On July 11, 2011, we collected 30 pooled bat urine samples from a pteropid bat roost at Bicentennial Park, Boonah, Queensland, Australia, for an HeV surveillance project. We screened samples using quantitative real-time PCR for HeV. Ten samples were negative for HeV and inoculated onto Vero (African green monkey kidney) and primary Pteropus alecto kidney (PaKi) cell monolayers. We observed no viral cytopathic effects (CPE) after 7 days; however, the Vero tissue culture supernatant (TCSN) from 1 sample (BO13) tested positive when we employed generic reverse transcription PCR primers for paramyxovirus and henipavirus/morbillivirus (12). Sequencing of PCR products revealed a novel henipavirus sequence. Further passage of this virus in Vero cells yielded no CPE, and the virus did not replicate. However, when we inoculated TCSN from Vero cells onto PaKi cells, we observed replication and CPE. We then propagated a working stock in Vero cells, resulting in viral CPE demonstrating small syncytia, cell fusion, and rounded up cells (Appendix Figure). We designated the novel virus Salt Gully virus (SGV) based on the collection location.

Figure 1

Figure 1. Phylogenetic analysis of members of the genus Henipavirus from a study investigating a novel henipavirus, Salt Gully virus, isolated from pteropid bats, Australia.A) We aligned complete...

Next-generation sequencing of RNA extracted from TCSN on the Illumina platform (https://www.illumina.com), followed by genome assembly and analysis, revealed a large, complete genome of 19,884 nt (GenBank accession no. PV233879), adhering to the rule of 6 for paramyxoviruses (13). This genome included 6 distinct open reading frames that encoded 6 proteins: nucleocapsid (N), phosphoprotein (P), matrix (M), fusion (F), glycoprotein (G), and RNA polymerase (L). In addition, an alternative start codon within the P gene indicated the presence of a 7th open reading frame and a C protein, consistent with other henipavirus genomes. Whole-genome nucleotide alignment with other known henipaviruses showed that SGV shared 38% identity with HeV and NiV, 37% identity with Angavokely virus (AngV), and 35% identity with CedV and Ghana virus. We determined SGV to be phylogenetically most closely related to AngV virus (Figure 1).

Figure 2

Figure 2. Growth of Salt Gully virus in Vero cells compared with Hendra virus and Cedar virus from study investigating a novel henipavirus, Salt Gully virus, isolated from pteropid bats, Australia....

Investigating species tropism and growth kinetics, we found that SGV infected Vero, pteropid bat, and human cell lines, showing varying levels of CPE by 3 dpi. We observed no viral CPE in porcine or primary equine cell lines by 7 dpi. HeV replicated in all 5 cell lines, and CedV replicated in all except equine (Table). Initially, SGV displayed delayed replication in Vero cells; however, SGV reached maximum viral titer by 8 dpi at a titer comparable to HeV and CedV, which peaked by 2 dpi and then declined (Figure 2).

Figure 3

Figure 3. Ephrin-B2 and ephrin-B3 host cell receptor usage from a study investigating a novel henipavirus, Salt Gully virus (SGV), isolated from pteropid bats, Australia. Three cell lines—HeLa-USU, recombinant HeLa-USU expressing ephrin-B2,...

We used human epithelial (HeLa)-USU cells and recombinant HeLa-USU cell lines expressing ephrin-B2 or ephrin-B3 to assess SGV receptor usage. We observed viral CPE in HeLa cells expressing ephrin-B2 and ephrin-B3 for HeV and in ephrin-B2-expressing HeLa cells for CedV after 2 dpi. HeV and CedV did not infect HeLa-USU cells, as shown previously (14). In contrast, SGV caused CPE in all 3 HeLa cell lines after 5 dpi, indicating usage of an unknown receptor that is neither ephrin-B2 nor ephrin-B3 (Figure 3).

We detected SGV in bat urine samples collected in Australia in 2011 and successfully isolated the virus using pteropid bat cell lines. Initially, the CPE of inoculated Vero cells was minimal, requiring passage in pteropid kidney cells to achieve efficient virus replication. Despite the identification of multiple new henipa-like viruses in pteropid bats and small mammals globally, viral isolation is largely unsuccessful and remains a technical challenge. In our study, employing primary cell lines derived from the relevant species resulted in virus isolation.

Full-length genome sequencing of SGV revealed a genome organization consistent with other known henipaviruses, with all predicted henipavirus protein ORFs identified. Whole-genome alignments comparing SGV to other henipaviruses revealed 35%–38% identity at a nucleotide level. Of interest, phylogenetic analysis of the genome clusters SGV with AngV, a henipavirus that was detected fruit bats (Eidolon dupreanum) in Madagascar in 2022 (15).

We assessed the ability for SGV to infect various mammalian cell lines in vitro, including Vero, PaKi, HeLa, equine fetal kidney, and porcine kidney cells. SGV’s notable ability to infect human cells underscores its potential for human infection. Unlike HeV, which infected all 5 cell lines, SGV did not cause CPE in porcine or equine cells. SGV could grow to high titer in Vero cells, albeit slower than HeV. Collectively, these results indicate SGV may not have the broad species tropism of pathogenic henipaviruses but could pose a human risk.

Ephrin-B2 and ephrin-B3 are host cell receptors for HeV and NiV (14), and the sequence conservation of those ligands across many species supports the broad species tropism of classical HNV. The HeLa-USU cell line we used has been shown to lack expression of ephrin-B2 and ephrin-B3 and was used to determine the functional host cell receptor for CedV. In our study, SGV infected all 3 HeLa-USU cell lines, demonstrating that SGV cell entry is not reliant on ephrin-B2 or ephrin-B3, suggesting SGV uses a yet unidentified host cell receptor. In comparison, research has shown the glycoprotein G of Ghana virus could bind to ephrin-B2 (but not ephrin-B3), whereas predicted structure-based alignments suggest AngV is unlikely to use ephrin receptors for host cell entry (15). Further characterization is required to determine the functional host cell receptor for SGV and accurately assess pathogenicity in other species.

In summary, we identified, isolated, and characterized a novel henipavirus from pteropid bats in Australia. Amid the increasing discovery of novel henipa-like viruses in new locations and species, SGV is a true henipavirus and phylogenetically clusters with other bat henipaviruses. However, this virus’s pathogenicity remains unknown, making the susceptibility of human and animal populations in Australia uncertain.