Mass spectrometry-based stable-isotope tracing uncovers metabolic alterations in pyruvate kinase-deficient Aedes aegypti mosquitoes

Abstract

A recent in vitro characterization of a recombinant pyruvate kinase (PK) from Aedes aegypti mosquitoes demonstrated that the enzyme is uniquely regulated by multiple allosteric effectors. Here, we further explored PK gene and protein expression, and enzymatic activity in key metabolic tissues of mosquitoes maintained under different nutritional conditions. We also studied the metabolic effects of PK depletion using several techniques including RNA interference and mass spectrometry-based stable-isotope tracing. Transcriptional analysis showed a dynamic post-feeding PK mRNA expression pattern within and across mosquito tissues, whereas corresponding protein levels remained stable throughout the time course analyzed. Nevertheless, PK activity significantly differed in the fat body of sucrose-, blood-fed, and starved mosquitoes. Genetic silencing of PK did not alter survival in blood-fed females maintained on sucrose. However, an enhanced survivorship was observed in PK-deficient females maintained under different nutritional regimens. Our results indicate that mosquitoes overcame PK deficiency by up-regulating the expression of genes encoding NADP-malic enzyme-1, phosphoenolpyruvate carboxykinase-1, phosphoglycerate dehydrogenase and glutamate dehydrogenase, and by decreasing glucose oxidation and metabolic pathways associated with ammonia detoxification. Taken together, our data demonstrate that PK confers to A. aegypti a metabolic plasticity to tightly regulate both carbon and nitrogen metabolism.

Graphical Abstract

1. Introduction

Aedes aegypti is the main vector of viruses that cause diseases such as Zika and dengue fever, which present major challenges to the public health (Lessler et al., 2016; Weaver et al., 2018; Messina et al., 2019). Because there are no specific treatments or effective vaccines currently available, the prevention and control of these diseases depends entirely on controlling the mosquito vectors. However, it is extremely difficult to control A. aegypti due to its ability to adapt to human environmental conditions (Lounibos and Kramer, 2016; Achee et al., 2019). Given these challenges and the failure of current methods to effectively control mosquito populations, a comprehensive understanding of mosquito metabolism is needed for the development of better vector mitigation strategies.

As an anautogenous mosquito species, female A. aegypti needs to obtain at least one blood meal to produce eggs. Previous studies have demonstrated that blood-fed A. aegypti utilizes several metabolic pathways to detoxify ammonia, a by-product of amino acid (AA) oxidation (Scaraffia et al, 2005; Scaraffia et al., 2006; Scaraffia et al, 2008; Scaraffia et al., 2010; Isoe and Scaraffia, 2013; Mazzalupo et al., 2016; Petchampai and Scaraffia, 2016; Isoe et al., 2017). Recently, a positional stable-isotope tracer analysis revealed a tight link between glucose and ammonia metabolism in A. aegypti. Specifically, several metabolites are synthesized from the carbon skeleton of glucose to facilitate ammonia detoxification and nitrogen waste disposal in blood-fed mosquitoes (Horvath et al., 2018). The synthesis of these metabolites occurs through multiple metabolic pathways including glycolysis, pentose phosphate pathway (PPP), Krebs cycle, and ammonia fixation, assimilation and excretion pathways (Horvath et al., 2018).

The final and rate-limiting step of glycolysis is catalyzed by pyruvate kinase (PK, EC 2.7.1.40), an enzyme that catalyzes the transfer of a phosphate group from phosphenolpyruvate to ADP, yielding pyruvate and ATP. Recently, two spliced variants of A. aegypti PK, designated as AaPK1 and AaPK2, were identified in the A. aegypti genome. The three-dimensional structure and kinetic properties of recombinant AaPK1 were also reported (Petchampai et al., 2019). In spite of the similarity at the AA sequence and structural levels with the human non-allosteric isoform of PK (PKM1), AaPK1 exhibited allosteric behavior (Petchampai et al., 2019). The allosteric nature of AaPK1 suggests that the enzyme has the ability to respond to different metabolic signals. The unique regulatory property of AaPK1 observed in in vitro assays encouraged us to study the total PK (both PK1 and PK2 isoforms) in vivo, using several techniques including RNA interference (RNAi), high-resolution accurate-mass (HRAM) liquid chromatography-mass spectrometry (HRAM-LC/MS) and HRAM ion chromatography-mass spectrometry (HRAM-IC/MS) methods. Our present data reveal that PK is modulated in the fat body in response to the nutritional status of the females. Survival of PK-deficient mosquitoes is dependent on whether females are maintained on sucrose, water, blood/sucrose or blood/water diets. Depletion of PK by RNAi significantly increased transcript levels of several genes that encode enzymes related to glucose and nitrogen metabolism, and impacted glucose oxidation and ammonia metabolism at specific time points during blood meal digestion. Our data provide evidence that PK plays a key regulatory role in the metabolic homoeostasis of A. aegypti females.

