Ploidy dynamics in aphid host cells harboring bacterial symbionts

Aphids have evolved bacteriocytes or symbiotic host cells that harbor the obligate mutualistic bacterium Buchnera aphidicola. Because of the large cell size (approximately 100 μm in diameter) of bacteriocytes and their pivotal role in nutritional symbiosis, researchers have considered that these cells are highly polyploid and assumed that bacteriocyte polyploidy may be essential for the symbiotic relationship between the aphid and the bacterium. However, little is known about the ploidy levels and dynamics of aphid bacteriocytes. Here, we quantitatively analyzed the ploidy levels in the bacteriocytes of the pea-aphid Acyrthosiphon pisum. Image-based fluorometry revealed the hyper polyploidy of the bacteriocytes ranging from 16- to 256-ploidy throughout the lifecycle. Bacteriocytes of adult parthenogenetic viviparous females were ranged between 64 and 128C DNA levels, while those of sexual morphs (oviparous females and males) were comprised of 64C, and 32–64C cells, respectively. During post-embryonic development of viviparous females, the ploidy level of bacteriocytes increased substantially, from 16 to 32C at birth to 128–256C in actively reproducing adults. These results suggest that the ploidy levels are dynamically regulated among phenotypes and during development. Our comprehensive and quantitative data provides a foundation for future studies to understand the functional roles and biological significance of the polyploidy of insect bacteriocytes.


General observation and methods for ploidy analysis on aphid bacteriome cells. Consistent
with previous observations 9, 21,22,40 , the bacteriome of viviparous aphids consisted of two types of cells: bacteriocytes and sheath cells (Fig. 2). Bacteriocytes contained Buchnera cells and were much larger than sheath cells. Sheath cells exhibited a flattened morphology and surrounded the bacteriocytes. Both cell types possessed a single nucleus. Bacteriocytes had a single prominent nucleolus, which was not stained using DAPI, but using "Nucleolus Bright Red" staining ( Fig. 2). Most sheath cells also had a single nucleolus, yet a small number had two. "Nucleolus Bright Red" also stained the peripheral region of Buchnera, probably because of the richness of RNA around Buchnera cells.
To determine the most suitable methods for ploidy analysis of aphid bacteriocytes, three types of methods, flow cytometry, Feulgen densitometry, and fluorometry were compared. First, flow cytometry successfully detected the nuclei of bacteriome cells and heads, and distinct peaks were present (Fig. S3). There were several peaks, which can be categorized as ploidy classes based on head peaks, assuming that the smallest peaks correspond to a diploid population. We recognized peaks up to 256C (256-ploidy) cells but could not distinguish cell types (i.e., bacteriocytes or sheath cells) in this method due to a lack of cytological information. Note that "C" means haploid genome size, for example, 2C = diploid and 8C = octoploid. Second, Feulgen densitometry also showed several ploidy levels of up to 128C (Fig. S4) in bacteriocytes. Sheath cells mainly consisted of 16-32C cells. However, we found that many cells were lost during the experimental procedures, probably due to the repeated washing processes and the long incubation time.  www.nature.com/scientificreports/ We found the third method, image-based fluorometry for isolated nuclei, the best for quantitative ploidy analysis of aphid bacteriocytes (Fig. 3). Fluorometry showed distinct peaks of integrated fluorescence intensity, and they could be categorized as each ploidy class based on the intensity of the smallest peak in head cells (diploid population). The results were consistent with other methods; ploidy levels were 32C-256C in bacteriocytes and 16C-32C in sheath cells. In this analysis, the nucleolus size was used to discriminate between cell types. During cytological observation, we obtained the size distribution of the nucleolus, and it was revealed that the nucleolus of bacteriocytes was always larger than that of sheath cells (Fig. S5). Based on the results, we determined the threshold of the size of the nucleolus. More specifically, in viviparous females, nuclei that have nucleoli larger than 20 μm 2 were categorized into bacteriocytes. Note that the peaks of sheath cells were not distinct or reliable for categorizing their ploidy class; therefore, we showed results focusing on bacteriocytes in the following sections.

