The newly-colonized Aedes aegypti Chetumal strain was characterized for its vector competence for DENV-2 [16,38]. Chetumal mosquitoes (F3 – F6) were originally collected from in the Yucatan Peninsula, Mexico. Mosquitoes were hatched and reared as reported previously [16,17,38]. After hatching female pupae were separated from males by size, since males are significantly smaller; transferred in plastic cups to cardboard cartons and allowed to emerge. Adult mosquitoes were maintained at 28°C, 70–80% relative humidity, with a 12–12 h light-dark photoperiod. Mosquitoes used as controls in tropism experiments were the Puerto Rico-Rex-D strain (a long colonized laboratory strain)  and the D2S3 strain (a laboratory selected, highly susceptible mosquito strain lacking a midgut escape barrier) .
A high passage DENV-2 strain (Jamaica 1409)  was obtained from the Centers for Diseases Control and Prevention (CDC- Fort Collins, CO, USA). The virus had been routinely passaged (> 25 times) in C6/36 cell culture, but there were only two relevant changes in comparison to the low passaged virus as the full-length sequence analysis indicated . These changes were a conservative amino acid substitution in E6 protein (Ile→Met) and a nucleotide change in the 3'UTR143 (G→T). The C6/36 cell cultures were infected at a multiplicity of infection (MOI) of 0.001 and incubated at 28°C for a week. At day 7, the medium was replaced with fresh medium, and at 12 days post-infection virus was harvested to prepare infectious bloodmeals [16,42].
Oral infection of mosquitoes with DENV-2
Aedes aegypti mosquitoes were challenged with DENV-2 Jam1409 using an artificial bloodmeal . Briefly, 100–150 Aedes aegypti female mosquitoes, 4 – 7 days post-emergence, were deprived of sugar for 24 h and water for 8 h prior to the bloodfeeding. The virus-infected cell monolayers were scraped from the flask surface, and 2 ml of this suspension was added to an equal amount of washed sheep erythrocytes to prepare the infectious bloodmeal, which was provided to the mosquitoes using a membrane feeding device. The meal was maintained at 37°C by circulating warm water though an external jacket. Mosquitoes were not allowed to feed longer than 45 min to prevent a decline in viral titers. Blood meals contained viral titers of 1.7 ± 0.7 × 107 PFU/ml. Blood fed mosquitoes were separated using a chill table, transferred to clean cartons, and maintained at as noted above.
Indirect immunofluorescence assays (IFA)
To determine viral tropisms, mosquito (n = 30–40) tissues were examined for the presence of virus antigen after dissecting midguts and salivary glands and squashing their thoraxes, heads, and abdomens on slides.
Midguts were dissected in PBS and fixed in 4% paraformaldehyde (Electron Microscopy, Hatfield, PA) for at least 4 h. The fixative was then removed, rinsed once in PBS, and then incubated for 1 h at room temperature with constant rocking in a solution containing: 1× PBS-Ashburner's solution (13 mM NaCl, 0.7 mM Na2HPO4, 1 mM NaH2PO4 at pH = 7.2), 1% Bovine Serum Albumin, and 0.1% Triton X-100, called PBT-0.1 (PBS-BSA-Triton). Typically to detect DENV-2 antigen by IFA, the primary antibody was the serotype-specific mouse 3H5–1 monoclonal antibody (ATCC number HB-46). To confirm the decline in DENV-2 antigen in midguts, a polyclonal human anti-dengue specific antibody (Biodesign International, Saco, ME) was also used as the primary antibody. Midguts were then incubated with a 1:400 dilution of goat-antimouse antibody conjugated with Alexa-488 fluorochrome (Molecular Probes Inc, Eugene, OR). Midguts were washed twice with PBT-0.1 with constant rocking at room temperature for 2 hr and incubated with a 1:40 dilution of Phalloidin-Alexa 546 (Molecular Probes Inc, Eugene, OR) for 20 min. Midguts were rinsed with PBS and placed onto slides, covered with Vectashield® (Vector Laboratories Inc, Burlingame, CA), and a coverslip was sealed with nail polish on the slide. Preparations were examined using the Olympus BH2-RFCA epifluorescence microscope or a confocal microscope (FVX-IHRT Fluoview Confocal, LSM, Olympus).
