental stage of each ovariole is opportune, being found in vitellogenesis, choriogenesis, and mature eggs from day 8 pupae [43,44]. A large amount of yolk proteins and no chorion proteins exist in oocytes during vitellogenesis. Chorion proteins appear just after vitellogenesis and continue throughout the whole of choriogenesis and until the formation of the eggshell of mature eggs [43,44,[52][53][54]. The expression of the 68 potential maternal genes in the vitellogenesis, choriogenesis, and mature eggs in day 8 pupae was analyzed by RT-PCR. The result showed that for 66 genes, transcriptional signals were detected, whereas BGIBMGA003296 and BGIBMGA002069 also had no transcriptional signals (Figure 3). Me31B and the other 31 genes (in total 32) showed consistent expression levels in the vitellogenesis, choriogenesis, and mature eggs in day 8 pupae (Figure 3). Hip14 (ZDHHC17) and the other 33 genes (in total 34) had transcriptional signals and presented different trends in expression in the vitellogenesis, choriogenesis, and mature eggs in day 8 pupae (Figure 3). transcriptional signals were detected, whereas BGIBMGA003296 and BGIBMGA002069 also had no transcriptional signals (Figure 3). Me31B and the other 31 genes (in total 32) showed consistent expression levels in the vitellogenesis, choriogenesis, and mature eggs in day 8 pupae (Figure 3). Hip14 (ZDHHC17) and the other 33 genes (in total 34) had transcriptional signals and presented different trends in expression in the vitellogenesis, choriogenesis, and mature eggs in day 8 pupae (Figure 3).

Transcriptional Degradation during the Maternal-to-Zygotic Transition
To identify the transcriptional degradation patterns of the 66 maternal genes during different developmental stage embryos, six time-series samples were collected at 0, 0.5, 1.0, 1.5, 2.0, and 3.0 h after fertilized embryo spawning, and were analyzed by reverse transcription-quantitative PCR (RT-qPCR). The specific primers for each gene that was subjected to RT-qPCR are shown in Table S1. In total, temporal control of their transcript clearance presented three different maternal transcript clusters during the maternal-to-zygotic transition (Figures 4-6, Table 2).
Regarding the other 58 maternal genes, the RT-qPCR results showed that their transcripts significantly changed with two main characteristics during the maternal-to-zygotic transition (MZT). In cluster 2 (Table 2), the transcripts of 17 genes were significantly decreased from 0.5 to 3.0 h

Transcriptional Degradation during the Maternal-to-Zygotic Transition
To identify the transcriptional degradation patterns of the 66 maternal genes during different developmental stage embryos, six time-series samples were collected at 0, 0.5, 1.0, 1.5, 2.0, and 3.0 h after fertilized embryo spawning, and were analyzed by reverse transcription-quantitative PCR (RT-qPCR). The specific primers for each gene that was subjected to RT-qPCR are shown in Table S1. In total, temporal control of their transcript clearance presented three different maternal transcript clusters during the maternal-to-zygotic transition (Figures 4-6, Table 2). In cluster 1 (Table 2), the transcript levels of 8 maternal genes (Sod2, Eif-4a, eIF4AIII, bai, Pabn2, Bin1, Chc and tud) showed no change from 0 to 3 h (Figure 4). The tissue expression levels of Eif-4a (sw22934), eIF4AIII (sw12663), Bin1 (sw1118), and Sod2 (sw21871) were high and uniform overall in 10 tissues at day 3 of the silkworm fifth instar (Figure 2). For a closer examination, we used RT-PCR to investigate these eight maternal genes and their transcript temporal control from 0 to 18 h after fertilized embryo spawning ( Figure S1). The transcripts of Sod2, Eif-4a, eIF4AIII, Bin1, Chc, and tud kept consistent levels from 0 to 18 h after fertilized embryo spawning, respectively ( Figure S1). Pabn2 and bai presented changing trends in expression at the transcriptional level ( Figure S1).
Regarding the other 58 maternal genes, the RT-qPCR results showed that their transcripts significantly changed with two main characteristics during the maternal-to-zygotic transition (MZT). In cluster 2 (Table 2), the transcripts of 17 genes were significantly decreased from 0.5 to 3.0 h ( Figure 5). In cluster 3 (Table 2), the transcripts of 41 genes were decreased significantly after 1.5 h (Figure 6). This indicates that the maternally supplied mRNAs of most maternal genes were universally degraded during MZT. Unlike other genes, the transcript of the wbl gene was decreased significantly from 0.5 to 2.0 h, and increased sharply at 3.0 h. This transcript belongs to cluster 2 and is an exception.

