Characteristic and Functional Study of Intersex, a Gene Related to Female Fertility in Bemisia tabaci

The intersex (ix) gene acts in concert with doublesex (dsx) at the end of the sex determination hierarchy to control somatic sexual differentiation in Drosophila melanogaster. Here, we report the Drosophila ix homolog in Bemisia tabaci (Btix) with differential splicing events. Four isoforms were found in B. tabaci adults, including two sex-specific transcripts (BtixF and BtixM). Knockdown of Btix had no measurable effects on female morphological phenotypes but reduced the expression of the vitellogenin gene and resulted in the production of significantly fewer eggs, a lower eclosion rate, and a shorter body size of female progeny in comparison with control females. These results increase our understanding of the genes underlying sex determination in B. tabaci and reveal a potential target for RNA interference-based pest management.


INTRODUCTION
Insect sex determination, which affects the development and reproduction of insects, is the result of long-term natural selection. Through long-term natural selection, insects have developed a great variety of sex determination mechanisms. Even in closely related species, the primary signals are different, for example, dosage of the X-signaling element (XSE) in Drosophila melanogaster (Erickson and Quintero, 2007), maternal input of transformer (tra) messenger RNA (mRNA) in Ceratitis capitata (Pane et al., 2002), and also in Musca domestica (Hediger et al., 2010) and Nasonia vitripennis (Verhulst et al., 2010), heterozygosity of the complementary sex determiner (csd) locus in Apis mellifera (Beye et al., 2003), female-specific piRNA in Bombyx mori (Kiuchi et al., 2014), a dominant male-determining factor (M factor), the Nix gene in Aedes aegypti (Hall et al., 2015), the Guy1 gene in Anopheles stephensi (Criscione et al., 2016), the Yob gene in Anopheles gambiae (Krzywinska et al., 2016), the Mdmd gene in M. domestica (Sharma et al., 2017), the MoY gene in Ceratitis capitate, and the MoY orthologs in Bactrocera oleae and Bactrocera dorsalis (Meccariello et al., 2019). Nevertheless, recent studies have indicated that some of the downstream sex determination genes are generally conserved, such as doublesex (dsx) and intersex (ix) (Sánchez, 2008).
Dsx, a downstream effector of the sex determination cascade, produces male-or femalespecific proteins via sex-specific splicing, controlling sexually dimorphic characters in different insect species (Shukla and Nagaraju, 2010). Ix, a gene required for female sexual development in Drosophila, acts together with dsx to regulate terminal differentiation. The gene ix has been characterized in Drosophila virilis, Megaselia scalaris, B. mori, Oncopeltus fasciatus, and Cyclommatus metallifer (Garrett-Engele et al., 2002;Sánchez, 2008;Aspiras et al., 2011;Gotoh et al., 2016). In D. melanogaster, ix is intronless and produce only one mRNA in all tissues of both males and females (Garrett-Engele et al., 2002). However, in Maruca vitrata and B. mori, ix is a multiexon gene with alternative splicing (AS) (Cavaliere et al., 2009;Arunkumar and Nagaraju, 2011). Dmix Bmix and Mvix mRNAs encode highly related and similarly sized proteins. Basic Local Alignment Search Tool (BLAST) analysis revealed 71-74% similarity among these three proteins. The IX proteins exhibit conserved organization at the amino (N)-terminal region similar to known transcriptional activation domains (Garrett-Engele et al., 2002). Functional conservation of ix has been demonstrated through heterologous expression of other Diptera and Lepidoptera ix homologs in transgenic D. melanogaster (Siegal and Baker, 2005;Cavaliere et al., 2009). RNA interference (RNAi) experiments with O. fasciatus (Hemiptera) and C. metallifer (Coleoptera) also indicated that ix has a conserved role in the sex determination pathway (Aspiras et al., 2011;Gotoh et al., 2016).
RNAi is a posttranscriptional gene-silencing mechanism triggered by the introduction of double-stranded RNAs (dsRNAs) and is ubiquitous in many eukaryotes, including insects (Wilson and Doudna, 2013). RNAi has become an effective technique for studying gene function in insects and provides a new method for insect pest control. The RNAi-mediated insect pest management strategy is environment friendly and highly specific. It has great application potential in both species-specific biopesticides and transgenic plants (Burand and Hunter, 2013).
The whitefly Bemisia tabaci (Hemiptera: Aleyrodidae) is a worldwide agricultural pest with a haplodiploid reproductive system (Blackman and Cahill, 1998). Our laboratory previously identified 26 genes putatively associated with sex determination in B. tabaci and found that da, mle, and PSI undergo sexspecific splicing in B. tabaci, while Imp, dsx, and tra2 do not (Liu et al., 2016(Liu et al., , 2019Guo et al., 2018). Using RNAi experiments, we also found that Btdsx and Bttra2 affect each other and are important for male genitalia formation (Guo et al., 2018). Btdsx regulates the expression of Vg in B. tabaci females (Guo et al., 2018). Bttra directly regulates the expression of Btfru and Btdsx. Bttra and Btdsx affect each other, and autoregulation of tra may exist in B. tabaci (unpublished). In this study, we cloned and characterized the ix gene in B. tabaci, explored the AS events of Btix in adult insects, and investigated the functions in regulating Vg expression and reducing female fecundity. The results suggest the potential of Btix as a target for RNAi-based pest management.

