Epithelial STAT3: A New Twist in Inducible Antimicrobial Resistance

Lower respiratory tract infections continue to be a leading cause of mortality, morbidity, and health care expenditures globally, affecting more than 300 million individuals and representing the sixth leading cause of death (1, 2). The term “lower respiratory tract infections” refers to a broad range of diseases, including acute bronchitis, pneumonia, and acute exacerbation of chronic lung diseases; pneumonia affects 24.8 per 10,000 adults annually (3). According to a World Health Organization report in 2019, pneumonia is the single largest infectious cause of mortality in children worldwide, killing 740,180 children under 5 years of age and accounting for 22% of all deaths of children 5 years of age and younger. Pseudomonas aeruginosa is an opportunistic gram-negative bacterium shown to be associated with infections in the lung (hospital-acquired and community-acquired infections), blood, and other parts of the body after surgical intervention (3). Although empirical treatment for P. aeruginosa infections includes antibiotics, P. aeruginosa is becoming more difficult to treat because of its intrinsic antimicrobial resistance to β-lactams, adaptability to different environments, and wide metabolic flexibility (4, 5). Therefore, there is an increased interest in understanding and modulating the immune response for the development of host-targeted therapies to better combat these infectious diseases in the lung.

The lung epithelium has developed a multifaceted defense mechanism to deal with dynamic interactions with microbes via the sensing of pathogens by germline-encoded cytosolic and membrane-bound pattern recognition receptors such as Toll-like receptors (6) and Nod-like receptors (7) and the launching of innate immune defenses by the release of antimicrobial peptides, reactive oxygen species (ROS), and cytokines (8, 9). The innate immune defense against acute bacterial infection in the lung involves alveolar transmigration and activation of neutrophils, which include autocrine and paracrine cross-talk between myeloid cells and parenchymal cells. The production of CXC chemokines, including CXCl1, CXCL2, and CXCL5, is the first step in a multistep process resulting in neutrophil-dependent host protection (10). Whereas CXCL1 and CXCL2 can be produced by myeloid and structural cells, CXCL5 is primarily produced by the structural (epithelial) cells during inflammation (10). STAT3 (signal transducer and activator of transcription 3) is a critical regulator of tissue resilience that is essential for preventing injury, and lung epithelial STAT3 has been shown to drive pulmonary defense through signaling from IL-6 and IL-22 (11). For the classical cascade of signal transduction, STAT3 lies dormant in the cytoplasm of resting cells until they are activated by common cytokines and growth factors that are often ligand- and cell type–specific (12). The best characterized pathway for activation of STATs is through phosphorylation by JAKs (Janus kinases) (12). However, STAT1 and STAT3 have been shown to be activated by EGFR (epidermal growth factor receptor) and PDGFR (platelet-derived growth factor receptor), which are receptors with intrinsic tyrosine kinase activity, and by other non–receptor tyrosine kinases besides JAKs (13). Mutations in the STAT3 gene in humans can result in hyper-IgE or Job’s (immunodeficiency) syndrome, which leads to augmented innate immune responses and recurrent pneumonia, and mice with defective STAT3 in alveolar epithelial cells exhibited decreased neutrophil recruitment and higher bacterial burden in the lungs compared with control mice during Escherichia coli–induced pneumonia (14). Using the murine model of gram-negative (E. coli) pneumonia, select members of the IL-6 cytokine family were shown to increase the production of CXCL5 by activation of lung epithelial cells in a STAT3-dependent fashion (15). The mRNA expression level of a member of the IL-6 cytokine family, oncostatin-M (OSM), was increased in the lung during E. coli infection, and OSM promoted neutrophil infiltration through STAT3-dependent production of CXCL5 from lung epithelial cells (15, 16), implying an important role of autocrine signaling in the inflammatory milieu in pneumonia.

