While immune cells play a critical role in early innate immune responses and microbial clearance during infection, excessive immune cell activity and the inflammatory host responses this activation may elicit, can also lead to inadvertent host tissue injury. The host has therefore developed mechanisms to control and coordinate immune cell activity. The programed death 1 (PD-1) signaling pathway is one such mechanism. The development of PD-1 expression on lymphocytes and its stimulation by PD-1 ligand (PD-L1) on other immune cells is associated with a suppressed immune phenotype. Although the suppressed state associated with PD-1 expression is the hosts attempt to keep an excessive immune response in check, if it occurs too early during infection or if it is too great during later infection, PD-1 signaling may actually impair host defense and microbial clearance and worsen outcome. Studies have now suggested that the blockade of PD-L1/PD-1 signaling can restore lymphocyte host defense function and may be protective under some conditions. In one study, PD-1 knockout animals were resistant to cecal-ligation and puncture (CLP)-induced septic shock. In other studies, provision of an anti PD-1 or anti PD-L1 antibody both very early after the onset of infection (3 hours after induction of peritonitis) and much later (up to 5 days after induction of peritonitis) provided protection. One group has also shown that PD-1 blockade leads to a reduction in peritoneal lavage bacteria colony counts in conjunction with higher plasma levels of inflammatory cytokines. Thus, modulating the PD-1/PD-L1 pathway holds promise in the treatment of severe infection. However, understanding under which conditions such inhibition will have beneficial as opposed to harmful effects is important for determining how widely PD-1 inhibition might be applied during clinical infection. Because the severity of underlying infection is one condition that varies substantially in patients presenting with sepsis and has an important impact on the efficacy of other immune-modulators, determining whether PD-1/PD-L1 inhibition has beneficial effects with both mild and severe infection is a critical question. Also, because, PD-1/PD-L1 inhibition has been studied exclusively in peritoneal infection models, determining whether it is also beneficial in the context of pulmonary infection, the most common cause of sepsis clinically, is important. To test these questions, the present study was designed to compare the effects of an anti-PD-L1 mAb versus placebo in mice challenged with intratracheal (IT) doses of S. aureus (SA) bacteria (i.e. a pneumonia model) designed to produce either low or high mortality rates (25% mortality, mild infection vs. 75%, severe infection respectively). The study was designed and completed in 2 parts. The first part (Part 1) included experiments to determine the doses of bacteria that will produce the low or high mortality rates that we seek and to also confirm that IT infection alters PD-1/PD-L1 expression on circulating and tissue immune cells. The second part (Part 2) included two groups of experiments. One group compared the effects of anti-PD-L1 mAb versus control on survival in mice challenged with IT SA low or high doses and observed for up to 168h. The other group compared the effects of these two treatments on blood and lung bacteria counts, cytokine levels, PD-1/PD-L1 tissue expression, tissue injury scores and other measures. Recent Progress 1. Since determining the two doses of SA leading to high or low mortality, we have interrogated the local immune response to low or high dose SA. We have also determined that innate immune cells are recruited to the lungs in large numbers and that PD-L1 and PD-1 are modulated differently on lung immune cells from animals infected with low vs high dose SA. 2. We have analyzed lungs from these animals by histology and there are changes related to infection that are consistent between the two groups. 3. We have analyzed the serum samples collected from the animals over the past year as a batch. We have evaluated circulating cytokines as well as soluble PD-1 and PD-L1. We designed a custom protein detection platform for the latter assay. 4. We have also conducted treatment studies evaluating mortality following blockade of the checkpoint pathway with an anti-PD-L1 antibody over a range of doses of bacteria and at various times after the onset of infection. 5. We have evaluated the levels of circulating antibody administered, the degree of checkpoint molecule blockade on immune cells, the effects of antibody administration on bacterial colony counts in the lungs of infected animals, and on histological changes. An abstract with data generated from this study was presented at the annual American Thoracis Society meeting in Dallas, TX in May 2019. We have also completed a systematic review of the literature surrounding checkpoint molecule expression on circulating immune cells from patients with sepsis compared to critically ill non septic patients or healthy volunteers. Inhibitory Immune Checkpoint Molecule Expression in Sepsis: a Systematic Review of the Literature. Busch L.M., Sun J., Eichacker P.Q., Torabi-Parizi P. Manuscript submitted to Critical Care Medicine. Additionally, we have performed a systematic review with meta-analysis of all studies employing animal models of sepsis and severe infection that have examined the mortality effects of checkpoint blockade. A Critical Evaluation of Preclinical Sepsis Models Testing the Efficacy of Checkpoint Inhibitors: A Systematic Review and Meta-analysis. Busch L.M., Sun J., Cui X., Welsh J., Eichacker P.Q., Torabi-Parizi P. Manuscript submitted and under review at Intensive Care Medicine.