2. Materials and methods

2.1. Reagents, chemicals, antibodies, and others

All the primers, and chemicals for PK activity assays were acquired from Millipore Sigma (Burlington, MA, USA). Oligo-(dT)₂₀ primer, reverse transcriptase, and GoTaq® DNA Polymerase were from Promega (Madison, WI, USA). PerfeCTa SYBR Green FastMix was obtained from Quanta BioSciences (Gaithersburg, MD, USA). TRIzol® reagent was from Thermo Fisher Scientific (Waltham, MA, USA). Bovine blood was from Pel-Freez Biologicals (Rogers, AR, USA). [1,2-¹³C₂]-glucose was from Cambridge Isotope Laboratories, Inc. (Tewksbury, MA, USA). Optima LC/MS-grade water, acetonitrile (ACN), and formic acid (FA) were from Thermo Fisher Scientific. A combined AA Standard H solution was from Thermo Fisher Scientific. All other authentic reference standards and chemicals were from Millipore Sigma. The 3-micron Intrada AA analytical column (2.1 x 150 mm) was from Imtakt USA (Portland, OR, USA). The 5-micron SeQuant ZIC®-pHILIC LC analytical column (2.1 x 150 mm) was from Millipore Sigma. The 4-micron Dionex IonPac AG-11-HC guard column (2 x 50 mm) and 4-micron IonPac AS11-HC analytical IC column (2.0 x 250 mm) were from Thermo Fisher Scientific. The Vanquish Ultra-High Performance Liquid Chromatography system, the Dionex ICS-5000+ capillary IC system, and the HRAM Orbitrap Fusion Mass Spectrometer were from Thermo Fisher Scientific. Custom-made rabbit polyclonal antibody against AaPK was from GenScript USA Inc. (Piscataway, NJ, USA). Rabbit polyclonal antibody against A. aegypti serine protease (Aa5G1) was provided by Dr. Miesfield (The University of Arizona, Tucson, AZ, USA). Aa5G1 was originally identified as one of the major serine proteases by Kalhok et al., (1999). Anti-α-tubulin was purchased from Developmental Studies Hybridoma Bank (The University of Iowa, Iowa City, IA, USA). Secondary antibodies, including IRDye 800CW goat anti-rabbit and IRDye 680RD donkey anti-mouse, were from LI-COR Biosciences (Lincoln, NE, USA).

2.2. Insect rearing, feeding, and microinjection of double-stranded RNA (dsRNA)

An A. aegypti colony (NIH-Rockefeller strain; National Institutes of Health, Bethesda, MD, USA) was reared and maintained as previously described (Isoe and Scaraffia, 2013; Mazzalupo et al., 2016). Newly emerged female mosquitoes were used for dsRNA microinjection or maintained on 3% sucrose for 4 days before being used in other experiments. To determine the levels of PK mRNA, protein, and enzymatic activity in response to a sucrose solution or a blood meal, 4-day-old mosquitoes were fed only on 3% sucrose or a bovine blood meal and maintained on 3% sucrose. For quantitative real-time PCR (qPCR) and western blotting experiments, thorax, fat body, midgut, ovaries, and Malpighian tubules were dissected from sucrose-and blood-fed female mosquitoes at different times after feeding. For PK activity assays, mosquito tissues were collected at 24, 48, and 72 h post blood meal (PBM). To examine the effect of sucrose deprivation on PK activity, 4-day-old female mosquitoes were fed the following diets: 1) 3% sucrose, 2) water, 3) a blood meal, and maintained on 3% sucrose after blood feeding, and 4) a blood meal, and maintained on water after blood feeding. Fat body and thorax were collected at 72, 96, and 120 h after feeding and stored at −20°C until used. For dsRNA microinjection, newly emerged female mosquitoes were injected with 1 µg of dsRNA targeting PK (dsRNA-PK), and firefly luciferase (dsRNA-FL) as a control using a Nanoject II microinjector (Drummond Scientific Company, Broomall, PA, USA) as previously described for other genes (Scaraffia et al., 2008). After injection, the females had access to 3% sucrose for 4 days. Four groups of 4-day-old injected female mosquitoes were then fed different diets as described above, and survival was assessed daily for 25 days. For HRAM-LC/MS and HRAM-IC/MS based carbon-13 isotope tracer metabolite analysis, 4-day-old dsRNA-injected females were fed a blood meal supplemented with 100 mM [1,2-¹³C₂]-glucose as previously described in non-injected wild-type A. aegypti mosquitoes (Horvath et al., 2018).

2.3. qPCR

Transcriptional profiles of PK in mosquito tissues were determined by qPCR as previously described for other genes (Isoe and Scaraffia, 2013). Briefly, total RNA was extracted using TRIzol® Reagent. cDNA was then synthesized using oligo-(dT)₂₀ and reverse transcriptase. Relative abundance of each gene examined was normalized using ribosomal protein S7 expression level (Table S1) as previously described (Isoe and Scaraffia, 2013). In addition, the transcriptional levels of genes encoding glucose-6-phosphate dehydrogenase (G6PDH), NADP-malic enzyme-1 (NADP-ME1), phosphoenolpyruvate carboxykinase-1 (PEPCK1), phosphoglycerate dehydrogenase (PHGDH), glutamate dehydrogenase (GDH), alanine aminotransferase-1 (ALT1), ornithine decarboxylase (ODC), and xanthine dehydrogenase-1 (XDH1) in response to PK knockdown were determined in fat body and thorax at 72 h PBM. Primer sequences for qPCR and dsRNA synthesis are shown in Table 1.

Table 1.

