More Research Updates

Atopic dermatitis (AD), or eczema, is one of the biggest challenges within the field of dermatology both in the clinic and in the research laboratory.

Read more

In the last several years, there has been a great deal of interest in the relationship between stem cells and cancer. In this article, Dr. Stephen Lyle considers several aspects of stem cell biology that apply to cancer, while focusing on the skin. Dr. Lyle believes that the application of advances in stem cell biology to cancer research can lead to a better understanding of what drives carcinogenesis, tumor progression, recurrence and metastasis, and that the localization of keratinocyte, melanocyte and neural crest-derived stem cells in the bulge region of the skin provides an excellent model system to study the relationship of stem cells and cancer.

Read more

Dermatologists are constantly exposed to clinical trials data at meetings, in the news, and in the medical literature.

Read more

Despite the adverse effects of ultraviolet (UV) exposure, the indoor tanning business is booming.

Read more

Research Updates

Gene Therapy or Other Strategies for Incurable Genetic Skin Diseases?

David Woodley, Mei Chen

Tuesday, August 07, 2007

There are a number of genetic diseases in dermatology that are devastating, life-threatening, and incurable. One prominent example is hereditary dystrophic epidermolysis bullosa (DEB). The patients are born with a defect in the gene that encodes type VII (anchoring fibril) collagen. Without the ability to synthesize and secrete normal type VII collagen alpha chains, DEB patients do not have an adequate complement of normal anchoring fibrils - large structures that hold the epidermis and its underlying basement membrane zone (BMZ) onto the papillary dermis. Without the normal complement of fully functional anchoring fibrils, DEB patients have poor adherence of their epidermis onto their dermis and experience marked skin fragility and blistering where the epidermis disadheres to the dermis. These blisters form erosions and heal with scarring and small pinpoint white milia cysts. There is no cure for DEB. After 2 to 4 decades of dealing with painful, disfiguring wounds, DEB patients often die from an aggressive squamous cell carcinoma arising in a chronic wound site and metastasizing to other organs, especially bone.

When one thinks of a treatment strategy for DEB, the first thing that comes to mind is likely to be gene therapy. How can we get a DEB patient's skin cells - keratinocytes or fibroblasts - to make functional type VII collagen alpha chains and form anchoring fibril structures?

If one examines keratinocytes or fibroblasts from DEB patients in the laboratory, they are very different from normal keratinocytes and normal fibroblasts. The DEB cells appear to be odd morphologically as they are enlarged and elongated. In addition, the DEB cells have a considerably reduced proliferative capacity compared with normal cells, and, compared with normal cells, the DEB cells do not attach well to their matrix substratum (tested by attachment assays) and have a hypomotility.

Using a lentiviral vector that expressed the full-length human type VII collagen alpha chain, we "gene-corrected" DEB skin cells from DEB patients who had severe recessive DEB (Hallopeau Simons) and made virtually no type VII collagen. (Lentiviral vectors, a type of retrovirus, can penetrate the intact membrane of the target cell and can change the expression of the target cell's gene for up to 6 months.) We found that by gene correcting these cells in the laboratory, all of their abnormal parameters (morphology, doubling time, matrix attachment, and motility) were completely normalized and identical to normal skin cells.

We combined DEB cells, gene-corrected DEB cells, or normal skin cells into skin equivalents (SEs). These are three-dimensional organotypic cultures that are organized into a well-developed epidermis and dermis. We then grafted these human SEs onto hairless mice that were immunocompromised and would not reject the human tissue. As predicted, the SEs composed of DEB cells showed poor epidermal-dermal adherence and no type VII collagen or anchoring fibrils at the dermal-epidermal junction (DEJ). In contrast, SEs composed of gene-corrected DEB cells or normal human cells, expressed type VII collagen and anchoring fibrils at the DEJ and exhibited excellent dermal-epidermal adherence.