Cellular features of bacteriome cells in viviparous and oviparous females, and males.
The cellular features were generally consistent among young adults (within 5 days of adult eclosion) of three morphs, viviparous and oviparous females, and males (Fig. 2). Nevertheless, Buchnera-absence zones in the cytoplasm of bacteriocytes, which are considered to be degeneration of Buchnera 45 , and bacteriocytes degeneration 46 were both observed more frequently in male bacteriocytes than in females (Fig. 2). The cell size of bacteriocytes was significantly different among morphs (LM with type II test, F = 286.15, df = 2, p < 0.001, Fig. S6). Viviparous females had significantly larger bacteriocytes (Tukey's test, p < 0.05, Fig. S6). The size of nucleoli was significantly different between bacteriocytes and sheath cells, regardless of aphid morphs (LM with type II test; viviparous females, χ 2 = 618.4, df = 1, p < 0.001, oviparous females, χ 2 = 1,430.4, df = 1, p < 0.001, males, χ 2 = 261.37, df = 1, p < 0.001, Fig. S5). There was no overlap in the nucleolus size between cell types (Fig. S5). Based on these data, we determined the threshold of the size of the nucleolus to discriminate between bacteriocytes and sheath cells. Specifically, in viviparous and oviparous females, and males, nuclei that have nucleoli larger than 20    presented. An image of DAPI-stained nuclei was also shown (the blue channel was extracted). Isolated nuclei of bacteriome cells were stained using DAPI, image-captured with a CCD camera, and their integrated fluorescence intensity was measured using ImageJ software. Nuclei were categorized into "bacteriocytes" or "sheath cells, " based on the size distribution of nucleolus (see "Materials and Methods"). Relative ploidy levels were calculated based on the data from head cells which are mainly diploid. Bacteriocytes of adult viviparous aphids consisted of 16C-256C cells, and 64-128 cells were dominant, while sheath cells exhibited lower ploidy levels (mainly 16C). "C" means haploid genome size, for example, 2C = diploid and 8C = octoploid. In oviparous females, first oviposition and mating with males were observed on days 3-4. They actively laid eggs until day 14, but their death was observed almost simultaneously ( Fig. S7).  Fig. S8a]. In particular, N1, N2, N3, and N4 periods lasted for 3.2 ± 0.11, 3.0 ± 0.09, 3.3 ± 0.12, and 4.2 ± 0.14 days (mean ± SEM, n = 16, Fig. S8a], respectively. Adult aphids started reproducing 2 or 3 days after eclosion (molt for an adult) and continued larviposition for approximately 4 weeks (Fig. S7). Based on these data, A7 aphids (7 days after eclosion) could be categorized as actively reproducing individuals. A21 aphids (21 days after eclosion) were categorized as senescent individuals, although they continuously produced offspring. During the nymphal stages of viviparous aphids, the morphology of bacteriome cells was generally consistent; all bacteriocytes and most sheath cells were uninuclear ( Fig. S8b), but very few of the latter cells had several small nuclei. Notably, there were drastic morphological changes in adult stages; bacteriocyte and sheath cell nuclei of A21 individuals were irregularly shaped in comparison with those of young (A0) and reproducing (A7) individuals. Furthermore, in A21, we frequently observed bacteriocytes in which the signals of DAPI and Nucleolus Bright Red signals on Buchnera were weak (Fig. S8b). These changes were consistent with symptoms of Buchnera degeneration and cell senescence, which have been previously reported 45,46 . Developmental stages had a significant effect on bacteriocyte size (LMM with type II test, χ 2 = 338.73, df = 6, p < 0.001). During postembryonic development, the size of bacteriocytes consistently increased (Tukey's test: N1 = N2 = N3 ≤ N4 < A0 = A7 = A21, p < 0.05, Fig. S9). The volume of bacteriocytes was positively correlated with those of their nuclei (Simple correlation analysis with LM; p < 0.001, R 2 = 0.83, Fig. S10). The size of nucleoli was significantly different between bacteriocytes and sheath cells, regardless of the post-embryonic developmental stages (LMM with type II test; N1, χ 2 = 891.82, df = 1, p < 0.001, N2, χ 2 = 294.04, df = 1, p < 0.001, N3, χ 2 = 842.31, df = 1, p < 0.001, N4, χ 2 = 817.18, df = 1, p < 0.001, old adults, χ 2 = 1,405.6, df = 1, p < 0.001, Fig. S11). There was no overlap in the nucleolus size between cell types (Fig. S11). Based on these data and the data from young adults, we determined the threshold of the size of the nucleolus for ploidy analysis (in N1 and N2, 10 μm 2 , and later stages, 25 μm 2 ).