Gene-specific primers used for RNAi and qPCR

Genes	Accession No.		Primer sequence (5’ to 3’)
Gene-specific primers used for RNAi
Pyruvate kinase, PK	AAEL014913	Forward	GATGGTCTGGGTGTCCGATTA
		Reverse	CTCAACCGTACAGGTCAGAGTG
Firefly luciferase, FL	U47295	Forward	AGCACTCTGATTGACAAATACGA
pGL3-basic vector		Reverse	AGTTCACCGGCGTCATCGTC
Gene-specific primers used for qPCR
Pyruvate kinase, PK	AAEL014913	Forward	CTGGTGGTTGACAGTATTAGCGG
		Reverse	CTTATCCTTTTCGGAGACGGC
Phosphoenolpyruvate carboxykinase-1, PEPCK1	AAEL000006	Forward	GGTCAACTGCGAGCCATCAA
		Reverse	GAGGCGACGTTGGTAAAGAG
Phosphoglycerate dehydrogenase, PHGDH	AAEL005336	Forward	CTGATTTCGGCTACCACGTT
		Reverse	CTCCACCGCATTGACCAGT
NADP-dependent malic enzyme-1, NADP-ME1	AAEL005790	Forward	GTCATGGCTATGCGACAGGA
		Reverse	CCAGAGATACCTCCCTTCG
Ornithine decarboxylase, ODC	AAEL007880	Forward	CGCAGCATGAACCTAGACGT
		Reverse	TGCTTGGCGTAGTCGAACAG
Alanine aminotransferase-1, ALT1	AAEL028144	Forward	CAACTGCCGGAGAAGGCAAT
		Reverse	TGATAGGTTCCATCCTTCTGACC
Xanthine dehydrogenase-1, XDH1	AAEL002683	Forward	GCGATTGACATTGGACAGATCGA
		Reverse	CCAGGAATGTCGGCGAAACC
Glutamate dehydrogenase, GDH	AAEL010464	Forward	GGCGAGAACCTGATGTACGA
		Reverse	GAGCAGGTGGTAGTTGGACT
Glucose-6-phosphate dehydrogenase, G6PDH	AAEL009507	Forward	TCACAGATGCACTTTGTCCGAGC
		Reverse	GGTCCACGAGAACCGTGCA
Ribosomal protein S7	AAEL009496	Forward	ACCGCCGTCTACGATGCCA
		Reverse	ATGGTGGTCTGCTGGTTCTT

T7 promoter sequence (5’ TAATACGACTCACTATAGGGAGA 3’) was added at 5’-end of each RNAi primer.

2.4. Western blotting

Total proteins were extracted from thorax, fat body, midgut, ovaries, and Malpighian tubules as previously described (Isoe and Scaraffia, 2013; Mazzalupo et al., 2016). Cytosolic proteins were isolated from fat body and thorax using Mitochondria Isolation Kit for Tissue (Thermo Fisher Scientific) according to the manufacturer’s instructions with minor modification. Approximately 200 μg of each mosquito tissue was homogenized using Model 125 Fisher Scientific™ Laboratory Homogenizer (Thermo Fisher Scientific). Proteins were stored at −80°C until used. The concentration of proteins was determined using Pierce™ Microplate BCA Protein Assay Kit according to the manufacturer’s protocol (Thermo Fisher Scientific). The proteins (20 µg) were resolved by SDS-PAGE on 4–15% gradient gels (Bio-Rad, Hercules, CA, USA) and subjected to western blotting as previously described (Isoe and Scaraffia, 2013). The dilutions of the primary antibodies were as follows: custom-made AaPK antibody and α-tubulin antibody (1:2,000), and Aa5G1 (1:1,000). The secondary antibodies, IRDye 800CW goat anti-rabbit and IRDye 800CW goat anti-mouse, were diluted at 1:10,000. The protein bands were visualized with an Odyssey infrared imaging system (LI-COR Biosciences) and densitometry was performed with the ImageJ software (Schneider et al., 2012).

2.5. PK activity assays

Mosquito tissues were homogenized in 5 parts (w/v) of chilled 20 mM triethanolamine (pH 7.5) using a hand-held pestle. The suspension was centrifuged twice at 14,000 g for 5 min. The supernatant was used to perform PK enzymatic assays as previously described (Bakszt et al., 2010; Petchampai et al., 2019).

2.6. Metabolite analysis in whole body and excreta of mosquitoes injected with dsRNA-FL and dsRNA-PK

2.6.1. Analyte stocks, intermediates, and calibration standard preparations.

Individual stock solutions for Gln (not contained in the combined AA Standard H solution) were prepared in 0.1% FA in water. Krebs cycle, PPP, and glycolytic metabolite stock solutions were prepared in Milli-Q water. A uric acid stock solution was prepared in 1 M NaOH. The nominal concentration for each stock solution was typically 50 mM or 1 mg/mL (PPP and glycolytic metabolites), and corrections were made for salt form, purity and water content reported in the product literature. Stock solutions were stored at −80°C when not in use. Individual sets of calibration standards (calibrators) were prepared as follows: i) calibrators for AAs were prepared at 0.40, 2, 10, and 50 µM in 90/10 ACN/water and 0.1% FA; ii) calibrators for organic acids (OAs), except uric acid, were prepared at 0.39, 0.78, 1.56, 3.13, 6.25, 12.5, 25, and 50 µM in 90/10 ACN/water and 0.1% FA; iii) calibrators for uric acid were prepared at 0.8, 4, 20, and 100 µM in 0.1 M NaOH; and, iv) calibrators for glycolytic and PPP metabolites were prepared at 0.078, 0.313, 1.25, 5, and 20 µg/mL in Milli Q water.

2.6.2. Sample reconstitution procedures.

For AA and OA determinations (except uric acid), dried mosquito whole body samples were prepared as previously described (Horvath et al., 2018), reconstituted in 100 µL or 50 µL of 90/10 ACN/water and 0.1% FA, respectively, vortex-mixed briefly, centrifuged at 17,000 g for 2 min, and the supernatants were transferred into polypropylene autosampler vials. A 10 µL volume of sample extract was injected on the HRAM-LC/MS system. The final dilution factor for the mosquito whole body samples was 20-fold for the AAs, and no dilution for the OAs. For uric acid determinations, dried mosquito whole body samples were reconstituted in 200 µL of 0.1 M NaOH, vortex-mixed briefly, at 17,000 g for 2 min, and the supernatants were transferred into polypropylene autosampler vials. Dried mosquito excreta samples were collected as previously described (Isoe et al., 2017), reconstituted in 1.5 mL of 0.1 M NaOH, vortex-mixed for 5 min, and heated to 105°C for 5 min. After cooling, the excreta samples were centrifuged at 17,000 g for 5 min. The supernatants were diluted 100-fold in 0.1 M NaOH, and transferred into polypropylene autosampler vials. A 5 µL volume of mosquito whole body and excreta extracts were injected on the HRAM-IC/MS system. The final dilution factor for the mosquito whole body samples was 40-fold. For Krebs cycle and glycolytic metabolite determinations, dried mosquito whole body samples were reconstituted in a 50 µL volume of Milli-Q water, vortex-mixed briefly, centrifuged at 17,000 g for 2 min, and the supernatants were transferred into polypropylene autosampler vials. A 5 µL volume was injected on the HRAM-IC/MS system without dilution.

