The hair follicle as a target for gene therapy

ARTICLE

The hair follicle plays a pivotal role in cutaneous biology. As an epidermal appendage, it possesses a full complement of epidermal cell types, including keratinocytes, melanocytes, and Langerhans cells, which all reside in the follicle outer root sheath that is contiguous with the epidermis. When these cell types are lost from the epidermis (for example, through wounding, ultraviolet light, thermal or chemical injury), precursor cells proliferate and migrate out of the hair follicle to repopulate the epidermis. The ultimate source of keratinocytes and possibly melanocytes for repopulating the epidermis is thought to reside in the hair follicle bulge [1-4]. The bulge consists of a cluster of biochemically distinct outer-root sheath cells which are located at the lowermost portion of the permanent follicle [2, 5]. Keratinocytes within the bulge have the characteristic properties of epithelial stem cells in that they are the longest-lived epithelial cells in the hair follicle [6], and they are biochemically distinct because they preferentially express cytokeratin 15 [5]. The bulge is also thought to give rise to the lower hair follicle matrix cells [2]. Thus, the targeting of cells within the hair follicle, and particularly the bulge, is a logical goal for those interested in gene therapy for the treatment of a wide variety of cutaneous disease.

Because of its accessibility, the skin may seem like an attractive target for topically delivered gene therapy. However, since the epidermis and stratum corneum are designed as barriers to prevent ingress of material such as bacteria and other infectious agents, the challenges of introducing genes into human epidermal cells are formidable. These difficulties are avoided by using an ex vivo approach in which cells are removed, transfected in vitro ­ usually with a viral vector, and reintroduced to the host. The major disadvantages of ex vivo strategies include the need to culture cells, which results in loss of the tissue architecture and alteration of the cell phenotype, and the surgical grafting that can lead to scarring and loss of appendageal structures such as the hair. The major advantages include the ability to expand the number of cells in vitro, the higher efficiencies of transfection achieved in vitro, the ability to screen the transduced cells prior to their introduction to the recipient and avoidance of an immune response to viral vectors, since the reintroduced cells should not produce viral proteins.

In vivo approaches involve direct gene delivery to intact skin through either viral or non-viral methods. Non-viral methods include direct injection of plasmid DNA or RNA:DNA oligonucleotides (RDO), particle bombardment, topical application of liposomes containing plasmid DNA (lipoplex), and topical application of naked DNA. In general, the major problems with in vivo methods include transient expression, low efficiency of gene transfer compared with in vitro techniques [7], and immune responses against viral vectors.

The constant proliferation within the epidermis and hair follicle is both an advantage and disadvantage for gene therapy. Retrovirus, RDO and probably adeno-associated virus (AAV) require cell proliferation for transgene expression and integration to occur. However, cells within the proliferative pool, called transient amplifying cells in the epidermis and hair follicle, generally have short life spans (the one notable exception to this may be the matrix keratinocytes in human scalp follicles which are thought to proliferate continuously for several years). Therefore, to achieve long term expression and permanent correction, stem cells must be targeted. Since stem cells rarely divide, targeting these cells is challenging and requires a thorough understanding of their proliferative behavior.

Because stem cells divide after a proliferative stimulus such as a wound, Ghazizadeh et al. were able to transduce long-lived cells within the epidermis and hair follicle in a series of ingenious experiments. They injected a retrovirus underneath the scab formed following removal of the epidermis by superficial dermabrasion. The wounding stimulated stem cell proliferation and allowed for stable integration of the transgene into progenitor cells in both the epidermis and hair follicle. This study showed that long-term retrovirus-mediated transgene expression is possible in the absence of immunological responses to the transgene products. While feasible in mouse models, retrovirus -mediated gene transfer in vivo may be limited with respect to its applicability in immunocompetent human patients.

Although permanent gene correction more than likely will be necessary to treat single gene mutations, for example in structural proteins, transient gene therapy may play an important role in the treatment of disorders caused by polygenic factors (Table 1). Transgenes would encode proteins that modulate hair growth or alter the immune response and essentially function similarly to pharmacological agents. They would not necessarily repair or correct mutated genes. Furthermore, because the lower hair follicle determines the characteristics of the hair, for example, its size and pigmentation, targeting the follicle at anagen onset, when the lower follicle is regenerating, may be a viable strategy for treating hair disorders.