The size of the nuclei and nucleolus, and ploidy levels in aphid bacteriocytes. The size of bacteriocyte nuclei was positively correlated with the ploidy class in all aphid categories [adult viviparous females, adult oviparous females, adult males, and all stages of viviparous/oviparous females (N1-N4 and A0, A7, A21 were pooled)] (Simple correlation analysis with LM; p < 0.001) (details in Fig. S13). There were significant effects of ploidy class on the size of the nucleolus in adult bacteriocytes of each morph (LM with type II test; viviparous females, F = 62.94, df = 2, p < 0.001, oviparous females, F = 23.97, df = 2, p < 0.001; males, F = 6.44, df = 3, p < 0.001, Fig. 6a). Note that 16C and 256C viviparous bacteriocytes were excluded from the analysis due to their small number. Similarly, 16C and 8C cells of females and males, respectively, were excluded from the analysis. In viviparous females, the size of the nucleolus consistently increased from 32 to 128C (Tukey's test, p < 0.001).
In oviparous females, 128C cells had larger nucleoli than 32C and 64C cells (p < 0.001 each), yet the difference between 32 and 64C cells was marginally non-significant (p = 0.06). In males, the size of the nucleolus of 128C cells was significantly larger than that of 16C and 32C cells (p < 0.001 each), but we did not find any significant difference among other comparisons (16C vs. 32C, p = 0.71, 16C vs. 64C, p = 0.10, 32C vs 64C, p = 0.08, 64C vs 128C, p = 0.18) (Fig. 6a). A significant effect of ploidy class on the nucleolus size was also detected in the data from each developmental stage of viviparous aphids (LMM with type II test; χ 2 = 788.83, df = 5, p < 0.001, Fig. 6b). Note that 4C and 512C bacteriocytes were excluded from the analysis because of the small number of observations. The size of the nucleolus consistently increased from 8 to 256C (Tukey's test, p < 0.05, Fig. 6b).

Discussion
This study presented quantitative data on ploidy levels and ploidy dynamics in the bacteriocytes, which are pivotal cells in aphid/Buchnera endosymbiosis. The method developed for ploidy analysis of aphid bacteriocytes in this study (Fig. 3) revealed the hyper polyploidy of the bacteriocytes ranging from 16-to 256-ploidy throughout the lifecycle. We also found significant differences in the ploidy levels among morphs in adult stages (viviparous females > oviparous females > males, Fig. 4) and developmental stages in viviparous females (reproducing adults > senescent adults > pre-reproducing adults > nymphs, Fig. 5). Considering that viviparous adults exhibited a high rate of reproduction, which was at a maximum in the first three weeks (Fig. S7), it would be reasonable that more metabolically active (actively reproducing) individuals show higher polyploidy in their bacteriocytes. We observed a similar pattern of bacteriocyte polyploidy in oviparous aphids (Fig. S12). These results provide fundamental information to understand the functional significance of polyploidy in aphid bacteriocytes. Our findings in this study raise the possibility that bacteriocyte polyploidy can enhance nutritional symbiosis between aphids and B. aphidicola, wherein both supply each other with nutrients that they cannot synthesize on their own 10,19,20 . Specifically, we predict that genes involved in amino acid metabolism, related to transport, and for symbiont regulation (e.g., lysozymes or cysteine-rich secreted proteins targeting bacteria), which have been reported to be highly expressed in the aphid bacteriomes 24,25,38,40 , may be upregulated in a ploidy-dependent manner 6,7,47 . On the other hand, it is commonly known that cell volumes change depending on ploidy levels 6,48 .