2.6.3. Targeted HRAM-LC/MS methods for measuring carbon-13 isotopologs of AAs, and OA metabolites.

The targeted HRAM-LC/MS methods used to quantify the carbon-13 labeled isotopologs of the AAs and OA metabolites were described previously (Horvath et al., 2018). Although HRAM-LC/MS method was used to measure uric acid and its carbon-13 labeled isotopologs previously (Horvath et al., 2018), we found the HRAM-IC/MS system to be better suited for the measurement of these compounds. This method is described in Section 2.6.4.

2.6.4. Targeted HRAM-IC/MS method for measuring carbon-13 isotopologs of Krebs cycle, PPP, glycolytic and uric acid metabolites.

Briefly, the targeted method for the measurement of the carbon-13 labeled isotopologs of Krebs cycle, PPP, glycolytic and uric acid metabolites was performed using an IC separations module coupled to a HRAM OrbiTrap Fusion MS system operated in negative ion mode with IonPac AG-11-HS guard and IonPac AS11-HS analytical columns installed. The Orbitrap Fusion was operated in full scan mode (m/z 80 – m/z 800) at a resolution of 240,000 (at m/z 200). A full description of the HRAM-IC/MS method parameters is provided in Supplemental Materials.

2.7. Statistical analysis

One-way ANOVA, and 2-way ANOVA were used to determine the differences in PK mRNA levels, PK activity, metabolite concentration and relative abundance. The statistical analysis of densitometric values, and gene expression in response to PK knockdown were performed with unpaired Student’s t test. The Kaplan-Meier method with the Gehan-Breslow-Wilcoxon test was used to evaluate the significance of differences in mosquito survivorship. All statistical analyses were conducted using GraphPad Prism version 6.0 for Mac OS X (GraphPad Software, La Jolla, CA, USA). A value of P < 0.05 was considered significant. All experiments were replicated at least 3 times with at least 3 independent samples.

3. Results

3.1. PK mRNA level is up-regulated in response to blood feeding

To assess the transcriptional profiles of PK in response to sucrose and blood feeding, we performed qPCR in mosquito tissues before and after blood feeding (Fig. 1). The expression of PK was up-regulated in response to blood feeding. In the fat body, the level of PK mRNA dramatically increased ~200 fold at 6–12 h PBM and decreased considerably thereafter. At 18–48 h PBM, PK mRNA abundance was relatively constant, but significantly higher (~60–80 fold) than sucrose-fed mosquitoes. At 72–96 h PBM, PK transcript abundance was not significantly different compared to that of mosquitoes prior to blood feeding. In the midgut and thorax, PK transcripts were induced at 12 h PBM. In both tissues, the transcripts remained significantly greater than control (~3–11 fold) until 48 h PBM, and subsequently they gradually decreased. Unlike the other tissues, the levels of PK transcripts in ovaries and Malpighian tubules were stable until 36 h PBM. In the ovaries, the transcripts reached a maximum at 72 h PBM and remained significantly higher at 96 h PBM than sucrose-fed mosquitoes. In the Malpighian tubules, however, PK mRNA had a peak at 48 h PBM and then returned to that of pre-blood feeding level at 96 h PBM (Fig. 1). To evaluate PK protein expression, we analyzed mosquito tissues by western blotting using anti-AaPK antibody. Tissues were dissected before and after feeding (3, 6, 12, 18, 24, 36, 48 and 72 h PBM). Although changes in mRNA expression were observed in all mosquito tissues examined (Fig. 1), the PK protein expression was relatively stable over the 72 h after blood feeding (Fig. 2). These results are different from those reported by Hou et al., (2015), in which an approximately 5-fold increase of PK protein expression was observed in A. aegypti fat body at 36 h after blood feeding, while the PK transcript level was up-regulated over 200-fold.

Figure 1. Transcriptional profiles of pyruvate kinase (PK) in Aedes aegypti tissues.

Tissues were dissected and analyzed individually from sucrose (S)-and blood-fed females at the indicated time points during the first gonotrophic cycle. PK mRNA levels were normalized to mRNA levels of the ribosomal protein S7. Data are presented as mean ± SEM of 10 mosquitoes (*P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 when compared to S).

Figure 2. Pyruvate kinase (PK) protein expression patterns in Aedes aegypti tissues.

Tissues were dissected from sucrose (S)-and blood-fed females at different time points during the first gonotrophic cycle. PBM: post blood meal. PK protein levels were analyzed by western blotting using an anti-A. aegypti PK (AaPK) antibody. Each lane contains 0.5 tissue equivalent of the protein extracts. Anti-α-tubulin antibody was used as an internal control for protein loading. Western blots are representative of 3 replicates.

