ARTICLE Three
types of moncocyte/macrophages can be recognized in the normal human dermis:
perivascular macrophages, intervascular or connective tissue macrophages,
and flat-shaped macrophages that line the epidermis or adnexal structures
[1]. The third type of macrophages, which lie directly beneath the dermo-epidermal
junction and are aligned horizontally along the epidermal axis, are documented
by various terms, such as dermal dendrocytes [2], lining cells [3] and epithelium-lining
macrophages [4]. The term "dermal dendrocyte" was first introduced by Headington
[5] in 1986 to describe polydendritic cells in normal human dermis previously
dismissed as bipolar dermal fibroblasts. Dermal dendrocytes express coagulation
factor XIIIa, an intracellular form of fibrin stabilizing factor [2]. Dermal
dendrocytes are concentrated in three defined dermal strata; i) directly
beneath the dermo-epidermal junction as mentioned above (subepidermal dendrocytes);
ii) around superficial dermal venules (perivascular dendrocytes), and iii)
diffusely within the deeper dermis (reticular dermal dendrocytes) [2]. Subepidermal
dendrocytes express factor XIIIa but not hematopoietic progenitor antigen
CD34, while perivascular dendrocytes express both factor XIIIa and CD34
[2]. Immunoelectron microscopy has shown that factor XIIIa and CD34 are
not specific for dermal dendrocytes, but are present on heterogeneous cell
types such as dermal perivascular dendritic macrophages and phagocytic macrophages
[2]. CD34 also labeled the inner surface of vascular endothelial cells [2].
Inflammatory dermatoses such as atopic eczema and psoriasis reveal an increase
in the number of dermal dendrocytes [6]. The terms "epithelium-lining macrophages"
and "lining cells" were employed for the flattened cells forming an almost
continuous single-cell row at the dermo-epidermal junction in the psoriatic
skin lesion [3, 4]. Epithelium-lining macrophages strongly express CD14
and CD11a, but variably express factor XIIIa [4]. Although these cells are
inconspicuous in normal skin, their number is increased in psoriasis, covering
about 80% of the dermo-epidermal junction [3, 4]. Our recent anatomical
study on haired guinea pig skin versus hairless guinea pig skin revealed
that the number of dendritic/spindle-shaped dermal interstitial cells was
significantly greater in the latter [7]. This data raises the possibility
that the large number of dermal interstitial cells, such as macrophages,
dermal dendrocytes and fibroblasts in hairless animals, may supplement the
defensive function of the hairs against the mechanical and chemical injuries.
The purpose of this study was to determine the number and distribution of
these interstitial cells that line the epidermis of normal human skin from
various anatomical locations, including haired and glabrous skin.
Materials and methods
Samples
For morphometric analysis in light microscopy 5 to 7 normal human skin
specimens per each anatomical site were collected from surgical specimens,
mostly with pigmented nevi. Ten anatomical locations were classified;
palm, sole, buttock, posterior trunk, anterior trunk, limb, neck, face,
lip and scalp. In total, 65 normal skin specimens were evaluated. The
total number of patients was 65. Hence, one specimen was obtained from
each patient. For immunohistochemistry, 5 each of normal sole and buttock
skin specimens (total 10) were selected from the 65 specimens. For electron
microscopy, four normal buttock skin specimens were obtained from four
additional male volunteers (29.4 + 1.2 years) after informed consent.
Morphometric analysis
The number of interstitial cells lining the epidermis in each x 100
microscopic field of 65 haematoxylin- and eosin- stained (H&E), 4
mum sections was quantified, using a computer-assisted image analyzer
(Olympus-Avio SP 500). The long axis of the nucleus (x), and maximum perpendicular
axis (y) were generated, and the y/x ratio was calculated. Interstitial
cells containing nuclei with a y/x ratio of < 0.5 were classified as
spindle-shaped cells, while those with a ratio of >= 0.5 were classified
as oval cells [8]. Over 2 mm of dermo-epidermal junction was evaluated
for each specimen. Cells lining follicular epithelium were not counted.
In normal skin from the palm and the sole (n = 10), the density of spindle-shaped
cells located beneath the apices of rete ridges was compared with that
of those along the dermal papillae.