In fact, our data demonstrated that the volume of bacteriocyte was positively correlated with those of the nuclei, which can be used as a proxy of the ploidy level ( Fig. S10 and S13). Bacteriocyte enlargement may simply increase the number of B. aphidicola that can be harbored, which should lead to enhanced nutritional symbiosis. In order to examine the effect of polyploidization on the aphid/Buchnera symbiosis, an integrated approach including gene expression analysis on a per-nucleus basis, and monitoring cell phenotypes such as ploidy-level and cell/ nuclear size is required.
In this study, we also found positive correlations between the ploidy class of bacteriocytes and the size of their nucleoli, regardless of the morphs and developmental stage (Fig. 6). The main function of the nucleolus is ribosomal biogenesis, and the size and morphology of nucleoli are linked to nucleolar activity, such as transcription and ribosomal RNA production rates [49][50][51] . In plants, there is evidence that more polyploid nuclei not only exhibit larger nucleolar size but also exhibit increased transcription of rRNA and mRNA; a positive correlation between DNA content and transcriptional activity has been identified in the polyploid tomato fruit pericarp 8 . It would be interesting to test the hypothesis that aphid bacteriocytes with higher ploidy levels produce more ribosomal RNA, leading to higher metabolic activity. A direct relationship between ploidy levels and nucleolar activity needs to be examined in future studies.
Aphid bacteriocytes have long been considered to be polyploid [Myzus persicae (Sulzer) 18 , Pemphigus spyrothecae Passerini 22 , and Cinara species 52 ]. Furthermore, it has been concluded that bacteriocytes are polyploid in many other insects, such as psyllids, whiteflies, scale insects, weevils, bark beetles, termites, and cockroaches 11,12,15,[53][54][55][56] , although quantitative data were lacking, except for psyllids 15 . In this study, by comprehensively describing the patterns of polyploidization, we demonstrated, for the first time, that high metabolic demand such as active reproduction is associated with higher polyploidy levels in aphid bacteriocytes. It would be valuable to investigate this relationship in intracellular symbiosis in various insect species. Accumulating information on bacteriocyte polyploidy will help us gain a better understanding of the maintenance and evolution of mutual relationships between host and symbionts because polyploidization in the symbiotic host cells is a common rule in insect-microorganism intracellular symbioses 12,13,57 .
In the present study, we compared three methods for ploidy level quantification and found an image-based fluorometry the best for the analysis of aphid bacteriomes (Fig. 4), because it could distinguish between cell types (e.g., bacteriocytes and sheath cells) by the size of nucleolus, unlike flow cytometry (Fig. S3). In addition, it was timesaving compared with Feulgen densitometry (Fig. S4). Moreover, it identified discrete data unlike the densitometry which returned more continuous value ( Fig. 3 and S4). Our results from fluorometry showed that sheath cells mainly consisted of 16C and 32C in adult viviparous aphids, although peaks in the histogram were not clear (Fig. 3); therefore, nuclei exhibiting more than 32C can be reasonably assumed to be bacteriocytes. These approach, the combination of several methods such as the fluorometry and flow cytometry would be applicable to other symbiotic systems of insects with microorganisms, wherein the bacteriome frequently contains several types of cells (e.g., primary or secondary bacteriocytes, and sheath cells) 13,14,21 .
In conclusion, we comprehensively described the patterns of polyploidization in aphid bacteriocytes, which has long been assumed to be polyploid, yet there have been no quantitative studies 9,10,14 . Based on the patterns and cytological features observed in this study, we suggest that hyper polyploidy may enhance gene expression levels and increase cell size, contributing to the nutritional symbiosis with the bacterial symbiont B. aphidicola. This study provides a foundation for further molecular-level analysis of the functions and underlying mechanisms of polyploidy in insect symbiotic host cells.