Salivary glands were dissected in PBS onto Vectabond® (Vector Laboratories Inc, Burlingame, CA) pretreated slides, fixed in cold acetone or fixed in 4% paraformaldehyde- 2% glutaraldehyde, and maintained at 4°C until immunosassay. Abdomens, thoraxes and heads were squashed on glass Vectabond®-pretreated slides, fixed in cold acetone for 15 min, and kept at 4°C
IFA was used to detect DENV-2 antigen in salivary glands, thoraxes, abdomens and heads. The primary serotype-specific monoclonal DENV-2 antibody 3H5-1 was diluted 1:200 in PBS and incubated for 1 h on the tissue. Slides were then washed twice for 5 min with PBS, and goat-anti-mouse avidin-conjugated secondary antibody containing 0.01% Evans Blue (1:200; Amersham Pharmacia Biotech, Piscataway, NJ) was added. After 40 min, samples were washed 2× in PBS and incubated for 20 min with a 1:200 dilution of fluorescein-streptavidin (Amersham Pharmacia Biotech, Piscataway, NJ). Following two additional PBS washes of 5 min each and one with distilled water, DABCO oil was added to the tissues and coverslips were added. All the incubations were performed at 37°C. An Olympus BH2-RFCA fluorescence microscope was used to examine slides.
Determination of midgut infection and dissemination rates
The midgut infection rate (MIR) was determined as the number of midguts containing DENV-2 antigen divided by the number of midguts examined. The dissemination rate (DR) was determined as the number of mosquitoes with detectable DENV-2 antigen in non-midgut tissues (eg, head squash tissues, salivary glands, fat body, etc.), divided by the number of mosquitoes with detectable virus antigen in the midgut.
Viral titers in midguts were determined by end point titrations (TCID50) in C6/36 cells. Titers were established using the microtiter plate titration assay described by Schoepp and Beaty , and endpoints were calculated by the Karber method  and expressed as log10 TCID50/ml. To determine viral titers in whole mosquitoes as well as in some midguts, plaque assay titrations in LLC-MK2 cells were used modifying the protocol originally reported by Malewicz and Jenkin .
DENV- 2 quantitative real time RT-PCR (Q-PCR)
DENV-2 genome copy number was determined Q-PCR in homogenized and filtered tissue from individual mosquitoes or midguts . RNA for PCR assays was extracted using QIAamp® Viral RNA Kit (Qiagen Sciences, Maryland, MA). The reaction mixture was prepared with 1× DyNAzyme® buffer (Finnzymes, Espoo, Finland), 0.2 mM dNTP mix, 0.5× SYBRGreen I (Molecular Probes Inc, Eugene, OR), 2.5 mM MgCl2, 10 μM of forward and reverse primers (5'CAAGTCGAACAACCTGGTCCAT3' and 5'GCCGCACCATTGGTCTTCTC3' respectively), 0.4U DyNAzyme II Recombinant DNA Polymerase (Finzymmes, Espoo, Finland), and 2 μl cDNA in a final volume of 20 μL. The SYBR Green I 10,000× stock was diluted in DMSO to a working solution of 100×, which was stored at -20°C in 15 μL aliquots to minimize exposure to light and freeze-thaw cycles. Reactions were performed in a microseal 96 microplates covered with optically clear caps (MJ Research, Waltham, MA) and placed in a Opticon 2 thermal cycler. Settings were: 95°C for 10 min, followed by 40 cycles of 95°C for 10 sec, 64°C for 20 sec, 72°C for 30 sec, and 84°C for 1 sec for fluorescence measurement. After a final extension at 72°C for 10 min, a melting curve was ascertained using the program: 70°C to 95°C, 0.2°C/read, 1 sec hold to confirm product specificity.
The primers used targeted a 177 bp region of the DENV-2 NS5 gene and quantified (+) strand DENV-2 RNA in the sample. Standard curves were generated on each plate by analyzing 2 × 102 to 2 × 108 copies/reaction of DENV-2 from plasmid standards.
This research was supported by NIH AI45430 and AI48740. MIS was granted a Fogarty International Center scholarship (TW1130).