AmorePacific R&D Center, Yongin-si, Korea
Role of androgen in mesenchymal epithelial interactions in human hair follicle.
J Investig Dermatol Symp Proc. 2005 Dec;10(3):209-11.
There are many more steroidogenic enzymes involved in the onset and development of AGAJ Invest Dermatol. 2005 Apr;124(4):675-85.
Circulating MMP9, vitamin D and variation in the TIMP-1 response with VDR genotype: mechanisms for inflammatory damage in chronic disorders?
The most important study on hair loss listed to date:
| Androgenetic alopecia and microinflammation | |
Yann F. Mahé,PhD, Jean-François Michelet,MSc, Nelly Billoni, MSc, Françoise Jarrousse, BTS , Bruno Buan, BTS , Stephane Commo, BTS, Didier Saint-Léger,PhD and Bruno A. Bernard,PhD
International Journal of Dermatology |
|
|
| |
Today, androgenetic alopecia (AGA) is considered to be an alteration of hair growth and/or a premature aging of the pilosebaceous unit with a multifactorial and even polygenic etiology. 1 The fact that the success rate of treatment with either antihypertensive agents, or modulators of androgen metabolism, barely exceeds 30% means that other pathways may be envisioned. The implication of various activators of inflammation in the etiology of AGA has progressively and recently emerged from several independent studies. 2,3,4,5,6,7,8,9,10,11 A fibroplasia of the dermal sheath, which surrounds the hair follicle, is now suspected to be a common terminal process resulting in the miniaturization and involution of the pilosebaceous unit in AGA. 2 8 We review here several observations underlining the possible implication of a slow, silent, and painless process in AGA. Because we think that it should not be confused with a classical inflammatory process, we have called it microinflammation. An early study referred to an inflammatory infiltrate of mononuclear cells and lymphocytes in about 50% of the scalp samples studied. 2 Another more recent study by Jaworsky et al . 3 confirmed an inflammatory infiltrate of activated T cells and macrophages in the upper third of the hair follicles from transitional regions of alopecia (i.e. which are characterized by actively progressing alopecia). This study also reported the occurrence of a developing fibrosis of the perifollicular sheath, together with the degranulation of follicular adventitial mast cells. The miniaturization of the hair follicles was found to be associated with a deposit of so-called "collagen or connective tissue streamers" beneath the follicle, 2,7 as well as a 2 2.5 times enlargement of the follicular dermal sheath composed of densely packed collagen bundles. 3 This thickening of the dermal sheath in progression zones of AGA has also recently been observed in our laboratory using immunohistochemical staining ( Fig. 1 ). Horizontal section studies of scalp biopsies indicate that the so-called perifollicular fibrosis is generally mild, consisting of loose, concentric layers of fibrotic collagen that must be distinguished from cicatricial alopecia. 4 It is unclear whether or not the fibrosis seen in follicular streamers (stelae or fibrous tracts) is permanent and/or alters the downgrowth of anagen hair follicles. Only 55% of male pattern AGA patients with microinflammation had hair regrowth in response to minoxidil treatment, which was less than the 77% of patients with no signs of inflammation, 4 suggesting that, to some extent, perifollicular microinflammation may account for some cases of male pattern AGA which do not respond to minoxidil. 4 Another study on 412 patients (193 men and 219 women) confirmed the presence of a significant degree of inflammation and fibrosis in at least 37% of AGA cases. 5 The upper location of the infiltrate near the infrainfundibulum 2 7 clearly distinguishes AGA from alopecia areata (AA), the latter disease being characterized by infiltrates in the bulb and dermal papilla zone. 12 The aim of this review is to determine the location and chronology of the microinflammation process within the complex pathophysiology of the human pilosebaceous unit in order to improve the possible approaches for the reduction or prevention of the development of AGA. |
|
| Classically, an inflammatory process is ascribed to a central major mediator or pathway. Such a monofactorial vision has been historically well exemplified by the famous interleukin-1 (IL-1) scheme developed by Oppenheim et al . 13 which is still valid even after 13 years. In fact, many inflammatory agents are at the center of a huge array of effects, involving cells, enzymes, adhesion molecules and other biological mechanisms. The identification of the effects of isolated factors is only part of the problem; it may be more important to determine when and where the individual factors are involved in the complex sequence. This pathway has been clearly identified, and several inhibitory anti-inflammatory drugs acting on this aspect of inflammation have been developed and clinically evaluated. 14,15 , 16,17 |
The cytokine/chemokine side of microinflammation
Why
does microinflammation take place in the
pilosebaceous unit and for what benefit
and purpose? Fig. 2
and Fig. 3
show, in a simplified sequence, that
inflammation is a multistep process which
may start from a primary event. Let us look
at the clues at the "crime scene" of AGA:
we observe a perifollicular infiltrate in
the upper follicle near the infundibulum.
