Bats are increasingly recognized as reservoirs of emerging zoonotic pathogens, with Egyptian rousette bats (ERBs) being the known reservoir of Marburg virus (MARV), a deadly filovirus. ERBs harbor MARV asymptomatically due to a coadapted host immunity-pathogen relationship. Recent studies show that ERBs use a disease-tolerant strategy to limit pro-inflammatory gene induction, preventing Marburg virus disease (MVD) linked immunopathology. However, the host-resistant strategy by which ERBs actively limit MARV burden remains unclear. This study used dexamethasone to suppress ERB pro-inflammatory responses and assessed MARV replication, shedding, and disease. Results showed that ERBs naturally mount coordinated pro-inflammatory responses at liver foci, involving recruited mononuclear phagocytes and T cells. When these responses are suppressed, ERBs display heightened MARV replication, oral/rectal shedding, and severe MVD-like liver pathology, demonstrating that ERBs balance immunoprotective tolerance with discreet MARV-resistant pro-inflammatory responses. These findings suggest that natural ERB immunomodulatory stressors like food scarcity and habitat disruption may potentiate viral shedding, transmission, and outbreak risk. ERBs, like other bats, have distinct genomic and transcriptomic features indicative of a coevolutionary arms race with ancient viruses. ERBs show expanded MHC loci, diversified natural killer receptor loci, and weaker IFN gene induction. ERBs inoculated with MARV mount virus-specific immune responses, including anti-MARV antibody development and immunological memory to reinfection, although these antibodies are non-neutralizing. Previous studies concluded that the lack of immunopathology in MARV-infected ERBs was explained by "disease tolerance," a strategy focused on controlling inflammatory response activation rather than actively controlling virus burden. This was supported by tissue-level transcriptional analysis showing innate immune gene responses upon MARV infection but mostly canonical IFN signaling genes, a striking difference from the severe cytokine-mediated pathology seen in humans and NHPs with MVD. This study shows that ERBs have coordinated pro-inflammatory responses at liver foci, involving recruited mononuclear phagocytes and T cells. When these responses are suppressed, ERBs display heightened MARV replication, oral/rectal shedding, and severe MVD-like liver pathology. These findings suggest that immunomodulatory ecological stressors like food scarcity and habitat disruption may potentiate viral shedding, transmission, and outbreak risk. The study also highlights the importance of immunological balance in ERBs, as a system pushed too far toward an immune tolerant and unresponsive state can dampen essential early inflammatory signaling, leading to serious virus-mediated pathology. The study demonstrates that ERBs, when deprived of their ability to mount appropriate inflammatory responses, are vulnerable to MVD-like pathology even with MARV loads several orders of magnitude lower than that found in sick NHPs preceding mortality. This suggestsBats are increasingly recognized as reservoirs of emerging zoonotic pathogens, with Egyptian rousette bats (ERBs) being the known reservoir of Marburg virus (MARV), a deadly filovirus. ERBs harbor MARV asymptomatically due to a coadapted host immunity-pathogen relationship. Recent studies show that ERBs use a disease-tolerant strategy to limit pro-inflammatory gene induction, preventing Marburg virus disease (MVD) linked immunopathology. However, the host-resistant strategy by which ERBs actively limit MARV burden remains unclear. This study used dexamethasone to suppress ERB pro-inflammatory responses and assessed MARV replication, shedding, and disease. Results showed that ERBs naturally mount coordinated pro-inflammatory responses at liver foci, involving recruited mononuclear phagocytes and T cells. When these responses are suppressed, ERBs display heightened MARV replication, oral/rectal shedding, and severe MVD-like liver pathology, demonstrating that ERBs balance immunoprotective tolerance with discreet MARV-resistant pro-inflammatory responses. These findings suggest that natural ERB immunomodulatory stressors like food scarcity and habitat disruption may potentiate viral shedding, transmission, and outbreak risk. ERBs, like other bats, have distinct genomic and transcriptomic features indicative of a coevolutionary arms race with ancient viruses. ERBs show expanded MHC loci, diversified natural killer receptor loci, and weaker IFN gene induction. ERBs inoculated with MARV mount virus-specific immune responses, including anti-MARV antibody development and immunological memory to reinfection, although these antibodies are non-neutralizing. Previous studies concluded that the lack of immunopathology in MARV-infected ERBs was explained by "disease tolerance," a strategy focused on controlling inflammatory response activation rather than actively controlling virus burden. This was supported by tissue-level transcriptional analysis showing innate immune gene responses upon MARV infection but mostly canonical IFN signaling genes, a striking difference from the severe cytokine-mediated pathology seen in humans and NHPs with MVD. This study shows that ERBs have coordinated pro-inflammatory responses at liver foci, involving recruited mononuclear phagocytes and T cells. When these responses are suppressed, ERBs display heightened MARV replication, oral/rectal shedding, and severe MVD-like liver pathology. These findings suggest that immunomodulatory ecological stressors like food scarcity and habitat disruption may potentiate viral shedding, transmission, and outbreak risk. The study also highlights the importance of immunological balance in ERBs, as a system pushed too far toward an immune tolerant and unresponsive state can dampen essential early inflammatory signaling, leading to serious virus-mediated pathology. The study demonstrates that ERBs, when deprived of their ability to mount appropriate inflammatory responses, are vulnerable to MVD-like pathology even with MARV loads several orders of magnitude lower than that found in sick NHPs preceding mortality. This suggests