Stem Cells that Control Skin and Hair Color


A research team at NYU Langone Medical Center has uncovered a pair of molecular signals that control the hair and skin color in mice and humans. Manipulation of these very signals may lead to therapies or even drugs to treat skin pigment disorders, such as vitiligo.

Vitiligo
Vitiligo

Vitiligo is somewhat disfiguring condition characterized by the loss of skin pigmentation, leaving a blotchy, white appearance. Finding ways to activate these two signaling pathways may provide clinicians with the means to mobilize the pigment-synthesizing stem cells that place pigment in skin structures and, potentially, repigment the pigment-bearing structures that were damaged in cases of vitiligo. Such treatment might also repigment grayed hair cells in older people, and even correct the discoloration that affects scars.

Workers in the laboratory of Mayumi Ito at the Ronald O. Perelman Department of Dermatology and the Department of Cell Biology showed that a skin-based stem cell population of pigment-producing cells, known as “melanocytes,” grow and regenerate in response to two molecular signals. The Endothelin receptor type B (EdnrB) protein is found on the surfaces of melanocytes. EdnrB signaling promotes the growth and differentiation of melanocyte stem cells (McSCs). Activation of EdnrB greatly enhances the regeneration of hair-based and epidermal-based melanocytes. However, EdnrB does act alone. Instead, the effect of EdnrB depends upon active Wnt signaling. This Wnt signal is initiated by the secretion of Wnt glycoproteins by the hair follicle cells.

This work was published in Cell Reports, April 2016 DOI: http://dx.doi.org/10.1016/j.celrep.2016.04.006.

Previous work on EdnrB has established that it plays a central role in blood vessel development. This work by Ito and his team is the first indication that pigment-producing melanocytes, which provide color to hair and skin, are controlled by this protein.

A lack of EdnrB signaling in mice caused premature graying of the hair. However, stimulating the EdnrB pathway resulted in a 15-fold increase in melanocyte stem cell pigment production, and by two months, the mice showed hyperpigmentation. In fact, wounded skin in these mice became pigmented upon healing.

Overexpression of Edn1 Promotes Upward Migration of McSCs and Generation of Epidermal Melanocytes following Wounding (A–D) Whole-mount image of X - gal - stained wound area of Dct-LacZ (control; A and B) and Tyr-CreER ; EdnrB fl/fl ; Dct-lacZ (C and D) at indicated days after re-epithelialization. (E–J) Double immunohistochemical staining of Dct and Ki67 in the bulge (E and H), upper hair follicle (F and I), and inter-follicular epidermis (G and J ) in control (E–G) and K14-rtTA ; TetO-Edn1-LacZ (Edn1; H–J) mice. (K–N) Whole - mount analyses of wound site (K and L) and de novo hair follicles (M and N) within wound site from control (K and M) and Edn1 mice (L and N) at 8 days after re-epithelialization. (O–R) Quantification of the number of Dct - LacZ+ cells in wound site (O), the percentage of Ki67+/Dct+ cells (P), the number of pigmented cells in wound site (Q), and the percentage of pigmented de novo hair (R), respectively. Dashed lines indicate periphery of wound site in (A) and (D) and boundary between epidermis and dermis in (E)–(J). Arrowheads show Dct - LacZ + cells in wound area in (A)–(D) and Ki67+/Dct+ cells (H)–(J). IFE, inter-follicular epidermis; UF, upper follicle. Data are presented as the mean ± SD. *p < 0.01; **p < 0.02; ***p < 0.05. The scale bar represents 1 mm in (A), 50 m m in (E), 200 m m in (K) and (L), and 100 m m in (M) and (N).
Overexpression of Edn1 Promotes Upward Migration of McSCs and Generation of Epidermal Melanocytes following Wounding (A–D) Whole-mount image of X-gal – stained wound area of
Dct-LacZ (control; A and B) and Tyr-CreER; EdnrB fl/fl; Dct-lacZ (C and D) at indicated days after
re-epithelialization. (E–J) Double immunohistochemical staining of Dct and Ki67 in the bulge (E and H), upper hair follicle (F and I), and inter-follicular epidermis (G and J) in control (E–G) and K14-rtTA;
TetO-Edn1-LacZ (Edn1; H–J) mice.  (K–N) Whole-mount analyses of wound site (K and L) and de novo hair follicles (M and N) within wound site from control (K and M) and Edn1 mice (L and N) at
8 days after re-epithelialization.  (O–R) Quantification of the number of Dct-LacZ+ cells in wound site (O), the percentage of Ki67+/Dct+ cells (P), the number of pigmented cells in wound site (Q),
and the percentage of pigmented de novo hair (R), respectively.  Dashed lines indicate periphery of wound site in (A) and (D) and boundary between epidermis and dermis in (E)–(J). Arrowheads show Dct-LacZ+ cells in wound area in (A)–(D) and Ki67+/Dct+ cells (H)–(J). IFE, inter-follicular epidermis; UF, upper follicle. Data are presented as the mean ± SD. *p < 0.01; **p < 0.02; ***p < 0.05.
The scale bar represents 1 mm in (A), 50 micrometer in (E), 200 micrometer in (K) and (L), and 100
micrometer in (M) and (N).