Insect Strains
The Q B. tabaci (Mediterranean, MED) population was originally collected from the fields in Beijing, China, in 2009, and was reared on cotton plants (Gossypium herbaceum L. cv. Zhongmian 49) in a glasshouse under natural light after isolation. The purity of the strain was monitored every two to three generations using a mitochondrial cytochrome oxidase I (mtCOI) marker (Chu et al., 2005). Samples of newly emerged adult females and males (within 1 h) were separately collected using a glass tube (5.0 × 0.5 cm), and the sex of each individual was determined with a stereomicroscope.

Gene Cloning and Identification of Alternative Splicing Variants
The annotated sequences obtained from our previous study (Liu et al., 2019) were used to design full-length primers to amplify the Btix gene ( Table 1). The PCRs were performed using Es-Taq Master Mix (CWBIO, Beijing, China) under the following conditions: initial denaturation at 95 • C for 5 min, followed by 35 cycles of 95 • C for 30 s, 55 • C for 30 s, and 72 • C for 1 min, and a final extension at 72 • C for 10 min. The products were purified by a Monarch DNA Gel Extraction Kit (NEB, Beijing, China), cloned into the pEASY-T1 vector (TransGen, Beijing, China), and sequenced. For AS analysis, complementary DNA (cDNA) from newly emerged female or male samples was used. The products were purified, cloned, and sequenced as described previously.

Phylogenetic Analysis and Sequence Alignment
Sequences were aligned using the program ClustalW with the following default gap penalty parameters: a gap opening of 10 and an extension of 0.2. A neighbor-joining (NJ) tree was constructed using the program MEGA 6.0 with a p-distance model and pairwise deletion of gaps (Tamura et al., 2013). The bootstrap support of tree branches was assessed by resampling amino acid positions 1,000 times.
The open reading frame (ORF) of the target nucleotide sequence was predicted by the ORF Finder tool on the National Center for Biotechnology Information (NCBI) website. The deduced protein domains were determined using the NCBI Conserved Domain Search Service. Molecular weights (MWs) and isoelectric points were analyzed via the ExPASy ProtParam tool. Sequence-similarity analyses were performed Frontiers in Physiology | www.frontiersin.org with DNAMAN 6.0. The amino acid sequences of IX were aligned using the ClustalW program and then shaded by the ExPASy BOXSHADE tool.

RNA Isolation
Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, United States) according to the manufacturer's instructions. RNA was quantified using a NanoDrop 2000 (Thermo Scientific, Wilmington, DE, United States), and integrity was checked with 1% Tris/borate/ethylenediaminetetraacetic acid (EDTA) (TBE) agarose gel electrophoresis. For gene cloning and quantitative PCR (qPCR) analysis, first-strand cDNA was synthesized using 1 µg of total RNA with the PrimeScript R RT reagent kit (TaKaRa Biotech, Kyoto, Japan) following the manufacturer's recommendations. The synthesized first-strand cDNA was used immediately or stored at −20 • C for later use.