3.2. PK activity is modulated in fat body in response to nutritional changes

To determine whether PK activity changes depend on the nutritional status of A. aegypti, we first examined the activity of the enzyme in different tissues. A schematic representation of the experimental time course for feeding conditions is shown in Fig. 3A. In the fat body and Malpighian tubules, PK activity decreased at 24–48 h PBM and returned to that of control at 72 h PBM (Fig. 3B and Fig. S1A). In the thorax and ovaries, PK activity remained stable throughout the time course analyzed (Fig. 3C and Fig. S1B). Next, we measured PK activity in the fat body and thorax of non-blood-and blood-fed mosquitoes that were provided with sucrose or starved with access to water. In the fat body of the non-blood-fed females, PK activity significantly increased in females starved for 72 h (Fig. 3D) whereas the enzymatic activity decreased in blood-fed mosquitoes that were supplied with water for 96 and 120 h (Fig. 3D). Nevertheless, PK activity in the thorax did not significantly change in response to nutritional alteration compared to control (Fig. 3E). To assess PK protein expression in non-starved and starved mosquitoes, we performed western blotting. As shown in Figs. 3F–G, PK protein expression increased in the fat body of females starved for 72 h, whereas PK expression remained unchanged in the thorax. These results indicate that PK is regulated in a tissue-specific manner.

Figure 3. Pyruvate kinase (PK) activity and protein expression in fat body and thorax of non-starved and starved Aedes aegypti mosquitoes.

A) Schematic representation of experimental time course for feeding conditions. B-C) PK activity in the fat body and thorax dissected from sucrose (S)-and blood/sucrose (B-S)-fed mosquitoes. D-E) PK activity in the fat body and thorax of sucrose (S)-, water (W)-, blood/sucrose (B-S)-, and blood/water (B-W)-fed mosquitoes. F-G) PK cytosolic protein expression, and protein band intensity in fat body and thorax of sucrose (S)-and water (W)-fed females. Levels of PK protein were analyzed by western blotting using an anti-A. aegypti PK (AaPK) antibody. Anti-α-tubulin antibody was used as a loading control. The protein band intensity was measured using the ImageJ software (Schneider et al., 2012). For PK activity assays, data are presented as mean ± SEM of three independent experiments (*P < 0.05, **P < 0.01, ***P < 0.001 and ****P < 0.0001 when compared to control). For western blotting, data represent the mean fold change ± SEM of three independent experiments (*P < 0.05 when compared to S).

3.3. Genetic knockdown of PK impacts nitrogen and carbon metabolism in A. aegypti females

To better understand the functional role of PK in A. aegypti females, we used dsRNA-mediated RNAi to silence the gene encoding this enzyme. A validation of the PK knockdown was carried out by western blotting probed with A. aegypti-specific AaPK primary antibody using fat body and midgut from individual mosquitoes (Fig. S2). Tissues were dissected from sucrose-fed mosquitoes at day 5 post dsRNA-FL or dsRNA-PK injection, and from blood-fed mosquitoes at day 3 and 7 PBM (day 7 and 11 post dsRNA injection, respectively). The single mosquito tissue analysis showed that levels of PK protein in RNAi-PK mosquitoes were significantly reduced or depleted in all samples examined compared to RNAi-FL control mosquitoes, showing the durable effect of RNAi on mosquito fat body and midgut tissues (Figs. S2A–B).

Next, we monitored survival of mosquitoes injected with dsRNA-FL or dsRNA-PK. A schematic representation of the experimental time course for dsRNA injections and feeding conditions is shown in Fig. 4A. We found that silencing of PK significantly enhanced survivorship of mosquitoes fed on 3% sucrose, water, or a blood meal followed by water (Fig. 4B, 4C, 4E). However, RNAi-mediated PK reduction in blood-fed females that were continuously supplied with sucrose did not affect mosquito survival (Fig. 4D). It is interesting to note that mosquitoes (both non-blood-and blood-fed) that were provided with water (Fig. 4C, 4E) had a shorter life-span than sucrose-fed mosquitoes (Fig. 4B, 4D). In sucrose-fed mosquitoes, mortality rates of dsRNA-FL and dsRNA-PK injected females reached 100% at 21-and 25-days post injection, respectively (Fig. 4B). The RNAi-FL and RNAi-PK mosquitoes that were water-fed all died at 11-and 14-days post injection, respectively (Fig. 4C). For blood-fed mosquitoes that were maintained on 3% sucrose, a 100% mortality was observed at 23-and 25-days post dsRNA-FL and dsRNA-PK injection, respectively (Fig. 4D). In addition, percent survival of the blood-fed females that were kept on water decreased to 0% at 12-and 15-days post dsRNA-FL and dsRNA-PK injection, respectively (Fig. 4E).

Figure 4. Effect of pyruvate kinase (PK) deficiency on Aedes aegypti survival.

A) Schematic representation of experimental time course for dsRNA injections and feeding conditions. Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and had access to 3% sucrose for 4 days. B-C) Percent survival of injected mosquitoes that were continually maintained on 3% sucrose or water after day 4 post injection. D-E) Percent survival of injected mosquitoes that were fed a blood meal at day 4 post injection, and maintained on 3% sucrose or water after blood feeding. Mosquito survival was monitored daily for 25 days after injection. Data are expressed as mean percentages ± SEM (*P < 0.05 and **P < 0.01 when compared to dsRNA-FL). At least 30 females were used for each dsRNA treatment.

To examine whether depletion of PK affected the transcriptional expression of genes involved in glucose and nitrogen metabolism, qPCR was performed on cDNA derived from dsRNA-PK and dsRNA-FL injected mosquitoes that were fed a blood meal at day 4 post injection and maintained on 3% sucrose after feeding. The expression of genes encoding G6PDH, NADP-ME1, PEPCK1, PHGDH, GDH, ALT1, ODC, and XDH1 were analyzed in the fat body and thorax from individual mosquitoes. G6PDH, ALT1, ODC and XDH1 transcript expression was similar in tissues dissected from dsRNA-FL and dsRNA-PK mosquitoes. However, NADP-ME1 and PEPCK1 transcripts were up-regulated in the fat body in response to PK silencing. In addition, the level of PHGDH mRNA was increased in the thorax, and GDH in both tissues of PK knockdown mosquitoes (Fig. 5). To determine whether blood meal digestion is affected by PK depletion through RNAi, we monitored the expression of Aa5G1 in the midgut of dsRNA-injected females by western blotting. As shown in Fig. S3, the amount of Aa5G1 in the midgut of PK-deficient females remained abundant at 72 h after blood feeding compared to control.