If the Wnt signaling pathway was blocked, stem cell growth and maturation sputtered and stalled and never got going, even when the EdnrB pathway was working properly. These mice had unpigmented fur (see E in figure below).

Loss of beta-catenin Function Suppresses Edn1-Mediated Effects on McSC Proliferation, Differentiation, and Upward Migration (A) Experimental scheme for treatment of Tyr-CreER ; b -catenin fl/fl ; K14-rtTA ; TetO-Edn1-LacZ ( b -cat cKO; Edn1 ) mice and control K14-rtTA ; TetO-Edn1- LacZ ( Edn1 ) mice. (B–E) Gross appearance of Edn1 (B and D) and b -cat cKO; Edn1 mice (C and E) at second (B and C) and third telogen (D and E). (F–K) Immunohistochemistry for indicated markers (F, G, I, and J) and bright- field image (H and K) of bulge/sHG region in skin sections from Edn1 mice (F–H) and b -cat cKO; Edn1 mice (I–K) at anagen II. (L–Q) Bright-field image (L–N) and Dct immunostaining of whole-mount wound site (O–Q) from Tyr-CreER ; b -catenin fl/fl ( b -cat cKO; L and O), Edn1 (M and P), and b -cat cKO; Edn1 mice (N and Q). (R and S) Quantification of the percentage of Dct+ cells positive for Ki67, Tyr, and pigmentation (R) and the number of Dct+ cells in wounded site (S). Dashed lines indicate border between hair follicle and dermis. Arrow- heads indicate double positive cells for indicated markers (F and G) and pigmented cells (H). Data are presented as the mean ± SD.*p<0.05;**p< 0.02; ***p < 0.001. The scale bar represents 1 cm in (B)–(E), 10 m min(F), and 200 m min(L). 1
Loss of beta-catenin Function Suppresses Edn1-Mediated Effects on McSC Proliferation, Differentiation, and Upward Migration
(A) Experimental scheme for treatment of
Tyr-CreER; beta-catenin fl/fl; K14-rtTA; TetO-Edn1-LacZ (beta-cat cKO; Edn1) mice and control K14-rtTA; TetO-Edn1-LacZ (Edn1) mice.
(B–E) Gross appearance of Edn1 (B and D) and
beta-cat cKO; Edn1 mice (C and E) at second (B and C) and third telogen (D and E). (F–K) Immunohistochemistry for indicated markers (F, G, I, and J) and bright-field image (H and K) of bulge/sHG region in skin sections from Edn1 mice (F–H) and beta-cat cKO; Edn1
mice (I–K) at anagen II. (L–Q) Bright-field image (L–N) and Dct immunostaining of whole-mount wound site (O–Q) from
Tyr-CreER; beta-catenin fl/fl (beta-cat cKO; L and O), Edn1 (M and P), and beta-cat cKO; Edn1 mice (N and Q). (R and S) Quantification of the percentage of Dct+ cells positive for Ki67, Tyr, and pigmentation (R) and the number of Dct+ cells in wounded site (S).
Dashed lines indicate border between hair follicle and dermis. Arrow-heads indicate double positive cells for indicated markers (F and G) and pigmented cells (H). Data are presented as the mean ± SD.*p<0.05;**p<
0.02; ***p < 0.001. The scale bar represents 1 cm in (B)–(E), 10 micrometers in (F), and 200 micrometer  in (L).

However, perhaps the most exciting finding for Ito and his colleagues was that Wnt-dependent, EdnrB signaling rescued the defects in melanocyte regeneration caused by loss of the Mc1R receptor. This is precisely the receptor that does not function properly in red-heads, which causes them to have red hair and very light skin that burns easily in the sun. These data suggest that Edn/EdnrB/Wnt signaling in McSCs can be used therapeutically to promote photoprotective-melanocyte regeneration in those patients with increased risk of skin cancers due to their very lightly colored skin.

Melanocyte Stem Cell Modeld

New Technology Reprograms Skin Fibroblasts


Fibroblasts are one of the main components of connective tissue, which is the main reason scientists typically exploit them for experiments. A collaborative team of scientists from the University of Pennsylvania, Boston University, and the New Jersey Institute of Technology have invented a way to reprogram fibroblasts without going through a pluripotent stage.