Real-Time Quantitative PCR
Quantitative PCR was performed using SuperReal PreMix Plus (SYBR Green) (Tiangen, Beijing, China) according to the manufacturer's instructions. Each reaction contained 1 µl of cDNA template, 10 µl of SuperReal PreMix Plus, 0.4 µl of ROX reference dye, and 0.5 µl of specific primers in a total volume of 20 µl. PCR was performed under the following conditions: denaturation at 95 • C for 10 min, followed by 40 cycles of 95 • C for 15 s, 60 • C for 30 s, and 72 • C for 30 s. The specific primers are shown in Table 1. Three independent biological replicates were included for each treatment. The results were standardized to the expression level of the B. tabaci SDHA gene (Li et al., 2013), and the relative differences in transcript levels were analyzed by the 2 − Ct method (Livak and Schmittgen, 2001).

RNA Interference
Primers (Table 1) containing a T7 promoter sequence were designed to amplify dsRNA targeting Btix. The dsRNA for enhanced green fluorescent protein (EGFP) was used as the negative control. The primers are listed in Table 1. All dsRNAs were prepared using the T7 Ribomax TM Express RNAi system (Promega, Madison, WI, United States). RNAi was achieved by directly feeding dsRNA to B. tabaci adults in a feeding chamber. A 0.20-ml drop of diet solution [5% yeast extract and 30% sucrose (wt/vol)] containing 0.5 µg/µl dsRNA was placed in the chamber (Upadhyay et al., 2011;Yang et al., 2017). Approximately 40 newly emerged females were introduced into an environmental chamber at 25 • C under a photoperiod of L14/D10 and a relative humidity (RH) of 70%. Each sample was represented by three technical replicates. After 2 days of feeding, mortality was recorded, and B. tabaci specimens were collected.
To determine the effects of silencing Btix on Vg expression and female development, we observed the RNAi females for 3 days. The phenotype of B. tabaci was observed by a stereomicroscope (Leica, M205C, Germany), and the expression levels of targeted genes were analyzed by real-time qPCR (RT-qPCR). The detail was as follows: first, ∼1,000 newly emerged B. tabaci females were observed to make sure that all individuals were wild type. Second, ∼500 newly emerged females were fed on dsBtix, and the others were fed on dsEGFP. Third, after 2 days of feeding, females were observed immediately or at 24, 48, 72 h with a stereomicroscope, and then were collected for qPCR analysis. Each replicate contained 40 females and 3 replicates per treatment. One-way ANOVA with Tukey's test (P < 0.05) were used to evaluate differences among treatments.

Fecundity of B. tabaci
To determine the effects of silencing Btix on B. tabaci reproduction, after 2 days of dsRNA feeding, females were placed in a plastic bottle (5.5 cm diameter, 15 cm height) with a cotton seedling (Supplementary Figure S1). Each bottle contained five RNAi females and five newly emerged males (20 replicates per treatment). After 3 days of oviposition, the number of B. tabaci individuals in each bottle was determined, and their eggs were counted using a stereomicroscope. B. tabaci reproduction was reported as the mean number of eggs per surviving whitefly over 3 days (7 days of oviposition was also measured). The eclosion rates were calculated as the total number of emerging adults divided by the number of eggs in each bottle. Newly emerged progenies (females and males) from each treatment were randomly selected and frozen (approximately 90 females and 90 males per treatment). Thereafter, the randomly selected adults were laid on their back, and their length was measured from the top of their head to the end of their abdomen with the aid of a stereomicroscope (Leica, M205C, Germany). Student's t-test (P < 0.05) were used to evaluate differences among treatments. SPSS version 19.0 (SPSS Inc., Chicago, IL, United States) was used for statistical analyses.

Cloning and Characterization of Btix
Based on our previously annotated sequence from the Q/MED B. tabaci genome database Liu et al., 2019), the full-length cDNA sequence of the Btix gene was obtained from PCR amplification, using adult cDNA as a template. The ORF encoded a protein of 181 amino acids with a predicted MW of ∼19.85 kDa and an isoelectric point (pI) of 5.65. In contrast to ix in D. melanogaster and hymenopterans, which is a single-exon gene, Btix possesses four exons ( Figure 1A and Supplementary Figure S2). BtIX contains glutamine-, glycine-, proline-, and serine-rich regions in the N-terminus (corresponding to positions 4-41), and in the C-terminus, there are several stretches of high amino acid identity (with the most striking such block corresponding to residues 92-112), which is a conserved feature of the IX protein ( Figure 1B and Supplementary Figure S3).

