Australian bat lyssavirus (ABLV) is a rhabdovirus of the genus Lyssavirus endemic in Australian bat populations that is capable of causing a fatal neurological disease in humans indistinguishable from clinical rabies. There are two genetically distinct variants of ABLV: the Pteropus strain (ABLVp) which circulates in all four species of frugivorous flying foxes (genus Pteropus) present on mainland Australia, and the Saccolaimus strain (ABLVs) present in the insectivorous yellow-bellied sheathtail bat (genus Saccolaimus). ABLV has caused three fatal human infections, the most recent in February 2013, and each manifested as acute encephalitis; however, incubation periods were variable, ranging from approximately 5 weeks to over 2 years [1–3]. In May 2013, ABLV infected two horses, representing the first spillover of ABLV into a terrestrial species other than humans [4, 5]. Although the number of ABLV spillover events have thus far been scarce and limited to only two terrestrial species, in vitro tropism studies indicate that the unknown ABLV receptor(s) is broadly conserved among mammals and suggests that terrestrial species other than humans and horses may be susceptible to ABLV infection .
ABLV is an enveloped, bullet-shaped, non-segmented, negative sense RNA virus belonging to the genus Lyssavirus of the family Rhabdoviridae within the order Mononegavirales. There are currently 12 species of lyssaviruses and 3 additional species that have not yet been classified ; ABLV is most closely related to classical rabies virus (RABV), the prototype member of the Lyssavirus genus. Lyssaviruses have a single envelope glycoprotein (G) that mediates all internalization steps from cell attachment to pH-dependent fusion with the host cell membrane . The lyssavirus G glycoprotein is a key determinant of the neurotropic and neurovirulent properties of lyssaviruses .
Rhabdoviruses gain access to the host cell cytoplasm, the site of viral replication, by receptor-mediated endocytosis and subsequent low pH-dependent fusion [10, 11]. Viruses internalized by receptor-mediated endocytosis predominantly use trafficking pathways mediated either by clathrin or macropinocytosis, but several alternative pathways have been reported (reviewed in [12, 13]). Different viruses utilize different pathways for entry and the endocytic pathway taken by a given virus largely depends on the host receptor it interacts with and the size of the virus. Clathrin-mediated endocytosis (CME) is the most commonly utilized entry pathway for viruses small and intermediate in size  and is the only pathway reported to be utilized by different rhabdoviruses, including VSV, RABV, and infectious hematopoietic necrosis virus (IHNV) [14–16]. The initial virus-host cell receptor interaction induces de novo clathrin-coated pit (CCP) formation at the site of viral binding [17, 18]. Both VSV and RABV were shown to internalize through partially coated clathrin pits that require actin to complete the internalization process [15, 18]. Entry of IHNV was also shown to be actin-dependent . After invagination, the CCP buds from the plasma membrane forming a clathrin-coated vesicle. The clathrin coat is then shed, enabling the uncoated vesicle to traffick to and fuse with the early endosome. The entry pathway of ABLV has not been examined; however, we recently demonstrated that disruption of lipid rafts via sequestration of cholesterol by methyl-β-cyclodextrin (MβCD) treatment significantly reduced ABLV G-mediated entry into HEK293T cells . This finding is compatible with a clathrin- or caveolae -dependent (CavME) entry pathway for ABLV; acute cholesterol depletion with MβCD has been shown to inhibit CCP budding  and caveolae formation is strictly dependent on cholesterol .
Once internalized by endocytosis, the acidic environment of the endosome triggers the conformational changes in the G glycoprotein that lead to fusion of the viral and endosomal membranes and the subsequent release of the viral genome into the cytosol. For most enveloped viruses, membrane fusion occurs either in early endosomes (EEs) (pH 6.5-6.0) or late endosomes (LEs) (pH 6.0-5.0) (reviewed in ). Recycling endosomes (REs) are responsible for directing vesicular cargo back to the plasma membrane and have been shown to be involved in some virus budding pathways  and particle assembly pathways (reviewed in ). The use of dominant negative (DN) Rab GTPases specific for EEs, LEs, and REs, have been used extensively to determine virus exit points from the endosomal network [23–25]. Overexpression of the mutant Rab proteins in host cells elicits a dominant negative effect over the endogenous wild-type protein; thus, a virus that requires a specific endosome for trafficking and fusion will be adversely affected by the overexpression of a DN Rab protein specific for that endosome. Lyssaviruses require vacuolar acidification for G glycoprotein-mediated viral and endosomal membrane fusion , but few studies have examined the endosomal trafficking of these viruses. RABV G was shown to co-localize with Rab5a, a marker for EEs, but not Rab9, a marker for LEs; this same study demonstrated that internalization of an anti-RABV G antibody/RABV G complex was dependent on the presence of functional Rab5a suggesting that RABV G fuses with EEs . Furthermore, the optimal pH for RABV G-mediated fusion is pH 5.8-6.0, which correlates with the pH of EEs . The membrane fusion requirements for ABLV have not been investigated.
In the present study, we used fluorescent protein-encoding recombinant vesicular stomatitis viruses (rVSV) that express ABLV G envelope glycoproteins to examine the ABLV entry pathway using both chemical and molecular approaches. Results indicate that ABLV is internalized into HEK293T cells by a clathrin-dependent pathway that like VSV, IHNV, and RABV, is dependent upon actin. Moreover, we show that ABLV G-mediated entry is Rab5-dependent, but Rab7-independent, indicating that ABLV G-mediated fusion occurs within the early endosomal compartment.