β-Defensins are cationic peptides with broad-spectrum antimicrobial activity that are produced by epithelia at mucosal surfaces. Two human β-defensins, HBD-1 and HBD-2, were discovered in 1995 and 1997, respectively. However, little is known about the expression of HBD-1 or HBD-2 in tissues of the oral cavity and whether these proteins are secreted. In this study, we characterized the expression of HBD-1 and HBD-2 mRNAs within the major salivary glands, tongue, gingiva, and buccal mucosa and detected β-defensin peptides in salivary secretions. Defensin mRNA expression was quantitated by RNase protection assays. HBD-1 mRNA expression was detected in the gingiva, parotid gland, buccal mucosa, and tongue. Expression of HBD-2 mRNA was detected only in the gingival mucosa and was most abundant in tissues with associated inflammation. To test whether β-defensin expression was inducible, gingival keratinocyte cell cultures were treated with interleukin-1β (IL-1β) or bacterial lipopolysaccharide (LPS) for 24 h. HBD-2 expression increased ∼16-fold with IL-1β treatment and ∼5-fold in the presence of LPS. Western immunoblotting, liquid chromatography, and mass spectrometry were used to identify the HBD-1 and HBD-2 peptides in human saliva. Human β-defensins are expressed in oral tissues, and the proteins are secreted in saliva; HBD-1 expression was constitutive, while HBD-2 expression was induced by IL-1β and LPS. Human β-defensins may play an important role in the innate defenses against oral microorganisms.

Gene transfer with recombinant murine leukemia viruses (MuLV) provides the potential to permanently correct inherited lung diseases, such as cystic fibrosis (CF). Several problems prevent the application of MuLV-based recombinant retroviruses to lung gene therapy: (i) the lack of cell proliferation in mature pulmonary epithelia, (ii) inefficient gene transfer with a vector applied to the apical surface, and (iii) low titers of many retroviral preparations. We found that keratinocyte growth factor (KGF) stimulated proliferation of differentiated human tracheal and bronchial epithelia. Approximately 50% of epithelia divided in response to KGF as assessed by bromodeoxyuridine histochemistry. In airway epithelia stimulated to divide with KGF, high-titer ampho- and xenotropic enveloped vectors preferentially infected cells from the basal side. However, treatment with hypotonic shock or EGTA transiently increased transepithelial permeability, enhancing gene transfer with the vector applied to the mucosal surfaces of KGF-stimulated epithelia. Up to 35% of cells expressed the transgene after gene transfer. By using this approach, cells throughout the epithelial sheet, including basal cells, were targeted. Moreover, the Cl− transport defect in differentiated CF airway epithelia was corrected. These findings suggest that barriers to apical infection with MuLV can be overcome.

Cystic fibrosis mice have been generated by gene targeting but show little lung disease without repeated exposure to bacteria. We asked if murine mucosal defenses and airway surface liquid (ASL) Cl− were altered by the ΔF508 cystic fibrosis transmembrane conductance regulator mutation. Naive ΔF508 −/− and +/− mice showed no pulmonary inflammation and after inhaled Pseudomonas aeruginosa had similar inflammatory responses and bacterial clearance rates. We therefore investigated components of the innate immune system. Bronchoalveolar lavage fluid from mice killed Escherichia coli, and the microbicidal activity was inhibited by NaCl. Because β-defensins are salt-sensitive epithelial products, we looked for pulmonary β-defensin expression. A mouse homolog of human β-defensin-1 (termed “MBD-1”) was identified; the mRNA was expressed in the lung. Using a radiotracer technique, ASL volume and Cl− concentration ([Cl−]) were measured in cultured tracheal epithelia from normal and ΔF508 −/− mice. The estimated ASL volume was similar for both groups. There were no differences in ASL [Cl−] in ΔF508 −/− and normal mice (13.8 ± 2.6 vs. 17.8 ± 5.6 meq/l). Because ASL [Cl−] is low in normal and mutant mice, salt-sensitive antimicrobial factors, including MBD-1, may be normally active.

Na+absorption via amiloride-sensitive Na+ channels is of critical importance in the transition between fetal and neonatal life in several tissues, including the colon, lung, and kidney. To characterize and contrast the mRNA expression of each of the three epithelial Na+ channel complex (ENaC) subunits, we conducted RNase protection assays (RPA) and in situ hybridization in colon and lung in fetal (17, 19, 20, and 21 days) and postnatal (1, 3, 9, 15, and 30 days) rats (r). In the colon the α-, β-, and γ-rENaC subunits showed quantitatively different but qualitatively similar expression. All three subunits gradually increased in abundance from fetal day 19 through day 30 of life. The amount of each subunit on day 30 was approximately three times the amount at day 1. In situ hybridization showed that each subunit was localized to the surface epithelial cells with minimal expression in the crypts. The lung showed a completely different pattern. In contrast to the colon, the total amount of α-rENaC mRNA (by RPA) in the lung increased dramatically from fetal day 19 to 21, whereas β- and γ-rENaC showed modest prenatal increases. The amounts of all three mRNAs fell after birth through day 9 (to about 75% of the day 1 value). On days 15 and 30 the amount of mRNA rose to approach the values on day 1. α-rENaC mRNA abundance always exceeded β- and γ-rENaC, and the quantitative expression was different for α- than for β- and γ-rENaC. In situ hybridization studies showed that all three subunits were expressed in epithelial cells of the bronchi, bronchioles, and alveoli and not in blood vessels. These studies show striking developmental heterogeneity in rENaC mRNA expression between lung and colon, probably reflecting different developmental regulatory mechanisms in these organs.