Material and methods
Aphids. In this study, we used a long-established parthenogenetic clone of the pea-aphid, A. pisum, ApL strain, which was originally collected in Sapporo, Hokkaido, Japan (referred to as Sap05Ms2 in 58 ). We confirmed that this strain only harbors the primary endosymbiont B. aphidicola by diagnostic PCR, as described in a previous study 59 . Viviparous insects were maintained on young, broad bean plants (Vicia faba L.) in a 16 °C incubator, with a photoperiod of 16 h light:8 h dark (long-day conditions). Sexual morphs (oviparous females and males) were induced by short-day conditions (e.g., 8 h light:16 h dark) ( Fig. S1 and S2, modified from 58,60,61 ). Viviparous females were randomly selected from the synchronized source populations and reared on young broad bean plants at 16 °C under long-day conditions. Newly larviposited aphids (G0, first-instar nymphs, Fig. S2)  www.nature.com/scientificreports/ transferred onto the leaves of the bean plants in a 15 °C incubator with a photoperiod of 8 h light:16 h dark. After these G0 nymphs grew into adults, they started to produce G1 offspring, which are morphologically identical to apterous/viviparous females but produce sexual individuals (Fig. S2). G1 adults produce G2 offspring that contains sexual individuals (oviparous females and males), but they also produce a few viviparous females (Fig. S2).
In this study, all oviparous females and males were G2 offspring, and all viviparous individuals were apterous. To reveal the fecundity and longevity of both viviparous and oviparous aphids, female aphids were reared separately and observed daily (see supplementary information "SI Methods").
Size and morphology of aphid bacteriome cells. Aphid bacteriomes consist of two types of cells: bacteriocytes containing Buchnera cells in their cytoplasm and sheath cells without them (Fig. 1b, 9,21,40 ). To gain more detailed cellular features of aphid bacteriocytes, we performed a morphological analysis of bacteriomes using a confocal laser-scanning microscope (CLSM; FV1000, Olympus, Japan) on three adult morphs (viviparous/oviparous females and males). These adults were within 5 days of eclosion. We also observed the bacteriome at each stage of viviparous females [nymphs N1 (first-instar nymph), N2 (second-instar nymph), N3 (third-instar nymph), and N4 (fourth-instar nymph), and young adults (3-5 days after eclosion), and old adults (approximately 21 days after eclosion)]. Each stage of viviparous females was used on the day of molting, but teneral insects were not used). Aphids were dissected in phosphate-buffered saline (PBS: 33 mM 143 KH 2 PO 4 , 33 mM Na 2 HPO 4 , pH 6.8) under a stereomicroscope (SZ61, Olympus, Japan), with fine forceps, and their bacteriome cells (bacteriocytes and sheath cells) were surgically isolated from the aphid abdomen. Totally, five adults were used for each morph, and three to six individuals were used for each stage of viviparous females. Bacteriome from adult morphs were pooled just after dissection, while those from each stage of viviparous females were treated individually. The cells were fixed with 4% paraformaldehyde in PBS for 30 min. Fixed bacteriome cells were washed three times in 0.3% Triton X-100 in PBS (PBS-T) for 15 min for permeabilization. The cells were then stained with 4,6-diamidino-2-phenylindole (DAPI) (1 μg/mL; Dojindo, Japan) for the nuclei and Alexa Fluor™ 488 phalloidin (66 nM; Thermo Fisher Scientific, USA) for the cytoskeleton (F-actin), respectively. The nucleolus, which is the site of both ribosomal RNA (rRNA) synthesis and the assembly of ribosomal subunits 49 , was visualized using Nucleolus Bright Red (1 mM; Dojindo, Japan). Nucleolus Bright Red dyes are small molecules that electrostatically bind to RNA in the nucleolus to emit fluorescence. After 1 h at room temperature (from 20 to 25 °C), the cells were washed three times with PBS-T for 15 min and mounted with VECTASHIELD antifade mounting medium (Vector Laboratories, USA). The morphology of the bacteriocytes and sheath cells was visualized using fluorescent staining and differential interference contrast microscopy. The captured images were processed using the image analysis software ImageJ (NIH, http:// rsb. info. nih. gov/ ij/). The diameter (D) of bacteriocytes and the nuclei was measured at its widest point. The approximate bacteriocyte and their nuclei volume (V) was calculated using the standard formula: V = π 6 D 3 . The nucleolus size (area) of both the bacteriocytes and sheath cells was also recorded.

