Chronic depletion of vertebrate lipids in Aedes aegypti cells dysregulates lipid metabolism and inhibits innate immunity without altering dengue infectivity

Aedes aegypti is the primary vector of dengue virus (DENV) and other arboviruses. Previous literature suggests that vertebrate and invertebrate lipids and the nutritional status of mosquitoes modify virus infection. Here, we developed a vertebrate lipid-depleted Ae. aegypti cell line to investigate if chronic depletion of vertebrate lipids normally present in a blood meal and insect cell culture medium would impact cell growth and virus infection. Chronic depletion of vertebrate lipids reduced cell size and proliferation, although cells retained equivalent total intracellular lipids per cell by reducing lipolysis and modifying gene expression related to sugar and lipid metabolism. Downregulation of innate immunity genes was also observed. We hypothesized that chronic depletion of vertebrate lipids would impact virus infection; however, the same amount of DENV was produced per cell. This study reveals how Ae. aegypti cells adapt in the absence of vertebrate lipids, and how DENV can replicate equally well in cells that contain predominately vertebrate or invertebrate lipids.


Unfunded studies
Enter: The author(s) received no specific funding for this work. Describe where the data may be found in full sentences. If you are copying our sample text, replace any instances of XXX with the appropriate details.
If the data are held or will be held in a public repository, include URLs, accession numbers or DOIs. If this information will only be available after acceptance, indicate this by ticking the box below. For example: All XXX files are available from the XXX database (accession number(s) XXX, XXX.).
• If the data are all contained within the manuscript and/or Supporting Information files, enter the following: All relevant data are within the manuscript and its Supporting Information files. of [XXX]. Data are available from the XXX Institutional Data Access / Ethics Committee (contact via XXX) for researchers who meet the criteria for access to confidential data.
The data underlying the results presented in the study are available from (include the name of the third party and contact information or URL). This text is appropriate if the data are owned by a third party and authors do not have permission to share the data.

Establishment of a lipid-depleted (LD) Aedes aegypti cell line 116
We established a lipid-depleted (LD) Ae. aegypti cell line by feeding cells with LD cell 117 culture media. LD cell culture media was made with fetal bovine serum (FBS) that had been 118 extracted with fumed silica [24,31,32]. Fumed silica removes cholesterol, phosphatidylcholines, 119 and triacyl glycerides from serum with minimal removal of fatty acids and essential proteins 120 such as albumin [31,32]. We confirmed that fumed silica removed cholesterol from FBS by 121 quantifying cholesterol concentration in complete (C) and LD cell culture media. Cholesterol 122 concentration was significantly lower in LD cell culture media ( Figure 1A). Additionally, we 123 showed that intracellular cholesterol concentration rapidly declined over the course of one week 124 when cells were treated with LD cell culture media ( Figure 1B). Free fatty acid (FFA) 125 concentration was quantified in C and LD cell culture media, and FFAs were reduced in LD cell 126 culture media by approximately 50 percent ( Figure 1C). Triacylglycerol (TAG) concentration 127 was quantified in C and LD cell culture media, and TAGs were reduced in LD cell culture media 128 by approximately 15 percent ( Figure 1D). Cell proliferation stalled for two months while 129 feeding with LD cell culture media, although the morphology of cells treated with C and LD cell 130 culture media were equivalent after this time frame (Figure 1E), and LD-Aag2 cells began to 131 proliferate, albeit at a lower rate than C-Aag2 cells ( Figure 1F). LD-Aag2 cells were expanded 132 approximately three months after initiation of this cell line and passage zero cells were frozen at 133 -80°C. LD-Aag2 cells were rarely grown past passage 6. 