Ex Vivo Gene Therapy

These in vitro and in vivo experiments suggest that gene therapy might help DEB patients. The idea would be to take a small 4-8 mm skin biopsy from the DEB patients, place their keratinocytes into culture, gene-correct the cells with our lentiviral method, grow large sheets of gene-corrected keratinocyte autografts, and then transplant them onto the DEB patient's wounds. We call this strategy "ex vivo gene therapy".

Our laboratory used a similar technology successfully on severe burn patients and showed that we could transplant sheets of cultured keratinocyte autografts onto burn victims and cover their wounds. Nevertheless, the experience with burn patients demonstrated that the use of transplanted cultured keratinocyte autografts was technically and logistically difficult. First, to optimize autograft take, we had to excise the burn patient's wounds down to muscle fascia in the operating room. The autograft was then placed on the dry fascia and carefully bandaged and the patient immobilized. Even under the best circumstances, a significant number of transplanted cultured autografts were lost, probably because they were such thin, delicate membranes that were readily traumatized by patient movement and the hostile protease-rich wound environment. Nevertheless, perhaps rather than using thin cultured keratinocyte sheets for transplantation onto DEB patients, one could construct SEs in the laboratory composed of the patient's gene-corrected keratinocytes and fibroblasts and then transplant these SEs onto the DEB patient's wounds. These SEs would have both an epidermal and dermal component and would be heartier and more resilient than this cultured keratinocyte autograft.

One remarkable thing about collagens is that cells make the alpha chains and export them outside of the cell. In the extracellular environment, the alpha chains wind up into a triple helix and make the collagen molecules. We found that both human keratinocytes and dermal fibroblasts could make type VII collagen alpha chains and export them into the extracellular space where type VII collagen molecules then organize themselves into anchoring fibril structures. Exactly how this happens is not known. We have shown that part of the type VII collagen alpha chain has affinity for other basement membrane components such as laminin-5 and type IV collagen. It might be that type VII collagen molecules in the extracellular space bind and aggregate on the existing basement membrane zone and this nidus of type VII collagen then allows self-assembly into anchoring fibril structures.

Intradermal Protein Therapy

Knowing this remarkable ability for self-assembly in the extracellular space, we wondered whether delivering the type VII collagen protein to DEB skin would allow it to incorporate into the BMZ of the skin, make anchoring fibrils and correct the DEB phenotype. We grafted DEB skin equivalents to immunocompromised mice as mentioned above and then injected recombinant human type VII collagen into the high papillary dermis of the DEB skin. We found that the injected collagen aligned itself in a linear fashion along the BMZ beneath the epidermis and formed anchoring fibrils. Moreover, the poor epidermal-dermal adherence of the DEB skin was reversed and normalized. We called this strategy "intradermal protein therapy," and these experiments suggested that protein therapy might be an easier and more direct treatment for DEB than ex vivo gene therapy

One advantage of protein therapy for DEB is that dermatologists have a lot of experience injecting type I collagen into the papillary dermis of patients who wish to improve their rhytids caused by photoaging and chronological aging. If we had approval from the U.S. Food and Drug Administration (FDA) for the use of recombinant type VII collagen, we could try this immediately on a DEB patient. How long would it last? We do not know. The turnover time of type VII collagen and anchoring fibrils in human skin is not known. We do know that we need to re-inject human dermal type I collagen into patients about every 6 months to keep their rhytids adequately plumped up. We also know from our grafting and animal experiments that the injected human type VII collagen remains in a linear configuration at the BMZ for at least 5 months, the longest time period we have examined to date.

Intravenous Protein Therapy

Severely affected DEB patients often have multiple wounds, particularly over trauma-prone sites such as the shoulders, lower back, elbows, knees, feet and hands. Therefore, the physician would need to do many intradermal injections of type VII collagen to treat these sites if the above outlined protein therapy were to work in DEB patients. Recently, however, we have been experimenting with delivering type VII collagen intravenously. If we wound the skin of a mouse or if we wound human skin that has been transplanted onto a mouse and then inject human type VII collagen into the tail vein of the mouse, the injected type VII collagen "homes" to the wound, incorporates into the healing wound's DEJ, forms anchoring fibrils, and markedly promotes wound closure. We call this strategy "intravenous protein therapy".

