Nutritional Relationship between Bemisia tabaci and Its Primary Endosymbiont, Portiera aleyrodidarum, during Host Plant Acclimation

Simple Summary Plant sap-sucking insects commonly have established mutualistic relationships with bacteria that live within their bodies and often provide nutrients that are lacking in the insect’s diet. The sweet potato whitefly (Bemisia tabaci) harbors one primary and up to seven secondary endosymbiotic bacteria. The primary endosymbiont of B. tabaci is already known to play a critical role in providing necessary nutrients for B. tabaci. Our objective was to study the relationship among B. tabaci, its primary endosymbiont, and the host plant through the effects of host plant shifting and acclimation, that is, physiological adjustments as an insect becomes accustomed to a new host plant over several generations. The results showed that host shifting from Chinese kale to cotton plants led to a decrease in the fecundity of B. tabaci in the first generation, which was restored after 10 generations of acclimation, and that its developmental time was also decreased by the tenth generation. Furthermore, essential amino acid biosynthesis genes of its primary endosymbiont were differentially regulated after B. tabaci had become acclimated to cotton plants. We speculate that the primary endosymbiont has a close nutritional relationship with B. tabaci during host plant acclimation. Abstract Plant sap-sucking insects commonly have established mutualistic relationships with endosymbiotic bacteria that can provide nutrients lacking in their diet. Bemisia tabaci harbors one primary endosymbiont, Portiera aleyrodidarum, and up to seven secondary endosymbionts, including Hamiltonella defensa and Rickettsia sp. Portiera aleyrodidarum is already known to play a critical role in providing necessary nutrients for B. tabaci. In the present study, the relationship among B. tabaci, its primary endosymbiont, and the host plant were examined through the effects of host plant shifting and acclimation. Bemisia tabaci was transferred from Chinese kale to four different host plants, and the effects on both its performance and the expression levels of nutrient-related genes of P. aleyrodidarum were analyzed. The results showed that host shifting from Chinese kale to cotton plants led to a decrease in the performance of B. tabaci in the first generation, which was restored after 10 generations of acclimation. Furthermore, the expression levels of essential amino acid biosynthesis genes of P. aleyrodidarum were found to be differentially regulated after B. tabaci had acclimated to the cotton plants. Host plant shifting and acclimation to cucumber, poinsettia, and tomato plants did not affect the fecundity of B. tabaci and the expression levels of most examined genes. We speculate that P. aleyrodidarum may help B. tabaci improve its performance and acclimate to new hosts and that P. aleyrodidarum has a close nutritional relationship with its host during host plant acclimation.


Introduction
Hemipteran insects commonly harbor endosymbiotic bacteria [1], which can be categorized into two groups. Primary endosymbionts, which can synthesize necessary nutrients lacked by their hosts, The host plants of B. tabaci used for the present study were Chinese kale, cotton (Gossypium barbadense), cucumber (Cucumis sativus cv. hualu), poinsettia (Euphorbia pulcherrima), and tomato (Solanum lycopersicum cv. ANT22), which are common host plants of B. tabaci MEAM1 from different families. Chinese kale, cotton, cucumber and tomato plants were grown from seeds, and poinsettia plants were propagated by cutting. All plants were grown at 28 ± 2 • C and under a photoperiod of L:D 16:8 h in environmental chambers. For the experiments, Chinese kale, cotton and tomato plants with 4-6 leaves, cucumber plants with 3-4 leaves, and poinsettia with 10-15 leaves were used.

Host Plant Shifting and Acclimation
Groups of 100 whitefly individuals (sex ratio 1:1) from the original colony reared on Chinese kale were transferred to insect-proof net cages (22 × 22 × 22 cm,Megaview). A total of five cages were used, each containing two plants of either Chinese kale, cotton, cucumber, poinsettia, or tomato. Two weeks later, all adult whiteflies were removed from the plants using a mouth aspirator, such that eggs and nymphs remained on the leaves. After the nymphs emerged to adults, 100 of the newly-emerged adult whiteflies were transferred to new net cages, each enclosed with two new host plants of the same species to breed the next generation. The whitefly colonies were reared on each host plant for 10 continuous generations.

