Following De Novo Triglyceride Dynamics in Ovaries of Aedes Aegypti Mosquitoes

Understanding the molecular and biochemical basis of egg development is a central topic in mosquito reproductive biology. Lipids are a major source of energy and building blocks for the developing ovarian follicles. Ultra-High Resolution Mass Spectrometry (UHRMS) combined with in vivo metabolic labeling of follicle lipids with deuterated water ( 2 H 2 O) can provide unequivocal identication of de novo lipid species during ovarian development. In the present study, we followed de novo triglyceride (TG) dynamics during the ovarian previtellogenic (PVG) stage (2-7 days post-eclosion) of female adult Aedes aegypti. The incorporation of stable isotopes from the diet was evaluated using liquid chromatography (LC) in tandem with the high accuracy (<0.3 ppm) and high mass resolution (over 1M) of a 14.5 Tesla Fourier Transform Ion Cyclotron Resonance Mass Spectrometer (14.5 T FT-ICR MS) equipped with hexapolar detection. LC-UHRMS provides effective lipid class separation and chemical formula identication based on the isotopic ne structure. The monitoring of stable isotope incorporation into de novo incorporated TGs suggests that ovarian lipids are consumed or recycled during the PVG stage, with variable time dynamics. These results provide further evidence of the complexity of the molecular mechanism of follicular lipid dynamics during oogenesis in mosquitoes.


Introduction
Nutrient availability and their allocation towards life processes play a critical balance during insect life cycle. [1][2][3] For example, lipid allocation follows these trade decisions: female mosquitoes mobilized lipids towards the ovaries as a source of oocyte maturation while been also essential for energy homeostasis. [4][5][6] In Aedes aegypti mosquitoes, the ovary development and energy allocations during a gonotrophic cycle are typically divided in three major periods: previtellogenesis (PVG), ovarian resting stage (ORS) and vitellogenesis (VG). 7,8 Lipids and glycogen are the primary energy reserves for egg development during the immature stages. 5,9,10 These teneral reserves are partially consumed during the PVG period; moreover, nectar-feeding adds critical reserves during the ORS, and a blood meal triggers VG. 8,[10][11][12][13] Previous studies revealed in Aedes aegypti, more than 80% of lipids found in eggs originate from sugars consumed before a blood meal. 14,15 Sugar-feeding is a critical source of nutrients during the PVG stage. Sugar digestion starts in the crop, from which part of the meal is transferred periodically into the midgut ( Figure 1). [16][17][18] Enzymes from the saliva ingested with the sugar meal and the midgut convert sucrose in glucose and fructose. 16,19 The biosynthesis of lipids begins in the midgut; sugars are used as precursors for fatty acid (FA) and triglycerides (TG) synthesis. 20,21 Experiments using radiolabeled sugars con rmed that they are utilized as substrate for TG synthesis. 22,23 While TG is the principal lipid produced in the fat body [24][25][26] , most of the lipids found in the hemolymph are in the form of diglycerides (DG). 24,25,27 The transport of lipids among tissues is mediated by Lipophorin (Lp) and the Lipid Transfer Particle (LTP). 16,18,28,29 Previous studies suggested that the ovaries can only synthesize small amounts of lipids 30,31 , therefore the majority of ovarian lipids must be produced elsewhere (e.g., fat body). 15,16,27,32 While TG reserves can be carried from the larva stage, de novo TG synthesis play undoubtedly a key role in Ae. aegypti oogenesis. 14,26,33 In the present work, we studied ovarian TG dynamics during the PVG stage (2-7 days post-eclosion) of sugar-fed female Ae. aegypti. Stable isotopes from deuterated water were incorporated into the mosquito sugar diet, and TGs were detected using liquid chromatography coupled to Ultrahigh Resolution Mass Spectrometry (LC-UHRMS). 28,[34][35][36][37] When a sugar diet with 2 H 2 O was provided (Figure 1), de novo synthesized TGs were labeled with 2 H along the fatty acyl tails. [38][39][40] This procedure permitted the identi cation of de novo synthesized TGs, as well as the analysis of the dynamics of ovary TG incorporation. A 14.5 T Fourier transform ion cyclotron resonance mass spectrometer (FT-ICR MS) equipped with hexapolar detection was utilized for effective isotope separation and chemical composition assignment. TGs were classi ed based on the fatty acid length and number of unsaturated bonds. Their relative abundances and degree of stable isotope incorporation (e.g., deuterium) were measured as a function of the time after adult eclosion.

Mosquito rearing conditions
Aedes aegypti of the Rockefeller strain were reared at 28 °C and 80% relative humidity. After eclosion, adult female mosquitoes were fed two different diet regimens: a) 20% Sucrose/ultrapure water (HPLC); b) 20% Sucrose/heavy water ( 2 H 2 O). During the experiments, insects were fed daily by wetting a 1 × 1 inch cotton pad with each individual diet. Containers were loosely covered with polyethylene wrap to prevent rapid evaporation of the feeding pad.

