Enhanced antiviral innate immunity supports the bat reservoir of Marburg virus

Abstract

Basal innate immunity with enhanced innate immune activation and response in a bat reservoir of Marburg virus underlies virus tolerance and zoonotic potential.

Defining how specific animals serve as viral reservoirs is paramount to understanding mechanisms of virus control and zoonotic potential for a virus to jump to humans. In this issue of Nature Immunology, Kellner et al.¹ applied single cell transcriptomic and tissue organoid analyses of entire respiratory and intestinal epithelium of the Egyptian fruit bat Rousettus aegyptiacus, a reservoir of Marburg virus (MARV). Direct comparison with human cells and organoids revealed that constituent cells of the R. aegyptiacus epithelia, but not human epithelia, are in an ongoing state of innate immune activation, marked by constitutive production of interferon (IFN)-ε, a non-canonical type I IFN, linking to basal innate immune effector gene expression¹. Moreover, when challenged with MARV, R. aegyptiacus cells and organoids rapidly induced type III IFN and high expression of IFN-stimulated genes (ISGs) within a global response that suppresses MARV replication and inflammation, while the human innate immune response to infection was comparatively blunted and instead polarized to inflammatory signaling¹. Thus, an enhanced innate immune response lacking inflammatory output is linked with MARV control in reservoir R. aegyptiacus to tolerate ongoing infection, allowing for inter-host virus dissemination and dangerous zoonotic potential to humans.

Bats belong to the mammalian order Chiroptera with over 1400 species worldwide. They are a natural reservoir for pathogenic human viruses and themselves are uniquely resilient to viral infection. R. aegyptiacus is a known reservoir of the filovirus MARV and when spillover occurs into humans, it causes deadly hemorrhagic fever disease². The basis for tolerance to viral infection in bats has been largely unknown, due to the limited capacity of bats as a model organism and an insufficient system for genetic manipulation. During infection viral pathogen associated molecular patterns (PAMPs) trigger pathogen recognition receptors that signal innate immune activation to induce type I and III IFN, culminating in the expression of hundreds of ISGs. ISG products have antiviral and immune modulatory activity to control infection³. Intriguingly, bats have evolved a heightened level of basal innate immune gene expression, including specific induced ISGs, making them distinct among other mammals in exhibiting immune tolerance to viral infection⁴.

Engineered organoids are a tissue-derived in vitro model that have started to fill the knowledge gap in understanding tissue function⁵. These self-organized three-dimensional cultures resemble the physiologic state of tissues and their multicellular composition in vivo. Organoid development requires a combination of factors and fine-tuned culture conditions to demonstrate composition, structure and functionality, as well as promote long-term growth. Kellner et al.¹ produced both bat airway and intestinal epithelial organoids derived from cryopreserved input tissues that were constructed to evaluate innate immune responses to viral infection and compared to similar human organoids cultured under identical conditions. Single cell RNA-sequencing was applied to identify distinct clusters of immune, stromal and epithelial cell types and specific single cell transcriptome from R. aegyptiacus epithelial tissues under different conditions, thus producing a high-resolution single cell transcriptomic atlas of cell type responses. The bat lung organoids were confirmed to have the cell-type diversity of regional airway epithelium. These bat lung organoids were comprised of cells expressing lineage-specific markers showing that alveolar type I (AT1) progenitor cells, alveolar type II (AT2) pneumocytes and other cells types could be enriched to establish bat alveolar organoids, and that these organoids can be expanded for more than six months, using specific cytokine and factor treatments. In addition to cell type diversity, the bat organoids retained in vivo characteristics, including basolateral polarity and cilia formation at the apical surface, as well as differentiated goblet cells.

For the bat intestinal organoids there was a high concordance of expected cell types between organoids and primary intestinal tissues, including intestinal stem cells, goblet cells, enterocytes (early and mature) and rare epithelial cell types, such as tuft cells, enteroendocrine cells and Paneth cells¹. Compared to the human organoids, R. aegyptiacus organoid intestinal enterocytes have high levels of IFN-ε¹. The expression of this IFN subtype is not driven by viral PAMP and pathogen recognition receptor signaling and it remains unknown how bat IFN-ε expression and production occur. However, the ISG response driven by bat IFN-ε clearly protects the epithelia against high level MARV replication and virus spread, providing a level of tolerance to MARV infection (Fig. 1).

Fig. 1. Schematic of bat and human airway and intestinal organoids and their response to Marburg virus infection.

R. aegyptiacus airway (a) or intestinal (b) epithelium organoids tolerate and control MARV infection with no apparent inflammatory signaling. Kellner et al.¹ show that basal levels of IFN-ε production in both airway and intestinal bat organoids, respectively by basal respiratory cells and small intestinal enterocytes, associate with enhanced levels of interferon-stimulated gene (ISG) expression. This work shows that IFN-ε signaling (IFN-ε +++) serves to induce and maintain a level of ISG expression that links with control of MARV. When exposed to RNA virus infection bat organoids were shown to rapidly induce type III IFN (IFN-λ+++)) and high ISG expression in the absence of inflammatory gene expression¹. This response effectively controls MARV infection and disease in the bat host. The human response to MARV is comparably blunted (IFN-ε -, IFN-λ+) and inflammatory¹ in both airway and intestinal organoids. Krt5+ basal respiratory cells, goblet cells, and ciliated cells (a), and tuft cells, Paneth cells, and crypt base columnar (CBC) cells (b) are depicted in the tissue architecture.

The R. aegyptiacus organoids expressed cellular receptors for multiple zoonotic viruses, including the MARV cellular receptor NPC1, and were clearly susceptible to MARV infection¹. Type III IFNs (IFNL1-like and IFNL3-like genes) were the most strongly induced IFN genes in the bat organoids, demonstrating a key role of type III IFN signaling in bat antiviral immunity¹. Type III IFN induction was not specific to MARV infection and was also seen in response to organoid infection by vesicular stomatitis virus, Middle East respiratory coronavirus (MERS), and influenza A virus. It is notable that bat organoids showed minimal proinflammatory gene expression induction in response to MARV infection or treatment with bat IFNL1 or universal IFNα2, as compared to the high induction of pro-inflammatory genes in human air liquid interface epithelial cell cultures and human intestinal organoids in response to similar treatment¹. Thus, the R. aegyptiacus organoid resilience to stimulus by pro-inflammatory signals and the induction of specific type III IFNs in response to viral infection differentiates them from the human organoid counterpart, and type III IFN provided long-lasting protection against RNA viruses in the R. aegyptiacus epithelium. These results, along with the single cell transcriptomic atlas produced by Kellner et al.¹, set the stage for characterization of other pathogens known to be harbored by bats, such as rabies virus, coronaviruses, paramyxoviruses, and others, for further understanding of viral control and insight into possible zoonotic transmission.

Organoids provide a tractable system for genetic manipulation using the gene editing tool CRISPR-Cas9 to perform mechanistic studies in evaluating the contribution of IFN-ε and type III IFN in bat antiviral immunity. However, technical limitations remain, including access to fresh bat tissue and the lack of bat-specific molecular reagents, such as bat type III IFNs. Kellner et al.¹ clearly showed that bat organoids can be produced from cryopreserved tissues, and they produced, and demonstrated the use, of bioactive bat type III IFN, thus enabling the organoid technology and bat type III IFN reagents to be shared and co-studied among different laboratory groups. This type of organoid systems have far-reaching applications, including antiviral drug discovery, infection and innate immunity biomarker identification, and genetic analysis of innate immune defense or host factors of viral infection and replication.