| Sign In to gain access to subscriptions and/or personal tools. |
Fascination with Epithelia: Architecture, Proteins, and FunctionsDept. of Oral Biology, Box 357132, University of Washington, Seattle, WA 98195-7132; bdale{at}u.washington.edu
Key Words: mucosa • keratin • filaggrin • innate immunity • antimicrobial peptides INTRODUCTION My scientific journey has been one filled with surprises, continuous discovery, and wonderful colleagues. For approximately 30 years, my scientific work has focused on epithelia, both oral epithelia and epidermis. During this time, a significant evolution has occurred in the way in which we think about epithelia and their functions, coupled with the obvious revolution in experimental methods. The primary mentor who started my fascination with epithelia was Irving B. Stern, a periodontist and electron microscopist who was the first to identify that hemidesmosomes are involved in attachment of the epithelium to the tooth surface as well as to the basement membrane (reviewed in Stern, 1981). He realized the necessity of identifying a biochemical basis to understand more fully the epithelial differentiation and stratification that were so clearly evident from the histology and ultrastructure of oral epithelium. Thus, I began an exploration of structural proteins associated with differentiation and keratinization—working in the epidermis, a much easier tissue to deal with when working with newborn rodents!
My own research focus has moved in much the same way that the field has changed over time (see Fig. 1
EVOLUTION OF FOCUS AND EMPHASIS When I began my studies in the 1970s, the epithelium (for example, gingival epithelium or epidermis) was viewed primarily as an independent entity that functioned as a protective covering for the oral cavity or the body as whole; an important aspect of that function was re-epithelialization during wound healing. The beautiful architecture of these stratifying epithelia could be appreciated by excellent morphological studies, several of which were conducted by my colleagues Irving Stern, George Odland, and Karen Holbrook at the University of Washington. This tissue structure invited exploration of the proteins that were associated with differentiation and stratification. At that time, the emphasis was on structural proteins that contributed to the protective function. Subsequent studies led to the initial understanding of keratin biochemistry, particularly through the work of Peter Steinert (Steinert, 1993), and my work on the characterization of an unusual protein, filaggrin, named for its ability to cause keratin filament aggregation—a remarkable and rapid reaction that was investigated in collaboration with Steinert and Holbrook (Dale et al., 1978). A conceptual change in approach occurred in the early 1980s, when the process of differentiation became the emphasis of several types of studies. These included investigation of the influence of connective tissue on the morphology of oral epithelia, by colleagues Ian Mackenzie and Murray Hill (Mackenzie and Hill, 1984), and functional differences in oral epithelium vs. epidermis, particularly permeability, by Christopher Squier (reviewed in Squier, 1991). In addition, calcium and retinoids were shown to have profound influences on differentiation of epithelial cells (Hennings et al., 1980; Fuchs and Green, 1981). It was also realized that keratins are a family of intermediate filament cytoskeletal proteins whose expression changes with differentiation. This work was a direct result of the development of monoclonal antibodies to the keratins and detailed analysis of keratins separated by two-dimensional gel electrophoresis (Sun et al., 1983). Through the use of these antibodies, we now understand that the family of keratins is expressed in a tissue-specific and differentiation-specific manner (reviewed by Presland and Dale, 2000). These reagents are now a mainstay in cell biology and pathology. A simultaneous development was the realization that the protective, but dead, keratinized surface cell layers contained proteins that were all produced in the underlying living cell layers but were modified to perform their functions. Involucrin was the first of a group of proteins identified that are produced as soluble precursors that are crosslinked by transglutaminase into the tough cornified envelope—a structure that is still the subject of biochemical and functional studies associated with the barrier properties of epithelia (Kalinin et al., 2002). Concurrently, my group found that the protein filaggrin was derived from a large insoluble precursor that had multiple filaggrin subunits that differed biochemically by phosphorylation from the functional product (Lonsdale-Eccles et al., 1980). This work resulted from one of those instantaneous insights that occurs in a dramatic flash—the light bulb above the head in a cartoon. I attended a seminar on protein phosphorylation by Ed Krebs, who later received the Nobel Prize with Edmond Fisher for their work on protein phosphorylation. I walked out of the seminar saying "that’s it—that has to be one of the major differences between profilaggrin and filaggrin." Profilaggrin was localized to keratohyalin granules by the use of newly developed antibodies (Dale and Ling, 1979) and processed by phosphatases and proteases during the transition of the living cells to the dead, cornified layer. This started us tracking enzymatic post-translational modifications associated with the terminal stages of the