Partly because of the problems associated with viral vectors for in vivo gene therapy, our research has focused on developing the use of topical liposomes for introducing plasmid DNA into hair follicle cells. Li and Hoffman pioneered these techniques when they transfected DNA into mouse hair follicle cells using topical lipoplexes (DNA/liposome mixtures) in vivo [8]. However, the applicability of these findings to human skin was not clear because of the marked differences between mouse and human skin and hair. In particular, mouse hair follicles cycle synchronously during the first two to three months of life, and are predominantly in telogen in the absence of proliferative stimuli. In the normal human scalp, the majority of the follicles are in anagen, while approximately 10% are in telogen, and a small percentage (5-10%) is in catagen or anagen-onset.

Additionally, in humans, the epidermis is thicker, hair follicles are much larger, and characteristics of dendritic epidermal cells are different from those in mice. In order to overcome the limitations of studying topical transfection only in mouse follicles, we utilized human scalp grafted to immunodeficient mice [9]. These grafts retain the characteristics of human skin and generate abundant normal-appearing human hair while remaining viable for many months. Histological analysis reveals that the human skin/SCID mouse chimera also parallels the growth of hair in the normal human scalp with respect to the proportion of follicles found in each stage of the cycle.

Recombinant genes are relatively large and polar molecules that must be guided to the nucleus of the target cell. Delivery vehicles are therefore critical to the success of gene therapy targeted to the hair follicle. The binding of the delivery vehicle to the target cell, the subsequent internalization of the DNA, the transport of the DNA from the cytoplasm to the nucleus, and finally, expression of the transgene all constitute potential limitations of this process, as well as targets for increasing efficiency of transfection.

Liposomes have a number of characteristics that make them suitable for topically applied carriers of drugs: they are composed of natural body constituents (e.g. lipids, sterols) and are therefore biodegradable, and they are relatively non-toxic and non-immunogenic [10]. Lipoplexes have been used routinely to transfect cells with plasmid DNA in vitro. Multiple variables affect transfection efficiency, including the ratio of liposomes to DNA, the absolute concentration of liposomes and DNA, and the liposome composition [11]. We found that similar variables affect transfection of human hair follicles in explant culture [9]. In particular, we found a wide range of transfection efficiencies (defined as number of cells expressing beta-galactosidase per follicle) based on the composition of the liposomes. In general, liposome preparations that were predicted to compact DNA were better for transfecting human hair follicles in vitro. Similarly, the same liposome preparation, pFx^TM-1, which is commercially available, yielded the highest transfection efficiency in vivo in mouse skin after topical application.

In vitro studies suggest that proliferating cells express plasmid DNA more efficiently than quiescent cells [12]. Because hair follicles generally only proliferate during anagen, we designed experiments to examine the influence of the stage of the hair cycle on the expression of topically-applied plasmid. By applying lipoplex (composed of pFx^TM-1 and pCMVbetagal) to the back skin of mice at different time points after depilation, we determined that hair follicles take up and express plasmid DNA only during anagen onset, which occurs within three days of depilation.

Our results may explain the variability of the findings of other investigators, who noted successful transfection in only one third of their experiments [13]. These investigators transfected mouse skin at 4 weeks of age when follicles in the same mouse may be in telogen, anagen onset or full-blown anagen. Only the follicles in anagen onset are transfected by this technique. By synchronously inducing anagen, we avoided this problem, and we conclude that the timing of lipoplex application to hair follicles just entering the anagen stage is a critical parameter for successful in vivo transfection.

At anagen onset, progenitor cells within the hair follicle are proliferating and they are accessible, probably because the follicle lacks an inner root sheath which normally prevents ingress of material from the environment later in anagen (Fig. 1). Even transient expression of transgenes at this time alters hair follicle cycling [14]. Our findings should pave the way for straightforward topical transfection experiments geared toward evaluating the in vivo effects of candidate genes on hair follicle growth and differentiation.