We found that the injected human type VII collagen did not home to un-wounded skin or to the other organs of the mouse. This strategy might be useful for DEB patients who have multiple skin wounds all over their body. Perhaps, recombinant type VII collagen could be injected into the vein of DEB patients and it would home to the entire DEB skin wounds and correct the DEB skin phenotype at those sites. This would be highly convenient to the patient and would also allow treatment of inaccessible sites such as the esophagus.

Recent experiments in our laboratory suggest that it may not be too far-fetched to envision protein therapy for DEB patients. A knock-out mouse model of DEB was made in the laboratory of Dr. Jouni Uitto. These mice have keratinocytes and fibroblasts that cannot make type VII collagen. They have severe skin fragility and blistering of their skin and no anchoring fibrils at their DEJ. These mice can only live a matter of a few days. We have intradermally injected human recombinant type VII collagen to these mice. It incorporates into the DEJ of the murine skin, forms anchoring fibrils, promotes the healing of wounds and causes the cessation of new blister formation. Moreover, the mice can then live months rather than days.

Intradermal Cell Therapy

Although intradermal protein therapy and intravenous protein therapy are direct attractive strategies for the treatment of DEB, there is the unknown question of how long the administered protein would last in the patient and how often re-administration of the exogenous type VII collagen would need to be done. Our laboratory and the laboratory of Dr. Paul Khavari at Stanford University have both taken gene-corrected DEB fibroblasts and intradermally injected them into DEB skin grafted onto mice and showed that the injected fibroblasts synthesize and secrete type VII collagen into the high dermis where it organizes into a linear configuration along the DEJ, forms anchoring fibrils, and normalizes the DEB phenotype. In this case, one would anticipate that the injected fibroblasts remain in the dermis and continue to synthesize and secrete type VII collagen. We call this strategy "intradermal cell therapy."

It may have the advantage of creating a more durable therapy than protein therapy since the exogenous cells in the dermis would be capable of continually synthesizing the needed protein. Another advantage is that dermal fibroblasts are easy to culture and grow into large quantities. Furthermore, the intradermal administration of gene-corrected cells is very direct and technically easy. It is not known, however, how long these exogenous cells would last and be functional in the DEB skin.

Vector Therapy

As mentioned above, to gene-correct DEB skin cells, we infect these cells with a lentiviral vector that expresses the full-length gene for human type VII collagen. The infected DEB cells then express the transgene and their phenotype normalizes. Using our DEB skin equivalent/mouse model, we tried simply intradermally injecting the lentiviral vectors into the DEB SEs. We found that the injected lentiviral vectors infected the resident endothelial cells and resident fibroblasts in the high dermis of the SE. These cells then began to synthesize and secrete human type VII collagen into the extracellular space. The extracellular type VII collagen then organized itself in a linear fashion along the DEJ, formed anchoring fibrils and corrected the DEB phenotype. We call this strategy "vector therapy."

This strategy bypasses the use of injecting exogenous gene-corrected cells, but rather uses the host's resident cells in the skin and turns them into cell factories that synthesize and secrete type VII collagen into the extracellular dermal space. Compared with protein therapy, vector therapy may offer more durability because fibroblasts are thought to have a long lifespan in human skin. Lentiviral vectors can be introduced into both dividing and non-dividing cells. Therefore, with vector therapy using lentiviral vectors, there is the possibility of infecting the host's resident stem cells in the skin and inducing a permanent gene-correction.

In working with these various strategies for DEB, we are very much struck by the plasticity of skin biology. We can cross species and introduce human type VII collagen intravenously into a wounded mouse and the result is healed mouse skin with both human and murine anchoring fibrils. No matter how we get type VII collagen alpha chains into the high dermis of skin, they somehow know how to bind to the DEJ and organize into anchoring fibrils spontaneously. That is why intradermal protein therapy, intravenous protein therapy, cell therapy, and vector therapy all work!


