Reproductive pattern in the solanum mealybug, Phenacoccus solani: A new perspective

Oviposition characteristics

The oviposition process of female adults of P. solani was observed on two transparent, single-well, concave slides (Fig. 1) (length * width = 76.2 mm * 25.4 mm; the thickness of each concave slide was 1.3 mm; the diameter and depth of the circular hole were 15 mm and 0.5 mm, respectively). The steps of the procedure were as follows: (1) a concave slide was placed face up on a flat table and a female adult was gently placed into the middle of the hole of the concave slide with a brush; (2) quickly another concave slide was placed face down, on top of the bottom concave slide; and (3) finally, rubber bands were used to bind the ends of the two concave slides to secure them. Preliminary experiments showed that the thickness of female adults was 0.91 ± 0.03 mm, so the above observation method did not harm the bodies of female adults. The egg hatching process was further tracked and observed under a NikonSMZ1500 zoom stereomicroscope (Nikon, Japan), and photographs were taken every 5 min from the start of the egg laying process.

Egg morphology

The number, length, width and hatching rate of eggs with eyespots (hereafter eggs with-ES) or eggs without eyespots (hereafter eggs without-ES) were further observed in the same batch of eggs under a NikonSMZ1500 zoom stereomicroscope. There were two kinds of treatments: one employed the concave slide method (placing a female adult between two concave slides as mentioned above), and the other employed the blade method, i.e., placing a female adult on detached potato leaves, with the petiole wrapped with defatted cotton to maintain leaf freshness and the placing the whole treated leaves in Petri dishes (diameter = 9.0 cm, thickness = 1.4 cm). The oviposition of female adults was observed every 30 min from 9.00 a.m. to 4.00 p.m, and the numbers of the two kinds of eggs were counted. Each female adult was biometrically tested once, and each treatment was repeated 15 times. A total of 25 eggs were randomly selected from the two kinds of eggs, and their lengths and widths were measured. After 72 h, the hatching of the two kinds of eggs was observed. Eggs with-ES hatched 114 individuals, while eggs without-ES hatched 45 individuals. The experiment was carried out in an artificial climate chamber with a temperature of 27 ± 1 °C, a humidity of 70% ± 5% and a photoperiod of 16 L: 8D.

Does the mother’s body affect the hatching of eggs?

Two treatments were established: (1) eggs with-ES were incubated under the mother’s body (Fig. 2A), and (2) eggs with-ES were artificially removed from the mother and kept on the concave slide (Fig. 2C). Then, the female adults were continuously observed through a NikonSMZ1500 zoom stereomicroscope every five minutes. Continuous stretching of the abdomen by female adults indicated that they were about to lay eggs. For treatment 2, the upper concave slide was removed immediately, and the mother was carefully removed with an insect pin. Hatching time was recorded as the time from when the female adult laid eggs to the time when the eggshell was completely detached. Each egg was bioassayed once; experiment 1 included 30 replicates, and experiment 2 included 47 replicates.

Microscopic differences between the two kinds of eggs

Unigene de novo assembly, functional annotation and data analysis

Transcriptome sequencing and analysis were conducted by OE Biotech Co., Ltd. (Shanghai, China). The functions of the unigenes were annotated by alignment of the unigenes with the NCBI nonredundant (NR), SwissProt, EuKaryotic Orthologous Groups (KOG), Gene Ontology (GO) and Kyoto-Encyclopedia of Genes and Genomes (KEGG) databases. Differentially expressed unigenes (DEGs) were identified using the DESeq (Anders & Huber, 2013) functions “estimate size factors” and “nbinom test”. A p value <0.05 and fold change >2 or <0.5 were set as the thresholds for significant differential expression. Hierarchical cluster analysis of DEGs was performed to explore transcript expression patterns, and KEGG pathway enrichment analysis of DEGs was performed using R based on a hypergeometric distribution.

Oviposition characteristics

Female adults of P. solani produced eggs in one generation through thelytokous parthenogenesis, and the eggs were long and oval, similar to those described by Zhi et al. (2018). Female adults secreted silken wax, but it never formed an ovisac. After egg laying, the eggshell was detached from the nymph, followed by the appearance of 1st-instar nymphs that quickly crawled away from the lower part of the mother. Moreover, in a batch of eggs laid by female adults of P. solani, two kinds of eggs with distinct morphological differences appeared (Figs. 3A/3B): two reddish-brown eyespots were seen on one type of egg (eggs with-ES; Fig. 3D), while the other type did not display these eyespots (eggs without-ES; Fig. 3C). The hatching process of eggs with-ES was as follows: (1) in the first 5 min, the eggs began to show considerable peristalsis beneath the mother, and at 10–15 min, the eggshell was gradually detached, and wax powder appeared on the surface of the body; and (2) at approximately 20 min, antennae and feet began to appear, and the hatching process was generally completed within 25 min, while the detached eggshell could be seen at the abdominal end of the 1st instar nymph (Fig. 4).