Results
De novo synthesis of ovarian lipid reserves.
The use of stable isotope labelling combined with ultrahigh resolution mass spectrometry has shown signi cant advantages for the analysis of biological pathways. 6,26,28,41,42 Different from other labelling techniques, the incorporation of stable isotopes does not change the chemical properties of the lipid and allows the detection of the labeled molecule by their isotopic pro le. This approach enables direct analysis of nutrient distribution, mobilization and metabolism. 4,23,37,[43][44][45] While the concept is simple, for the analysis of complex biological samples it requires: i) the use of complementary pre-separation techniques to diminish matrix effects and increase the sensitivity of the analysis 27,37,46-53 and ii) the use of high magnetic elds to achieve ultrahigh mass resolution with short transient duration for compatibility with LC separation (e.g., resolving power greater than m/Dm 50% > 1000000 and mass accuracy better than 1 ppm). 27,36,[41][42][43][46][47][48] More recently, in addition to the use of higher magnetic elds, alternative detection strategies (e.g., hexapolar detection, 3W+) has allowed better sensitivity and shorter analysis time, making more e cient the coupling of LC-FT-ICR MS. 41 Table 1 displays a summary of the TGs detected using LC-FT-ICR MS with a mass accuracy lower than 300 ppb. TGs extracted from ovaries of females fed 2 H 2 O labeled sugar diets revealed similar EIC pro les; although, the addition of 2 H creates distinct isotopic patterns ( Figure S1). The use of ultrahigh resolution FT-ICR MS allowed the observance of the isotopic ne structure. This permits direct discrimination of the signals containing 2 H from that of naturally occurring isotopes (e.g., 13 C).
The de novo synthesized TGs have the characteristic presence of 2 H in their elemental composition (Figure 3). For example, the mass pro le of de novo synthesized TG 48:2 showed the substitution of 30 hydrogens by deuterium atoms ( Figure 3A). These substitutions were absent in ovarian samples from females that received a non-labeled sucrose diet ( Figure 3B). A closer inspection to the 820 -834 m/z range highlights the importance of ultrahigh resolution MS for this type of analysis ( Figure 3C) The trends of changes and quantities of TG were similar in ovaries from females raised on non-labeled diet and labeled diet (Figure 4). In good agreement with previous reports, TGs containing 48, 50, 52, and 54 carbons were the most abundant species. 27 As a general tendency, the amount of TGs increased with the days after eclosion, reaching a steady state towards the last days. Most signi cantly, deuterium incorporation was observed in all TGs species, but with unique dynamics for each of them. In terms of total lipid amounts, three-quarters of TG lipids were labeled during the six-day experiment, with the percentage of 2 H incorporation increasing as a function of time (Figures 4 and S2).

Dynamics of de novo TG incorporation into the ovary
The investigation of the dynamic changes of ovarian lipids during the PVG period was assisted by the incorporation of stable isotopes, providing an additional dimension of absolute quantitative values. The analysis of the number of deuterium atoms incorporated into eighteen different TG species at different days after adult eclosion, indicated a dynamic constant adjustment of ovarian TG stores during the experiment. In Figure 5, we display the distribution of the number of deuterium incorporated into each TG as a function of time.

Discussion
Oogenesis is energetically costly. Ae. aegypti females can lay over 120 eggs in a gonotrophic cycle; therefore, a tightly regulated control of nutrient allocations to the ovaries is critical. 5,7,8,10,27 The ovarian resting stage in Ae. aegypti, is a period marked by constant adjustment of the reproductive output based on nutritional and hormonal status; this adjustment typically occurs through follicular resorption by apoptosis. 5,7,30 The link between an increased oocyte lipid content and a successful reproductive output has been reported in invertebrate and vertebrate systems. 11,13,23 In this study, we started to address a fundamental question: "Can we use stable isotopes to establish time pro les of TGs incorporation into the developing ovarian follicles (composition, abundance and dynamic changes), which could lead us to a better understanding on the effect of nutrients on the dynamic importation of follicular lipids during oogenesis? " In the current experiments we focused in addressing the issue of the distinct contributions of teneral reserves (those already carried at adult eclosion) and adult sugar feeding to the TGs imported into the ovarian follicles during the previtellogenic stage. To discriminate between these two nutritional sources, we labeled the lipids that were de novo synthesized using a 2 H 2 O labeled sugar diet. Taking the advantage of stable isotope labelling; our studies con rmed that a signi cant portion of the ovary lipids were isotopically labeled with 2 H, indicating an active de novo TG synthesis and transport into the ovaries. The dynamic changes on the ovarian TG lipid pro les suggested that ovarian TG lipids were utilized and replaced by de novo synthesized lipids. An increase of 2 H-labelling was correlated with increases in carbon number, molecular size, and number of unsaturated bonds of TGs. Overall, the replacement of hydrogen atoms by deuterium is also in uence by enzymatic activities and kinetic parameters.
Based on lipids dynamics, labeled 2 H-TGs were incorporated into the ovaries during the entire experiment, but the total lipid amount was either reduced or stable by day 7, suggesting that some of the follicles might have been resorbed.
In summary, our studies showed de novo TG lipid mobilization and storage into the developing ovaries. The use of a stable isotope labeled-sugar diet enabled the dynamic study of lipid incorporation into ovarian follicles at the TG species level. The LC-UHRMS work ow provides the isotopic pro le of characteristic TG species and the number of deuterium incorporated. Multiple TG species were detected in the 800 to 900 m/z range with a mass accuracy lower than 300 ppb.