In this issue of the Journal, Kulkarni and colleagues (pp. 679–688) demonstrate how STAT3 activation can be induced by a combination of a TLR2/6 agonist (Pam2CSK4) and a TLR9 agonist (ODN M362) (together referred to as Pam2ODN) and outline a redox-based mechanism for STAT3 activation in the epithelium for inducible antimicrobial resistance (17). Although their previous studies showed that Pam2ODN-induced protection is mediated by lung epithelial cells (18, 19), Kulkarni and colleagues expanded on the signaling pathway underlying this process by using an in vitro human epithelial cell model and an in vivo mouse model to show that the combination (Pam2ODN) treatment results in STAT3 activation in lung epithelial cells. Following an elegant approach, using alveolar type II epithelial cell–specific STAT3-knockout mice and challenging them with P. aeruginosa following Pam2ODN treatment, the authors confirmed this pathway and show that the resulting induced protection is STAT3-dependent. Despite previous reports showing members of the IL-6 family of cytokines to be powerful activators of STAT3 (11, 15, 16), the authors provide evidence that epithelial STAT3 activation, which is required for inducible resistance against P. aeruginosa pneumonia, can be IL-6–independent. The discrepancy between these reports is likely to reflect a difference between the baseline STAT3 activation investigated in prior studies and the Pam2ODN-induced STAT3 activation in the report of Kulkarni and colleagues. The authors outline this inducible pathway by restricting signaling from IL-6 family cytokines in vitro or by using IL-6–knockout mice treated with Pam2ODN before P. aeruginosa infection. Mechanistically, the authors illustrate that Pam2ODN leads to ROS generation by the nicotinamide adenine dinucleotide phosphate oxidase enzymes, which are required for STAT3 activation. Moreover, the authors demonstrate that nicotinamide adenine dinucleotide phosphate oxidase–generated ROS activate the EGFR, which then phosphorylates/activates STAT3 for inducible resistance. To highlight the importance of EGFR in the activation of STAT3, the authors reveal that inhibition of EGFR by an EGFR kinase inhibitor, erlotinib, results in loss of activation of STAT3 in vitro and in vivo. The authors further show that transcripts for OSM and LIF (leukemia inhibitory factor) are not increased following Pam2ODN treatment, suggesting that, like the IL-6 family members, these molecules are also not responsible for activation of STAT3.

Additional contributions of STAT3 from other nonepithelial cell types to the protective response were not specifically investigated in the report of Kulkarni and colleagues. Nonetheless, the authors have previously depleted STAT3 in various myeloid and lymphoid cell populations, and Pam2ODN was able to induce protection despite the loss of these cell types (18, 20–22). Interestingly, the deletion of STAT3 in the lung epithelium did not completely impair bacterial clearance in Pam2ODN-treated mice, suggesting: 1) that Pam2ODN-induced protection against pathogens encompasses pathways other than alveolar type II epithelial cell–derived STAT3 activation and/or 2) the involvement of other transcription factors or signaling cascades. There have been numerous reports of oxidative stress influencing the STAT3 cascade, although it was unclear whether ROS-upregulated or -downregulated STAT3 activation was involved (8). The authors show that ROS plays an essential role in activating the EGFR, which then upregulates STAT3 phosphorylation/activation. Overall, the study by Kulkarni and colleagues expands our understanding of the mechanism of Pam2ODN-induced STAT3 activation that leads to inducible resistance against gram-negative (P. aeruginosa) pneumonia (Figure 1). These findings also provide more targets for therapeutic interventions that may help to overcome disease burden in susceptible populations. With the implications of epithelial-derived STAT3 for Pam2ODN-induced protection against P. aeruginosa pneumonia, this study sheds new light on a novel role of the alveolar epithelium in addition to its conventional functions, which include gas exchange and microbial recognition. However, several questions remain regarding the downstream signaling cascades by which activated epithelial transcription factor STAT3 contributes to protection against pulmonary pathogens. Future work is therefore required to fill these knowledge gaps by targeting cell-specific STAT3 in inflammatory disease settings in the lung. It is imperative to note that this study is likely to have broad implications in the field of pneumonia because Pam2ODN induces host protection against different categories of microbes, including bacterial, fungal, and viral pathogens (9, 18, 20).

Figure 1.

Schematic illustration describing the steps by which Pam2ODN induces host resistance in Pseudomonas aeruginosa pneumonia. A combination of a Toll-like receptor (TLR) 2/6 agonist (Pam2CSK4) and a TLR9 agonist (ODN M362) termed Pam2ODN activates TLR2/TLR6 on cell and TLR9 on endosomal membranes of epithelial cells, which results in the transcription of NOX that leads to the formation of reactive oxygen species (ROS). In turn, ROS induce the phosphorylation of STAT3 on EGFR through the kinase to induce target genes to ultimately cause inducible protection against Pseudomonas pneumonia through induction of target genes in the nucleus. EGFR = epidermal growth factor receptor; NOX = nicotinamide adenine dinucleotide phosphate oxidase; P = phosphorylation.