Ae. albopictus was reared at 27 ± 0.5 °C and 80% relative humidity (rh) under a photoperiod of 14 h light: 10 h dark. Larvae and adults were fed with ground fish food (TetraMin Baby, Melle) and 10% sucrose, respectively. For colony maintenance and RNAi experiments, adult females were artificially blood-fed with sheep blood (Carolina Biological Supply Company) by using a glass mosquito feeder (Chemglass).

The Ae. albopictus ortholog of D. melanogaster (DmY-y; accession number: NP_476792) was identified by performing a BLAST search of the Ae. albopictus genome. To clone a AalY-y cDNA, total RNA was isolated from females 48 h post-blood-feeding (n = 10) by using the RNeasy Mini Kit (Qiagen). First-strand cDNAs were synthesized with the SuperScript III First-strand Synthesis System (Invitrogen), using an oligo-(dT) primer according to the manufacturer’s instructions. Primers 5′-ATG TGG AAG TCG GTG GTC TG-3′ and 5′-CTA GTA CAG CTG CTT CCA G-3′ that included the predicted start and stop codons, respectively, were used to amplify the coding sequence for AalY-y (1,662 bp) by PCR. The cDNA fragment was cloned into the pGEM-T vector (Promega) and sequenced. The GenBank accession number of the AalY-y clone is MN702767.

Yellow-y-like proteins in insect genomes were identified via a BLAST search in the NCBI database, using the AalY-y protein sequence as query. Protein sequences were analyzed for signal peptides, using the SignalP 5.0 server (http://www.cbs.dtu.dk/services/SignalP). Major Royal Jelly Protein (MRJP) domains were identified, using the Conserved Domain Database (CDD, http://www.ncbi.nlm.nih.gov). N-glycosylation sites were predicted, using the NetNGlyc 1.0 server (http://www.cbs.dtu.dk/services/NetNGlyc/). Multiple sequence alignment of deduced amino acid sequences of the Yellow-y proteins was made, using the ClustalW software tool (http://www.ebi.ac.uk/Tools/msa/clustalw2). See Supplementary Table S1 for the accession numbers of Yellow-y proteins used for the amino acid sequence alignment.

To analyze expression patterns of AalY-y during development, total RNA was isolated from pools of whole insects (n = 10, except for embryos and larvae) at various developmental stages, including embryos, larvae, pupae, sugar-fed adult males and females, and blood-fed females (36–48 h after blood feeding), followed by cDNA synthesis as described above. To analyze induction patterns of AalY-y in response to the ingestion of blood, total RNA was isolated from sugar-fed females or from blood-fed females collected at 12, 24, 36, 42, 48, 60, 72, 84, and 96 h following blood or sugar feeding (n = 10 for each time point). To analyze tissue-specific expression of AalY-y, total RNA was isolated from ovaries and carcasses (whole body minus ovary) dissected from 48 h post-blood-fed females (n = 10). Total RNA was independently isolated for each of the three biological replicates. Real-time PCR were performed using the primers 5′-TCG AGC ACA GCT TCT TCT TC-3′ and 5′-GAC AGG GAC ATT CCG AAG ATA C-3′ in a 40 μL reaction volume containing 1 μL of template cDNA, 20 μL TB green™ Premix Ex Taq (TAKARA), 0.25 μM of each primer, using the Thermal Cycler Dice real-time PCR system (TAKARA). Real-time PCR was carried out with an initial denaturation at 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 60°C for 30 s. At the end of the PCR reaction, a melting curve was generated to evaluate the possibility of undesirable side products. The threshold cycle number (Ct) was determined and used for comparative quantitative (2^−ΔΔCt) analyses. The primer set consisting of 5′-ACA AGC TGC GTC ACT TCT ACG ACA-3′ and 5′-CTT GTC GTT TCC ACC GGC AAT CTT-3′ was used to amplify transcripts for Ae. albopictus ribosomal protein S6 (AalRpS6) (accession number: AF154066) as the reference gene to normalize for differences between the concentrations of cDNA templates (Noh et al., 2020).

dsRNA for AalY-y was synthesized according to the protocol described previously (Arakane et al., 2005). The template for synthesis of dsAalY-y (464 bp) was amplified by PCR using the primers 5’-(T7)-CAC ACT ACA TCT CGA GTG-3′ and 5’-(T7)-GAT GTG TAC TTG GTC ACT-3’. T7 indicates the T7 polymerase recognition sequence. dsAalY-y was synthesized by using the AmpliScribe™ T7-Flash™ Transcription kit (Epicentre Technologies). dsRNAs for AalY-g, AalY-g2, and for the enhanced green fluorescent protein (dsEGFP) were synthesized as described previously (Noh et al., 2020).

