Minoxidil: Mechanisms of Action on Hair Growth

A.G. Messenger; J. Rundegren


The British Journal of Dermatology. 2004;150(2) 

In This Article

The Cellular Response to Minoxidil

Whatever the mechanism whereby minoxidil modulates hair growth, there must be a primary effect on cell function ( Table 2 ). The hair follicle is a complex structure comprising epithelial, dermal, pigment and immune cells, and a perifollicular vasculature and neural network. Interactions between these cells are involved in regulating epithelial growth and differentiation and the hair cycle. Several of these cell types have been used in isolation to study minoxidil action, but attempts to localize minoxidil or a minoxidil metabolite binding to a specific cell population within the hair follicle have been unsuccessful.[42] Uptake studies in mouse vibrissae follicles showed that minoxidil and minoxidil sulphate concentrated in melanocytes and pigmented epithelial cells in the suprapapillary region of the follicle. However, this was probably due to nonspecific binding to melanin as there was no evidence of minoxidil binding in nonpigmented follicles yet pigmented and nonpigmented follicles showed a similar growth response to minoxidil.[43]

Several studies have examined the effect of minoxidil on cell proliferation in vitro. A variety of cell types have been used including epidermal keratinocytes, hair follicle keratinocytes and skin fibroblasts from humans, mice and macaques. In some studies, established keratinocyte and fibroblast cell lines have been used. The results have been variable and, to some extent, contradictory.

Boyera et al.[44] studied the effect of minoxidil on human keratinocytes of epidermal and hair follicle origin using a range of different culture conditions and proliferative markers. They found that micromolar concentrations of minoxidil stimulated proliferation in both cell types and in all culture conditions, whereas millimolar concentrations inhibited cell growth. In cells cultured from the stumptail macaque, minoxidil stimulated thymidine uptake by follicular keratinocytes but not by epidermal keratinocytes.[45] O'Keefe and Payne[46] also failed to show a stimulatory response to minoxidil in cultured human epidermal keratinocytes, although Baden and Kubilus[47] reported that minoxidil prolonged the time after confluence that keratinocytes could be subcultured.

Studies using fibroblasts have yielded similarly variable results. Murad and Pinnell[48] reported that high concentrations of minoxidil inhibited growth of human skin fibroblasts. On the other hand, thymidine uptake was increased in macaque follicular fibroblasts cultured in micromolar concentrations of minoxidil, but not in nonfollicular fibroblasts.[45] Sanders et al.[39] proposed that the variable results of cell culture experiments may be explained by the potassium channel-blocking activity of aminoglycoside antibiotics, routinely incorporated into cell culture media. Minoxidil stimulated growth of NIH 3T3 fibroblasts cultured in the absence of aminoglycosides but not in their presence, and the proliferative response of 3T3 cells to minoxidil was prevented by the potassium channel blockers tolbutamide and tetraethylammonium. In cultured human keratinocytes, aminoglycoside antibiotics partly suppressed the proliferative response to minoxidil but did not abolish it.[44]

The variations in the cell types and experimental protocols used mean that it is difficult to compare the results from these studies. On balance, they suggest that minoxidil can have a stimulatory effect on cell growth at clinically relevant concentrations, or delay cell senescence, and there is limited evidence that this is mediated by its action as a potassium channel opener.

Two groups have studied the effect of minoxidil on collagen synthesis. Murad and Pinnell[48] showed that minoxidil suppressed activity of the enzyme lysyl hydroxylase in human skin fibroblast cultures at concentrations down to 25 µmol L,-1 leading to production of a collagen deficient in hydroxylysine.[49] This appeared to be specific for lysyl hydroxylase as the activity of prolyl hydroxylase, which shares the same substrates and cofactors as lysyl hydroxylase, was unaffected. Minoxidil (0.5 mmol L-1) also suppressed collagen synthesis by rat vibrissae dermal papilla cells, both in monolayer cultures and in cells grown in collagen gels.[50] The concentrations of minoxidil used in these studies were quite high and the relevance of the results to hair growth is unknown.

