For all of us, the image of a young girl playing with her dog or learning to play the piano is a happy one. But, imagine that a dog licking the young owners face is so painful that it becomes her inspiration to stand up and begin walking. Or, consider that the joy of playing the piano must overcome the physical pain of blistering fingers caused by lightly striking the keys. Finally, couple these symptoms with the knowledge that the underlying disease has no cure. Given all of this, our young girl will surely face obstacles greater than most of us will confront in a lifetime.
The above scenario highlights what a child would face growing up with the disease Epidermolysis bullosa, a group of inherited disorders in which the skin and mucous membranes are so fragile that the slightest touch may cause painful blistering. In this paper, we will be looking at the cell biology of epidermolysis bullosa. Beginning with a discussion of the symptoms, we move on to detail the affected cellular organelle and recent findings related to the molecular and genetic basis of the disease. We close the paper on an optimistic note, looking toward the future use of gene therapy.
Description and Symptoms
It is easy to forget that the skin constitutes the largest human organ. Our skin serves us well, acting as a barrier against the environment and offering mechanical strength coupled with flexibility. In addition, the skin is an integral component in the regulation of body temperature. All of these functions are fundamental to healthy survival in the world. So, when the health of the skin is compromised, the consequences can be quite serious.
Epidermolysis bullosa refers to a group of skin diseases that involve blistering as a result of minor physical trauma. This category of diseases can be divided into three major classes, depending on which level of the skin is involved in blistering. To set the stage, let us digress a moment on the basic anatomy of the skin. In the cartoon diagram below, we are looking at the vertical organization of the skin. (A more detailed version of this can be found in [1,fig. 22-21].) The epidermis is a stratified tissue involving several differentiated layers and lies above a thin extracellular matrix layer called the basal lamina. Below this layer lies the connective dermis tissue. Cells in the first layer above the basal lamina are called basal epidermal cells. If blistering occurs below the basal lamina, within the dermis, we refer to this as dystrophic epidermolysis bullosa. Blistering that involves a split within the basal lamina is referred to as junctional epidermolysis bullosa. Finally, if the blistering involves the basal epidermal cells, the disease is called epidermolysis bullosa simplex, or EBS for short. EBS will be the focus of this paper.
To diagnose EBS, a complete physical exam should be performed with an emphasis on inspection of all skin and mucous membranes. The key symptom would be easy blister formation after minimal physical stress and/or exposure to increased temperature. Blisters would typically heal without scarring. After careful examination of the pathology of blistered areas, you would find cytolysis in the basal cell layer (see figure above). The amount and severity of blistering can vary and leads to three different clinical phenotypes summarized below; a more detailed account can be found at the interactive web-site "The Electronic Textbook of Dermatology" [7]:
Cell Biology of Keratin Intermediate Filaments
Each basal epidermal cell contains a network of protein filaments forming a key organelle called the cytoskeleton. The cytoskeleton contains 3 types of protein filaments: actin filaments, microtubules, and intermediate filaments [1,Figure 16-2]. Each type of filament is a polymer which is composed of different monomeric protein subunits. The intermediate filaments function to provide mechanical stability and resistance to stress. Our previous paper [2] gives a full account of the various components of the cytoskeleton, focusing on the intermediate filaments.
For basal epidermal cells, it is important to keep in mind that a primary protein subunit from which intermediate filaments are synthesized comes from a group of fibrous proteins called the keratins. More than 30 different types of keratin protein are expressed and used in making up the intermediate filaments of epidermal cells.
A quick review of the intermediate filament polymerization process, cellular location and anchoring of these filaments will help to set the stage for a fuller understanding of EBS. Keratin proteins all have a central rod domain which naturally adopts an alpha helix conformation. Upon close examination, a hydrophobic stripe of amino acid residues will wind around the alpha helix. Two different monomer types, a keratin of acidic type I and another keratin of alkaline type II, come together forming a coiled-coil dimer (usually called a heterodimer). The formation of heterodimers is favored by stability, since the coiled-coil simultaneously hides the hydrophobic stripes on each keratin alpha helix. These dimers now combine into tetrads which stack into helical arrays and form keratin intermediate filaments; called KIF's for short. This process is schematically summarized in [1,fig. 16-14] and further molecular details can be found in [5]. Finally, KIF's are attached to specialized proteins involved in anchoring the cell to either the basal lamina (hemidesmosomes) or to adjacent epidermal cells (desmosomes); see [1,fig. 19-18]. The schematic diagram below integrates all of these remarks in cartoon form:
It is apparent that the structure of KIF's leads to an internal framework within a basal epidermal cell. To understand the basic function of KIF's, one can imagine applying an external force on the cell. In a well designed system, the force load will be distributed through the network of KIF's and junctions, dissipating the force. This is completely analogous to the manner in which force loads on an airplane are distributed through an internal framework. Without such a design feature, the entire load would be applied to a small region of the plasma membrane; such a system could only tolerate very small forces. Interestingly, cytoplasmic intermediate filaments seem to be dispensible for cells growing on a flat plastic surface in tissue cultures; this is in contrast to the essential roles of actin filaments and microtubules. This suggests that nature must have had in mind a function for KIF's suited to the needs of a eukaryotic cell in it's external environment.