RNAi was carried out as described previously (Noh et al., 2020). dsAalY-y (approximately 1 μg per insect) were injected into the lateral thorax of 3 days-old females ∼2 h following blood-feeding (n = 15 for each of the three independent experiments). dsEGFP was injected to serve as a negative control. After injection, dsRNA-treated females were maintained at 27 ± 0.5°C and 70% rh and allowed to lay eggs 4 days after treatment. Eggs were collected immediately after oviposition for further experimentation. To analyze the knockdown level of AalY-y transcripts, cDNAs were synthesized from total RNA isolated from adults 48 h after dsRNA treatment (n = 5). Total RNA was independently isolated for each of the three biological replicates.

The ultrastructure of the chorion was analyzed by TEM as described previously (Noh et al., 2020). Eggs obtained from dsRNA-treated females were collected 1, 3, 24 and 48 h after oviposition (HAO) and fixed in a mixture of 0.1% glutaraldehyde and 4% paraformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4) for 24 h at room temperature. Samples were rinsed three times for 15 min with 0.1 M sodium cacodylate buffer (pH 7.4), and then dehydrated in a progressive ethanol gradient of 50, 60, 70, 80, 90, 95, and 100% for 20 min each. The tissues were infiltrated in LR white resin (Electron Microscopy Sciences) (1:1 ethanol:resin for 4 h, 1:2 ethanol:resin overnight and 100% resin for 4 h). The tissues were vacuum-infiltrated for 2 h, embedded in gelatin capsules (Electron Microscopy Sciences), and then the resin was polymerized at 55°C for 12–14 h, followed by ultrathin sectioning. Ultrathin sections (∼90 nm) were stained with 4% aqueous uranyl acetate for 10 min and then imaged, using a transmission electron microscope (JEM-1400, JEOL).

Desiccation resistance of eggs was analyzed according to the protocol described previously (Valencia et al., 1996; Rezende et al., 2008). Approximately 300–400 eggs obtained from dsRNA-treated females (n = 13–15) at 48 HAO were transferred to filter paper and then air-dried at room temperature and 40–50% rh for 30 min. The number of collapsed eggs was counted under a stereomicroscope (Leica M80).

Statistical analyses were performed using PSPP-1.4.0. A one-way ANOVA followed by the Tukey’s honestly significant difference (HSD) test was used to compare data from multiple samples. The Student t-test was used to analyze statistical difference between two samples (pairwise comparison).

The Ae. albopictus ortholog (AalY-y) of D. melanogaster Yellow-y protein (DmY-y) was identified by performing a BLAST (tblastn program) search of the Ae. albopictus genome database. Primers encoding the predicted start and stop codon regions were used to amplify a 1,662 bp cDNA that included the full-open reading frame. The predicted amino acid sequence is 593 residues with a theoretical molecular mass of 62.1 kDa and a predicted pI of 5.75. Consistent with our expectation that Yellow is a secretory protein, AalY-y contains a putative signal peptide sequence in addition to a conserved ∼285 amino acid-long MRJP domain (Supplementary Figure S1). Y-y appears to be a single-copy gene in all species analyzed except for Bactrocera species such as B. tryoni, B. dorsalis and B. latifrons that have two genes encoding Y-y-like proteins (denoted Y-y1 and Y-y2 by Nguyen et al. (2021)) (Supplementary Figure S2). The amino acid sequence identities and similarities of AalY-y with other insect homologs range from 51 to 95% and 76–98%, respectively (Supplementary Table S2). As shown in Supplementary Figure S3, the multiple amino acid sequence alignment of the AalY-y sequence with its orthologs from several species belonging to different orders of insects indicated that Yellow-y proteins from Diptera and Lepidoptera analyzed, including D. melanogaster (DmY-y), Musca domestica (MdY-y), B. tryoni (BtY-y1 and -y2), B. dorsalis (BdY-y1 and -y2), B. latifrons (BlY-y1 and -y2), Aedes aegypti (AaeY-y), Anopheles gambiae (AgaY-y), Bombyx mori (BmY-y), Papilio xuthus (PxY-y), S. litura (SlY-y), S. frugiperda (SfY-y), P. xylostella (PxyY-y) and A. ipsilon (AiY-y) have a C-terminal extension of 85–151 amino acids, whereas this extension is absent in Yellow-y orthologs from Coleoptera, Hymenoptera and Hemiptera, such as Tribolium castaneum (TcY-y), Tenebrio molitor (TmY-y), Leptinotarsa decemlineata (LdY-y), Apis mellifera (AmY-y), Nasonia vitripennis (NvY-y), Oncopeltus fasciatus (OfY-y) and Platymeris biguttatus (PbY-y), suggesting that Yellow-y proteins in Diptera and Lepidoptera have gained the additional C-terminal stretch rather recently in evolutionary time. All of Yellow-y proteins analyzed have one to three putative N-glycosylation sites in the MRJP domain and/or the C-terminal stretch (Supplementary Figure S3).