We now wish to translate this laboratory work into the clinic and attempt to help DEB patients. As we pondered the various strategies outlined above, we concluded that the first attempt should be with intradermal protein therapy. This approach would likely have an easier time passing through the regulatory bodies such as the FDA because it is so direct and easy and also does not involve any virus being introduced into the patient.

Any new protein to be used for therapy in human beings must be a Good Manufacturing Practice (GMP)-quality protein. Although we can manufacture milligram quantities of human recombinant type VII collagen in our University of Southern California (USC) laboratories, the purified collagen has not gone through the rigors of GMP processing. USC does not have a GMP facility, so we need a partner. It is estimated that this step will cost about $300,000. When this is accomplished, we believe we can begin clinical trials of treating DEB patients with intradermal human recombinant type VII collagen.

Lastly, what about the possibility of invoking an autoimmune disease when one injects a large protein into a person who is genetically naïve to that protein? Would this DEB patient's immune system view the newly introduced protein as a foreign protein and begin to make autoantibodies against it? Would our above strategies all result in converting a patient with genetic DEB into a patient with epidermolysis bullosa acquisita (EBA)? The answer is possibly yes.

It appears that our DEB knockout mice treated with type VII collagen protein therapy begin to make antibodies against the injected protein. To prevent this, it may be that DEB patients will require some level of immunosuppression. Nevertheless, a paper by Dr. Susana Ortiz-Urda from the laboratory of Dr. Paul Khavari provides some hope that treated DEB patients might not develop autoimmunity to type VII collagen and develop EBA. In their study, they examined the gene defects in many DEB patients from the National EB Registry. They found that the majority of these patients had a gene defect that allowed some of the amino-terminal end of the type VII collagen alpha chain to be made. This end of the alpha chain contains the non-collagen 1 (NC1) domain of molecule. We and others have shown that the NC1 domain is the most antigenic part of type VII collagen. Therefore, if the immune systems of the majority of DEB patients have seen NC1, it may be that these DEB patients are tolerant to the most antigenic part of the molecule and would not develop an autoimmune EBA-like condition.


  1. Chen M, Kashahara N, Keene DR, et al. Restoration of type VII collagen expression and function in dystrophic epidermolysis bullosa. Nat Genet. 2002;32:670-5.
  2. Woodley DT, Krueger GG, Jorgensen CM, et al. Normal and gene-corrected dystrophic epidermolysis bullosa fibroblasts alone can produce type VII collagen at the basement membrane zone. J Invest Dermatol. 2003;121:1021-8.
  3. Woodley DT, Atha T, Huang Y, et al. Intradermal injection of lentiviral vectors corrects regenerated human dystrophic epidermolysis bullosa skin tissue in vivo. Mol Ther. 2004;10:318-26.
  4. Woodley DT, Keene DR, Atha T, et al. Injection of recombinant human type VII collagen restores collagen function in dystrophic epidermolysis bullosa. Nat Med. 2004;10:693-5.
  5. Woodley DT, Remington J, Huang Y, et al. Intravenously injected human fibroblasts home to skin wounds, deliver type VII collagen and promote wound healing. Mol Ther. 2007;15:628-35.
  6. Ortiz-Urda S, Thyagarajan B, Keene DR, et al. Stable nonviral genetic correction of inherited human skin disease. Nat Med. 2002;8:1166-70.
  7. Oritz-Urda S, Lin Q, Green CL, et al. Injection of genetically engineered fibroblasts corrects regenerated human epidermolysis bullosa skin tissue. J Clin Invest. 2003 Jan;111:251-5.
  8. Woodley DT and Chen M. Epidermolysis bullosa: then and now. J Am Acad Dermatol. 2004;51:S55-7.