Using the human skin/SCID mouse chimera, it was possible to define parameters important for transfection, including liposome composition, timing of liposome application to the onset of a new hair cycle, and pretreatment with depilation and retinoic acid [9]. Both depilation, and depilation with retinoic acid pretreatment, increase the percentage of follicles in anagen onset as well as the efficiency of transfection. It is still unclear whether the effects of depilation and retinoic acid pretreatment are due to enhancement of transfectability of different subpopulations of follicles, or whether each treatment has additive effects on all of the follicles. In our studies, the transfection efficiency, calculated as a percentage of follicles within the population of follicles at anagen onset expressing beta-galactosidase was 48 ± 10% in xenografts treated with both depilation and retinoic acid, in contrast to only 6 ± 9% without pretreatment.

Topical gene therapy using liposomes with plasmid DNA to target the hair follicle in anagen onset may be useful for multigenic disorders such as alopecia areata and androgenetic alopecia (Table 1). In both of these conditions, the preponderance of affected hair follicles are in telogen or anagen onset, thus the majority of follicles should be susceptible to transfection with topical lipoplex. As mentioned, even transient expression of plasmid DNA at this time could result in long lasting changes in the phenotype of the hair by altering the characteristics of the lower hair follicle.

One topical approach for permanent correction of at least single base pair mutations stems from work on RNA-DNA oligonucleotides (RDO). Alexeev and Yoon [15] demonstrated that RDOs encoding for the correction of a single base pair mutation in tyrosinase can correct this mutation in albino mouse melanocytes in culture [15]. We then showed that this same RDO administered by topical application using liposomes can correct the albino mutation in vivo [16] in mice. After application, several pigmented hairs are detected in treated areas. These results demonstrate that topical lipoplex can also target melanocytes within the follicle. Ultimately, for RDO technology to become useful for therapeutic purposes in skin, its low efficiency will have to improve, and keratinocytes will have to be targeted.

In the future, by using different liposome preparations in combination with different gene promoters, it may be possible to target discreet cell populations within the cutaneous epithelium. For example, stem cells in the bulge might be targeted by the use of the keratin 15 promoter. The timing of expression could also be controlled by addition of inducible promoter elements to the transgenes. This would allow for a safe, convenient method of long-term transgene expression without reactive immune modulation. The major difficulties will be in manipulating the hair cycle to allow for transfection and increasing levels of expression. Levels of expression may be enhanced by increasing uptake of DNA by cells [17] and increasing transport of DNA into the nucleus [18].

As the entire field of gene therapy advances, perhaps through the advent of safer vectors and newer delivery vehicles, as well as adjunctive treatments that increase delivery (for example electroporation), cutaneous gene therapy may become a modality for treating skin and hair disorders as well as systemic disorders. Epidermolysis bullosa (EB) lamellar ichthyosis (LI) and X-linked ichthyosis are all monogenic recessive skin diseases in which promising advances suggest gene therapy will eventually play a role in treatment. Dominant disorders, or those that require elimination of a dominant negative protein such epidermolytic hyperkeratosis and epidermolysis bullosa simplex, represent more of a challenge [19].

Gene therapy targeted to the skin may also be used for systemic treatments, such as immunization and treatment of deficiencies in soluble factors. Shi and colleagues [20] and Fan and colleagues [21] have demonstrated a specific immune response to protein products encoded by topically-applied naked DNA in aqueous solution to mouse skin. This work suggests that topical approaches can target dendritic antigen presenting cells in the skin. Other potential systemic applications of cutaneous gene therapy include treatment of hemophilias as well as growth hormone deficiency by using epidermal keratinocytes as synthetic, secretory cells [22].

REFERENCES

1. Cotsarelis G, Kaur P, Dhouailly D, Hengge U, Bickenbach J. Epithelial stem cells in the skin: definition, markers, localization and functions. Experimental Dermatology 1999; 8: 80-8.

2. Cotsarelis G, Sun TT, Lavker RM. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis. Cell 1990; 61: 1329-37.

3. Jordan SA, Jackson IJ. MGF (KIT ligand) is a chemokinetic factor for melanoblast migration into hair follicles. Developmental Biology 2000; 225: 424-36.