epithelial differentiation process (Resing et al., 1989). Of course, this was only part of the story, and the molecular studies by my associate Richard Presland revealed additional fascinating aspects, including a calcium-binding domain in this protein (Presland et al., 1992). Advances in molecular biology led to great strides in the understanding of epithelial proteins and to investigations of the molecular mechanisms associated with gene regulation. Molecular biology, coupled with genetics, held great fascination as we began to probe defects in epithelial structural proteins, following the lead of UW pioneers in collagen disorders, Peter Byers and Paul Bornstein. We found that filaggrin and profilaggrin are poorly expressed or absent in the dominantly inherited skin disorder, ichthyosis vulgaris (Sybert et al., 1985). In collaboration with Philip Fleckman and Virginia Sybert, colleagues in Dermatology, we classified the most severe recessive ichthyotic skin disorder, Harlequin ichthyosis, into three biochemically distinct types, suggesting multiple possible underlying genetic defects (Dale et al., 1990). Nevertheless, the mutated genes have not yet been identified in either of these disorders. Meanwhile, those working with keratin progressively identified mutations in individual keratins that result in epithelial cell fragility. This work led to the understanding of blistering epithelial disorders that affect both skin and oral mucosa (reviewed by Corden and McLean, 1996), and the concept that the phenotype of the disorder reflects the tissue-specific expression of the mutated keratin. During the 1990s, the research emphasis shifted again. It was realized that epithelial cells were active participants in cell-cell communication via production of cytokines, chemokines, receptors, growth factors, etc., that affect the proliferation, differentiation, and wound healing of epithelial cells and other cell types. Several of these discoveries occurred through observations of transgenic mice over-expressing particular proteins with keratin gene promoters used to ensure expression in particular epithelial compartments. EPITHELIA AND INNATE IMMUNITY; EVOLUTION IN UNDERSTANDING OF EPITHELIAL FUNCTIONS Until a few years ago, the epithelial response to infection was viewed mainly as one of wound healing and re-epithelialization. Development of the concepts of innate immunity and the subsequent discovery of pattern recognition receptors in the late 1980s and early 1990s (Janeway, 1989; Medzhitov, 2001) were critical in the transition to the way we now think about epithelia. While the epithelial compartment does provide a physical barrier to infections, it is now clear that it also has an active role in innate host defense. Epithelium, especially mucosal epithelium, is constantly in contact with microbes. Epithelial cells actively respond in an interactive manner; they secrete interleukin-8 and other chemokines and cytokines to alert various cell types and to attract neutrophils (Darveau et al., 1997), and they produce natural antimicrobial peptides in response to bacterial products.
My own transition in thinking about epithelial functions began in 1995 when I saw a publication in Science from Michael Zasloff’s group on the expression of an antimicrobial peptide in cow tongue (Schonwetter et al., 1995). As I recall, I went to find our departmental microbiologist, Aaron Weinberg, and found him in the hallway holding the same article, and said, nearly in unison, that human gingiva must have these peptides! This was the beginning of several years of collaborative studies on oral antimicrobial peptides, the beta-defensins. In the transition to a new area of research, there is always a bit of luck along with hard work and good ideas that are essential to get a foothold. These things came together with some important local transition funding, encouragement from Tomas Ganz, a pioneer in the area of antimicrobial peptides, and a very talented PhD student, Suttichai Krisanaprakornkit. We found that human beta-defensin-1 was constitutively expressed, but that beta-defensin-2 was inducible in cultured oral epithelial cells exposed to bacterial or pro-inflammatory stimulants (Krisanaprakornkit et al., 1998, 2000). We also showed that the inducible hBD-2 was expressed in every sample of normal gingival tissue examined—probably because of the constant exposure of oral mucosa to bacteria (Dale et al., 2001). By contrast, hBD-2 is expressed in skin only in response to infection and inflammation. Although Weinberg and I have somewhat different emphases in our current work, this initial work together was an important start in a new direction, working on the hypothesis that these antimicrobial peptides are part of the innate immune system, which is a complex set of responses that helps to keep the oral microbial flora in check. Further developments by others also showed clearly that these peptides signal additional host responses (Yang et al., 1999). Thus, the epithelium participates actively in responding to infections, and integrating innate and acquired immune responses, as summarized diagrammatically in Fig. 2