134 135 Ae. aegypti are sterol auxotrophs and derive this lipid from vertebrate blood and other 136 sources [20,24,33]. We expected that depriving cells of vertebrate lipids would reduce 137 intracellular cholesterol concentration. We hypothesized that LD-Aag2 cells would adapt by 138 synthesizing fatty acids to compensate for the loss of sterols and key vertebrate lipids in their 139 nutrition. To test this hypothesis, we quantified the total intracellular lipid content of C and LD-140 Aag2 cells using a lipophilic and fluorescent Nile Red stain and nuclear Hoechst counter stain 141 [34]. Hoechst counter stain was used to normalize total intracellular lipid signal per cell. 142 Additionally, we fed LD-Aag2 six times (6X) normal glucose concentration to see if LD-Aag2 143 cells utilized glucose for de novo lipid biosynthesis. Representative Nile Red fluorescent images 144 revealed that LD-Aag2 and LD-6X-Aag2 cells contained at least as much total lipid per cell as 145 C-Aag2 cells (Figure 2A). We quantified the area of each cell and determined that LD-Aag2 146 cells were significantly smaller than C-Aag2, although treatment with LD-6X glucose cell 147 culture medium reverted cell size to normal ( Figure 2B). We then used pixel intensity as a proxy 148 for lipid concentration and determined that C-Aag2 and LD-Aag2 cells had the same amount of 149 total intracellular lipid per cell. Interestingly, addition of LD-6X cell culture media significantly 150 increased the total intracellular lipid per cell ( Figure 2C). We hypothesized that LD-Aag2 cells 151 maintain similar total intracellular lipids per cell by reducing lipolysis of triglycerides stored in 152 lipid droplets. We tested this hypothesis by performing a glycerol analysis of C-Aag2 cell free 153 supernatants after treatment with C or LD cell culture media for seven days. Lipolysis converts 154 triglycerides into one glycerol and three fatty acid molecules. Production of glycerol fell sharply 155 at days 5 and 7 post-treatment with LD cell culture media ( Figure 2D). 156 Cell proliferation of LD-Aag2 cells was reduced despite cells containing equivalent 158 amounts of total intracellular lipids per cell. We hypothesized that LD-Aag2 cells used glucose 159 and remaining FFAs for de novo lipid biosynthesis rather than for ATP generation. To test this 160 hypothesis, we quantified intracellular ATP concentration in cells three days post-treatment with 161 C and LD cell culture media. There was no difference in intracellular ATP concentration per cell 162 (Figure 3). Similarly, there was no difference in the total concentration of intracellular ATP. However, we 168 did find that cell proliferation in LD-Aag2 cells was reduced. We hypothesized that Aag2 adapt 169 to lipid-depleted conditions by changing expression of metabolism genes, which would reduce 170 mitotic rate to match the reduction in catabolic substrate and promote de novo lipid biosynthesis. 171

172
We tested this hypothesis by performing next generation RNA sequencing (RNA-Seq) 173 analysis and investigated differential gene expression between C and LD-Aag2 cells that were 174 uninfected or infected with DENV for three days. Cluster analysis showed significant differences 175 in gene expression between each of these groups of cells ( Figure 4A). Volcano plot analysis 176 revealed 1,828 differentially expressed genes, with 648 upregulated and 1,180 downregulated 177 genes between uninfected C and LD-Aag2 cells ( Figure 4B). Raw RNA-Seq data comparing 178 each of the different experimental conditions can be seen in Supplementary Tables 1-4. 179

180
The g:Cost Functional Profiling tool at g:Profiler was used to quantify related genes that 181 were upregulated and downregulated in LD-Aag2 cells. Only unique identifiers and not "novel" 182 genes identified by RNA-Seq were used for analysis. The molecular function, biological process, 183 and cellular compartment categories with the highest number of intersecting upregulated LD-184 Aag2 genes were catalytic activity, DNA replication, and membrane, respectively (Figure 5). 185 The molecular function, biological process, and cellular compartment categories with the highest 186 number of intersecting downregulated LD-Aag2 genes were catalytic activity, small molecular 187 metabolic process, and membrane, respectively ( Many differentially expressed genes were present that did not fall within a significantly 201 altered functional gene category. We noted that some innate immunity genes were also 202 dysregulated in LD-Aag2 cells ( Table 1). We validated RNA-Seq data using many of these 203 targets, and simultaneously revealed that LD-Aag2 failed to upregulate expression of a number 204 of innate immunity genes during DENV infection (Figure 8). These genes also failed to respond 205 to infection in the supplementary RNA-Seq datasets. 206 207

DENV infection in C and LD-Aag2 cells 226