Concerning KIF's in basal epidermal cells, we can ask: What might go wrong? In a situation where the polymerization process was faulty at some stage, fibres may be defectively weak or perhaps large numbers congregate in clumps not having properly formed. This would lead to a scenario in which an applied force could lead to a rupture of the plasma membrane, lysing of the cell and blister formation. This sort of phenomena is illustrated clearly in [1,fig. 16-19C]. Depending on the extent of the KIF defect or inadequate attachment to anchoring proteins, the severity of blistering could vary.
These comments establish a strong correlation between the blistering symptoms of EBS and the consequences of defective KIF's in epidermal cells. In fact, according to [7] part of the standard diagnosis of blistering diseases (such as EBS) is an electron microscopic study of a skin biopsy. This allows one to determine, for example, if the blistering occurs in the basal epidermal layer of the skin. Examinations of this sort have also revealed that clumping of the KIF's in the cytoplasm preceeds cytolysis and blister formation. In addition, in both transgenic mice and humans with EBS, basal cells often rupture in a zone beneath the nucleus and above the basal lamina anchoring proteins. This is the longest portion of the cell and would be expected to be the most fragile if KIF's are defective.
In summary, on the cellular level, evidence strongly supports the hypothesis that EBS is a disease caused by defective KIF's in the basal epidermal cell layer.
Molecular Biology and Genetics
Knowing the molecular basis of EBS helps with both prognosis and treatment. Up to this point, we have argued that EBS is caused by defective KIF's. The next step is to link defective KIF's with some defective gene(s). To this end, a large and recent body of work has led to a sequence of observations that ultimately offers the hypothesis that EBS is caused by defects in keratins K5 or K14 in basal epidermal cells. The papers [3] and [6] offer an extensive bibliography to the original sources. Here is a brief discussion of each major line of evidence:
Given that EBS is caused by keratin K5 and K14 gene mutations, one can try to link the mutation site with the actual EBS phenotype; i.e. Dowling-Meara, Koebner or Weber-Cockayne. In the figure below, we have indicated the division of these two genes into various domains. Recall that all keratin proteins contain an alpha helical rod domain. Following the cartoon below, this domain is divided into four regions (1A, 1B, 2A, 2B) interrupted three times by short non-helical linker sequences (L1, L12, L2). We have indicated the location of mutations that correlate with particular types of EBS. Note that the "hatched" portion of the cartoon denotes highly conserved regions near the start of segment 1A and near the end of segment 2B. Here are some conclusions that follow from this schematic:
Deletion mutation studies have demonstrated that the conserved regions (hatched in diagram) play a special role in proper KIF assembly. This suggests that minor changes in these regions may have the most deleterious effects, which supports the fact that Dowling-Meara is the most severe form of EBS.
It is important to comment on the manner in which EBS is inherited. In the dominant negative genotype, the mutation is expressed as a defective K5 or K14 gene product. This means that even if one good copy of the gene is present, the mutated keratin gene product will interfere with the normal gene product and lead to a loss-of-function phenotype. In contrast, in the rare recessive forms such as Koebner EBS, the gene product is not expressed because of the absence of either K5 or K14, leading to a loss-of-function.
Finally, keratin genes and keratin intermediate filaments are quite ubiquitous. In multicellular animals, such as mice or cow, keratin intermediate filaments will play a similar important role in providing basal skin cell mechanical strength.
Theraputic Outlook
Once we know the genetic defect for any form of epidermolysis bullosa (either simplex, junctional or dystrophic), genetic counseling can take place and gene therapy becomes a possibility.
Before discussing gene therapy, we should quickly recall that the health of the epidermis is directly dependent on the basal cell layer. This is because a certain small population of basal cells act as stem cells from which all epidermal cells arise, differentiate, migrate upward and ultimately lose their nuclei to form the keratinized carcuses (the squames layer) at the outer layer. The entire process from mitotic birth at the basal level to appearance at the skin surface takes roughly 2 to 4 weeks. (In fact, flaking off of this outer keratinized layer produces a large amount of common household dust.) See [1,p.1158] for more details on this process.