The senior author of this study, Xiaowei Xu, associate professor of pathology and laboratory medicine at the University of Pennsylvania School of Medicine, said, “Through direct reprogramming, we do not have to go through the pluripotent stem cell stage, but directly convert fibroblasts to melanocytes . So these cells do not have tumorigenicity” (the ability to cause tumors).

Melanocytes are found in the skin and they are responsible for the pigment in our skin. They are in the uppermost layer of the skin, known as the epidermis, and produce melanin, a brown pigment that helps screen against the harmful effects of UV light.

Turning a fibroblast into a melanocytes might seem trivial for a stem cell scientist; just reprogram the fibroblast into an induced pluripotent stem cells and then differentiate it into a melanocytes. However, this procedure utilized direct reprogramming, in which the fibroblast was converted into a melanocytes without traversing through the pluripotent stage. The difficultly with direct reprogramming is finding the right cocktail of genes and/or growth factors that will accomplish the deed.

Xu and his colleagues began their search by examining the genes that are specific to melanocytes. They found 10 different transcription factors that are important for melanocytes development. Next they screened these ten genes for their ability to convert a fibroblast into a melanocytes. They found that of the ten melanocytes-specific genes, three of them, Sox10, MITF, and PAX3 could do the job effectively. They called this gene combination “SMP3.”

When Xu and others tested SMP3 on mouse embryonic fibroblasts, they quickly expressed melanocytes-specific genes. When Xu’s group used SMP3 on human fetal dermal cells, once again, the cells rapidly differentiated into melanocytes. Xu and his team referred to these cells as hi-Mel, which is short for human, induced melanocytes.

When hi-Mel were grown in culture they produced melanin a plenty. When they were implanted into the skin of pigmentless mice, once again they rose to the challenge and made a great deal of pigment. Thus hi-Mel express the same genes as melanocytes and they behave for all intents and purposes as melanocytes.

Xu and his colleagues think that their procedure might be able to treat human patients with a condition called vitiligo in which the skin has patches that are devoid of pigment.

Another potential use of this technology is a way to effectively study melanoma, one of the most dangerous skin cancers known to human medicine. My good friend and SAU colleague died over a year ago from melanoma and having better ways to treat this monster would have been marvelous for Charlie, and his family, who miss him dearly. By generating melanocytes from the fibroblasts of melanoma patients, they can “screen not only to find why these patients easily develop melanoma, but possibly use their cells to screen for small compounds that can prevent melanoma from happening.”.

Also, because so the body contains so many fibroblasts in the first place, this reprogramming technique is well-suited for other cell-based treatments.

iPSCs from other sources


Nature News has a fascinating article on induction of pluripotent stem cells by means of genetic reprogramming.  Typically, skin cells have been used for reprogramming experiments.  Cells called fibroblasts, which are prevalent in skin and help heal the skin when injured, have been the cell of choice for induced pluripotent stem cell (iPSC) production.

Recently, fat cells and pigmented skin cells appear to produce iPSCs much more efficiently and quickly.  Reprogramming procedures with skin fibroblasts are rather inefficient and slow.  Reprogramming human skin cells takes about a month for 1 in 10,000 fibroblasts to form iPSCs. Other cell types can do the job besides fibroblasts.  For example blood, hair, bone marrow, and neural stem cells can be converted into iPSCs, but their conversion rate is not any better than that of skin fibroblasts.  However, foreskin fibroblasts from a baby are better at making iPSCs (see Aasen, T. et al. Nat. Biotechnol. 26, 1276-1284), but this is hardly a good source of cells for adults.

Is there are better way?  Apparently there is.  Joseph Wu and Michael Langaker at Stanford University School of Medicine in California have converted fat tissue into iPSCs, and it took them only two days to acquire enough material for reprogramming.  Compare that to about a month of biopsies to get enough fibroblasts.  Additionally, cellular reprogramming of fat cells took only two more weeks and was 20-times more efficient than fibroblast reprogramming.  By using fat cells, they were able to reduce the time required for the procedure by six to eight weeks (Sun, N. et al. Proc. Natl. Acad. Sci. USA, doi:10.1073/pnas.0908450106).  Likewise, Konrad Hochedlinger and his co-workers at the Massachusetts General Hospital in Boston reprogrammed melanocytes, the skin cells that produce pigmented skin.  Melanocytes undergo reprogramming after just 10 days and with five-fold greater success rates compared with fibroblasts (Utikal, J., Maherali, N., Kulalert, W. & Hochedlinger, K. J. Cell Sci., doi:10.1242/jcs.054783).

These findings suggest that making iPSCs is easier than previously thought.  The therapeutic uses of iPSCs just became even more attractive.