© 2007 American Thoracic Society
Correspondence The Pulmonary Source of Hepatocyte Growth Factor in Non–Small Cell Lung CancerNational Chung Hsing University, Taichung, Taiwan To the Editor: Hepatocyte growth factor (HGF), a multifunctional cytokine that behaves as a motogen, mitogen, and morphogen for a variety of cells, has been detected in various cell types, including alveolar epithelial cells, hepatocytes, mesenchymal cells, neutrophils, macrophages, hematopoietic stem cells, endothelial cells, dendritic cells, myofibroblasts, and cancer cells (1). In patients with idiopathic pulmonary fibrosis, Hojo and coworkers (2) showed that HGF was mainly detected in lung epithelial cells. Interestingly, in these patients serum level of HGF correlated well with that of neutrophils elastase. Moreover, in patients with acute respiratory failure, Jaffre and colleagues (3) showed that HGF was predominantly identified in neutrophils. In Pseudomonas aeruginosa–diseased lung of mice, HGF, however, was not detected in neutrophils, but in neutrophil-scavenging macrophages (4). The major difference between these two studies, other than study subjects, human and mouse, is the pathological method used to recognize cell types that are expressing HGF. Although both groups used indirect immunostaining technique to identify HGF-expressing cells, Morimoto and coworkers (4) used horseradish peroxidase–conjugated streptavidin of LSAB, and Jaffre and colleagues (3) used alkaline phosphatase–conjugated streptavidin of ABC method. The advantage of pathology examination is the ability to localize cells that are expressing the specific genes, which are particularly expressed under different physiological or pathological conditions. Due to safety and convenience, chromogenic immunohistochemistry (IHC) and in situ hybridization (ISH) were developed to replace radioimmunohistochemistry and ISH that used radioisotope probes. For chromogenic development, two enzymes, horseradish peroxidase and calf intestinal alkaline phosphatase, were commonly used as the conjugate, which was directly linked to the probe, the primary or the secondary antibodies that could in turn catabolize the specific enzyme substrate into visible precipitates. Removal of the endogenous alkaline phosphatase and peroxidase, which were largely expressed in leukocytes, became a key task for precise cell identification (5, 6)—in particular, the endogenous leukocyte alkaline phosphatases that were predominantly expressed in neutrophils and myeloperoxidase that was richly present in monocytes and macrophages (3, 4). Pretreatment of pathological sections with 3% hydrogen peroxide (H2O2), a product of myeloperoxidase, may not be sufficient to completely inactivate the enzyme per se (4). In terms of endogenous leukocyte alkaline phosphatases, no attempt was made to remove them (3). As noted in our recent article (7) and previous reports, we used a mixture of sodium azide and hydrogen peroxide (6) to inhibit the endogenous peroxidase, and 50% formamide in 6x SSC (salt sodium citrate, pH 7.2) at 75°C for 5 min to inactivate leukocyte alkaline phosphatase. By applying these methods, we detected the presence of HGF mRNA in alveolar type II (ATII) epithelial cells in 61 of 77 cases with pathology-confirmed non–small cell lung cancer (NSCLC), who were also heavy cigarette smokers. In fact, in addition to morphological criteria, we have further analyzed the feature of cells by combining IHC for surfactant protein B and ISH for HGF mRNA, and we found no evident HGF expression in alveolar macrophages (7), or in neutrophils, which, on the other hand, indicated that patients might have bronchial infections. Moreover, by demonstrating HGF message in ATII cells, our data not only support what Sakai and coworkers (8) and Wislez and colleagues (9) have observed, but also confirm the nature of HGF and ATII cells in response to pulmonary stress, in which ATII cells may act as pulmonary stem cells under the influence of HGF to maintain intactness of alveolar conformation and proper gas-exchange function (10). In fact, our results also provide a more rational interpretation to the discrepancies observed by Jaffre and coworkers (3), in which only blood neutrophils respond to lipopolysacchrides (LPS) induction to secrete HGF, but not neutrophils from bronchoalveolar lavage (BAL). Since both blood and BAL neutrophils were isolated by an affinity binding (to pan HLA class II molecules) method to exclude CD3+, CD19+ and CD14+ cells, it is reasonable to suggest that white blood cells from whole blood could contain residual cell types, including hematopoietic stem cells and dendritic cells (11), which were also shown to be HGF-positive. Isolation of neutrophils from BAL, on the other hand, would not have problem of cell contamination. We do appreciate the precious suggestion by Bonay and colleagues and we did not forget neutrophils (12). Footnotes Conflict of Interest Statement: K.-C.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. References
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
