Arthropod-borne viruses (arboviruses) circulate in nature between arthropod vectors and vertebrate hosts. Arboviruses often cause devastating diseases in vertebrate hosts, but they typically do not cause significant pathology in their arthropod vectors. Following oral acquisition of a viremic bloodmeal from a vertebrate host, the arbovirus disease cycle requires replication in the cellular environment of the arthropod vector. Once the vector has become systemically and persistently infected, the vector is able to transmit the virus to an uninfected vertebrate host. In order to systemically infect the vector, the virus must cope with innate immune responses and overcome several tissue barriers associated with the midgut and the salivary glands. In this review we describe, in detail, the typical arbovirus infection route in competent mosquito vectors. Based on what is known from the literature, we explain the nature of the tissue barriers that arboviruses are confronted with in a mosquito vector and how arboviruses might surmount these barriers. We also point out controversial findings to highlight particular areas that are not well understood and require further research efforts.

In vivo targeted gene disruption is a powerful tool to study gene function. Thus far, two tools for genome editing in Aedes aegypti have been applied, zinc-finger nucleases (ZFN) and transcription activator-like effector nucleases (TALEN). As a promising alternative to ZFN and TALEN, which are difficult to produce and validate using standard molecular biological techniques, the clustered regularly interspaced short palindromic repeats/CRISPR-associated sequence 9 (CRISPR/Cas9) system has recently been discovered as a "do-it-yourself" genome editing tool. Here, we describe the use of CRISPR/Cas9 in the mosquito vector, Aedes aegypti. In a transgenic mosquito line expressing both Dsred and enhanced cyan fluorescent protein (ECFP) from the eye tissue-specific 3xP3 promoter in separated but tightly linked expression cassettes, we targeted the ECFP nucleotide sequence for disruption. When supplying the Cas9 enzyme and two sgRNAs targeting different regions of the ECFP gene as in vitro transcribed mRNAs for germline transformation, we recovered four different G1 pools (5.5% knockout efficiency) where individuals still expressed DsRed but no longer ECFP. PCR amplification, cloning, and sequencing of PCR amplicons revealed indels in the ECFP target gene ranging from 2-27 nucleotides. These results show for the first time that CRISPR/Cas9 mediated gene editing is achievable in Ae. aegypti, paving the way for further functional genomics related studies in this mosquito species.