The current emphasis in my laboratory is two-fold: first, focusing on the signaling pathways used by epithelial cells to respond to bacteria stimulants to express the inducible beta-defensins and the implications this has on informing other cell types about the environment; and second, how individuals may differ in their susceptibility to infection due to genetic differences in the defensin genes. With respect to signaling pathways, Krisanaprakornkit made one of those very surprising and interesting findings, the kind that causes us all to sit back and think. He found that hBD-2 was induced in oral epithelial cells by non-pathogenic bacteria (commensals) in a manner that did not utilize the expected NF-  B transcription factors that are usually associated with inflammatory response to bacteria via the important Toll-like receptor family. Instead, signaling utilized the JNK and p38 mitogen-activated protein kinase (MAPK) pathways (Krisanaprakornkit et al., 2002). This differed from results in other tissues—for example, tracheal epithelium (Diamond et al., 2000). But with further thought, it was very logical: Why should epithelial cells respond to a non-pathogen by utilizing a pathway associated with inflammatory signaling? Somehow, epithelial cells are capable of distinguishing a pathogen from a non-pathogen and staging a carefully executed and appropriate response. Molecular signaling mechanisms associated with this important distinction between bacteria are a new emphasis by Whasun Chung in our laboratory. I believe that this ability of epithelial cells to respond and to signal concerning the state of their environment is a topic that will help us understand how epithelia of various body sites can exist harmoniously with different populations of commensal organisms, even considering that a commensal for one site can be a pathogen for another. We are also pursuing the idea of cell-cell communication and cross-talk via defensins as a means of signaling the state of the environment—a process that puts the epithelium center stage in innate immunity. The second part of our current research focus, the relationship of defensins and an individual’s susceptibility to infection, posed special challenges to investigators. Because there are numerous beta-defensins, it is not possible to prove their importance unequivocally by knock-out technology in a mouse model system. Therefore, we sought another route—via genetics—to ask questions about the role of defensins in disease. The logic was that if there are mutations or, more likely, single nucleotide polymorphisms (SNPs) that alter the expression or function of defensins, they may lead to altered susceptibility to oral infection. The studies began with identification of multiple SNPs in the DEFB1 and DEFB2 (now renamed DEFB104) genes by colleague and PhD student Richard Jurevic (Jurevic et al., 2002). These polymorphisms, especially those that had a reasonably high frequency, occurred mainly in the regions of these genes that did not code for the functional protein. However, these SNPs could alter the amount of the protein expressed. The next step was the correlation of these SNPs with a disease situation. We reasoned that people who have slight immunosuppression might have greater reliance on innate immune defenses than would healthy individuals who have multiple ways to fight infection. As a specialist in oral medicine, Jurevic knew that people with Type 1 diabetes are susceptible to oral Candida infection and that they have marginal immunosuppression, so this population was chosen for initial testing. Much to our surprise, he found that a polymorphism in the gene encoding beta-defensin-1 was associated with protection from oral Candida carriage. In other words, people who had the SNP had very low levels of oral Candida, and this was seen in both diabetic and non-diabetic individuals (Jurevic et al., 2003). This interesting result supports the initial hypothesis of a relationship between epithelial defensins and oral health and opens a whole new area of investigation. How does beta-defensin-1 affect Candida carriage? Can this SNP be used as part of a diagnostic susceptibility profile in situations in which Candida infection could become life-threatening, such as transplant recipients? And finally, how does this SNP alter beta-defensin-1 expression? Page Fredericks in the lab has used a molecular approach to address the last question. Initial results strongly suggest that the SNP results in increased protein expression, and this may explain the basis of the apparent protection from oral Candida carriage. Are defensin SNPs correlated with other oral infections? Because defensins are secreted and may function synergistically with salivary antimicrobials, they may even protect against caries or, conversely, contribute to genetic susceptibility to dental decay. Therefore, we have begun a project on defensins in caries-prone children. I am an epithelial biologist, not a dentist, or a dermatologist, or an immunologist, or a geneticist. I have been privileged to have the help of experts in those fields, and the efforts of colleagues and talented students, each of whom contributed from his or her unique perspective to the overall success of our studies. I have had some wonderful experiences, learning and thinking about the role of epithelia in new ways. The academic research journey is full of puzzles whose pieces come together from unexpected directions. It has been a very rewarding path. DEDICATION This review of my career is dedicated to the memory of my daughter, Jessica Louise Dale. She was a wonderful young woman, full of adventure and courageous to the end.
Received for publication July 16, 2003. Accepted for publication July 25, 2003. REFERENCES
Journal of Dental Research, Vol. 82, No. 11,
866-869 (2003)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