Phylogenetic Analysis of Btix
We identified putative homologs of IX in three dipteran, two lepidopteran, and six hymenopteran species by searching the BeeBase (Elsik et al., 2016) or NCBI website. Protein sequences were aligned, and an NJ phylogenetic tree was constructed. The resulting tree clustered the IX proteins within each insect order, and BtIX was closely related to the IX from Hymenoptera ( Figure 1C).

Splice Variation of Btix
To analyze splice variations of Btix, we amplified mRNA products from newly emerged female and male adults using full-length primers ( Table 1). A total of four isoforms were found in B. tabaci adults: two non-sex-specific isoforms (Btix C ), one female-specific isoform (Btix F ), and one male-specific isoform (Btix M ) (Figure 2). The four isoforms encoded four proteins with different sizes, and among them, only Btix M encoded a truncated non-functional protein (Supplementary Figure S4).

Btix RNA Interference
To determine the function of Btix, we used RNAi technology to knock down the expression of Btix in the newly emerged female adults of B. tabaci. For dsRNA feeding, RT-qPCR results showed that Btix expression was significantly decreased at 2 days of feeding (77.2%, P < 0.01; Figure 3A), and the survival rate was more than 95% (Supplementary Figure S5). Furthermore, over the next 3 days, the Vg expression in the dsRNA-fed group decreased to varying degrees, particularly on the first (P < 0.01) and 3rd days (P = 0.018) after RNAi (Figure 3B). After 3 days of observation, no phenotypic (body color and morphological defects of genitalia) changes were observed compared with control females (Supplementary Figure S6).

Function of Btix in Female Reproduction
We found that knocking down the expression of the Btix gene affected female reproduction in B. tabaci. In Btix dsRNA-fed females, fecundity was significantly reduced by 19% (P < 0.01; Figures 4A,B), and the eclosion rate of progeny was reduced by 16% (P = 0.025; Figure 4C). The average body length of female     progeny was also markedly shorter than that of control females (shortened by 7%) (P < 0.01; Figures 4D, 5).