Previous literature suggests that intracellular lipids are critical for DENV infection. LD-227 Aag2 cells were depleted of cholesterol and LD cell culture media had significantly less FFAs. 228 We noted that LD-Aag2 responded to lipid-depleted conditions by upregulating gene pathways 229 related to lipid metabolism and de novo lipid biosynthesis. Further, LD-Aag2 cells had lower cell 230 proliferation, and key innate immunity genes failed to respond to DENV infection. We 231 hypothesized that chronic depletion of vertebrate lipids would influence DENV infection. To test 232 this hypothesis, we inoculated monolayers of C-Aag2, LD-Aag2, and LD-6X-Aag2 cells with 50 233 focus forming units (FFUs) of DENV. Importantly, DENV was inoculated onto cells in LD cell 234 culture media, and this media was replaced with either C, LD, or LD-6X cell culture media for 235 the next three days. This avoided any influence of vertebrate lipids during the early stage of 236 infection [8,13]. We first determined when virus is actively being shed from C-Aag2 cells by 237 collecting cell free supernatants 0, 1, 3, and 5 days post-infection (dpi) and quantifying viral 238 RNA (vRNA). DENV shedding was first detected 3 dpi and shedding began to plateau 5 dpi 239 ( Figure 9A). We then collected cell free supernatants from virus producing cells 3 dpi and 240 normalized vRNA to the number of virus-producing cells ( Figure 9B). There was no difference 241 in the amount of vRNA produced per cell. Normalizing vRNA to cell number allowed us to 242 control for differences in cell proliferation between cell types. We also took cell free 243 supernatants from each cell type 3 dpi and inoculated fresh C-Aag2 monolayers. Freshly infected 244 monolayers were incubated for three days and then fixed and stained with anti-DENV antibody. 245 FFUs were quantified and normalized by the total number of virus producing cells. There was no 246 significant difference in FFUs produced per cell (Figure 9C-D). act as a metabolic switch that communicates when a blood meal is available. 281 RNA-Seq revealed multiple pathways that were likely involved in deriving energy and 282 producing invertebrate lipids. Genes associated with FFA metabolism and synthesis were 283 upregulated (e.g., carboxylesterase, fatty-acyl-CoA reductase, phospholipase, and polyketide 284 synthase), which likely led to catabolic breakdown of remaining FFA acids present in LD cell 285 culture medium and production of ATP. Polyketide synthase and fatty acid synthase (FAS) genes 286 are also evolutionarily related [39]. We searched our dataset to see if FAS genes were induced in 287 LD-Aag2 cells. FAS1 (AAEL001194) was upregulated 9-fold. FAS2 (AAEL008160) was 288 upregulated 2.5-fold. We also note that pyruvate carboxylase (AAEL009691), which is the rate-289 limiting enzyme responsible for fatty acid synthesis, is upregulated 2-fold. This suggests that 290 LD-Aag2 cells produce fatty acids from precursor molecules. Genes related to production of 291 extracellular matrix (ECM)-related proteins were also upregulated (e.g., immunoglobulin-like 292 protein, fibronectin domain protein, and ECM protein), which may have been an adaptation to 293 changes in the cell membrane and a need to express different proteins to facilitate adherence to 294 the cell culture plate. LD-Aag2 cells down regulated genes related to glycolysis, 295 gluconeogenesis, and protein synthesis (e.g., pyruvate kinase, phosphoenolpyruvate 296 carboxykinase, phosphoglycerate mutase, and tRNA-related genes), which likely conserved 297 energy and reduced cell proliferation. Many genes were also downregulated that involved lipid 298 metabolism and synthesis (e.g., phosphoglycerate mutase 1, phosphoenolpyruvate 299 carboxykinase, sphingomyelin synthase, and fatty acid desaturase  downregulated and many that were tested failed to respond to infection. This dichotomy may 323 allow for DENV to persist in mosquitoes despite a fluctuation in vertebrate lipids and nutrition. 324 These data highlight the adaptability of both mosquitoes and DENV. Aag2 cells were able to 325 adapt to lipid depletion by leveraging sugar and lipid metabolic pathways, and DENV was able 326 to replicate to the same degree regardless of nutritional differences perhaps due to utilization of 327 invertebrate-specific lipids and/or suppressed innate immunity. This is not entirely surprising 328 given the challenges mosquitoes face in acquiring a blood meal and their ability to survive on 329 plant nectar and sugar alone. It is also not surprising given that the DENV life cycle requires 330 transmission between invertebrate and vertebrate species. 331 332 We also considered the reduced cell size and proliferation in LD-Aag2 cells and 333 hypothesized that this was due to the loss in vertebrate lipids as an energy source. Specifically, 334 fatty acids present in fetal bovine serum (FBS) may provide chemical energy via lipolysis. 