In the mosquito, the midgut epithelium is the initial tissue to become infected with an arthropod-borne virus (arbovirus) that has been acquired from a vertebrate host along with a viremic bloodmeal. Following its replication in midgut epithelial cells, the virus needs to exit the midgut and infect secondary tissues including the salivary glands before it can be transmitted to another vertebrate host. The viral exit mechanism from the midgut, the midgut escape barrier (MEB), is poorly understood although it is an important determinant of mosquito vector competence for arboviruses. Using chikungunya virus (CHIKV) as a model in Aedes aegypti, we demonstrate that the basal lamina (BL) of the extracellular matrix (ECM) surrounding the midgut constitutes a potential barrier for the virus. The BL, predominantly consisting of collagen IV and laminin, becomes permissive during bloodmeal digestion in the midgut lumen. Bloodmeal digestion, BL permissiveness, and CHIKV dissemination are coincident with increased collagenase activity, diminished collagen IV abundance, and BL shredding in the midgut between 24–32 h post-bloodmeal. This indicates that there may be a window-of-opportunity during which the MEB in Ae. aegypti becomes permissive for CHIKV. Matrix metalloproteinases (MMPs) are the principal extracellular endopeptidases responsible for the degradation/remodeling of the ECM including the BL. We focused on Ae. aegypti (Ae)MMP1, which is expressed in midgut epithelial cells, is inducible upon bloodfeeding, and shows collagenase (gelatinase) activity. However, attempts to inhibit AeMMP activity in general or specifically that of AeMMP1 did not seem to affect its function nor produce an altered midgut escape phenotype. As an alternative, we silenced and overexpressed the Ae. aegypti tissue inhibitor of metalloproteinases (AeTIMP) in the mosquito midgut. AeTIMP was highly upregulated in the midgut during bloodmeal digestion and was able to inhibit MMP activity in vitro. Bloodmeal-inducible, midgut-specific overexpression of AeTIMP or its expression via a recombinant CHIKV significantly increased midgut dissemination rates of the virus. Possibly, AeTIMP overexpression affected BL degradation and/or restoration thereby increasing the midgut dissemination efficiency of the virus. The biological nature of the midgut escape barrier in insects for arthropod-borne viruses has been a mystery for decades. Here we show that the basal lamina (BL) surrounding the mosquito midgut acts as a barrier for chikungunya virus, an alphavirus, which has emerged in the New World hemisphere around three years ago. The barrier became permissive for the virus during digestion of a viremic bloodmeal inside the midgut lumen. Concurrent with BL permissiveness, we observed that collagen IV, a major component of the BL became temporally degraded while the BL was visibly damaged. Based on previous findings, we hypothesized that matrix metalloproteinases such as Ae. aegypti (Ae)MMP1 may be involved in BL degradation. We confirmed that recombinant AeMMP1 exhibited strong gelatinase activity, which was profoundly reduced when recombinant AeMMP1 interacted in vitro with the recombinant Ae. aegypti tissue inhibitor of metalloproteinases (AeTIMP). When transgenically overexpressing AeTIMP in an attempt to temporally inhibit general MMP activity in the mosquito midgut, we observed that the dissemination efficiency of chikungunya virus became significantly increased, while its midgut infection was not affected. It is possible that AeTIMP overexpression affected BL degradation/restoration permitting increased quantities of virus to escape from the midgut.

Zika virus (ZIKV) drew worldwide attention when a recent epidemic was linked to fetal microcephaly. Here we used human embryonic stem cell derived trophoblasts as a model for primitive placental trophoblast to test the hypothesis that there are differences in how the two genetically distinct ZIKV lineages, African (AF) and Asian (AS), target the human placenta. Upon infection with three AF (ib-H30656, SEN/1984/41525-DAK, and MR-766) and three AS (FSS13025, MexI-44, and PANcdc259249) ZIKV strains, we observed that severe placental cell lysis was only induced after infection with AF strains, while viral replication rates remained similar between both lineages. Differences in cytopathic effects (CPE) were not observed in Vero cells, indicating that the AF strains were not inherently superior at cell lysis. Taken together, we propose that infection with AF strains of ZIKV early in pregnancy would likely result in pregnancy loss, rather than allow further fetal development with accompanying brain damage. Our results also suggest that the long term laboratory-adapted MR-766 strain does not behave aberrantly in cell culture relative to other AF lineage strains.

Chikungunya virus (CHIKV) is an emerging mosquito-borne virus belonging to the Togaviridae, which is transmitted to humans by Aedes aegypti and Ae. albopictus. We describe the infection pattern of CHIKV in two New World Ae. aegypti strains, HWE and ORL. Both mosquito strains were susceptible to the virus but showed different infection patterns in midguts and salivary glands. Even though acquisition of a bloodmeal showed moderate levels of apoptosis in midgut tissue, there was no obvious additional CHIKV-induced apoptosis detectable during midgut infection. Analysis of expression of apoptosis-related genes suggested that CHIKV infection dampens rather than promotes apoptosis in the mosquito midgut. In both mosquito strains, the virus was present in saliva within two days post-oral infection. HWE and ORL mosquitoes exhibited no salivary gland infection barrier; however, only 60% (HWE) to 65% (ORL) of the females had released the virus in their saliva at one week post-oral acquisition, suggesting a salivary gland escape barrier. CHIKV induced an apoptotic response in salivary glands of HWE and ORL mosquitoes, demonstrating that the virus caused pathology in its natural vector.