Figure 5. Effect of pyruvate kinase (PK) deficiency on transcript levels of specific genes in Aedes aegypti tissues.

Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and fed a blood meal at day 4 post injection. Fat body and thorax were dissected at 72 h after blood feeding. The transcript levels were normalized to the mRNA levels of the ribosomal protein S7. G6PDH, glucose-6-phosphate dehydrogenase; NADP-ME1, NADP-malic enzyme-1; PEPCK1, phosphoenolpyruvate carboxykinase-1; PHGDH, phosphoglycerate dehydrogenase; GDH, glutamate dehydrogenase; ALT1, alanine aminotransferase-1; ODC, ornithine decarboxylase; XDH1, xanthine dehydrogenase-1. Data are presented as mean ± SEM of 6 or 8 individual mosquitoes (*P < 0.05 and ***P < 0.001 when compared to dsRNA-FL).

To evaluate whether PK deficiency impacts the abundance or kinetics of metabolites involved in glycolysis, PPP, Krebs cycle, and pathways associated with ammonia fixation, assimilation and detoxification, we fed 4-day-old dsRNA-injected females a blood meal supplemented with [1,2-¹³C₂]-glucose, and examined the incorporation of the ¹³C-atoms from [1,2-¹³C₂]-glucose into several metabolites in mosquito whole body and excreta by HRAM-LC/MS and HRAM-IC/MS methods. As expected, the [¹³C]-atoms from the carbon skeleton of [1,2-¹³C₂]-glucose were mainly incorporated into one and two carbon atoms in most of the metabolites analyzed (Figs. 6–9). In addition, a fast rise with a maximum level of [¹³C]-atom incorporation was observed at 1–6 h, followed by a slower time decay at 12–24 h after feeding for almost all the metabolites analyzed in whole body of both dsRNA-FL (control) and dsRNA-PK mosquitoes (Figs. 6–8). [¹³C]-uric acid synthesis and excretion rate was high at the beginning and at the end of time course studied, respectively (Fig. 9). The kinetics data and abundance of the AAs and OAs of dsRNA-FL mosquitoes (black symbols and line in Figs. 6–9) are in agreement with our previous observations in non-injected wild-type A. aegypti mosquitoes (Horvath et al., 2018). PK deficiency affected the concentration of the majority of the metabolites monitored: AAs involved in ammonia metabolism, metabolites of glucose pathways and Krebs cycle as well as the concentration of uric acid synthesized and excreted during the time course studied (red symbols and line in Figs. 6–9). The significant changes in the abundance of metabolites (compared to the control) are summarized in Table S2. Decrease of metabolite abundance under RNAi-mediated PK were more notable 6 h after feeding where many of the metabolites are close to their maximum level. The metabolites of glycolysis and PPP (Fig. 7) exhibited the most pronounced decrease compared to the control, more than 5-fold for glucose-6-phosphate and fructose-6-phosphate, and about 3-fold for pyruvate and lactate, all of them mainly in the [¹³C₂]-isotopologs. In all the other groups examined, AAs (Fig. 6), Krebs cycle (Fig. 8) and uric acid (Fig. 9), the [¹³C₁]-isotopolog declines (approximately 2-fold or less) were typically observed taking place with a decline in the [¹³C₂]-isotopolog (Table S2). Taken together, these data disclose how genetic knockdown of PK significantly impacts both carbon and nitrogen metabolism in mosquitoes.

Figure 6. Effect of pyruvate kinase (PK) deficiency on [¹³C]-amino acid concentrations in mosquito whole body.

A) Kinetic studies of [¹³C]-Alanine. B) Kinetic studies of [¹³C]-Glutamic acid. C) Kinetic studies of [¹³C]-Glutamine. D) Kinetic studies of [¹³C]-Proline. Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and fed a blood meal supplemented with [1,2-¹³C₂]-glucose. The amino acids were analyzed in mosquito whole body extracts by HRAM-LC/MS. Data are presented as mean ± SEM of three independent samples (*P < 0.05, **P < 0.01 and ***P < 0.001 when compared to dsRNA-FL). N = 15 mosquitoes for each time point.

Figure 9. Concentration of uric acid (UA) [¹³C]-isotopologs in whole body and excreta of mosquitoes injected with dsRNAs.

Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and fed a blood meal supplemented with [1,2-¹³C₂]-glucose. [¹³C]-UA concentration in whole body was measured at 6, 12, and 24 h after feeding, while the [¹³C]-UA concentration in excreta was examined at 12 and 24 h after feeding. Measurements were made by HRAM-IC/MS. Data are presented as mean ± SEM of three independent samples (*P < 0.05 when compared to dsRNA-FL). N = 15 mosquitoes for each time point.

Figure 8. Relative abundance of [¹³C]-isotopologs of Krebs cycle metabolites in whole body of mosquitoes injected with dsRNAs.

Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and fed on a blood meal supplemented with [1,2-¹³C₂]-glucose. Measurements for citrate, isocitrate, malate and fumarate were made by HRAM-IC/MS, and the measurement of α-ketoglutarate was made by HRAM-LC/MS. Plots of the measured relative abundance of [¹³C]-isotopologs against the degree of [¹³C]-labeling for metabolites that exhibited kinetic enrichment of [¹³C]-labeling. Each column corresponds to a single metabolite isotopolog series, and each individual data plot corresponds to an individual time point, over the time course studied. Data are presented as mean ± SEM of three independent samples (*P < 0.05 and ***P < 0.001 when compared to dsRNA-FL). N = 15 mosquitoes for each time point.