Establishment of the method for ploidy analysis on aphid bacteriocytes. To establish a method
for ploidy analysis of aphid bacteriocytes, we used three methods: flow cytometry, Feulgen densitometry, and fluorometry. Young adult viviparous aphids (3-5 days after eclosion) or late instar (third-fourth instar) of nymphs were used. In all analyses, not only bacteriome cells but also head cells were used as diploid controls, which were confirmed as diploid by preliminary analysis of sperm cells (haploid). We first performed flow cytometry, which is an efficient and commonly used method for nuclear DNA-content analysis 62,63 . For this analysis, bacteriome cells dissected from five aphids were suspended and repeatedly pipetted in 250 μL of trypsin buffer (0.11% Nonidet P40, 0.1% sodium citrate, 0.05% spermine tetrahydrochloride, 0.01% Tris base, and 0.003% trypsin in distilled water). Heads were ground with tight-fitting pestles in the trypsin buffer. After incubation at room temperature (from 20 to 25 °C) for 10 min, 5 μL of trypsin inhibitor solution (trypsin inhibitor, from soybean, 25 mg/mL) and 1 μL of RNase A (100 mg/mL) were added. After the mixture was incubated at room temperature (from 20 to 25 °C) for 10 min, 250 μL of dilution buffer (0.11% Nonidet P40, 0.1% sodium citrate, 0.17% spermine tetrahydrochloride, and 0.01% Tris base in distilled water) was added, and the mixture was filtered through a 48 μm nylon mesh. Isolated nuclei were stained with DAPI (1 μg/mL). The filtered mixture was incubated at room temperature for at least 10 min and stored on ice until use. Stained nuclei were analyzed for DNA-DAPI fluorescence using a Cell Sorter SH800 (SONY, Japan) at an excitation wavelength of 405 nm. Each experiment was performed in triplicates.
Second, we conducted Feulgen densitometry, which have been widely used for DNA content analysis 64 . This method was used in the ploidy analysis of psyllid bacteriome cells 15 . This analysis was performed according to the protocol of Hardie et al. (2002) 64 . Briefly, the bacteriome cells and heads were dissected from an aphid and then smeared on glass slides. The smears were fixed in MFA (methanol, formalin, acetic acid = 85:10:5 v/v) for 24 h, hydrolyzed in 5.0 N HCl for 2 h, and stained with Schiff reagent for 2 h. Images of stained nuclei were captured with a BX-61 microscope (Olympus, Japan) and a DS-Fi1 CCD camera (Nikon, Japan). All the steps were performed at room temperature (from 20 to 25 °C). Using ImageJ, the green channel was extracted, and the integrated optical density (IOD) of the Feulgen stain in the nuclei was measured. Background signal intensity was measured in an area adjacent to each nucleus and deduced from the nuclear IOD. Each experiment was performed in triplicates. In this analysis, the nuclei of bacteriome cells were not isolated; therefore, we can categorize cell types based on their cytoplasmic status [Buchnera presence (bacteriocyte) or absence (sheath cells)].