335 Interestingly, C-Aag2 and LD-Aag2 cells contained the same levels of intracellular ATP, and we 336 did not see an improvement in cell proliferation when LD-Aag2 cells were provided with excess 337 glucose (data not shown). These data suggest that LD-Aag2 cells may use glucose and FFAs 338 present in LD cell culture media for energy production and that the reduced cell size and mitotic 339 rate may due to factors other than low levels of ATP. One possibility is that vertebrate sterols 340 serve as structural components for cell membranes and lipid containing organelles, and that 341 deprivation of these raw materials limits cell growth and replication [17,23,24,45,47]. Cholesterol was measured using the Amplex Red Cholesterol Assay Kit (Invitrogen). Cell 388 culture media and Aag2 cells that were treated with either C or LD cell culture medium for 1-7 389 days were lysed in PBS containing 1% Triton X-100 and protease inhibitor cocktail and used for 390 analysis. Equivalent volumes of cell culture medium or 2 ug of total protein was used for each 391 sample. The reactions were brought up to a total reaction volume of 50 ul using 1x reaction 392 buffer from the kit and applied to individual wells on a 96-well microplate. 50 uL of Amplex 393 Red reagent/HRP/cholesterol oxidase/cholesterol esterase working solution was added to each 394 well and the plate was incubated at 37 °C for 30 minutes protected from light. Fluorescence was 395 measured using an Epoch2 microplate spectrophotometer and Gen5 software (Agilent) with 396 excitation at 550nm and emission detection at 590nm [24]. Free fatty acids (FFAs) were 397 quantified in C and LD cell culture media using the Free Fatty Acid Fluorometric Assay Kit 398 according to manufacturer's instructions (Cayman Chemical). Triacylglycerols (TAGs) were 399 quantified in C and LD cell culture media using the Triglyceride Colorimetric Assay Kit 400 according to manufacturer's instructions (Cayman Chemical). Glycerol was quantified in C-401 Aag2 cell free supernatants after treatment with C and LD cell culture media 1, 3, 5, and 7 days 402 post-treatment using the Glycerol Colorimetric Assay Kit according to manufacturer's 403 instructions (Cayman Chemical). Each sample of cell culture media or cell free lysate was 404 measured in triplicate using the Epoch2 microplate spectrophotometer and Gen5 software 405 (Agilent). 406 407

Nile Red Total Intracellular Lipid Assay 408
The Nile Red lipid droplet stain was obtained from Santa Cruz Biotechnology. The Hoescht 409 nuclei counterstain was obtained from Life Technologies. Total intracellular lipids were stained 410 using Nile Red and the Hoescht stain was used to quantify and normalize data by cell number as 411 needed [34]. Briefly, 100,000 C, LD, and LD-6X-Aag2 cells were seeded in a 96 well plate in 412 triplicate for 48 hours. Cell were then fixed with 4% paraformaldehyde in PBS for 10 minutes 413 before staining cells with Hoescht (1 μg/mL in PBS) for 15 minutes, and Nile Red (1 μg/mL in 414 PBS) for 15 minutes. Ten representative digital photos were taken using an EVOS fluorescent 415 microscope at 20X magnification across the triplicate samples and photos were analyzed using 416 ImageJ software. Nile Red staining was used to measure the relative cell area of C, LD, and LD-417 6X-Aag2 cells, and Nile Red pixel intensity coupled with Hoescht staining was used to measure 418 the relative amount of total intracellular lipids per cell. 419 420

ATP Assay 421
The Colorimetric ATP Assay kit was purchased from BioVision and used with Vivaspin 422 500 10kDa MWCO spin filters purchased from GE Healthcare. Briefly, C and LD-Aag2 cells 423 were grown to confluence in 12-well plates in triplicate. This assay was conducted 3 days post 424 feeding of each cell line with their respective insect cell growth medium. 17,000 cells of each 425 cell line were pelleted at 1,700 RPM for 7 minutes, and then re-suspended in ATP lysis buffer. 426 Cell lysates were then passed through 10kDa MWCO spin filters before following 427 manufacturer's instructions. Each respective cell lysate sample was measured in duplicate at OD-428 570 nm using the Epoch2 microplate spectrophotometer and Gen5 software (Agilent). 429 430 RNA Sequencing (RNASeq) 431 RNA Sequencing (RNASeq) was conducted using a service offered by Novogene, USA 432 (California UC Davis) on both uninfected and DENV2-infected C and LD-Aag2 cells. Cells were 433 either treated with media alone or infected with ~100 focus forming units of DENV2. RNA was 434 extracted 3 dpi. RNA was extracted using the Qiagen RNeasy Plus RNA Extraction kit with 435 gDNA eliminator columns. RNA purity was ensured using 1% agarose gels and a 436 NanoPhotometer. RNA integrity was ensured using the RNA Nano 6000 Assay Kit of the 437 hybridization. In order to select cDNA fragments of preferentially 150~200 bp 450 in length, the library fragments were purified with AMPure XP system (Beckman Coulter). Then 451 3 μL USER enzyme (NEB) was used with size-selected, adaptor-ligated cDNA at 37°C for 15 452 