Figure 7. Relative abundance of [¹³C]-isotopologs of glucose pathway metabolites in whole body of mosquitoes injected with dsRNAs.

Newly emerged females were injected with dsRNA targeting A. aegypti PK (dsRNA-PK) or dsRNA against firefly luciferase (dsRNA-FL) and fed a blood meal supplemented with [1,2-¹³C₂]-glucose. Measurements for glucose-6-phosphate and fructose-6-phosphate were made by HRAM-IC/MS, and measurements for pyruvate and lactate were made with HRAM -LC/MS. Plots of the measured relative abundance of [¹³C]-isotopologs against the degree of [¹³C]-labeling for metabolites that exhibited kinetic enrichment of [¹³C]-labeling. Each column corresponds to a single metabolite isotopolog series, and each individual data plot corresponds to an individual time point, over the time course studied. Data are presented as mean ± SEM of three independent samples (*P < 0.05 and ***P < 0.001 when compared to dsRNA-FL). N = 15 mosquitoes for each time point.

4. Discussion

PK is an evolutionary conserved enzyme of the glycolytic pathway (Petchampai et al., 2019; Schormann et al., 2019). In humans, there are four isoforms of PK, which are PKM1, PKM2, PKR, and PKL. With the notable exception of PKM1, all of the PK human isoforms are allosterically regulated by fructose-1,6-bisphosphate (Mattevi et al., 1996; Israelsen and Vander Heiden, 2015). It was also reported that a single phosphate group on the sugar is essential for the activation of human PKL, whereas the second phosphate facilitates a firm binding of the effector molecule (Ishwar et al., 2015). Point mutations associated with allosteric regulation of the human PKL were also identified (Tang et al., 2019). Regulation of metabolism by allosteric enzymes is critical for controlling flux into metabolic pathways that support appropriate cell functions. Despite a growing body of PK research, little effort has been given to investigating the role of PK in insect metabolism. Our previous metabolomics study pointed out that the carbon skeleton of glucose contributes to ammonia clearance and nitrogen waste disposal in blood-fed A. aegypti female mosquitoes (Horvath et al., 2018). More recently, a kinetic and structural characterization of the recombinant PK1 from A. aegypti uncovered that the enzyme is allosterically regulated by fructose-1-phosphate, ribulose-5-phosphate, Ala, Gln, Pro, and Ser (Petchampai et al., 2019). Our present data support our previous findings and reveal that PK controls both carbon and nitrogen metabolism of sugar-, blood-fed and starved A. aegypti females. We also demonstrate that genetic disruption of PK slows down the overall mosquito carbon and nitrogen metabolism.

In A. aegypti tissues, we have observed that PK transcripts are up-regulated after blood feeding but PK protein expression remained highly and constitutively expressed throughout the time course analyzed. Although the lack of changes in PK protein expression could be because protein expression simply does not respond to PK mRNA levels, we cannot rule out the possibility of a rapid PK protein turnover during key physiological processes such as blood meal digestion, ammonia metabolism, vitellogenesis and reproduction. Fast glucose oxidation during the first 6 h after blood feeding is required to support ammonia detoxification (Horvath et al., 2018). The transient decrease of PK activity in the fat body of blood/sucrose-fed A. aegypti at 24 and 48 h after blood feeding compared to sugar-fed females is correlated with intense proteolysis of blood in the midgut lumen (Noriega and Wells, 1999; Isoe et al., 2009), nitrogen waste disposal (von Dungern and Briegel, 2001; Mazzalupo et al., 2016; Isoe et al., 2017; Horvath et al., 2018), biosynthetic demands for vitellogenesis and egg development (Attardo et al., 2005; Roy et al., 2018), gluconeogenesis (Zhou et al., 2004a, 2004b), glycogenesis and lipogenesis (Zhou et al., 2004a, 2004b; Hou et al., 2015). PK activity in fat body returned to that of sucrose-fed mosquitoes at 72 h after blood feeding. In contrast, our observation of increased PK protein expression and activity in the fat body of non-blood-fed females starved for 72 h, suggests a higher glycolytic rate due to energetic needs. The activation of PK during starvation was also reported in the whole body of the mealworm beetles (Papadopoulos et al., 2005). On the other hand, a significant decrease of PK activity in the fat body of the blood/water-fed females after 96 and 120 h starvation periods suggests a down-regulation of the glycolytic pathway resulted from a reduction in carbohydrate reserves. Our data provide evidence that PK activity is properly modulated in fat body, depending on the nutritional status of the female mosquitoes. Although PK activity remains relatively stable in thorax of non-starved and starved females, mosquitoes are expected to first deplete carbohydrate reserves during prolonged starvation (Foster, 1995) followed by catabolism of lipids and proteins, as occurs in other organisms (McCue et al., 2017; Rosendale et al., 2019), leading to a decrease of survival over time.

