After Sepsis, Macrophages Fall Silent: Mouse Study Maps the Immune Collapse Behind Deadly Secondary Pneumonia

Surviving a serious infection is only half the battle. In the days and weeks after sepsis, major surgery, or severe trauma, many patients enter a paradoxical state in which the immune system, having just fought a ferocious inflammatory war, becomes strangely unresponsive to new threats. This window of immunosuppression is when opportunistic pathogens such as Pseudomonas aeruginosa strike, causing pneumonias that are among the leading killers of hospitalized and immunocompromised patients worldwide. Pneumonia claims roughly 2.5 million lives each year according to World Health Organization estimates, and it hits hardest among children, the elderly, and the chronically ill. Now, a team of Dutch researchers has dissected exactly what happens to a key arm of innate immunity during this vulnerable period, and their findings point to macrophages as central players whose silenced alarm systems leave the lungs exposed.

The study, conducted by Suzanne Abbring, Bart Moerings, and colleagues at Wageningen University and partner institutions, used a well-established “two-hit” mouse model to reproduce the clinical sequence seen in intensive care units. The first hit was cecal ligation and puncture, or CLP, a surgical procedure in which the cecum is ligated and pierced with a needle, allowing gut bacteria to leak into the abdominal cavity and trigger polymicrobial sepsis. The second hit, delivered four days later, was an intranasal dose of Pseudomonas aeruginosa, a notorious hospital-acquired pathogen that causes pneumonia. The four-day interval was deliberately chosen because prior work shows the initial hyperinflammatory phase of sepsis resolves within two to three days, after which mice enter the immunosuppressive phase, mirroring the timing of nosocomial infections in humans, which typically emerge three to five days after hospitalization.

Importantly, the researchers designed their first hit to be relatively mild rather than lethal. The procedure produced less than six percent mortality over four days, with a maximum average Clinical Severity Score of just 2.5 out of 10, yet it was clearly sufficient to induce systemic inflammation. Four hours after surgery, CLP mice showed significantly elevated circulating levels of the pro-inflammatory cytokines TNF-alpha, IL-6, and MCP-1 compared with sham-operated controls, and they lost roughly fifteen percent of their body weight over the following days. This sublethal design was intentional: the goal was not to study death from sepsis itself, but to create a clinically relevant state of immune weakening and then examine, in fine detail, how the body responds when a second pathogen arrives.

To find the right dose for the second hit, the team tested a range of Pseudomonas inocula, from one million to one hundred million colony-forming units, and the results revealed a strikingly narrow therapeutic window. At the lowest dose, one million CFU, neither sham nor CLP mice showed any signs of infection at all. At one hundred million CFU, all animals, regardless of prior surgery, rapidly reached humane endpoints, meaning they succumbed to the overwhelming bacterial challenge itself rather than to any immunosuppression. Only at the intermediate dose of fifty million CFU did the difference between the groups emerge: CLP mice maintained persistently elevated clinical severity scores twenty hours after infection, while sham mice began to recover. This dose-dependence, the authors note, means the optimal inoculum cannot simply be extrapolated from previous studies, since differences in mouse strain, sex, bacterial strain, and route of administration can dramatically shift outcomes.

When the researchers examined the lungs twenty hours after infection, they uncovered the molecular signature of immunosuppression. Although the number of Pseudomonas colonies did not differ significantly between sham and CLP animals, the pulmonary cytokine response told a very different story. CLP mice mounted significantly weaker production of TNF-alpha and IL-6, the canonical pro-inflammatory signals that recruit and activate immune defenses, compared with sham mice given the same bacterial dose. Levels of interferon-gamma were also reduced in CLP animals, and this suppression was present even in animals instilled with saline rather than bacteria, indicating it reflected the lingering effects of the first hit rather than the second. Surprisingly, levels of IL-10, an immunosuppressive cytokine often elevated after sepsis in previous studies, were actually lower in the lungs of CLP animals, a discrepancy the authors attribute to differences in experimental design and the inclusion of proper saline controls.

The chemokine data reinforced the picture of a muted immune alarm. Chemokines such as CXCL9, CXCL10, and CCL22, which orchestrate the recruitment of macrophages and other leukocytes to sites of infection, failed to rise in CLP animals after Pseudomonas challenge, whereas sham mice showed clear increases. CXCL9 and CCL22 were significantly reduced across all bacterial doses tested, and CXCL10 and CXCL5 were reduced in most groups. Only CXCL1, a neutrophil-attracting chemokine, remained comparable between the two groups. Together, these findings suggest that after an initial septic insult, the lung’s ability to broadcast the molecular distress signals needed to summon cellular reinforcements is profoundly impaired, even when bacteria are physically present in the tissue.