Real-time PCR was performed to analyze the expression patterns of AalY-y during Ae. albopictus development. AalY-y transcripts were detected at all stages of development analyzed from embryo to adult stages, with trace levels detected in sugar-fed adult males, but the levels were substantially higher in blood-fed adult females (Figure 2A). Thus, we further analyzed the induction patterns of AalY-y in response to blood feeding. As shown in Figure 2B, the expression of AalY-y gradually increased, with the highest levels 36 h after blood feeding and then gradually decreased thereafter by the end of our sampling period (96 h after blood feeding). To assess the tissue-specific expression of AalY-y in the blood-fed females, we dissected 48 h post-blood fed-females to obtain ovary and carcass (whole body without ovary) tissues. The transcripts of AalY-y were detected in the ovaries but not in the carcass (Figure 2C), indicating that in adult females of Ae. albopictus, AalY-y is specifically expressed in the ovaries after blood-feeding.

Yellow-y transcripts have also been detected in ovaries from other mosquito species, and its protein products were present in extracts from the eggshells of Ae. aegypti and An. gambiae (Amenya et al., 2010; Marinotti et al., 2014). In addition, Yellow-g (Y-g) and yellow-g2 (Y-g2), which comprise one of the yellow gene subfamilies, appear to be specifically expressed in ovaries and may have a role in eggshell/chorion rigidity in many insect species, such as Ae. albopictus, D. melanogaster, Ae. aegypti, An. gambiae and the German cockroach, Blattella germanica (Claycomb et al., 2004; Irles et al., 2009; Amenya et al., 2010; Dissanayake et al., 2010; Marinotti et al., 2014; Noh et al., 2020). Our results suggest that, like yellow-g and -g2, ovary-specific expression of AalY-y in blood-fed adult females might participate in eggshell formation, particularly in the dark black melanin production of Ae. albopictus eggs.

RNAi was used to determine whether AalY-y is required for eggshell melanization. Injection of dsRNA for AalY-y (dsAalY-y) into adult females shortly after blood-feeding led to a substantial decrease in transcript levels of AalY-y compared to that in dsEGFP-treated control animals (Figure 3A) without any significant effect on their fecundity (Figure 3B). However, in contrast to a hatch rate of 84% for eggs from the dsEGFP-treated females, the hatching rate of eggs from dsAalY-y-treated females decreased significantly to only 53%. We recently reported that injection of either dsAalY-g or dsAalY-g2 into adult females had no effect on their fecundity, whereas the resulting eggs exhibited an obviously lower hatch rate (Noh et al., 2020). These results suggest that, like AalY-g and AalY-g2, AalY-y is important for egg hatching and/or embryo viability. Similarly, reduced egg hatch rates were also observed in Y-y knockout mutants of two Spodoptera species, S. litura (Shirai et al., 2021) and S. frugiperda (Han et al., 2021). In the former, SlY-y-deficient embryonic larvae grew and developed normally, once they were dissected out manually from eggshells, suggesting that SlY-y may play a role in rigidity of the mandibles, which is critical for chewing out of the eggshell/chorion during egg hatching in S. litura.