2
7 This suggests that the primary causal
event for the triggering of inflammation
might occur near the infundibulum. 3,7
Supporting this point of view, improvement
of the inflammatory aspect of AGA has been
reported in a pilot study with an antimicrobial
lotion.
7
One could speculate that several inhabitants
of the scalp, such as the "triad" ( Propionibacterium
sp.; Staphylococcus sp.;
Malassezia ovalis ) or other members
of the transient flora, could be involved
in this complex inflammatory process. 7
The presence of
porphyrins (produced by Propionibacterium
sp.) in the pilosebaceous duct of
58% of AGA patients (compared with 12% of
control subjects), which are able
to induce the production of complement (C5)
chemotactic factor, is considered to be
a possible cofactor of this initial pro-inflammatory
stress. 6,7
Keratinocytes
are also known to respond within minutes
to chemical stress, pollutants, UV irradiation
or even mechanical stress. 37
Not only are radical oxygen species,
38
NO, 39
PGs, and histamine 40
produced, but also intracellularly stored
IL-1 is released 37,41
(see Fig. 2
and step 1 of Fig. 3
). By itself, this pro-inflammatory
cytokine (as well as IL-1 which binds to
the same receptor) is able to inhibit the
growth of isolated hair follicles in culture
in vitro. 9
11 This concentration-dependent inhibition
of human hair elongation and survival indicates
a high sensitivity to IL-1 of the isolated
organ in culture in vitro (IC
50 = 10 pg/mL 11
). In vivo , transgenic mice
which overexpress IL-1 in the basal
epidermis and in the outer root sheath of
their pelage hair follicles exhibit a spontaneous
cutaneous phenotype characterized by a sparseness
of hair. 42
As a response to an IL-1 signal, adjacent
keratinocytes which express receptors for
IL-1 start to engage the transcription of
IL-1 responsive genes 41
( Fig. 3
, step 2). In
vitro , following IL-1 stimulation,
this transcriptional activation cascade
is induced within 6 h in plucked human hair
follicles. 11
Alternatively, skin keratinocytes, which
may also have antigen presenting capabilities,
could theoretically induce T-cell proliferation
in response to bacterial antigens. 51 These
antigens, once they have been "tagged,"
are then selectively destroyed by infiltrating
macrophages, Langerhans cells, or natural
killer cells. 50,52 On many occasions, however,
the causal agent persists, resulting in
sustained inflammation ( Fig. 3, step 4).
This corresponds partly to the situation
which has been pictured in the progression
zone of roughly one-third of alopecia cases:
infiltrating T lymphocytes, together with
mastocytes and macrophages, located in the
upper perifollicular adventitial dermal
sheath perpetuate a local inflammatory stage.
27 This phase of inflammation
often results in tissue remodeling, where
collagenases, such as matrix metalloproteinase
(MMP)-9 (transcriptionally activated by
pro-inflammatory cytokines) or MMP-8 (directly
produced by infiltrating cells), may play
an active role. 5355 Thus, collagenases
are suspected to contribute to the tissue
changes and the so-called "perifollicular
fibrosis" by "preparing"
tissue matrix and basal membranes for macrophages
and T-cell adhesion. Accordingly,
this scenario facilitates the secretion
of membrane-anchored cytokines, such as
TNF-. 55 Other factors, such as MCP-1, have
been directly suspected to contribute to
organ fibrosis in an experimental model
of renal inflammation. 56 As MCP-1, together
with other chemokines, was found to be expressed
in human hair follicles in vitro, 11 as
well as in the eccrine ducts of sebaceous
glands in vivo, 57 it might also be actively
involved in the progression of perifollicular
fibrosis detected in AGA. 26 The development
of perifollicular fibrosis might thus appear
as the signature of a disequilibrium between
pro-and anti-inflammatory pathways.
Relations
between inflammation and steroidogenesis:
the missing link
There is no question that androgens are
major modulators of hair loss. Recently,
it was shown that testosterone inhibited
the growth of outer root sheath keratinocytes
only when they were cocultured with dermal
papilla cells derived from the bald scalp
of an adult macaque, 58 reinforcing the
hypothesis of an androgen influence on hair
growth via the dermal papilla. 59
The potent metabolite of testosterone (i.e.
