Defining the Lung's Response to Endotoxin

Timothy S. Blackwell, M.D. and John W. Christman, M.D.

Vanderbilt University School of Medicine and the Nashville Veterans Affairs Medical Center, Nashville, Tennessee

	ARTICLE

TOP ARTICLE REFERENCES

Over the last few years, important advances have been achieved in understanding the signal transduction mechanisms underlying the cellular response to gram-negative bacterial endotoxin. In particular, the roles of lipopolysaccharide-binding protein, CD14, and toll-like receptors have been identified, as well as the pathway from cell surface binding of toll-like receptors to activation of the transcription factor complex, NF-B, and the production of proinflammatory cytokines. These events have been defined in isolated cell culture systems and investigated in animal models of endotoxin-induced inflammation.

Previously, carefully controlled studies of the effects of endotoxin in humans have been largely limited to investigation of the effects of systemic administration of endotoxin. The study by O'Grady and coworkers in this issue of the American Journal of Respiratory and Critical Care Medicine (pp. 1591- 1598) (1) represents an important contribution to endotoxin research in humans. In a controlled study, the investigators evaluated the local response when endotoxin was instilled in the lungs of human volunteers. At several time points after instilling Escherichia coli endotoxin, the investigators performed bronchoalveolar lavage (BAL) in lung segments that had previously received either endotoxin or saline. Cellular inflammatory response, cytokine production, and permeability were assessed in BAL fluid. A distinct pattern of inflammatory response was identified, with an early influx of neutrophils by 2-6 h, followed by a later influx of lymphocytes and monocytes. A variety of cytokines were found to be increased at early time points, including tumor necrosis factor- (TNF-), interleukin (IL)-1, IL-6, granulocyte-colony-stimulating factor (G-CSF), IL-8, monocyte chemotactic protein-1 (MCP-1), macrophage inflammatory protein-1 and  (MIP-1 and ), and extractable nuclear antigen-78 (ENA-78). Production of these cytokines peaked at 2-6 h after endotoxin exposure, and returned to normal levels by 24 h (except G-CSF). These data support the hypothesis that local challenge with endotoxin results in the discrete, limited production of inflammatory mediators and influx of immune cells, followed by resolution. Interestingly, the investigators also identified increased concentrations of counterinflammatory molecules in BAL, including TNFR1, TNFR2, and IL-1ra (but not IL-10), which potentially help limit or terminate the inflammatory response. BAL protein and albumin content were altered at the time of the inflammatory cell influx. This alteration in perrmeability and mild hypoxemia demonstrates the physiological relevance of the endotoxin-induced inflammation of the lungs. Although not evident in this report of O'Grady, the recent report of Kline and coworkers demonstrated substantial variability in the human response to a lipopolysaccharide (LPS) aerosol that was related to genetic factors, sex, comorbidities, and exposure history (2). Differences between the consistancy of the LPS response in these studies could be related to the mode of delivery or the concentration of endotoxin that reached the surface of airways and alveoli and the exact outcome measurements employed in the two studies.

Taken together, the findings of this study and previous studies by this group involving systemic endotoxin administration to humans demonstrate compartmentalization of the inflammatory response, at least at doses of endotoxin safe for human volunteers. With intravenous endotoxin, there is a pronounced systemic response with fever and cardiovascular alterations. Interestingly, neutrophil margination occurs in the lungs after intravenous endotoxin, but neutrophilic alveolitis is not detectable by BAL and neither albumin concentration nor cytokine levels are increased in BAL fluid. (3). Despite high levels of circulating cytokines, the lung is relatively spared in systemic inflammation induced by a single intravenous dose of endotoxin. The contrast between the effects of locally delivered endotoxin to the lungs and systemically delivered endotoxin highlights the important contributions of local production of cytokines in directing the inflammatory response in the lungs.

Direct instillation of endotoxin into human lungs produces a characteristic pattern of inflammation, with accumulation of mediators and inflammatory cells similar to that found in the early phase of the acute respiratory distress syndrome (ARDS) (4, 5). That airway instillation of endotoxin, but not systemic administration of endotoxin, produces an ARDS-like inflammatory response has at least three important implications. First, although ARDS frequently occurs as a result of systemic inflammatory conditions, such as sepsis, local mediator production by the lung is likely sufficient, and may be necessary, to produce the pathobiology seen in ARDS. Second, recruitment of inflammatory cells to the airspaces, where they can produce tissue injury, requires gradients of chemokines (like IL-8 and ENA-78) favoring migration of these cells into the lung parenchyma (6). Although intravenous endotoxin generates high levels of circulating IL-8, neutrophil recruitment to the airspaces does not occur. In contrast, accumulation of IL-8 (in association with other neutrophil chemotactic factors) in the lungs following airway instillation of endotoxin produces significant neutrophilic alveolitispresumably by creating a tissue gradient favoring immigration of neutrophils to the airspaces. Finally, if local production of inflammatory mediators and establishment of chemokine gradients for inflammatory cell recruitment are important for the pathogenesis of ARDS, then therapy directed toward limiting the local inflammatory response in the lungs could limit tissue injury and improve outcomes in humans with ARDS.

	References

TOP ARTICLE REFERENCES

1. O'Grady NP, Preas HL II,, Pugin J, Fuiza C, Tropea M, Reda D, Banks SM, Suffredini AF. Local inflammatory responses following bronchial endotoxin instillation in humans. Am J Respir Crit Care Med 2001; 163: 1591-1598 [Abstract/Free Full Text].

2. Kline JN, Cowden D, Hunninghake GW, Schutte BC, Watt JL, Wohlford-Lenane CL, Powers LS, Jones MP, Schwartx DA. Variable airway responsiveness to inhaled lipopolysaccharide. Am J Respir Crit Care Med 1999; 160: 297-303 [Abstract/Free Full Text].

3. Suffredini AF, O'Grady NP. Pathophysiologic resonpses to endotoxin in humans. In: Braude H, Opal SM, Vogel SN, Morrison DC, editors. Endotoxin in health and disease, 1st ed. New York: Marcel Dekker; 1999. p. 817-830.

4. Martin TR. Lung cytokines and ARDS: Roger S. Mitchell Lecture. Chest 1999; 28: 2S-8S .

5. Meduri GU, Kohler G, Headley S. , Tolley E, Stentz F, Postlewaite A. Inflammatory cytokines in the BAL of patients with ARDS. Persistent elevation over time predicts poor outcome. Chest 1995; 108: 1303-1314 [Abstract/Free Full Text].

6. Blackwell TS, Lancaster LH, Balckwell TR, Venkatakrishnan A, Christman JW. Chemokine gradients predict neutrophilic alveolitis in endotoxin-treated rats. Am J Respir Crit Care Med 1999; 159: 1644-1652 [Abstract/Free Full Text].