Cluster
No. of Maternal

Discussion
In our previous study, 76 potential silkworm maternal genes were identified by orthologous comparison [48]. In this study, 68 of the 76 potential silkworm maternal genes were initially identified through further sequence alignment analysis, and 2 of the 68 maternal genes were not expressed in the silkworm eggs of virgin moths. The mRNAs of maternal genes are produced by females and loaded into the embryos [2,6]. Thus, the expression of a gene was not detected in embryos that can be identified as a non-maternal gene. A total of 66 maternal genes were finally identified in silkworm.
The embryonic development of B. mori is significantly different from that of Drosophila. The silkworm has a pair of ovaries each composed of four ovarioles, each of which contains a chain of follicles [43,52]. The previous research on in vitro culturing of B. mori ovarian follicles showed that follicle development starts from middle vitellogenesis to late choriogenesis [55]. The follicles develop depending on an endogenous developmental program that does not require the presence of additional factors from tissues outside the ovariole [43,52]. Each follicle is composed of an oocyte and seven nurse cells surrounded by a single layer of the follicular epithelium [43]. The degenerated nurse cells are devoured by follicular epithelial cells at the end of vitellogenesis [43]. The 66 maternal genes were expressed in vitellogenesis on day 8 female pupae, which suggests that maternal mRNA is derived from the nurse cells.
The development of silkworm follicles is divided into 12 different stages [43]. The developmental stage of each ovariole is opportune, being found in vitellogenesis, choriogenesis, and mature eggs from day 8 pupae [43,44]. The rate of progression of vitellogenesis toward choriogenesis is estimated to be 2-2.5 h per follicle [56,57]. The eggs are considered mature upon finishing the formation of the eggshell in the choriogenesis period, and the mature eggs first appear in each ovariole proximal oviduct in day 8 female pupae [43,44,52]. The maternal genes of Hip14 (ZDHHC17) and the other 26 genes (27 in total) had higher expression levels in vitellogenesis than in choriogenesis and mature eggs, whereas their expression levels were similar in choriogenesis and mature eggs on day 8 female pupae. These 27 maternal genes may have biological functions in the developmental process from vitellogenesis to choriogenesis in B. mori.
The MZT is a tightly regulated process that is identified by the elimination of maternal mRNAs and the initiation of zygotic transcription. This process has been extensively studied within model species. Each model organism has a unique pattern of maternal transcriptional clearance classes during the MZT. Four subsets of transcripts were characterized in Drosophila: Stable mRNAs, mRNAs targeted solely by the maternal or the zygotic degradation pathway, and those targeted by both pathways [7,12,31,58]. In activated, unfertilized eggs of Drosophila, maternal decay activity is present but zygotic activity is absent because no zygotic genome activation (ZGA) occurs. Thus, the degradation rate is significantly reduced compared with zygotic activity [7,31,[59][60][61]. For maternal transcripts degradation during the development of zebrafish, a subclass of the cleared maternal mRNAs begins at fertilization, whereas others are mainly degraded after ZGA [62,63]. In Xenopus laevis, fertilization-induced deadenylation does not trigger decay immediately, but only after ZGA causing their deadenylation and degradation [40,64,65]. In the mouse, maternal mRNAs are degraded by both the maternal and the zygotic degradation pathways [41,66]. These are evolutionarily conserved mechanisms through which the mother provides gene products to the egg to drive the earliest stages of development.
Silkworms, like other insects such as Lepidoptera and Coleoptera, undergo superficial cleavage. In silkworm, the degradation of maternal gene mRNAs can be divided into three clusters during the MZT. Cluster 1 is stable mRNAs. In cluster 1, the mRNAs level of Tud is stable from zero to three hours. Tudor is a stress granule (SG) member that is activated upon various environmental stresses. Tudor (Tud) participates in posttranscriptional regulation in B. mori [67]. Silkworm Tudor depletion increases the levels of PIWI-interacting RNAs (piRNAs), which associate with PIWI proteins to protect genome integrity by silencing transposons in the germline [68]. Thus, cluster 1 includes stable mRNAs that perform essential housekeeping functions required during the MZT. The union of sperm and egg pronuclei occurs about two hours after silkworm eggs are laid. From about 2.0 to 2.5 h, the zygote divides repeatedly by mitosis and forms many cleavage nuclei [45][46][47]. The maternal gene mRNAs decay in clusters 2 and 3 is firstly initiated by maternal degradation activity. The initial time-point of degradation of cluster 2 is earlier than that of cluster 3. In unfertilized silkworm eggs, maternal decay activity is present, but zygotic activity is absent. Because no ZGA occurs, the degradation rate is significantly reduced in unfertilized eggs compared to that of fertilized eggs [48].
The study of transcriptional regulation has produced many discoveries that have improved our understanding of development. Understanding the post-transcriptional regulation of maternal mRNA is crucial to uncover the mechanisms that control the coordinated changes in zygotic transcription initiation [6]. The MZT represents an extreme scenario involving these mechanisms. In silkworm, according to the requirements of natural and programmed embryonic development [43,48,69], studying the establishment of transcriptional quiescence during oogenesis and identifying the first genes to be expressed during embryo (mature eggs) formation will continue to improve our understanding of transcriptional regulation during MZT.