minutes followed by 5 minutes at 95 °C before PCR. Then PCR was performed with Phusion 453 High-Fidelity DNA polymerase, Universal PCR primers and Index (X) Primer. At last, PCR 454 products were purified (AMPure XP system) and library quality was assessed on the Agilent 455 Bioanalyzer 2100 system. The clustering of the index-coded samples was performed on a cBot 456 Cluster Generation System using HiSeq PE Cluster Kit cBot-HS (Illumina) according to the 457 manufacturer's instructions. After cluster generation, the library preparations were sequenced on 458 an Illumina Hiseq platform and 125 bp/150 bp paired-end reads were generated. Raw data (raw 459 reads) of fastq format were firstly processed through in-house perl scripts. In this step, clean data 460 (clean reads) were obtained by removing reads containing adapter, reads containing ploy-N and 461 low-quality reads from raw data. At the same time, Q20, Q30 and GC content the clean data 462 were calculated. All the downstream analyses were based on the clean data with high quality. Unbound virus was removed and fresh C and LD cell culture media was added onto their 492 respective cell type. Total cellular RNA was extracted at 0, 12, and 24 hours post infection (hpi) 493 using an RNeasy RNA Extraction kit with gRNA eliminator columns (Qiagen). Total RNA was 494 analyzed using a singleplex format in 48-well plates with a total reaction volume of 10 μl using 495 an Eco Illumina instrument. Reverse transcription and quantitative PCR were performed in the 496 same closed tube with 50 ng of total RNA per reaction using the Power SYBR Green RNA-to-Ct 497 1-Step RT-qPCR Kit (ThermoFisher). All primers were used at a final concentration of 1 μM. 498 Primer sequences are available in Supplemental Table 7. Cycling conditions were 50 °C for 30 499 min (reverse transcription) and 95 °C for 15 min, followed by 45 cycles of 94 °C for 15 s, 55 °C 500 for 30 s and 72 °C for 30 s. Fold-differences in gene expression were determined using the Pfaffl 501 method and data were normalized based on total nanograms of cellular RNA. For DENV2 viral 502 RNA (vRNA) analysis, C, LD, and LD-6X-Aag2 cells were grown to ~70% confluence in 96 503 well plates and inoculated with 50 focus-forming units (FFUs) of DENV2 in LD cell culture 504 media for 1 hour at 27 °C in triplicate. Unbound virus was removed and fresh C, LD, and LD-6X 505 cell culture media was added onto their respective cell type. At 3 dpi, 10 µL of cell free 506 supernatants were harvested and spiked into RLT buffer from the RNeasy RNA extraction kit 507 (Qiagen). At this time, the total number of cells per well were quantified using a hemocytometer. 508 This was performed to account for any differences in cell proliferation between cell types. 509 RNA was then extraction as listed above, following by RT-qPCR using F 5′ CAG ATC TCT 510 GAT GAA TAA CCA ACG 3 and R 5′ CAT TCC AAG TGA GAA TCC TTT GTC A 511 3'primers. The same protocol as shown above was used for DENV2 vRNA RT-qPCR, and 512 relative vRNA levels were normalized per cell. 513 514 Focus Forming Unit (FFU) infectivity assays 515 C, LD, and LD-6X-Aag2 cells were grown to ~70% confluence in 96 well plates and 516 inoculated with 50 focus-forming units (FFUs) of DENV2 in LD cell culture media for 1 hour at 517 27 °C. Inoculation of DENV2 in LD cell culture media is important in order to avoid inhibiting 518 virus fusion with vertebrate extracellular vesicles [24]. Unbound virus was removed and fresh C, 519 LD, and LD-6X-Aag2 cell culture media was added onto their respective cell type. At 3 dpi, 70 520 uL of C, LD, and LD-6X-Aag2 cell-free supernatants were inoculated onto fresh ~70% 521 confluence monolayers of C-Aag2 cells. Unbound virus was removed after 1 hour and fresh C 522 cell culture media was added onto each of the monolayers. At 3 dpi, confluent C-Aag2 523 monolayers were fixed in 4% paraformaldehyde in phosphate buffered saline (PBS) for 10 524 minutes, cell membranes were permeabilized with 0.1% Triton X-100, and DENV2 antigen was 525 stained with 1:200 anti-DENV2 envelope antibody (3H5-1, EMD Millipore) in PBS with 1% 526 bovine serum albumin (BSA). The total number of DENV-positive foci were revealed using a 527 1:200 horseradish peroxidase-conjugated secondary antibody in PBS plus 1% BSA, and an AEC 528 Peroxidase Substrate kit (Vector Laboratories). DENV2-positive foci were quantified using the 529 Evos XL Core Imaging System. DENV2 FFUs were normalized to the total number of producing 530 C, LD, and LD-6X-Aag2 cells present 3 dpi. The total number of producing cells were quantified 531 using a hemocytometer after cell free supernatants were harvested. This was performed to 532 account for any differences in cell proliferation between cell types.