Performance of B. tabaci
To determine the effects of host plant shifting and acclimation on the performance of B. tabaci, adult females were collected from the first and tenth generations on each host plant. Adult females were enclosed individually with a leaf of each host plant in a gauze bag (6 × 15 cm, 109 mesh, Megaview) for oviposition. After 72 h, the female was removed and the number of eggs counted using a stereomicroscope to determine its fecundity. The number of newly-emerged adults developing from the eggs was also recorded daily to calculate the developmental time and the survival rate of "egg and nymphal stages". The performance assay was conducted with 20 replicates per host plant for original, G1, and G10 colonies. The original colony represented the colony reared on Chinese kale. The data of original and G1 colonies were compared to examine the effect of host plant shifting, and the data of G1 and G10 colonies were compared to examine the effect of acclimation.

Expression of Nutrient-Related Genes of P. aleyrodidarum
To interpret the relationship between host plant shifting and acclimation and the expression of P. aleyrodidarum nutrient-related genes, the expression levels of essential amino acid biosynthesis genes (trpA, trpB, leuB, leuC, ilvC, and ilvD), which are encoded only in the genome of P. aleyrodidarum [26], were analyzed via a reverse transcription quantitative polymerase chain reaction (RT-qPCR) assay. Tryptophan, leucine and isoleucine biosynthesis cannot be completed by B. tabaci itself, and most essential genes in biosynthesis pathways are only present in P. aleyrodidarum. Therefore, we chose to examine these six genes from three amino acid biosynthesis pathways. For both the first and tenth generations of whiteflies, 25 males and 25 females were collected from each host plant. Each set of 5 males or 5 females was pooled to make single samples. Males and females were analyzed separately because females need more nutrients than males for reproduction [32]. The samples were then homogenized with 10 µL of RLT buffer in a 1.5-mL eppendorf tube using a micropestle. Total RNA was extracted using a RNeasy Mini Kit (Qiagen, Valencia, CA, USA) following the manufacturer's instructions. Extracted RNA was then treated with a TURBO DNA-free Kit (Invitrogen, Carlsbad, CA, USA) to eliminate residual DNA. cDNA was synthesized from 500 ng of RNA using a PrimeScript RT Reagent Kit (Takara Bio Inc., Kusatsu, Shiga, Japan) according to the manufacturer's instructions. SYBR green real-time PCR was then performed in a 20-µL reaction volume, containing 10 µL of 2 × SensiFAST SYBR Hi-ROX Mix (Bioline, London, UK), 7 µL of ddH 2 O, 0.5 µL of each primer (10 µM), and 2 µL of the cDNA template. The primer sets used were gene-specific (Table 1) and designed in the present Insects 2020, 11, 498 4 of 13 study by Primer Express Software (Applied Biosystems, Foster City, CA, USA) using the default criteria. The cycling conditions were as follows: one cycle of denaturation (95 • C, 5 min), followed by 40 cycles of denaturation (95 • C, 5 s), annealing (60 • C, 10 s), and extension (72 • C, 20 s). The PCR efficiency of trpA, trpB, leuB, leuC, ilvC, ilvD, and gyrA specific primer sets were 98.9%, 104.0%, 104.0%. 100.4%, 98.5%, 102.7%, and 105.6%, respectively. Melting curves were generated at the end of the reaction to confirm the specificity of the amplification. The relative expression levels of the essential amino acid biosynthesis genes (trpA, trpB, leuB, leuC, ilvC, and ilvD) of P. aleyrodidarum were calculated using the 2 −∆Ct method. gyrA of P. aleyrodidarum was used as a normalization gene [33]. Each experiment was conducted using pooled samples of whiteflies from each host plant in the first and tenth generations with five biological replicates.