Returning to our earlier comments on the various types of EB, junctional and many dystrophic epidermolysis bullosa cases are inherited recessively and result from a lack of expression of functional protein. In addition, the rare recessive Koebner EBS cases fit this description. Cases such as these could be treated by gene replacement therapy. The web-site [4] provides an excellent description and graphics of the procedure involved. In a nutshell, two approaches are available. One approach uses a device called the gene gun which fires tiny gene coated gold spheres (carrying the correct DNA sequence) directly into the skin. The problem is that the method has proven to be inefficient for two reasons. On the one hand, only a few epidermal cells receive the gene. More imporantly, the duration of the treatment is shortlived, since the gene is not typically delivered to the stem cells. An alternate gene replacement approach, is more feasible. Roughly speaking, an initial biopsy of basal cells is taken and grown up in culture. A wild type copy of the good gene is transferred to the cultured cells. After gene transfer, cultured cells are grafted back onto the skin in areas of repeated blistering.
Dominantly inherited diseases such as EBS and some forms of dystrophic epidermolysis bullosa are caused by a protein defect that results in abnormal filament or fibril assembly. In these instances, the approach would be to knockout selectively the mutated allele while at the same time preserving the normal allele. At this time, research is still pending, thus there is no such therapy available.
The Future
Our discussion has focused entirely on a defective keratin intermediate filament disease of the epidermal skin. However, this body of work suggests a paradigm that could be applied to several other KIF containing cell types within the human body. A study of the relationship between KIF defects and disease in these other cell types is currently underway. Key examples would be epithelial cells of the gastrointestinal tract or certain keratins associated with hair and nail formation.
Summary
The study of keratin diseases of the basal epidermal cells has given us a clearer understanding of the structure and function of intermediate filaments and their associated adhesion/anchoring proteins. A picture has emerged in which keratin intermediate filaments play a central role in providing mechanical stability and resistance to stress for the cell. A hypothesis has emerged that defects in two keratin genes (K5 and K14) lead to the blistering disease epidermolysis bullosa simplex. A strong correlation between the actual site of mutation and the phenotype of EBS has also come to light.
Our approach on this project (and the last one) has been the following: One person presents the lecture, one person produces the handout and one person produces the paper. However, as a group we hash out the particulars of the outline and summary. In the case of this project, Jack agreed to speak, Dave prepared the paper and Dan prepared the outline and log. We started with a draft outline Jack prepared. We used this as a skeleton to which we added more material for the handout. This was then filled in to form the paper. It was a process of successive refinement with everyone involved. We had an eventful Saturday in Tacoma where we realized "meeting at the Starbucks near the Tacoma Mall" is not detailed enough; there are THREE Starbucks in this area! After running around between Seattle and Olympia, we did settle on a final form of the handout. Over the weekend, Dan implemented final handout revisions, Dave finished revising the paper and Jack worked on his presentation. We are a dedicated Biology 401 team!
References
[1] Alberts. et. al, Molecular Biology of the Cell, 3rd edition., Garland Publishing, 1994.
[2] Collingwood, D., Hwang, D., Mehn, J., Epidermolysis Bullosa Simplex: The Cytoskeleton. Biology 401, Summer 1997. A web viewable document located at the WWW address: http://www.biology.washington.edu/bsa/bio401-sum97/papers%20AB/AB%20Epidermolysis%201.html
[3] Corden, L., McLean, W. Human keratin diseases: Hereditary fragility of specific epithelial tissues. Exp. Dermatol. 1996:5, p. 297-307.
[4] Dystrophic Epidermolysis Bullosa Research Association, Inc. An interactive web-site with the WWW address: http://www.debra.org/.
[5] Fuchs, E. Intermediate filaments and disease: Mutations that cripple cell strength. J. Cell Bio., Vol. 125, No. 3, May 1994, p. 511-516.
[6] Korge, P, Krieg, T. The molecular basis of inherited bullous diseases. J. Mol. Med. (1996), 74, p. 59-70.
[7] Marinkovich, P., The Electronic Textbook of Dermatology. An interactive web-site with WWW address http://telemedicine.org/blister.htm#dejfig
Dave Collingwood _______________________________
Daniel Hwang _______________________________
Jack Mehn ________________________________
This document is web viewable at the WWW address:
http://www.math.washington.edu/~colling/Biology/paper2.html