Third, image-based fluorometry of the isolated nuclei was conducted. Bacteriome cells were dissected from three individuals in PBS, and the PBS droplets containing the cells were transferred onto glass slides by careful pipetting. Nuclear isolation buffer (10 μL), identical to "dilution buffer, " was added to the cells on the glass slide. The nuclei of these cells were isolated by agitation using fine needles (insect pins, Shiga Konchu Fukyusha, Japan) www.nature.com/scientificreports/ in the droplets. This step was performed under a stereomicroscope (SZ61) in order to confirm that cytoplasm was completely destroyed, and nuclei were isolated. The buffer containing the isolated nuclei was air-dried, allowing the nuclei to be tightly stuck on the glass slide. It is worth noting that in our preliminary experiments, this mounting process had the least cell loss. Then, they were fixed in MFA for 30 min at room temperature (from 20 to 25 °C). The slides were washed thrice with distilled water. Nuclei were stained with DAPI solution (1 μg/ mL DAPI and 2 mg/mL RNase A) for 1 h at room temperature (from 20 to 25 °C). The slides were washed three times with distilled water and mounted with VECTASHIELD antifade mounting medium. The same devices used for the Feulgen densitometry were used for image capturing. Using ImageJ, the blue channel was extracted, and the integrated fluorescent intensity (IFI), which is the integrated gray value in the region of interest, was measured for each nucleus. The background signal intensity was measured in an area adjacent to each nucleus and deduced from nuclear IFI. The experiment was performed in duplicates. In this analysis, the nucleolus size (area) was also measured from the image of DAPI-stained nuclei, and compared with the data from confocal microscopy to discriminate cell types (bacteriocytes and sheath cells).
Ploidy analysis for each morph, and each developmental stage of viviparous aphids. Bacteriome cells and heads were dissected in PBS and processed with the abovementioned "fluorometry" method. First, to assess the variation in the degree of polyploidy among morphs, young adult individuals (3-5 days after adult eclosion) of viviparous/oviparous females and males were used. Three individuals were pooled for each morph. Nucleolus sizes were recorded to discriminate cell types. Second, to elucidate the dynamics of polyploidization and post-embryonic development of aphids, bacteriocytes of viviparous aphids at different developmental stages were analyzed. Specifically, the following stages were used: nymphs N1, N2, N3, and N4, and adults at three distinct time points: A0 (0 days after eclosion) as pre-reproductive adults, A7 (7 days after eclosion) as actively reproducing adults, and A21 (21 days after eclosion) as senescent individuals. All viviparous individuals were apterous. Each stage of viviparous females was used on the day of molting, but teneral insects were not used.
To discriminate cell types, the nucleolus size (area) was measured from the image of DAPI-stained nuclei, and compared with the data obtained by confocal microscopy. Nucleolus sizes have also been used as indicators of the translation activity of cells 8,49 . We also recorded the size (area) of nuclei. Three individuals were used in each stage. Additionally, each developmental stage of oviparous females (nymphs N1, N2, N3, and N4, and A0 and A7 adults) was also analyzed. For discrimination of cell types in oviparous females, data from viviparous females were used, because we preliminarily confirmed that the size of the nucleolus of bacteriocytes in oviparous females was not significantly different from that in viviparous females. Three individuals were included in each stage.

Statistical analysis.
To compare the size of bacteriocytes among aphid morphs (young adults of viviparous females, oviparous females, and males), we used a linear model (LM), followed by Tukey's post hoc test. In this analysis, morphs were treated as fixed effects. For analysis of cell size among the developmental stages of viviparous females, we used linear mixed models (LMM), followed by Tukey's post hoc test. In the analysis, developmental stages and individuals were included as fixed and random effects, respectively. For analysis of the relationship between the volume of bacteriocytes and their nuclei in viviparous females, we used a LM. In the analysis, pooled data from all developmental stages of viviparous females (N1-N4 and A0, A7, A21) were used, because sample size was enough large (totally n = 181). For pairwise comparisons of ploidy levels of bacteriocytes among morphs and developmental stages of viviparous/oviparous females, we used the Brunner-Munzel test, which is a non-parametric test that adjusts for unequal variances. Significant p-values were adjusted using Bonferroni's correction. To compare the size of the nucleus and nucleolus in bacteriocytes, which was recorded during fluorometry, among ploidy classes, we used LMs and LMMs. In the analysis for nucleus size, ploidy class was treated as numerical factor and simple regression test was conducted on the log-transformed parameters. In the analysis for nucleolus size, ploidy classes were treated as factorial fixed effects and Tukey's HSD was used as a post hoc test, and individuals were included as a random effect. These analyses were performed for each morph.