Insects and Collection of Samples
B. mori (Dazao) larvae were reared under standard conditions (25 • C and 70% humidity). The larvae-pupae, pupae, moths, and eggs were maintained under a 12 h light/12 h dark photoperiod at 25 • C and 70% humidity. The developing oocytes (eggs) undergoing vitellogenesis, choriogenesis, and mature eggs were separately collected from ovarioles that were dissected from pharate adults eight days after larval-pupal ecdysis, according to previous studies [43,44]. The eggs of virgin moths were collected from ovarioles that were dissected from the freshly hatched female moths. Freshly hatched moths were immediately mated for 3 h, and the female moths were subsequently gathered for spawning for 15 min. Zero h is defined as the 15th minute after most female moths spawn. Then, the eggs were collected at specific points-in-time (0, 0.5, 1.0, 1.5, 2.0, 3.0, 6, 12, and 18 h) under the same conditions as previously described [48].

Identification of B. mori Maternal Genes
In our earlier study, we obtained 76 potential maternal genes in the B. mori genome [48]. We obtained their sequences from the silkworm database (SilkDB), which were used to search the NCBI database and silkBase [49]. As for the maternal genes that were similar or overlapping between the NCBI database and silkBase, the longer of the two was selected. The newly annotated protein sequences were obtained and applied to predict signal peptides by SignalP 4.1 Server.

Transcript Detection Reverse Transcription-PCR
Reverse transcription-PCR (RT-PCR) was used to analyze the expression patterns of maternal genes. Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) from the samples including developing oocytes (eggs) of vitellogenesis, choriogenesis, and mature eggs in day 8 pupae, the eggs of virgin moths, and after spawning at specific points-in-time (0, 0.5, 1.0, 1.5, 2.0, 3.0, 6, 12, and 18 h). Total RNA concentrations were quantified, and single-stranded cDNA was synthesized by using a PrimeScript™ RT kit (TaKaRa, Dalian, China) according to the manufacturer's instructions. A 25 µL PCR reaction system was established by initial denaturing at 94 • C for 5 min, 35 cycles of denaturing at 94 • C for 30 s, annealing at 58 • C for 30 s, and extension at 72 • C for 30 s. This was followed by a final extension at 72 • C for 10 min before storing at 12 • C. BmRPL3 was used as an internal control [70]. A pair of specific primers for each gene was used in RT-PCR, as shown in Table S1. The RT-PCR product of each gene was separated by 1.2% agarose gel electrophoresis.

Tissue Expression Patterns Based on Microarray Database
We downloaded the microarray data from the SilkMDB to analyze tissue expression patterns of the maternal genes in 10 silkworm tissues on day 3 of the fifth instar [50]. A genome-wide microarray with 22,987 probes was designed and constructed in the silkworm genome, and each probe is also provided in this database [50]. The probes of SPE, BAEE, and Pabn2 were not found in the database, as shown by the attached BLAST search. The microarray data of the other 63 maternal genes are provided in Table S2. The expressed genes are defined as previously described [51]. GeneCluster 2.0 software was used to visualize the expression levels [71].

Transcript DecayDetection by Reverse Transcription-Quantitative PCR
Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) from the eggs collected at specific points-in-time (0, 0.5, 1.0, 1.5, 2.0, and 3.0 h). A fraction of the RNA was treated with DNase. After verifying the quality, the RNA was used to synthesize the first-strand cDNA using the PrimeScript™ RT Master Mix (Perfect Real Time; TaKaRa, Dalian, China) according to the manufacturer's instructions. Reverse transcription-quantitative PCR (RT-qPCR) was performed as previously described [48]. A pair of specific primers for each gene was used in RT-qPCR, as shown in Table S1.

Conclusions
In the current work, 66 maternal genes in silkworm were characterized through bioinformatics analysis and expression detection. The expression of these genes in vitellogenesis, choriogenesis, and mature eggs in day 8 pupae was analyzed using RT-PCR. We analyzed the maternal gene mRNAs decay in fertilized eggs in B. mori from six points-in-time by RT-qPCR. The 66 maternal genes formed three clusters of degradation patterns during the MZT. The maternal gene mRNAs of cluster 1 were stable. The initial time-point of degradation of cluster 2 was earlier than that of cluster 3. The maternal gene mRNAs decay of clusters 2 and 3 was firstly initiated by maternal degradation activity. Our findings expand upon the identification of silkworm maternal genes and provide a perspective for the embryo development in B. mori.