Statistical Analyses
The performance data of B. tabaci were analyzed by Kruskal-Wallis test with post-hoc Dunn's test. The gene expression data were first performed with normality transformation (logarithm transformation) and were then analyzed by one-way analysis of variance (ANOVA). When significant differences were observed, the Tukey test was employed as a post-hoc test after ANOVA. All data analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).

Effects of Host Plant Shifting and Acclimation on the Performance of B. tabaci
After the whiteflies had been reared on new host plants for 10 generations, they were considered acclimated to the new host plant. Thus, their performance at the first and tenth generations was evaluated and compared with that of the original colony. The survival rates of "egg and nymphal stages" of the first-generation whiteflies reared on each new host plant were not significantly different to those of the whiteflies from the original colony ( Figure 1  Regarding developmental time of "egg and nymphal stages", the transfer from Chinese kale to cucumber plants resulted in a significant decrease for the first generation, which slightly lengthened by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the group transferred from Chinese kale to cotton plants, developmental time for the first generation was not significantly different to that of the original colony (Dunn's test, p > 0.05) but was significantly decreased by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the groups transferred from Chinese kale to poinsettia and tomato plants, respectively, developmental time was significantly longer in comparison to the original colony for both generations ( Figure 2, Dunn's test, p < 0.05). The fecundity of the whiteflies was significantly reduced after they had been transferred from Chinese kale to cotton plants (Dunn's test, p < 0.05) and then reverted to its original level after they had fed on cotton plants for 10   Regarding developmental time of "egg and nymphal stages", the transfer from Chinese kale to cucumber plants resulted in a significant decrease for the first generation, which slightly lengthened by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the group transferred from Chinese kale to cotton plants, developmental time for the first generation was not significantly different to that of the original colony (Dunn's test, p > 0.05) but was significantly decreased by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the groups transferred from Chinese kale to poinsettia and tomato plants, respectively, developmental time was significantly longer in comparison to the original colony for both generations ( Figure 2, Dunn's test, p < 0.05). Regarding developmental time of "egg and nymphal stages", the transfer from Chinese kale to cucumber plants resulted in a significant decrease for the first generation, which slightly lengthened by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the group transferred from Chinese kale to cotton plants, developmental time for the first generation was not significantly different to that of the original colony (Dunn's test, p > 0.05) but was significantly decreased by the tenth generation ( Figure 2, Dunn's test, p < 0.05). For the groups transferred from Chinese kale to poinsettia and tomato plants, respectively, developmental time was significantly longer in comparison to the original colony for both generations ( Figure 2, Dunn's test, p < 0.05). The fecundity of the whiteflies was significantly reduced after they had been transferred from Chinese kale to cotton plants (Dunn's test, p < 0.05) and then reverted to its original level after they had fed on cotton plants for 10 generations (  The fecundity of the whiteflies was significantly reduced after they had been transferred from Chinese kale to cotton plants (Dunn's test, p < 0.05) and then reverted to its original level after they had fed on cotton plants for 10 generations (Figure 3, Dunn's test, p < 0.05). However, for transfers from Chinese kale to the other three host plants, no significant effects were observed (Figure 3, Kruskal-Wallis test, Chinese kale: p = 0.34; cucumber: p = 0.70; poinsettia: p = 0.08; tomato: p = 0.37).
Taken together, it was found that host plant shifting and acclimation affect the performance of B. tabaci, especially on cotton plants.

