Hematopoietic Stem and Progenitor Cells Improve Survival from Sepsis By Boosting Immunomodulatory Cells

Sepsis is a dysregulated inflammatory syndrome that accounts for as many as 20% of deaths worldwide. Elevated production of pro-inflammatory cytokines during sepsis, such as IL-1, IL-6, interferons (IFNs), and tumor necrosis factor contribute to the development of fever, vasodilation, and multiorgan failure. Novel therapies to treat sepsis are urgently needed.

Hematopoietic stem and progenitor cells (HSPC) are responsible for the day-to-day production of blood and immune cells. Recent work from our group and others indicates that during emergency hematopoiesis, inflammatory signals including cytokines, chemokines, and pathogen-derived molecules direct HSPCs to differentiate into effector immune cells. While these signals are essential for a proper immune response, excessive signaling in HSPCs can be detrimental and lead to their depletion. Thus, the interactions between HSPCs and their inflammatory environment may play a deterministic role in immune responses and sepsis.

We used a mouse model of Group A Streptococcus (GAS) infection to examine the role of HSPCs in pathogenic infection and sepsis. GAS is a common pathogen that can cause a plethora of diseases from mild skin infections to life-threatening necrotizing fasciitis. We infected mice with 10 ⁶ cfu GAS by intramuscular injection, which typically results in sepsis and death within 7 days, and examined the impact of this infection on peripheral blood (PB) and bone marrow (BM) populations. In just 24 hrs after GAS infection, BM myeloid and HSPC populations are significantly depleted, with myeloid cells being heavily trafficked into circulation following increased levels of monocyte chemoattractant protein-1 (MCP-1). Lineage tracing experiments using KRT18-CreERT2:Rosa26-lox-STOP-lox-TdTomato demonstrated that endogenous HSPCs differentiate toward the myeloid lineage after GAS infection. Based on these data, we hypothesized that the inflammatory environment of GAS infection drives rapid HSPC differentiation resulting in a depletion that could be rescued by the infusion of new HSPCs. To test this hypothesis, we infused GAS-infected mice with 10 ⁴ naïve HSPCs (1.7x10 ⁷ cells per m ²) and evaluated pathogen load and overall survival. This number of HSPCs infused is very low in comparison to the current granulocyte therapies that use ~10 ¹⁰ cells per m ² cells per infusion. BM and PB analysis showed that HSPC infusion restored HSPC levels and significantly increased myeloid progenitors and circulating myeloid cells. Strikingly, HSPC infusion in GAS-infected mice significantly increased survival, with 50-75% of mice surviving as opposed to 0-10% of controls. Despite the restoration of hematopoietic populations, surprisingly, GAS-infected mice infused with HSPCs did not show a reduction in pathogen load.

Given that HSPC infusion significantly increased survival without impacting pathogen clearance, we sought to determine whether infused HSPCs served an immunomodulatory role. Analysis of BM and PB did not show any changes in lymphocyte populations, suggesting that Tregs and Bregs were not strongly affected. However, BM and PB MDSC populations were severely depleted during GAS sepsis, and HSPC infusion led to a dramatic restoration of these MDSC populations. Interestingly and in accordance with MDSC numbers, the overall cytokine levels of GAS-infected mice are lower after HSPC infusion. Notably, serum levels of cytokines known to drive the symptoms of sepsis, like TNF, IL-12, MIP-1a, IL-6, and IL-1b were dampened in HSPC-rescued mice.

In conclusion, while HSPC infusion did not reduce bacterial load, it conferred a significant survival advantage to GAS-infected mice. Our data showing restoration of MDSCs and lower cytokine levels after HSPC infusion suggest that HSPC infusion supports the development of immunomodulatory cells that can prevent sepsis-related hyperinflammation and death. Current work is directed at defining specific HSPC subpopulations that mediate this effect. Importantly, the rescue potential of such low numbers of infused HSPCs highlights the feasibility of this technique and its potential applications. Overall, the information gained in this project may contribute to a new therapeutic strategy to use HSPCs to fight bacterial infections and sepsis where granulocyte infusions have so far produced only mixed results.