Eggs from either dsEGFP- or dsAalY-y-treated Ae. albopictus females were pale by 1 h after oviposition (HAO). However, the former became dark black by 2–3 HAO, while the latter had a light yellow-brown color at 2 HAO and were only reddish-brown even at 48 HAO, which was significantly lighter in color than its counterpart dsEGFP eggs (Figures 3C,D). As we reported previously (Noh et al., 2020), both dsAalY-g and dsAalY-g2 eggs exhibited a delayed initial melanization by 2 HAO with the eggshell eventually becoming dark black by ∼3 HAO, as had been observed in dsEGFP control eggs (bottom two panels in Figure 3C). These results indicate that although AalY-g and AalY-g2 are required for rapid eggshell melanization soon after oviposition, AalY-y also plays a major role in melanin production for black eggs of Ae. albopictus. In contrast, the shape of dsAalY-y eggs, like that seen in the dsEGFP control ones, was spindle-like, whereas dsAalY-g and dsAalY-g2 eggs were abnormally crescent- and blimp-shaped, with the outermost exochorion being more fragile and partially peeled off (Figure 3C). These results suggest that, unlike AalY-g and AalY-g2, AalY-y is not important for the rigidity of the eggshell or the integrity of the exochorion of the Ae. albopictus egg.

The Ae. albopictus eggshell, like that seen in many other mosquito species, is composed of two maternally-derived layers, exochorion and endochorion, secreted by follicle cells, as well as the innermost chitinous serosal cuticle layer that is secreted by the zygotically-derived extra-embryonic serosal cells during embryogenesis (Rezende et al., 2008; Goltsev et al., 2009; Farnesi et al., 2015; Noh et al., 2020). In Ae. albopictus, the endochorion consists of two morphologically distinct layers, a highly electron-dense outer-endochorion (OE) and a less electron-dense monotonous inner-endochorion (IE) (Noh et al., 2020).

We tested for ultrastructural changes of dsAalY-y eggs by TEM. In the dsEGFP control eggs, as seen in the wild-type strain of Ae. albopictus, the OE showed an electron-lucent spotty morphology at 1 HAO, which became electron-dense at 3 HAO and very electron-dense with obscure morphology at 24 HAO and beyond (Figure 4A), probably due to a progressive tanning of this layer. Similarly, the electron-lucent spotty OE was evident in 1 HAO dsAalY-y eggs (Figure 4A). However, unlike the result observed with the dsEGFP eggs, the OE was still electron-lucent at 3 HAO and exhibited an abnormally electron-lucent and electron-dense spotty morphology at 24 HAO and even at a later stage (48 h) of embryonic development (Figure 4A). There was no obvious difference in morphology between the IE layers of dsEGFP and dsAalY-y eggs.

We reported recently that Ae. albopictus eggs collected by 18 HAO collapsed shortly after transfer to low humidity conditions, whereas more than 90% of the 24 HAO and later eggs remained intact, indicating that the eggs acquired tolerance to desiccation by 24 HAO under our rearing conditions (Noh et al., 2020). We further analyzed the desiccation resistance of the dsAalY-y eggs at 48 HAO. Like that seen in eggs obtained from the wild-type strain of Ae. albopictus, dsEGFP-control eggs exhibited high desiccation resistance at 48 HAO, whereas nearly half of the dsAalY-y eggs collapsed when they were moved to a low-humidity condition (Figures 4B,C). In addition, the 48 HAO eggs from either dsAalY-g- or dsAalY-g2-treated females, in which little or no electron-dense OE was evident (Figure 4A), showed rather poor desiccation resistance, with approximately 70–80% of the eggs collapsing under a low-humidity condition (Figures 4B,C) (Noh et al., 2020). These results indicate that, like AalY-g and AalY-g2, AalY-y is required for normal morphology and the formation of the OE, a structure that is critical for egg resistance to desiccation in Ae. albopictus.

In this work, we demonstrated that AalY-y, involved in the tyrosine-induced melanin biosynthetic pathway, plays an important role in black melanization of the eggshell/chorion as well as functions in the maintenance of the proper morphology of the OE, egg hatch/embryo viability and egg resistance to desiccation in Ae. albopictus. Our TEM studies indicate that the key tissue/structure affected by the deficiency of AalY-y protein is the outer-endochorion layer of the egg, which presumably acts as a barrier against water loss. This is the first evidence that a Yellow-y protein functions in normal eggshell integrity and formation, which is critical for insect reproduction, in addition to cuticle pigmentation and/or morphology. Our results provide further support for the hypothesis that different members of the family of yellow genes affect not only melanin-type pigment production but also participate in different, but essential physiological roles in the maintenance of eggshell integrity, function and viability.