Relationship between Host Plant Shifting and Acclimation and the Expression of Nutrient-related Genes of P. aleyrodidarum
Since B. tabaci depends on P. aleyrodidarum to supplement several essential amino acids [27], we hypothesized that host plant acclimation is associated with this nutritional mutualistic endosymbiont. Therefore, we examined the expression levels of some tryptophan, leucine, and isoleucine biosynthesis genes (trpA, trpB, leuB, leuC, ilvC, and ilvD) of P. aleyrodidarum for the whiteflies reared on each new host plant in the first and the tenth generations. Figures 4-7 show the expression levels after B. tabaci was transferred to cucumber, cotton, poinsettia, and tomato, respectively.
For female whiteflies transferred to cucumber plants (Figure 4), ilvC expression was significantly downregulated in tenth-generation whiteflies (Tukey test, p < 0.05). For males transferred to cucumber plants (Figure 4), trpA expression was significantly downregulated in the first generation (Tukey test, p < 0.05), and leuB expression was significantly upregulated after 10 generations (Tukey test, p < 0.05). The expression of both leuC and ilvC in tenth-generation males was significantly downregulated compared to that in the original colony (Tukey test, p < 0.05). Taken together, it was found that host plant shifting and acclimation affect the performance of B. tabaci, especially on cotton plants.

Relationship between Host Plant Shifting and Acclimation and the Expression of Nutrient-Related Genes of P. aleyrodidarum
Since B. tabaci depends on P. aleyrodidarum to supplement several essential amino acids [27], we hypothesized that host plant acclimation is associated with this nutritional mutualistic endosymbiont. Therefore, we examined the expression levels of some tryptophan, leucine, and isoleucine biosynthesis genes (trpA, trpB, leuB, leuC, ilvC, and ilvD) of P. aleyrodidarum for the whiteflies reared on each new host plant in the first and the tenth generations. Figures 4-7 show the expression levels after B. tabaci was transferred to cucumber, cotton, poinsettia, and tomato, respectively.  For female whiteflies transferred to cotton plants ( Figure 5), most genes were differentially expressed between the first and the tenth generations (Tukey test, p < 0.05). The expression of both trpA and trpB was significantly upregulated (Tukey test, p < 0.05), while that of ilvC and ilvD was significantly downregulated (Tukey test, p < 0.05). For males transferred to cotton plants ( Figure 5), there were no significant differences in the expression of all examined genes (one way-ANOVA, trpA: p = 0.83; trpB: p = 0.37; leuB: p = 0.72; ilvD: p = 0.99) except for leuC and ilvC, which were downregulated   For female whiteflies transferred to tomato plants (Figure 7), ilvC expression was significantly downregulated in the first generation (Tukey test, p < 0.05), and leuC expression was significantly downregulated after 10 generations (Tukey test, p < 0.05). For males transferred to tomato plants (Figure 7), the expression of trpA, trpB, and ilvC was significantly downregulated after being transferred to tomato plants (Tukey test, p < 0.05), and trpA expression was further downregulated after 10 generations (Tukey test, p < 0.05). In addition, leuB expression was significantly downregulated after 10 generations (Tukey test, p < 0.05), and leuC expression was significantly downregulated in the tenth generation compared to that of the original colony (Tukey test, p < 0.05). In summary, the biosynthesis of the essential amino acids provided by P. aleyrodidarum was regulated during host plant shifting and acclimation, especially when B. tabaci acclimatedwas transferred from Chinese kale to cotton plants (Figure 8).