The prostaglandin PGH2 is formed from arachidonate by the action of a cyclooxygenase (COX), also known as prostaglandin endoperoxide synthase (PGHS). PGH2 is the substrate for subsequent enzymatic modifications leading to the prostaglandins (PGD2, PGE2, PGF2), prostacyclin (PGI2) and thromboxane A2. There are two isoforms of PGHS, a widely distributed constitutive form PGHS-1, and an inducible form PGHS-2. The PGHS-1 isoform has been immunolocalized to the dermal papilla of human hair follicles during anagen and catagen.[51] Immunostaining for PGHS-2 was also seen in the dermal papilla but staining was weaker than that for PGHS-1 and was present only in anagen follicles. Minoxidil (AC50 = 80 µmol L-1) stimulated the activity of purified ovine PGHS-1 in vitro and increased production of PGE2 in cultured human dermal papilla cells and mouse fibroblasts. Lachgar et al.[50] also found that minoxidil (12 µmol L-1) stimulated PGE2 production by cultured dermal papilla cells, in this case derived from rat vibrissae, as well as production of leukotriene B4. They also found that minoxidil inhibited prostacyclin synthesis by dermal papilla cells (measured as 6-keto-prostaglandin F1), as had an earlier study using bovine endothelial cells.[52] Prostanoids have many biological functions in different tissues, acting through specific G protein-coupled receptors[53] and, in some cases, via nuclear receptors.[54] We do not know whether prostanoids have a physiological role in regulating hair growth, although, latanoprost, a topical synthetic PGF2 analogue used in the treatment of glaucoma, causes hypertrichosis of the eyelashes.[55] Topical treatment with latanoprost also stimulates hair regrowth on the scalp in balding stumptail macaques.[56]

Nuck et al.[57] studied the antiandrogenic potential of minoxidil on androgen-dependent cutaneous structures of the flank organ of female golden Syrian hamsters. Neither 1% nor 5% minoxidil topical solution applied to one flank for 3 weeks prevented the androgen-dependent growth of the pigmented spot, sebaceous glands or hair follicle diameter induced by subcutaneous capsules filled with testosterone. However, significant inhibition was seen following topical application of 5% progesterone. The effect of minoxidil on human hair growth is not confined to androgen-dependent hair follicles and these findings are consistent with the conclusion that minoxidil does not act through androgen pathways. However, Sato et al.[58] reported that minoxidil stimulates 17ß-hydroxysteroid dehydrogenase (17ß-HSD) in cultured human dermal papilla cells and also has a small stimulatory effect on 5-reductase activity. 17ß-HSD catalyses the interconversion of testosterone and androstenedione and may therefore increase or reduce androgen responses. A high concentration of minoxidil (0.5 mmol L-1) was used in this study and the relevance of the results to hair growth in vivo is unclear.

The idea that minoxidil stimulates hair growth by increasing cutaneous blood flow has been the subject of two studies giving contradictory results. Wester et al.[59] studied the effect of topical minoxidil (1%, 3%, 5%) on blood flow in balding scalp using laser Doppler velocimetry (LDV) and photopulse plethymography. Both methods showed an increase in skin blood flow following application of minoxidil that was statistically significant with the 5% solution. On the other hand, Bunker and Dowd,[60] also using LDV, failed to find any change in skin blood flow following application of 3% minoxidil topical solution to the scalp in 10 balding men, whereas all but one showed an increase in blood flow after applying the vasodilator 0.1% hexyl nicotinate. The difference in results may have been due to the higher concentration of minoxidil used in the first study although, as Bunker and Dowd point out, 3% minoxidil topical solution is clinically effective. Sakita et al.[61] studied the effect of minoxidil topical solution on the hair follicle vasculature in the rat using transmission electron microscopy. In minoxidil-treated animals there was no difference in the total area of follicular capillaries compared with controls but there was an increase in capillary fenestrations. The authors suggested that the increase in fenestrations may be due to vascular endothelial growth factor (VEGF) (see below), but the functional significance of this observation was not discussed.

VEGF has a central role in promoting angiogenesis as well as influencing diverse cell functions including cell survival, proliferation and the generation of nitric oxide and prostacyclin.[62] The perifollicular capillary network is coupled to the hair cycle, increasing during anagen and then regressing during catagen and telogen. Yano et al.[63] found that capillary proliferation during anagen was temporally and spatially associated with expression of VEGF in the outer root sheath of murine hair follicles. Transgenic overexpression of VEGF in the outer root sheath increased perifollicular vascularization and led to accelerated hair growth following depilation and the growth of larger hairs. This effect was prevented by systemic administration of a VEGF antibody. Lachgar et al.[64] found that the expression of VEGF mRNA and protein in cultured human dermal papilla cells was stimulated by minoxidil in a dose-dependent fashion. A fivefold increase in VEGF protein occurred in extracts of cells incubated with 12 µmol L-1 minoxidil, and there was a similar increase in mRNA expression. A possible mechanism for minoxidil stimulation of VEGF has been proposed by Li et al. from experiments on cultured dermal papilla cells.[41] They found that adenosine also increases VEGF release and the VEGF response to minoxidil was prevented by pharmacological blockade of A1 and A2 adenosine receptors. mRNAs for the A1, A2A and A2B adenosine receptors, as well as the sulphonylurea receptor SUR2B, were detected by the reverse transcriptase-polymerase chain reaction. The authors suggested that binding of minoxidil to SUR2B promotes secretion of ATP, which is rapidly converted to adenosine and activates adenosine signalling pathways.