Egg morphology

In the present study, we observed that female adults could lay two types of eggs, those with-ES and those without-ES, in either treatment. The ratio of eggs with-ES was significantly larger than that of eggs without-ES (on concave slides, 91.56 ± 2.14 vs. 8.44 ± 2.14, respectively, χ2 = 86.01, n = 15, p < 0.001; on leaves, 87.31 ± 2.90 vs. 12.69 ±2.90, respectively, χ2 = 13.07, n = 15, p < 0.001) (Fig. 5A). The eggs were long and elliptical, with lengths of 0.320 ± 0.006 mm (with-ES) and 0.305 ± 0.008 mm (without-ES) (t = 1.42, n = 25; Fig. 5C) and widths of 0.146 ± 0.002 mm (with-ES) and 0.147 ± 0.003 mm (without-ES) (t = 0.24, n = 25; Fig. 5D). Furthermore, the hatching rate of eggs with-ES was 100%, and no eggs without-ES hatched (Fig. 5B).

Does the mother’s body affect the hatching of eggs?

To investigate the extent to which the mother’s body affected the hatching of eggs, eggs without-ES were removed from the area beneath the mother’s body after they were laid. Therefore, we further tested whether eggs with-ES hatched after they were removed from the area beneath the mother’s body. We found that eggs with-ES could hatch normally under any treatment (Figs. 2A–2C), but the hatching times were different. Inside the mother’s body, the hatching time was 24.30 ± 0.60 min (Fig. 2D), but when the mother’s body was removed, the hatching time was reduced by nearly 5 min to 19.47 ± 0.45 min (Fig. 2E). Therefore, we suggest that the mother’s body has no effect on the success of egg hatching, or it could be inferred that the failure of eggs without-ES to hatch was largely due to internal factors. Many species of scale insects secrete abundant wax and form an ovisac that covers eggs and prevents their adhesion, but in some species, ovisacs are not built, and the time of egg development outside of the maternal body is decreased (Gavrilov-Zimin, 2018). P. solani belongs to the latter group, secreting a small amount of wax and never forming an ovisac. Thus, it is not surprising that the eggs of P. solani hatched so quickly, especially when the eggs were isolated.

Microscopic differences between the two types of eggs

Egg surface and internal structure

The surface and internal structure of eggs with and without-ES were observed using SEM and TEM. The contour of appendages could be clearly seen across the eggshell, and the bristles, tubular glands and six conical receptors (which were symmetric, with three on each side) could be seen on the surfaces of eggs with-ES (Fig. 6A); moreover, complete blood cells, cytoplasm, mitochondria and myocutaneous filaments were observed inside (Fig. 6B). Eggs without-ES had a smooth surface (Fig. 6C) and contained only lipid droplets, endoplasmic reticulum and free ribosomes (Fig. 6D).

RNA sequencing and data analysis

Illumina sequencing generated approximately 45 M reads per sample after the removal of low-quality reads. These reads were assembled randomly and produced 55,558 unigenes with an N50 of 1,026 nt. After annotating unigenes with several databases and calculating the expression of unigenes as fragments per kilobase of exon model per million reads mapped (FPKM), correlation coefficients between samples were calculated and used to estimate biological repeatability and differences between groups. The correlation coefficient of 3 biological replicates in the group with-ES and the group without-ES was >0.8, and the sample correlation coefficient between these two groups was only 0.4, showing an obvious difference between these two groups (Fig. 7A). DEGs were identified and screened with the thresholds of p < 0.05 and fold change > 2 (or fold change < 0.5) (Figs. 7B/7C). There were 13,164 DEGs between the with-ES and without-ES groups, including 9,243 up regulated DEGs and 3,921 down regulated DEGs. The DEGs are shown with a volcano plot, and a heat map was generated from hierarchical cluster analysis to show the expression patterns of the DEGs (Fig. 7B). KEGG enrichment analysis of the DEGs was performed to determine the main pathways associated with these DEGs. We found that the JAK-STAT, Notch, Hippo, and Wnt signaling pathways and dorso-ventral axis formation, wax biosynthesis, progesterone-mediated oocyte maturation, cell cycle, eukaryote ribosome biogenesis, insulin secretion, and nitrogen metabolism pathways were significantly enriched (Fig. 8).

