Our results suggest that middle ear epithelial cells release IL-1α when stimulated by NTHi components and that this cytokine acts in an autocrine/paracrine synergistic manner with NTHi to up-regulate β-defensin 2. Furthermore, in contrast to IL-1α that acts via the ERK MAP kinase pathway to induce β-defensin 2 transcription, the synergistic effect of this cytokine on β-defensin 2 up-regulation by NTHi molecules appears to be mediated by the p38 MAP kinase pathway (Figure 6). Such synergism may be a way of boosting the epithelial cells' initial response to NTHi and its components.
The signaling mechanisms appear to be more complicated in the presence of both exogenous and endogenous inducers. NTHi, an exogenous inducer and TNF-α, an endogenous inducer synergize with each other to induce NF-κB activation and consequently up-regulate inflammatory mediators including IL-1β and IL-8 as well as TNF-α . NTHi first interacts with its receptors of host cells such as Toll-like receptor 2 (TLR2) , and leads to the activation of multiple signaling cascades resulting in the up-regulation of inflammatory cytokines such as TNF-α. The up-regulated TNF-α in response to NTHi synergize with NTHi to induce NF-κB via two distinct signaling pathways, the NIK-IKK-IκBα and the MEKK1-MKK3/6-p38 MAPK pathways .
Strains of NTHi, a small gram-negative bacterium, exist as commensal organisms in the human nasopharynx . Nasopharyngeal colonization of NTHi may persist for prolonged periods of time, even in the antibody-containing mucosal secretions since the bacteria undergo antigenic variation. This can occur by several molecular mechanisms such as point mutation, gene amplification, phase variation or horizontal gene transfer and homologous recombination . Although NTHi rarely causes life-threatening infections, it is nonetheless a clinically important pathogen since it causes otitis media in children and exacerbates chronic obstructive pulmonary disease in adults [35, 36]. The interactions between NTHi and the host are not well understood. These interactions determine chronic colonization with or without inflammation or disease. The interaction of NTHi antigens and specific host molecules are likely to be involved in the transition of NTHi from a commensal to a pathogenic organism. Further studies are, however, needed to elucidate these mechanisms.
Our results showed that the induction of β-defenin 2 by NTHi is not a pan-epithelial response, in that Hela and A549 cells do not highly respond to NTHi lysates, whereas HMEEC cells do. In addition, although NTHi treatment up-regulated IL-1α expression in Hela, A549 and HMEEC cells, treatment of the cells with IL-1α resulted in the up-regulation of β-defensin 2 only in A549 and HMEEC, but not in Hela cells. These results indicate that expression patterns of receptors or signaling molecules is cell-specific and thus determines the responses to extracellular stimuli. Our results are in agreement with earlier reports of the hypo-responsiveness of A549 cells to lipopolysaccharide (LPS), and of A549 and normal human bronchial epithelial cells to Gram-positive group B Streptococci [37, 38]. A deficiency in the expression of TLR2 and 4 may also explain the lack response of these cells to NTHi as well. Our experimental observation of the non-responsiveness of Hela cells to NTHi, is on the other hand different from previous observations demonstrating transcriptional up-regulation of β-defensin-2 by LPS  or activation of NF-κB by NTHi . This apparent difference, however may be explained by Mineshiba and colleagues' use of LPS and the fact that Shuto and coworkers studied the nuclear translocation of NF-κB, which may not be in itself sufficient for the transcriptional activation of β-defensin-2 in Hela cells.
Although the autolysis of NTHi has not been documented, it is likely that NTHi undergoes autolysis similar to that seen with Haemophilus influenzae type b , S. pneumoniae  or E. coli . Cytoplasmic, as well as membrane components of NTHi are thought to persist in the effusion and act as long lasting inducers of inflammation. This notion is supported by the fact that bacterial DNA is detectable in most sterile effusions , whereas endotoxin is detectable in 60% of the cases . While the molecules (possibly surface antigens) of intact NTHi have been demonstrated to induce proinflammatory cytokines in human respiratory epithelial cells , other studies have shown that the cytoplasmic fraction of lysed NTHi also contains important molecules for stimulating epithelial cells [46, 47]. The NTHi lysate used in this study was prepared by sonification of the bacteria and mimics normal pathophysiology.
Interleukin 1 is a central mediator of the inflammatory response and occurs in two forms. Although the acidic form, known as IL-1α, and the neutral form, IL-1β, only share 45% homology at the nucleic acid level, both proteins bind to the same receptor [48, 49]. Furthermore, both proteins are pleiotropic and affect processes such as inflammation, immunity and hemopoiesis. Moreover, while IL-1β is active only in its secreted form, IL-1α is active as an intracellular precursor, a membrane-associated cytokine, as well as a secreted molecule.
Our results suggest a role for IL-1α but not IL-1β in the up-regulation of β-defensin 2 transcription. These results are consistent with the observations that IL-1α and not IL-1β, released from epithelial cells infected with respiratory syncytial virus, enhanced expression of intercellular adhesion molecule-1 in pulmonary epithelial cells . Moreover, our data suggest that the induction of IL-1α production in middle ear epithelial cells and the autocrine/paracrine effect of this molecule may be one of the central events involved in the up-regulation of chemotactic factors for the recruitment of immune cells into the middle ear. It should be noted that although the concentration of IL-1α released into the culture medium was less than 100 pg/ml/106 cells, since the molecule is acting in an autocrine manner, its effective concentration is likely to be much higher locally.
In our previous studies, we demonstrated that IL-1α could up-regulate β-defensin 2 transcription via a Src-dependent MEK1/2-ERK1/2 signaling pathway . Interestingly however, the present study shows that the p38 MAP kinase, not the ERK1/2 pathway is involved in synergism of IL-1α and NTHi components. Moreover, our preliminary results suggest that the p38 MAP kinase pathway is involved in the activation of β-defensin 2 transcription by NTHi (unpublished data). We postulate that the activation of ERK and p38 MAP kinase pathways can result in a synergistic up-regulation of β-defensin 2. These results point to the complexity of the signaling pathways that control the expression of innate immune molecules such as β-defensin 2 and indicate the need for further investigation into the regulation of β-defensin 2 by NTHi and IL-1α.