In vitro and in vivo host range of Anopheles gambiae densovirus (AgDNV)

AgDNV is a powerful gene transduction tool and potential biological control agent for Anopheles mosquitoes. Using a GFP reporter virus system, we investigated AgDNV host range specificity in four arthropod cell lines (derived from An. gambiae, Aedes albopictus and Drosophila melanogaster) and six mosquito species from 3 genera (An. gambiae, An. arabiensis, An. stephensi, Ae. albopictus, Ae. aegypti and Culex tarsalis). In vitro, efficient viral invasion, replication and GFP expression was only observed in MOS55 An. gambiae cells. In vivo, high levels of GFP were observed in An. gambiae mosquitoes. Intermediate levels of GFP were observed in the closely related species An. arabiensis. Low levels of GFP were observed in An. stephensi, Ae. albopictus, Ae. aegypti and Cx. tarsalis. These results suggest that AgDNV is a specific gene transduction tool for members of the An. gambiae species complex, and could be potentially developed into a biocontrol agent with minimal off-target effects.

In vivo AgDNV host specificity. We next investigated viral host range among mosquito species in vivo using An. gambiae, An. stephensi, An. arabiensis, Ae. aegypti, Ae. albopictus and Culex tarsalis. 40-50 adult mosquitoes of each species were injected with 1 × 10 7 of vUTRAcGFP. At 7 days post injection, mosquitoes were visually examined for GFP expression using fluorescence microscopy. We defined a seven category scoring criteria (0-6) for the level of fluorescence expression in individual mosquitoes (Fig. 2). This scoring system allowed us to compare the viral infection levels semi-quantitatively and analyze the distribution of GFP expression level within each species. The known permissive mosquito, An. gambiae, exhibited scores ranging from 3 to 6 with an average score of 4.8 (Fig. 3A,G). The closely related species An. arabiensis exhibited scores ranging from 2 to 5 with an average score of 3.3 (Fig. 3C,G). In other mosquito species, the distributions were shifted and had statistically significantly lower ranges

Discussion
Although AgDNV was originally isolated from An. gambiae Sua5B cells 9 , no fluorescence or viral DNA was detected after vUTRAcGFP infection of this cell line. However, Sua5B cells are permissive to viral replication of the naturally occurring virus present in the cell line or if transfected with a recombinant infectious clone plasmid 9 . These results suggest that Sua5B cells may have been originally infected with AgDNV from the original mosquito colony from which the cell line was established, however, during development of the cell line and/or over long-term serial passage the cells lost essential host factors (such as receptors) required for new infection. In contrast, the An. gambiae MOS55 cell line retains these factors and is permissive to infection. Comparison of these two cell lines may help identify the specific receptors required for AgDNV entry into cells.
The An. gambiae species complex consists of at least seven morphologically identical mosquito species, to which both An. gambiae and An. arabiensis belong 19 . We had initially hypothesized that AgDNV would in general infect Anopheline mosquitoes better than species belonging to other genera. However, this was not the case. While An. gambiae, and to a lesser extent An. arabiensis were susceptible to infection, the congeneric species An. stephensi was refractory to infection (Fig. 3B,G). An. stephensi is the major Asian vector of human malaria and is not part of the gambiae complex 19 . These observations suggest that AgDNV is specifically adapted to infect An. gambiae and closely related species.
We unexpectedly observed intermediate to high levels of GFP expression (scores of 3 to 5) in a low percentage of Ae. albopictus individuals (Fig. 3D), leading to a significantly higher mean infection score for Ae. albopictus compared to Ae. aegypti or Cx. tarsalis (P < 0.05) (Fig. 3D-G). These results complement our cell line data, where C6/36 cells (derived from Ae. albopictus) were also minimally permissive to viral infection. To date, there are no reports detailing the molecular mechanisms underlying host specificity of mosquito densoviruses. Clathrin-mediated endocytosis has been shown to be important for infection for mammalian and insect parvoviruses such as canine parvovirus (CPV) and Junonia coenia densovirus (JcDNV) [20][21][22] . The clathrin-mediated endocytosis pathway is likely used by mosquito densoviruses as well. Structural variation of the receptors and downstream molecules could determine the host specificity of AgDNV among mosquito species and remains to be investigated. Unlike other densoviruses that are highly pathogenic to their hosts, AgDNV has a negligible impact on the life span of An. gambiae [1][2][3][4][5][6]11 . The highly infectious but non-pathogenic and specific nature of the interaction between An. gambiae and AgDNV suggests a history of co-evolution and host-specific adaptation in this system that is distinct from other studied mosquito densoviruses. Further experiments to elucidate the molecular mechanisms of this observed specificity will provide mechanistic insights into the evolution of host-specific pathogens, and will inform on the utility of using DNVs for targeted biocontrol of vector mosquitoes in the field. were produced by co-transfection of MOS55 cells with the recombinant virus plasmid pUTRAcGFP and the wild type AgDNV helper plasmid pBAgα as described [9][10] . Viral titer for these infection experiments was determined with qPCR using a standard curve as previously described 10 . Briefly, DNV samples were TURBO DNase (Ambion) treated to digest plasmid DNAs. Total DNA was extracted using DNEasy kits (Qiagen). qPCR was performed using the Quantitect SYBR Green Kit (Qiagen) on a Rotor-Gene Q (Qiagen) with EGFP primers: 5′ TCA-AGA-TCC-GCC-ACA-ACA-TC 3′ , 5′ TTC-TCG-TTG-GGG-TCT-TTG-CT 3′ . A standard curve was created using a dilution series of pUTRAcGFP ranging from 10 3 to 10 8 copies.
In vitro AgDNV infection quantitation. MOS55 (An. gambiae), Sua5B (An. gambiae), C6/36 (Ae. albopictus) and S2 (Drosophila melanogaster) cell lines were cultured in Schneider's media with 10% fetal bovine serum. Cells were infected with 1 × 10 9 virions of recombinant GFP-expressing AgDNV (vUTRAcGFP) 10 . Viral DNA level in infected cells was determined by qPCR as described above. GFP mean fluorescence intensity (MFI) was determined using flow cytometry with FlowJo software. Statistical differences between treatments were determined using analysis of variance (ANOVA) with Bonferroni's correction for multiple comparisons.
In vivo AgDNV infection quantitation. Mosquitoes  ) were held at 27 °C and 80% relative humidity and were maintained on expired human blood or commercially obtained bovine blood using a membrane feeding system, and were allowed access to 10% sucrose solution ad libitum through a cotton wick. For each species, 3-5 day old females were anesthetized by chilling and injected with 1 × 10 7 vUTRAcGFP using a glass capillary needle. Injected mosquitoes were held at 27 °C and 80% relative humidity with 10% sucrose. A 7 category scoring scale (Fig. 2) was used to visually quantify GFP expression using an Olympus BX40 epifluorescent microscope at 7 days post-injection. Statistical differences between treatments were determined using analysis of variance (ANOVA) with Bonferroni's correction for multiple comparisons. An. gambiae injected with media were used as a negative control.