4. Taylor G, Lehrer MS, Jensen PJ, Sun TT, Lavker RM. Involvement of follicular stem cells in forming not only the follicle but also the epidermis. Cell 2000; 102: 451-61.

5. Lyle S, Christofidou-Solomidou M, Liu Y, Elder D, Albelda S, Cotsarelis G. The C8/144B monoclonal antibody recognizes cytokeratin 15 and defines the location of human hair follicle stem cells. J Cell Science 1998; 111: 3179-88.

6. Morris RJ, Potten CS. Highly persistent label-retaining cells in the hair follicles of mice and their fate following induction of anagen. Journal of Investigative Dermatology 1999; 112: 470-5.

7. Ghazizadeh S, Harrington R, Taichman L. In vivo transduction of mouse epidermis with recombinant retroviral vectors: implications for cutaneous gene therapy. Gene Ther 1999; 6: 1267-75.

8. Li L, Hoffman RM. The feasibility of targeted selective gene therapy of the hair follicle. Nature Medicine 1995; 1: 705-6.

9. Domashenko A, Gupta S, Cotsarelis G. Efficient delivery of transgenes to human hair follicle progenitor cells using topical lipoplex. Nat Biotechnol 2000; 18: 420-3.

10. Fresta M, Puglisi G. Application of liposomes as potential cutaneous drug delivery systems. In vitro and in vivo investigation with radioactively labelled vesicles. Journal of Drug Targeting 1996; 4: 95-101.

11. Felgner PL, Gadek TR, Holm M, Roman R, Chan HW, Wenz M, Northrop JP, Ringold GM, Danielsen M. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proceedings of the National Academy of Sciences of the United States of America 1987; 84: 7413-7.

12. Nicolau C, Sene C. Liposome-mediated DNA transfer in eukaryotic cells: dependence of the transfer efficiency on the type of liposome used, and the host cell stage. Biochem Biophys Acta 1982; 721: 185-90.

13. Alexander MY, Akhurst RJ. Liposome-mediated gene transfer and expression via the skin. Human Molecular Genetics 1995; 4: 2279-85.

14. Sato N, Leopold PL, Crystal RG. Induction of the hair growth phase in postnatal mice by localized transient expression of Sonic hedgehog. J Clin Invest 1999; 104: 855-64.

15. Alexeev V, Yoon K. Stable and inheritable changes in genotype and phenotype of albino melanocytes induced by an RNA-DNA oligonucleotide. Nature Biotechnology 1998; 16: 1343-6.

16. Alexeev V, Igoucheva O, Domashenko A, Cotsarelis G, Yoon K. Localized in vivo genotypic and phenotypic correction of the albino mutation in skin by RNA-DNA oligonucleotide [see comments]. Nature Biotechnology 2000; 18: 43-7.

17. Rothbard JB, Garlington S, Lin Q, Kirschberg T, Kreider E, McGrane PL, Wender PA, Khavari PA. Conjugation of arginine oligomers to cyclosporin A facilitates topical delivery and inhibition of inflammation. Nat Med 2000; 6: 1253-7.

18. Subramanian A, Ranganathan P, Diamond SL. Nuclear targeting peptide scaffolds for lipofection of nondividing mammalian cells. Nature Biotechnology 1999; 17: 873-7.

19. Khavari PA. Therapeutic gene delivery to the skin. Molecular Medicine Today 1997; 3: 533-8.

20. Shi Z, Curiel DT, Tang DC. DNA-based non-invasive vaccination onto the skin. Vaccine 1999; 17: 2136-41.

21. Fan H, Lin Q, Morrissey GR, Khavari, PA. Immunization via hair follicles by topical application of naked DNA to normal skin. Nat Biotechnol 1999; 17: 870-2.

22. Fakharzadeh SS, Zhang Y, Sarkar R, Kazazian HH, Jr. Correction of the coagulation defect in hemophilia A mice through factor VIII expression in skin. Blood 2000; 95: 2799-805.