The Skin(ny) on Regenerating the Largest Organ to Save a Patient’s Life

Abstract

Stem cells hold enormous potential to regenerate an entire organ for organ replacement therapy. Recently, in Nature, Hirsch et al. (2017) restored the expression of laminin-332 in epidermal stem cells isolated from an individual with junctional epidermolysis bullosa and grafted the entire skin back to save the patient’s life.

An ultimate goal of regenerative medicine is to regenerate an entire organ and save patients who suffer from irreparable organ failure. Despite remarkable progress in stem cell biology, gene therapy, and genome editing in recent years, the complete regeneration and replacement of a human organ has been out of reach due to technical, ethical, and regulatory issues. However, a recently successful case (Hirsch et al., 2017), performed under some extraordinary conditions, provides a glimpse of the profound potentials of stem cell therapy, but also raises many questions for future studies.

Junctional epidermolysis bullosa (JEB) is a severe skin disease caused by loss-of-function mutations in genes encoding basement membrane components such as three genes that encode the subunit of laminin-332: LAMA3, LAMB3, and LAMC2. JEB is usually manifested as the failure of skin epithelium to attach to the basement membrane, causing extensive blistering and chronic wounds that severely compromise the barrier function of the skin. In this case, a 7-year-old child carried homozygous mutations within intron 14 of LAMB3. Shortly after admission in June 2015, he completely lost the epidermis on ~60% of his total body surface area (TBSA), and his condition continued to deteriorate. With the exhaustion of all other therapeutic options and a dire prognosis if left untreated, a stem cell therapy that combines ex vivo cell cultivation and retroviral infection to restore LAMB3 expression followed by an autologous skin graft (Mavilio et al., 2006) was approved by the regulatory authorities, the ethical review board, and the patient’s parents. A 4-cm² biopsy from a non-blistering area of the patient was used to establish primary keratinocyte culture on lethally radiated mouse 3T3J2 feeder fibroblasts—a method pioneered by James Rheinwald and Howard Green more than 40 years ago (Rheinwald and Green, 1975). In 8 days, ~5 million keratinocytes were expanded to ~21 million cells, which underwent retroviral transduction of LAMB3 cDNA to restore LAMB3 expression. These ‘‘corrected’’ cells were then further expanded and cultured on either plastic or fibrin substrate for skin transplantation. The first graft was performed about a month after the initial biopsy, followed by two more transplantations in the following months. In total, ~8,500 cm² transgenic skin grafts, a more than 2,000-fold expansion over the initial biopsy size, were generated to treat the complete epidermal loss on ~80% TBSA at the time of first transplantation. Remarkably, the patient was discharged only a few months after the first transplantation and returned to regular elementary school soon after his release. The transplanted skin is currently stable and shows no sign of deleterious effects or blistering.

The use of retroviral gene delivery raised concerns of a possibility for insertional mutagenesis and subsequent clonal selection, which could lead to adverse effects such as tumor development. Meanwhile, the random proviral integration also provided an opportunity for clonal analysis of skin stem cells because each unique integration can serve as a permanent marker for cell lineages. Using genome-wide deep-sequencing analysis of pre-graft transgenic cells (PGc) and primary keratinocytes isolated from 4- (4Mc) and 8-month (8Mc) post-graft biopsies, the authors detected 27,303 integrations in PGc, but only ~200–400 integrations in 4Mc and 8Mc. Most notably, the authors found that self-renewable skin lineages were established by a relatively small number of skin stem cells that could form holoclones. This observation supported the existence of a small percentage of long-lived skin stem cells, which give rise to holoclones, and many transit-amplifying progenitors, which give rise only to meroclones or paraclones (Barrandon and Green, 1987). Furthermore, the integration patterns within the clonogenic stem cells did not show preferences to different genomic regions such as intergenic, introns, exons, or promoters. Nor did the authors detect the enrichment of integration events near genes related to tumorigenesis. These genome-scale results were also supported by clinical observations, in which no adverse effects were observed in the transplanted skin. Thus, the regenerated skin has saved the patient’s life without adverse effects by the time of publication, 2 years after the initial biopsy.

This study demonstrates the important role of holoclone-forming long-lived stem cells. Therefore, it is critical for future studies and clinical treatment to preserve these stem cells. Remarkably, the lethally radiated 3T3J2 mouse fibroblast cells initially identified by Howard Green more than 40 years ago (Rheinwald and Green, 1975) are still the best option for this purpose. The same approach is also used in Epicel, an FDA-approved autologous epidermal cell graft procedure available in the U.S. With improved characterization of secreted proteins, exosomes, and the ability to modulate properties of extracellular matrix, it is possible to finally replace the mouse cells with defined factors and culture medium. On the other hand, future studies into purification and characterization of human skin stem cells are warranted to improve our knowledge and clinical application of these cells. To address the potential risk of proviral integration and genotoxicity arising from the use of retroviral gene delivery, it is now possible to use genome editing tools such as CRISPR-Cas9 to guide the correction of mutated genes directly (Ma et al., 2017) or to insert the therapeutic gene into a specific locus. Although Hirsch and colleagues found no evidence for clonal selection, it remains possible that random proviral integration will interfere with gene functions or negatively affect somatic cell evolution. One potential issue for such targeted approaches, however, is that the low efficiency of precise genome editing will prolong the time between the initial biopsy and the first treatment. Finally, a healthy skin not only functions as a protective barrier, but has additional functions such as secreting sweat, cooling the body, and sensing the environment. These functions are mediated by skin appendages such as hair follicles, sweat glands, and Merkel cells. Although regeneration of skin epidermis has been successful, our ability to regenerate skin appendages is limited. Because most skin appendages require interactions between epithelial and dermal cells, a better understanding of cell-cell communication among multiple cell types during normal development and wound healing and development of suitable culture conditions will be prerequisites to regenerate fully functional skin.

The success in this extraordinary case provides a blueprint for stem cell-based therapies in the future. First, it is essential to establish a productive ex vivo cell culture system to expand tissue stem cells that are capable of establishing self-renewable cell lineages. Second, a safe and effective gene therapy method should be used to deliver or correct desired genes in tissue stem cells. Third, the expansion of gene-targeted tissue stem cells and their differentiation into functional organoids (Clevers, 2016) will be required before potential clinical applications. Fourth, the regulatory authorities must also evolve with new discoveries in biology and biotechnology (Halme and Kessler, 2006). In conclusion, the scientific community still has much to learn about the biology of stem cells and how to enhance the safety and efficacy of stem cell-based therapy. However, with the success demonstrated by Professor De Luca and his team, a new hope is on the horizon to fulfill the promise of regenerative medicine for millions of patients.