DISCUSSION
In this study, we cloned and characterized the ix gene in B. tabaci (Btix; Figures 1A,B) and found that Btix is structurally conserved. First, BtIX shared several modular elements with other insect IX proteins, including the amino acid composition of the N-terminus and high amino acid identity in the C-terminus ( Figure 1B and Supplementary Figure S3). The former region resembles known transcriptional activation domains (Garrett-Engele et al., 2002), and the latter is considered an important region for DSX F binding (Yang et al., 2008). Second, the IX homologs were conserved in their short length. BtIX is only 181 amino acids long, and the D. melanogaster IX is 188 amino acids in length. The M. vitrata IX presents an additional region of 38 amino acids in the C-terminus (Cavaliere et al., 2009) and is the longest homolog, yet it is only 232 amino acids in length ( Figure 1C).
ix is a single-exon gene and not sex-specifically spliced in Drosophila (Garrett-Engele et al., 2002). However, in M. vitrata and B. mori, the ix orthologs present, respectively, five and six exons (Cavaliere et al., 2009). A female-specific and pupae-stage specific ix transcript is found in M. vitrata, and a testis-specific splice form of ix is found in B. mori (Cavaliere et al., 2009;Arunkumar and Nagaraju, 2011). In B. tabaci, the Btix gene contains four exons; a female-specific isoform (Btix F ) and a male-specific isoform (Btix M ) were found in B. tabaci adults (Figure 2). However, we did not find any sex-specific exons, so we were unable to design specific qPCR primers to reveal the developmental and temporal expression patterns of the four isoforms. In addition, since Btix M only has 186 bp, we did not find a suitable RNAi fragment. Therefore, we only investigated the function of Btix F mainly in this paper.
The gene ix is a conserved downstream gene at the end of the sex determination cascade, second to dsx. In D. melanogaster, IX specifically interacts with DSX F , forming a DNA-binding complex to activate Yolk protein (Yp) gene expression and control female sexual differentiation (Garrett-Engele et al., 2002). Female ix mutants of D. melanogaster display characteristic morphological defects that are highly similar to those in dsx mutants (Chase and Baker, 1995;Garrett-Engele et al., 2002). Ectopic expression of D. virilis and M. scalaris ix cDNA fully rescues the D. melanogaster female ix mutants, and expression of B. mori and M. vitrata cDNA partially restores the mutants (Siegal and Baker, 2005;Cavaliere et al., 2009). Knockdown of ix in O. fasciatus produced defects in both sexes, reducing the size of both male and female genitalia (Aspiras et al., 2011). Depletion of ix in C. metallifer affected sex-specific morphological characters such as mandible size and body color in females only (Gotoh et al., 2016). These results highlight the conserved function of ix in insect sex determination. However, silencing Btix did not affect the morphological changes in B. tabaci females (Supplementary Figure S6). This may be because we treated the adults. Our lab is working on a new RNAi method for B. tabaci nymph and egg stage. When the method is mature, the role of Btix in sex determination will be better studied.
In this paper, the expression of Vg decreased significantly on the 1st and 3rd day after silencing Btix ( Figure 3B). This result indicates that Btix plays an important role in regulating Vg expression. However, the detection of Vg expression immediately after RNAi had no effect. This may be because Vg is not a target gene for Btix. In D. melanogaster, yolk protein genes (Yp1 and Yp2) are the best characterized target genes of sex-specific DSX proteins and are affected by any change in the sex-determination pathway (Shukla and Nagaraju, 2010). DSX F regulates Yp expression and requires IX; similarly, the female-specific function of IX is mediated by DSX F (Waterbury et al., 1999;Garrett-Engele et al., 2002). To understand the interaction between Btix and Btdsx, we silenced Btdsx and Btix, respectively, to detect the expression of these two genes. The results showed that silencing Btdsx did not affect the expression of Btix and vice versa (Supplementary Figure S7). Since ix functions as a complex with DSX in Drosophila, the RNAi results can only indicate that there is no cascade control between Btix and Btdsx. Whether Btix combined with Btdsx participate in sex determination remain further explained by subsequent experiments.
The Vg gene plays a vital role in oocyte and embryo development in insects. Silencing of vitellogenin or vitellogenin receptor genes resulted in significant mortality and reduced fecundity in B. tabaci adults (Upadhyay et al., 2016;Grover et al., 2018). Therefore, it is reasonable to believe that Btix may affect reproduction in B. tabaci. In the present study, disruption of the expression of Btix in female adults reduced fecundity, eclosion rate of progeny, and shortened the body length of female progeny (Figure 4). These results suggest that Btix does have an effect on B. tabaci reproduction. In previous transgenic D. melanogaster tests, female ix mutants were sterile; despite expression of the B. mori ix, fertility was severely reduced (Siegal and Baker, 2005). These results showed that ix has an evolutionarily conserved function in the control of female fertility.
RNAi technology is promising for the control of insect pests; its main challenge is to find suitable target genes. In general, functional genes encoding essential proteins can be suitable RNAi targets. Downregulation of the expression of these genes can reduce the fitness or survival rate of insect pests (Katoch et al., 2013). For example, RNAi knockdown of the P450 CYP6CM1, GST, or HOT gene increased mortality in a B. tabaci-resistant strain (Grover et al., 2018). Knockdown of SexiCSP3 in Spodoptera exigua caused high mortality, low hatching rate, and reduced oviposition (Gong et al., 2012). Knockdown of genes in the ecdysone pathway reduced the survival and delayed the development of B. tabaci during nymphal stages (Luan et al., 2013). In this paper, we confirmed the important role of Btix in B. tabaci reproduction, indicating that Btix can be used as a potential target for RNAi-based pest management.

CONCLUSION
In the present study, we cloned and characterized the Btix gene from B. tabaci and provided evidence for the presence of alternatively spliced forms of ix in B. tabaci. Furthermore, silencing Btix downregulated Vg expression and had some negative effects on female reproduction in B. tabaci. Our findings indicate that ix is functionally conserved in the control of female fertility. In addition to expanding our understanding of the genes that may determine sex in B. tabaci, these results will help us develop new methods for the effective control of B. tabaci. The functional relationships of Btix and Btdsx, as well as the specific contribution of Btix in regulating the terminal differentiation of B. tabaci, remain to be further studied.

DATA AVAILABILITY STATEMENT
The datasets generated for this study can be found in the GenBank: QAB02863.1.