Discussion
In the present study, we investigated the effect of transferring B. tabaci from Chinese kale to four different host plants. Since the nutrient composition and defensive compounds of the phloem sap differ in each host plant, the performance of B. tabaci may be improved or impaired. This is reflected For female whiteflies transferred to cucumber plants (Figure 4), ilvC expression was significantly downregulated in tenth-generation whiteflies (Tukey test, p < 0.05). For males transferred to cucumber plants (Figure 4), trpA expression was significantly downregulated in the first generation (Tukey test, p < 0.05), and leuB expression was significantly upregulated after 10 generations (Tukey test, p < 0.05). The expression of both leuC and ilvC in tenth-generation males was significantly downregulated compared to that in the original colony (Tukey test, p < 0.05).
For female whiteflies transferred to cotton plants ( Figure 5), most genes were differentially expressed between the first and the tenth generations (Tukey test, p < 0.05). The expression of both trpA and trpB was significantly upregulated (Tukey test, p < 0.05), while that of ilvC and ilvD was significantly downregulated (Tukey test, p < 0.05). For males transferred to cotton plants ( Figure 5), there were no significant differences in the expression of all examined genes (one way-ANOVA, trpA: p = 0.83; trpB: p = 0.37; leuB: p = 0.72; ilvD: p = 0.99) except for leuC and ilvC, which were downregulated after 10 generations (Tukey test, p < 0.05).
For female whiteflies transferred to tomato plants (Figure 7), ilvC expression was significantly downregulated in the first generation (Tukey test, p < 0.05), and leuC expression was significantly downregulated after 10 generations (Tukey test, p < 0.05). For males transferred to tomato plants (Figure 7), the expression of trpA, trpB, and ilvC was significantly downregulated after being transferred to tomato plants (Tukey test, p < 0.05), and trpA expression was further downregulated after 10 generations (Tukey test, p < 0.05). In addition, leuB expression was significantly downregulated after 10 generations (Tukey test, p < 0.05), and leuC expression was significantly downregulated in the tenth generation compared to that of the original colony (Tukey test, p < 0.05).
Insects 2020, 11, 498 9 of 13 In summary, the biosynthesis of the essential amino acids provided by P. aleyrodidarum was regulated during host plant shifting and acclimation, especially when B. tabaci acclimated to cotton plants ( Figure 8). Figure 7. The expression levels of tryptophan, leucine, and isoleucine biosynthesis genes of P. aleyrodidarum in male and female B. tabaci reared on tomato plants in the first and the tenth generations. Expression levels in the original colony are also shown. Horizontal bars indicate standard errors. Different letters indicate significant differences (Tukey test, p < 0.05).
In summary, the biosynthesis of the essential amino acids provided by P. aleyrodidarum was regulated during host plant shifting and acclimation, especially when B. tabaci acclimatedwas transferred from Chinese kale to cotton plants (Figure 8).

Discussion
In the present study, we investigated the effect of transferring B. tabaci from Chinese kale to four different host plants. Since the nutrient composition and defensive compounds of the phloem sap differ in each host plant, the performance of B. tabaci may be improved or impaired. This is reflected