Oviposition and hatching characteristics

In the present study, we found eggs laid by parthenogenetic P. solani, but soon emerged 1st-instar nymphs following hatch. Because this hatching process is fast (<30 min), researchers might think that 1st-instar nymphs are produced by P. solani female adults. A similar phenomenon was reported in another invasive mealybug, Phenacoccus solenopsis, and because both eggs and 1st-instar nymphs were found in oocysts, Vennila et al. (2010) considered the scale insect to be able to reproduce in both ways, i.e., via ovoviviparity and oviparity. In Coccus hesperidum, naked nymphs appeared from the vulvar orifice, but thin eggshells were shown to remain in the female reproductive tract (Hagan, 1951), and in P. solani, the eggshells were kept beneath the mother.

In Matsucoccus matsumurae, there are two types of eggs, those with and without eyespots, but the eggs with eyespots are similar to those without eyespots, and the egg types show different developmental periods with normal hatching (Xie et al., 2014). However, P. solani laid two types of eggs in one batch, with no significant difference in apparent size: one with eyespots that hatched and another without eyespots that failed to hatch. Generally, eggs are under stress from external factors, which may prevent them from hatching properly. For example, some of the eggs laid by heat-treated females of Nilaparvata lugens were unable to hatch due to failure during blastokinesis (Lee & Roger, 1987). Further research revealed that yeast-like symbiotes in N. lugens play an important role in the embryonic and postembryonic development of eggs, especially the ventral differentiation of the embryo (Wilkinson & Ishikawa, 2001; Nan et al., 2016). The symbiotic bacteria in the bodies of most mealybug subfamily insects are Tremblaya princeps (Gruwell et al., 2010), play a substantive role in the host plant specificity of their hosts (Baumann, 2005; Hansen & Moran, 2014), and are correlated with host development (Huang, Zhao & Lu, 2015); therefore, we suggest that the absence of symbiotes might explain the presence of nondeveloping eggs. Another type of insect egg that does not hatch is the nutritive egg common in social insects, such as ants. The nutritive eggs of ants are unfertilized eggs that cannot hatch and are eaten in colonies containing a queen (Heinze, Trunzer & Oliveira, 1996; Heinze et al., 1999). However, for this parthenogenetic and thelyotokous species of mealybug, the factors causing P. solani to lay eggs that cannot hatch require further study.

Physiological and molecular differences between the two types of eggs

According to these observations of egg surface and internal structure, we suggest that eggs with-ES are alive, with features such as conical receptors, blood cells and myocutaneous filaments, and that they still have an eggshell and are close to hatching, even though they no longer resemble an egg on a microscopic level. Although some important organelles such as endoplasmic reticulum and free ribosomes were present in the eggs without-ES, mitochondria had never been found. Mitochondria are a special organelle that contain their own genomes and plays an important role in oocyte maturation and embryo development (Ferenz, 1993; Lieber et al., 2019). Moreover, mitochondria are inherited only in the maternal line, i.e., mitrochondrial DNA is passed only through the mother’s egg cell (Ma, Xu & O’Farrell, 2014; Lieber et al., 2019). Lack of mitochondrial redistribution in cytoplasm was a sign of immature oocyte and was closely related to low developmental of eggs (Bavister & Squirrell, 2000). If mitochondria were missing, it could be fatal to the development of egg cell and later embryonic development. Therefore, we hypothesized that the development of egg cell might be arrested for the eggs without-ES in the absence of mitochondria.

We further determined the differences between eggs with and without ES at the molecular level. The result revealed that the differences in terms of unigene expression between two types of eggs were highly significant, and we also found that the JAK-STAT, Notch, Hippo, and Wnt signaling pathways and some important pathways related to metabolism and nutrition were significantly enriched. Recently, the JAK-STAT, Notch, Hippo and Wnt signaling pathways were found to independently or interactively participate in the regulation of egg production (Hombria & Brown, 2002; McGregor, Xi & Harrison, 2002). For example, mutual antagonism between the Notch and JAK/STAT signaling pathways provides a crucial facet of follicle cell patterning and ultimately helps establish the polarity of the egg chamber (Assa-Kunik et al., 2007), and the Hippo pathway controls polar cell specification by repressing Notch activity (Chen et al., 2011). Moreover, some of these important signaling pathways are involved in aspects of cell development and metabolic function, such as dorsal-ventral axis formation, wax biosynthesis, insulin secretion and nitrogen metabolism.