S1PR2: A Fulcrum in the Balance of Type 1 and Type 2 Responses during Sepsis-induced Acute Lung Injury

The paradigm of type 1 and type 2 immune responses was first described nearly 40 years ago, with the discovery of T-helper type 1 (Th1) cells secreting inflammatory cytokines such as IFN-γ in response to bacterial and viral infection and Th2 cells releasing IL-4 in the context of extracellular parasite and helminth infection (1, 2). Since then, numerous other cells and cytokines have been incorporated into an increasingly complex model, with additional responses identified (e.g., Th17-associated type 3 responses), as well as newly discovered functions (e.g., the role of type 2 responses in tissue repair and homeostasis) (3). What remains clear, however, is the notion of equilibrium: Type 1 responses and type 2 responses counterbalance each other, and a perturbation of this balance is often associated with the development of disease (3).

It is through this lens that Gong and colleagues (pp. 215–225) approached their study of sepsis-induced acute lung injury (ALI) in this issue of the Journal (4). Sepsis is the most common cause of acute respiratory distress syndrome (ARDS), with over 200,000 patients/yr identified with sepsis-induced ARDS in the United States alone (5). Sepsis-induced ALI occurs when an infection (either within the lungs or extrapulmonary) triggers a systemic inflammatory type 1 or type 3 response, resulting in damage to the alveolar-capillary barrier and the canonical development of pulmonary edema and impaired oxygenation. Consistent with the aforementioned equilibrium hypothesis, a growing body of evidence suggests that type 2 responses (including eosinophils, type 2 innate lymphoid cells, and cytokines such as IL-4, IL-5, and IL-13) may be able to balance type 1 and type 3 responses and ameliorate this sepsis-induced ALI (6–9). In addition, prior work has implicated a role for the sphingolipid sphingosine 1-phosphate (S1P) in the maintenance of the alveolar-capillary barrier, acting through various S1P receptors to promote or inhibit ALI (10). Indeed, in mouse models, increased levels of the S1P receptor 2 (S1PR2) have been associated with ALI, whereas lack of S1PR2 reduced proinflammatory cytokine release and mortality during sepsis (11, 12). Given that S1PR2 has also been associated with type 2 allergic immune responses (13), the authors hypothesized that S1PR2 may play a critical role in regulating the balance between type 1 and type 2 responses during sepsis-induced ALI.

In comparing patients with sepsis with healthy control subjects, the authors observed activation of type 2 immune responses among patients with sepsis 24 hours after admission, with both cells and cytokines upregulated (4). Intriguingly, the magnitude of these responses was associated with higher arterial oxygen tension/fraction of inspired oxygen ratios and lower Sequential Organ Failure Assessment scores, suggesting that the host response of patients who were less severely ill was characterized by a stronger type 2 response. They also noted that expression of S1PR2 mRNA in circulating monocytes from these patients was inversely associated with type 2 responses, prompting a chicken-and-egg question: Does the type 2 response reduce S1PR2 expression, or does a reduction in S1PR2 expression potentiate activation of type 2 responses?

To address this question, the authors turned to a mouse cecal ligation and puncture model of sepsis, using S1PR2-null mice, mixed bone marrow chimeras, and inhibitors of S1PR2 signaling. In a series of elegant experiments, the authors identified that S1PR2 expression specifically on pulmonary macrophages regulated type 2 responses during sepsis, with lack of macrophage S1PR2 (either genetically or through pharmacologic inhibition) augmenting type 2 responses and resulting in reduced sepsis-induced ALI. Furthermore, the authors determined that this regulation was mediated by the type 2 alarmin cytokine IL-33: Lack of macrophage S1PR2 led to increased macrophage IL-33 expression and subsequent protection from ALI, consistent with the known capacity of IL-33 to protect against sepsis-induced lung injury (7).

These data raise several additional questions. First, how does S1PR2 regulate IL-33 secretion? Regulation of IL-33 is notoriously distinct from that of either traditional Golgi-trafficked, secreted cytokines but also from other IL-1 family cytokines that require caspase processing for activation and release; understanding the connection between S1PR2 and IL-33 during both homeostasis and disease would therefore be of significant interest (14). Second, although the presence of S1PR2 clearly suppresses beneficial type 2 immune responses, S1P signaling through other S1P receptors can also protect against lung injury through reinforcement of the alveolar-capillary barrier; what regulates the balance between S1P’s divergent actions in maintenance of the endothelial barrier while simultaneously preventing type 2 responses from repairing that barrier? Alternatively, does this mechanism exist to facilitate repair, wherein a reduction of S1P levels over the course of an inflammatory response will simultaneously lead to a reduction in barrier integrity but also facilitate IL-33 release, which will lead to barrier restoration? Third, the authors used a nonpulmonary model of sepsis to induce lung injury; however, the question whether this mechanism applies in pneumosepsis (which constitutes the majority of sepsis-associated ARDS) remains unanswered. Finally, is modulation of S1PR2 function a viable strategy for management of sepsis-associated ALI?

In keeping with the Sepsis-3 (Third International Consensus Definitions for Sepsis and Septic Shock) definition, sepsis-associated ALI may be defined as a dysregulated host response to infection that results in inflammatory injury in the lungs. Gong and colleagues have revealed an important role for macrophage-expressed S1PR2 in restricting endogenous IL-33–dependent type 2 immune responses during sepsis-induced ALI, potentially representing a form of “dysregulation.” Thus, the problem is not simply that a septic type 1/type 3 inflammatory response was mobilized to respond to a pathogen and incidentally caused lung injury; rather, the problem is also that the suppression of the type 2 response by S1PR2 prevented appropriate lung healing. By extension, restoration of a balanced, regulated response might result in better outcomes. In conducting this work, Gong and colleagues have provided additional evidence that the immune system in disease is best considered in terms of the balanced relationship between its varying elements rather than simply too much or too little of any single element.