5-dihydrotestosterone, 5-DHT) is considered
as a "culprit". 60 5-DHT is generated
from testosterone through the activity of
5-reductase (5-R). Two active isoforms of
5-R, which differ both in tissue site distribution
as well as in optimal pH for enzymatic activity,
have been identified and cloned. 61,62,
63 While the type II isoform is considered
to be the major isozyme in genital tissues,
61 the type I isoform is considered to be
the major isoform expressed in skin and
in the pilosebaceous unit. 64,65 Isoform
II, however, has recently been detected
in the inner root sheath of the pilosebaceous
unit by immunohistochemistry, 66,67 Northern
blotting, 67 and the pH dependence of optimal
enzymatic activity. 67 Thus, the contribution
of both isoforms in the regression of the
pilosebaceous unit is still a matter of
debate. Recently, a clinical study using
finasteride, a strong inhibitor of 5-RII
(and weak inhibitor of 5-RI), showed that
intervention in androgen metabolism could,
to some extent, modulate the progression
of AGA, when the drug was given by the oral
route, 68 but not topically. 69 After oral
ingestion, an improvement of hair growth
was observed, which was associated with
a drastic reduction of serum levels of 5-DHT,
corresponding to those observed in castrates.
68 Despite such a
reduction of circulating 5-DHT levels, however,
a number of individuals (60-70%) still remained
unresponsive to this treatment, indicating
again that simple dysregulation of 5-DHT
synthesis levels or a genetic polymorphism
of 5-R genes cannot account for all cases
of AGA, and a polygenic etiology should
be considered. 1
Thus,
to date, the only evident link that can
be established between androgen metabolism
and the complex inflammatory process is
sebum production which is controlled by
androgens.
70 As sebum harbors a large amount of microorganisms
which use lipids as nutrients, 8 it cannot
be excluded that, at least for some individuals,
androgen metabolism might facilitate the
colonization of the sebaceous infundibulum
and sebaceous ducts by such microorganisms
which may be involved in the first steps
of pilosebaceous unit inflammation.
We propose here working
hypotheses which do not invalidate the contribution
of a hereditary genetic androgen imbalance
in AGA, 60 but rather attempt to integrate
the neglected microinflammatory aspects
of alopecia into the complex etiology of
AGA. On the one hand, excessive local
and/or endocrine, genetically exacerbated
5-DHT synthesis results in sebaceous
gland enlargement; 2,60 as a consequence,
some scalps might
offer more comfortable niches to harbor
the previously mentioned pro-inflammatory
microorganisms. 6,7 On the other
hand, androgen imbalance and metabolism
may be locally exacerbated by pro-inflammatory
cytokines. For example, gingival fibroblasts
have been reported to modify their androgen
metabolism through the action of several
growth factors, such as epidermal growth
factor (EGF), transforming growth factor
beta (TGF-), and the pro-inflammatory cytokines
IL-1 and TNF-. 71 Therefore, one could speculate
that, once the inflammatory
process has been triggered, the androgenetic
mechanism of alopecia could subsequently
be locally amplified. This upregulation
of androgen metabolism by pro-inflammatory
cytokines remains, however, to be established
at the pilosebaceous unit level.
| Our visit to the "crime scene" of AGA yielded many clues ( Fig. 4 ). We know now that, at least in about one-third of cases, the tool which causes the lethal damage is a microinflammatory process. Several factors are present, however, which are suspected to have handled the tool: androgens, microbial flora, endogenous or exogenous stress, genetic imbalance, and possibly others. Although other suspects or tools are likely to be discovered in the future, it cannot be excluded that, for each individual, the causal agent, as well as the sequence of events or combined factors, may be different. The large number of molecules claimed to be active and patented in this field, 89 and their limited efficacy in offering a definite and extensive cure of AGA, confirm that the mechanism of AGA is highly complex. Accordingly, it appears that, due to the complexity and multiple interactivities and cooperations involved throughout the distinct inflammatory pathways (partly described in Fig. 2 ), an anti-inflammatory strategy should be targeted to the appropriate effector(s) at the right moment. For this purpose, we have developed a simple assay to evaluate individuals with potentially affected hair follicles. 11 We observed that plucked hair specimens of 33% of the 116 volunteers evaluated could be classified as highly inflammatory in terms of spontaneous IL-1 production. 11 Consequently, the identification of the "inflammatory alopecic individuals" may help to adapt the right answer to the right cause. Such a selective approach might be valuable for other parameters, such as an imbalance in 11 HSD activity, 5-DHT synthesis, or microorganism colonization. Encompassing individual diversity is thus a prerequisite for appropriately addressing the biological conditions contributing to AGA. Our findings and a review of the literature suggest that inflammation in its diversity is a potentially active player to consider in this approach. |