Due to the fact that PK activity significantly differs in fat body of sugar-, blood-fed and starved mosquitoes, we anticipated that genetic disruption of functional PK by RNAi would be lethal. In contrast, the life-span studies showed that survival rate increased in sucrose-, water-or blood/water-fed females, and did not change in blood/sucrose-fed females with PK deficiency compared to RNAi-FL control mosquitoes. Dysfunction in nutrient processing in PK-deficient mosquitoes could impact survival. Alternatively, reduced metabolism may suppress aging. We subsequently found that blood/sucrose-fed PK-deficient females up-regulated genes encoding NADP-ME1 and PEPCK1 in fat body, PHGDH in thorax, and GDH in fat body and thorax at 72 h after blood feeding. These enzymes catalyze reactions involved in several pathways related to carbon and nitrogen metabolism. NADP-ME catalyzes the oxidative decarboxylation of malate into pyruvate with the concomitant production of NADPH, a reducing agent essential for biomolecular synthesis, including nucleotides and lipid biosynthesis, and antioxidant pathways (Frenkel, 1975). PEPCK catalyzes the conversion of oxaloacetate to phophoenolpyruvate and plays a critical role in several pathways including gluconeogenesis, glyceroneogenesis and Ser synthesis (Yang et al., 2009). PHGDH catalyzes the first reaction in the Ser biosynthetic pathway by converting the glycolytic intermediate 3-phosphoglycerate into phosphohydroxypyruvate (Grant, 2018). GDH catalyzes the reversible interconversion of Glu to α-ketoglutarate, but during ammonia detoxification the GDH reaction is favored into the Glu synthesis to generate Gln and Pro (Scaraffia and Wells, 2003; Scaraffia et al., 2005, 2006, 2008, 2010). Thus, the up-regulation of NADP-ME1, PEPCK1, PHGDH, and GDH transcripts in RNAi-PK mosquito tissues is expected to re-direct some glycolytic and Krebs cycle intermediates into biosynthetic pathways to compensate for significant reduction or depletion of PK. In correlation with survival and gene regulation data, our stable isotope tracer analysis indicates that RNAi-PK knockdown alters glucose and nitrogen metabolism at the atomic level. The glycolytic flux in PK-deficient mosquitoes was disrupted mainly during the first hours after feeding a blood meal supplemented with [1,2-¹³C₂]-glucose. It led to a substantial reduction of metabolites generated in the glycolytic pathway, PPP as well as downstream pathways -a scheme of metabolic interactions at [¹³C] atomic levels induced by glucose on ammonia metabolism in non-injected wild-type A. aegypti mosquitoes was proposed previously (Horvath et al., 2018). The most pronounced decrease in metabolites abundance was identified in the glycolytic pathway and PPP in their very initial processes, where the [¹³C₂]-glucose-6-phosphate and [¹³C₂]-fructose-6-phosphate levels decreased by more than 5-fold under PK deficiency at 6 h after feeding. [¹³C₂]-pyruvate and [¹³C₂]-lactate levels were approximately 3-fold lower at the same time point. Most of the other [¹³C]-metabolites, including those involved in Krebs cycle, and ammonia fixation, assimilation and excretion pathways (synthesis of specific AAs and uric acid), exhibited under PK deficiency a decrease of up to a factor of 2 relative to control mosquitoes at specific times during the time course studied. In spite of the reduction of the intermediates rich in carbon, mosquitoes with PK deficiency were capable of synthesizing AAs needed to support ammonia detoxification (Fig. 6). [¹³C]-Ser concentrations did not significantly change during the time course studied (data not shown), suggesting a fast turnover of this AA. Overall, silencing of PK by RNAi resulted in lower conversion of [1,2-¹³C₂]-glucose into [1,2-¹³C₂]-glucose-6-phosphate that can be metabolized further through the glycolysis and PPP. It led to the aforementioned reduction of [1,2-¹³C₂]-fructose-6-phosphate production in the glycolytic pathway, and the reduction of [1,3-¹³C₂]-fructose-6-phosphate and [1-¹³C]-fructose-6-phosphate generated in the PPP that could feed the glycolytic pathway. The decrease of intermediate substrates in the glycolytic pathway and PPP led to a reduction of [¹³C]-metabolites in the Krebs cycle, [¹³C]-AAs and [¹³C]-uric acid (Fig. 6–9). The delay in blood meal digestion in mosquitoes with PK deficiency (Fig. S3) provides additional evidence that disruption of PK prevents ammonia toxicity by slowing down carbon and nitrogen metabolism. Taken together, our data demonstrate that PK confers to sugar-, blood-fed or starved female mosquitoes a metabolic flexibility to adapt to nutritional changes and energetic demands as well as to maintain carbon and nitrogen homeostasis. Genetic knockdown of PK in mosquitoes resulted in a non-lethal phenotype but significantly impacted both carbon and nitrogen metabolism, and triggered an increase of life-span in female mosquitoes under specific nutritional regimens. Our findings contribute to uncover a regulatory role of PK in the carbon and nitrogen metabolism of the arbovirus vector A. aegypti.

Highlights.

• Pyruvate kinase (PK) activity in fat body varies in response to nutritional changes

• PK silencing enhances survival in females maintained under specific nutritional regimens

• PK-deficient females increase transcript levels of specific genes

• PK knockdown reduces the abundance of metabolites involved in several pathways

Abbreviations:

AA

amino acid

AaPK

Aedes aegypti pyruvate kinase

Aa5G1

A. aegypti serine protease

ACN

acetonitrile

ALT1

alanine aminotransferase-1

dsRNA

double-stranded RNA

FA

formic acid

FL

firefly luciferase

GDH

glutamate dehydrogenase

G6PDH

glucose-6-phosphate dehydrogenase

HRAM

high-resolution accurate-mass

IC/MS

ion chromatography-mass spectrometry

LC/MS

liquid chromatography-mass spectrometry

NADP-ME1

NADP-malic enzyme-1

OA

organic acid

ODC

ornithine decarboxylase

PBM

post blood meal

PEPCK1

phosphoenolpyruvate carboxykinase-1

PPP

pentose phosphate pathway

PHGDH

phosphoglycerate dehydrogenase

PK

pyruvate kinase

PKM1, PKM2, PKR and PKL

human PK isoforms

qPCR

quantitative real-time PCR

RNAi

RNA interference

XDH1

xanthine dehydrogenase-1