Discussion
In the present study, we investigated the effect of transferring B. tabaci from Chinese kale to four different host plants. Since the nutrient composition and defensive compounds of the phloem sap differ in each host plant, the performance of B. tabaci may be improved or impaired. This is reflected in our results, whereby the developmental time of B. tabaci was significantly altered by the transfer from Chinese kale to cucumber, poinsettia, and tomato plants, with an impact on the first generation. Furthermore, the fecundity of B. tabaci was significantly reduced after being transferred from Chinese kale to cotton plants. A similar effect has also been recorded in other plant sap-sucking insects such as Ac. pisum, which displays lower fitness after it has been transferred from Vicia faba to V. villosa, Medicago sativa, and M. truncatula [34]. Zhang et al. [35] also reported that when Aphis gossypii is transferred from cotton to cowpea, pumpkin, and zucchini, the amount of Buchnera (the primary endosymbiont of Ap. gossypii) becomes unstable, and the survival rates of Ap. gossypii significantly decrease. Host plant shifting is, therefore, a challenge for both the directly-affected insects and their indirectly-affected endosymbionts; there are new diet conditions and defensive compounds that may require acclimation.
In the present study, the expression levels of selected amino acid biosynthesis genes of P. aleyrodidarum were differentially regulated when B. tabaci was transferred or acclimated to some host plants. This may be the result of the endosymbionts responding to physiological conditions in the insect host or the host manipulating its endosymbionts to correspond with its physiological conditions [36]. For example, the lysosomal system of Ac. pisum is involved with the degradation of its primary endosymbiont, Buchnera, and cereal weevils (Sitophilus spp.) also can digest and recycle their primary endosymbiont, Sodalis pierantonius, by cell apoptosis and autophagy activation after their exoskeleton is formed [37,38]. To overcome defensive compounds or to compensate for the different nutrient compositions in new host plants, B. tabaci may actively manipulate the amount and activities of its endosymbionts during host plant acclimation. The performance of B. tabaci may consequently be influenced.
Notably, the fecundity of B. tabaci was restored and the developmental time decreased after it had been reared on cotton plants for 10 generations. Also, the expression of trpA and trpB was upregulated, and that of ilvC and ilvD was downregulated after 10 generations. One possible explanation of the improved performance of B. tabaci is that it manipulates the expression of essential amino acid biosynthesis genes of P. aleyrodidarum, thus helping it to overcome the changing nutrient conditions in the phloem sap. This explanation is supported by a study revealing that none of the regulatory genes related to amino acid biosynthesis were found in P. aleyrodidarum, which suggests that it is B. tabaci, not P. aleyrodidarum, that can regulate the biosynthesis of amino acids [39]. Moreover, one generation may be insufficient for B. tabaci to respond to the different diet conditions between the phloem sap of cotton and Chinese kale, which could explain why the fecundity of first-generation whiteflies was lower than that of the original colony. Our findings suggest that the differential expression of essential amino acid biosynthesis genes of P. aleyrodidarum is positively associated with the performance of B. tabaci reared on cotton plants. Further research is warranted to investigate the mechanism of how P. aleyrodidarum improves the performance of B. tabaci on cotton plants.
The expression levels of few selected amino acid biosynthesis genes of P. aleyrodidarum were regulated after B. tabaci were transferred from Chinese kale to tomato and poinsettia for 10 generations. The differentially regulated genes varied depending on host plant and were different between males and females. Since the developmental time of B. tabaci increased after 10 generations, B. tabaci may not acclimate to poinsettia and tomato within 10 generations. The causes of differential expression between males and females are not clear. One possible reason is that females need more nutrients than males for reproduction [32]. Further research is warranted to investigate the mechanism of why the expression of the selected genes was different between males and females.
Although the expression levels of some selected amino acid biosynthesis genes of P. aleyrodidarum were significantly regulated after B. tabaci was transferred from Chinese kale to tomato, the performance of B. tabaci decreased. This suggests that nutritional regulation manipulated by B. tabaci may not be the only factor that affects whitefly performance. Other factors, such as mechanical defense and defensive compounds in plants, may also affect their performance and physiological condition. Rakha et al. [40] reported that a high density of type IV trichomes on tomato leaves reduces infestation by B. tabaci and increases their mortality, which suggests higher resistance toward whiteflies. In addition, we cannot rule out the possibility that the differences between the first and tenth generations could potentially be due to random fluctuations rather than actual acclimation.
Since endosymbionts have close relationships with their insect hosts in many aspects, the potential use of symbionts as tools to control insect pests and vector-borne diseases has received considerable attention in recent years [41]. Indeed, several strategies for controlling vector-borne plant diseases and insect pests of crops have been proposed to target insect endosymbionts [42]. Paratransgenesis, which modifies the symbiotic microorganisms by genetic transformation, is a conceivable approach. This has been shown, for example, by Arora et al. [43] in a study of Pantoea agglomerans, a symbiont of Homalodisca vitripennis that transmits Xylella fastidiosa causing Pierce's disease in grapevines. They genetically engineered P. agglomerans to express antimicrobial peptides, which successfully killed X. fastidiosa in its insect vector [43]. Therefore, a greater understanding of the interactions among B. tabaci, its endosymbionts, and the host plant may lead to novel strategies in the control of B. tabaci.

Conclusions
The present results demonstrate that host shifting from Chinese kale to cotton plants leads to a decrease in the performance of B. tabaci in the first generation, which is restored by the tenth generation. Furthermore, after B. tabaci has acclimated to the cotton plants, the expression levels of essential amino acid biosynthesis genes of its primary endosymbiont, P. aleyrodidarum, are differentially regulated. The present study suggests that P. aleyrodidarum may help B. tabaci improve its performance and acclimate to new host plants and that P. aleyrodidarum has a close nutritional relationship with its host during host plant acclimation.