Recruitment of neutrophils into lung by CXC chemokines.

Abstract

<italic>Introduction</italic>. Chemokines are capable of recruiting specific leukocytes. The CXC chemokines KC and MIP-2alpha, murine homologues for IL-8, attract neutrophils. Chemokine-driven recruitment of neutrophils occurs in many pulmonary disease processes and has been thought to be associated with neutrophil activation and neutrophil-mediated tissue injury. The <italic>de facto</italic> association of neutrophil recruitment and lung injury is contradicted by the common occurrence of bacterial pneumonia where significant neutrophil recruitment occurs but does not usually result in lung injury. <italic>Hypothesis</italic>. The studies undertaken in this thesis pursue the hypothesis that chemokine-driven neutrophil migration into the lung does not lead to subsequent neutrophil-mediated tissue and/or microvascular injury. <italic>Data</italic>. An <italic>in vivo</italic> model of neutrophil recruitment following aspiration of chemokine into the lung was established in BALB/c mice to assess lung injury and examine neutrophil function. The doses of chemokine used for this model (10--1000ng) encompassed and exceeded the physiological range measured in other studies of disease and injury models. Both KC and MIP-2alpha aspiration demonstrated a dose dependent recruitment of significant numbers of neutrophils into the lung airspace. This response was similar in different mouse strains. No significant injury was detected to the lung tissue after physiological levels (10--100ng) of CXC chemokine aspiration as measured by albumin and total protein leakage into the airspace, pulmonary function parameters, hemorrhage into the alveoli, pulmonary edema quantitated by wet/dry lung weights and light and electron microscopic examination of lung tissue. Only at supraphysiologic levels (1000 ng) was any evidence of injury seen by these parameters. A series of repeated aspirations (six doses at one week intervals) of the chemokines at physiological levels (10ng) also failed to produce lung injury confirming that no residual effect accumulates and resolution is complete following neutrophil recruitment. Compartmentalized examination of the lung tissue indicated that the KC and MIP-2alpha were interacting with the neutrophils directly either at the endothelial cell surface or interendothelial cell cleft. Assessment of the activational state of neutrophils recruited by KC and MIP-2alpha revealed that the neutrophils were only partially activated by chemokine-driven recruitment. Mac-1 (CD11b/18 complex) levels were elevated indicating secondary granule fusion. This was confirmed to be a direct effect of the chemokines by <italic>in vitro</italic> studies. Comparing <italic> in vivo</italic> derived BAL fluids with <italic>in vitro</italic> cell supernatants and lysates it was determined that gelatinase B was released from tertiary granules as a consequence of transmigration into the alveoli rather than direct chemokine exposure. Primary granule products myeloperoxidase and beta-glucuronidase were shown to be intact in KC and MIP-2alpha recruited neutrophils when compared to both unstimulated glycogen-elicited peritoneal and blood derived neutrophils. <italic>Conclusion</italic>. Taken together these data indicate that KC and MIP-2alpha exposure alone does not cause the release of the tissue damaging enzymes of the primary granules but that a second signal is required to stimulate the neutrophils to continue the inflammatory cascade. Our system may be used to successfully model neutrophil recruitment in disease conditions as well as conditions of routine immune surveillance and cell trafficking.