To isolate the role of macrophages specifically, the researchers turned to an ex vivo approach. They collected spleen-derived macrophages and adherent bone marrow cells from CLP and sham animals that had not been exposed to Pseudomonas, then stimulated these cells in culture with lipopolysaccharide, the bacterial endotoxin that robustly activates macrophages. The results were unambiguous: macrophages from CLP animals released significantly lower amounts of TNF-alpha and IL-6 in response to LPS than those from sham animals, while simultaneously producing significantly higher levels of the anti-inflammatory cytokine IL-10. This shift, in which pro-inflammatory output is blunted while IL-10 production is preserved or enhanced, is a hallmark of endotoxin tolerance and immune paralysis. Critically, the fact that CLP macrophages could still produce IL-10 demonstrated that the cells were not dead or completely inert; they retained functional capacity but had been reprogrammed toward a suppressed state.

That residual capacity opened the door to the study’s most intriguing experiment. Building on their earlier work showing that arabinoxylans, non-digestible polysaccharides found in cereal grains, can induce trained immunity in human macrophages, the team exposed spleen-derived macrophages from CLP and sham animals to five different arabinoxylan preparations before LPS stimulation. The results were preliminary but promising. A rice bran-derived preparation significantly enhanced TNF-alpha production in macrophages from CLP animals, partially reversing the immunosuppressed phenotype, while other rice-derived preparations showed moderate effects on TNF-alpha or IL-6 production. Rice hull and two other rice bran fractions showed tendencies toward restored cytokine output, though not all preparations were effective, and wheat bran had no impact. The authors stress that this pilot experiment lacked the statistical power to draw firm conclusions, but it suggests that dietary fibers may hold genuine potential as immune-supportive interventions for immunocompromised patients.

The study also carries important methodological lessons for the field. The researchers found that total bacterial load in the lungs of CLP animals was elevated even in saline-instilled controls, likely reflecting bacterial dissemination from the abdominal infection rather than impaired pulmonary clearance of Pseudomonas. This finding underscores the necessity of including saline-instilled CLP control groups when assessing susceptibility to secondary infection, a control that several previous studies omitted. The authors also acknowledge limitations: their macrophage cultures were derived from heterogeneous cell populations, the analysis of macrophage polarization states such as M1 and M2 phenotypes was not performed, and sample sizes were based on prior experience rather than formal power calculations. Only male mice were studied, so whether the findings extend to females remains unknown.

Nevertheless, the broader significance of the work is clear. By systematically mapping the cytokine, chemokine, and cellular responses across multiple bacterial doses, the study provides one of the most comprehensive characterizations to date of the immunosuppressive phase that follows sepsis-like insults. It confirms that macrophage dysfunction, rather than simply an inability to clear bacteria, lies at the heart of vulnerability to secondary pneumonia, and it demonstrates that these silenced cells can be coaxed back toward responsiveness with the right stimuli. The ex vivo LPS challenge model, the authors argue, offers a practical and ethically advantageous platform for screening immune-enhancing compounds before moving into animal or human studies, potentially reducing animal use while accelerating the search for therapies. As antibiotic resistance rises and populations age, strategies that restore immune resilience, whether through dietary fibers, targeted immunomodulators, or combinations thereof, may become essential tools for protecting the millions of patients each year whose weakened defenses leave them exposed to the opportunists waiting in the hospital ward.

Subject of Research: Macrophage immunosuppression and secondary Pseudomonas aeruginosa pneumonia susceptibility in a murine two-hit sepsis model

Article Title: Macrophage immunosuppression in a two-hit murine Pseudomonas aeruginosa -induced pneumonia model

Article References: Abbring, S., Moerings, B. G., Furber, M., Witkamp, R. F., Govers, C., van Bergenhenegouwen, J., Mes, J. J., & van Norren, K. (2026). Macrophage immunosuppression in a two-hit murine Pseudomonas aeruginosa-induced pneumonia model. Heliyon, 12(15), Article e45561. https://doi.org/10.1016/j.heliyon.2026.e45561

Image Credits: AI Generated

DOI: 10.1016/j.heliyon.2026.e45561

Keywords: macrophages, immunosuppression, sepsis, Pseudomonas aeruginosa, pneumonia, two-hit model, cecal ligation and puncture, cytokines, chemokines, arabinoxylans, endotoxin tolerance, innate immunity