Immunohistochemistry
The specimens were routinely processed, embedded in paraffin, and cut
into 4 mum sections. The sections were then deparaffinized, and rehydrated
in a graded ethanol series and finally in PBS. Prior to staining for FXIIIa,
sections were incubated with 0.1% trypsin in PBS for 10 min at 37°
C. Anti-human FXIIIa rabbit antiserum (Carbiochem Corp, diluted in 1:160)
or anti-human CD34 mouse monoclonal antibody (Becton Dickinson, diluted
1:20) was applied for 1 hr at room temperature in a moist chamber. Nonimmune
rabbit IgG(1:400) or mouse IgG (1:20) served as a negative control. For
the second layer, biotinylated species-specific IgG (1:200) was applied
for 30 min. After blocking endogenous peroxidase activity, specimens were
incubated with avidin-biotin peroxidase complex (ABC) for 30 min.
Electron Microscopy
Four normal buttock skin specimens were fixed overnight at 4° C
in 2.5% (v/v) glutaraldehyde, 2.0% (v/v) paraformaldehyde with 0.06% (w/v)
calcium chloride in 0.1 M sodium cacodylate buffer (pH 7.3). After rinsing
in cacodylate buffer, specimens were post-fixed in 2% (v/v) osmium tetroxide
for 2 hrs, dehydrated in a graded ethanol series, followed by propylene
oxide, and then embedded in Epon 812. Ultrathin sections, cut with a Porter-Blum
MT2B ultramicrotome, were stained with uranyl acetate and lead citrate,
and observed with a Hitachi H-7000 electron microscope.
Statistics
The data were subjected to statistical analysis by either Student's
t test or Mann-Whitney non-parametric test. The differences were considered
significant at p < 0.01.
Results
Histologically, two types of interstitial cells were recognized beneath
the dermo-epidermal junction of normal human skin: cells containing oval
nuclei and spindle-shaped cells containing elongated nuclei (Fig.
1).
Both oval cells and spindle-shaped cells beneath the interfollicular
epidermis expressed factor XIIIa, but not CD34. Factor XIIIa+
spindle-shaped cells were distributed predominantly beneath the apices
of rete ridges, while dendritic cells containing oval nuclei were predominant
in the dermal papillae (Fig.
2). A few spindle-shaped cells immediately beneath the acrosyringium
and around the intradermal eccrine ducts extending from the bases of rete
ridges, expressed CD34 (Fig.
3).
Ultrastructurally, oval cells in light microscopy had several dendritic
cytoplasmic processes and contained oval nuclei. Spindle-shaped cells
in light microscopy displayed thin (approximately 100 nm), elongated cytoplasmic
processes that parallelled the dermo-epidermal junction and extended up
to 40 mum into the dermal matrix. Both types of cells contained characteristic,
electron-dense, plasma-membrane associated plaques, i.e. fibronexus
[2] (Fig. 4).
When compared with the number of oval cells in the anterior trunk (mean
8.16 ± SE 0.57/mm), no significant difference was determined for
any anatomical location (Fig.
5A). In contrast, apart from the lip, spindle-shaped cells were focally
located beneath the dermo-epidermal junction in normal skin samples from
various anatomical sites. When compared with the number of spindle-shaped
cells in the anterior trunk (mean 0.60 ± SE 0.22/mm), significantly
greater numbers were determined in the palm (4.11 ± 1.24; p <
0.01), sole (3.52 ± 0.83; p < 0.001) and buttock (2.52 ±
0.49; p < 0.01). The numbers of spindle-shaped cells in the scalp,
face, neck, limb and posterior trunk were similar to that in the anterior
trunk (Fig. 5B). In normal
skin from the palm and the sole, the number of cells containing oval nuclei
between the apices of rete ridges and dermal papillae was not significantly
different (Fig. 5C). However,
the number of spindle-shaped cells located beneath the apices of rete
ridges (mean 7.35 + SE 1.56) was significantly greater than that along
the dermal papillae (1.39 + 0.39, p < 0.01; Fig.
5D).
Discussion
The present study demonstrated that the number of spindle-shaped cells
was greater beneath the apices of rete ridges than in the dermal papillae.
Moreover, regional variations were observed in the number of spindle-shaped
cells beneath the epidermis. The number of spindle-shaped cells was greatest
in the palm and sole, where the number of eccrine sweat glands was also
greater than at other locations. Spindle-shaped cells associated with
acrosyringium were CD34-positive, whereas those beneath the epidermis
were CD-34 negative. Hence, it is unlikely that spindle-shaped cells beneath
the epidermis are related to eccrine sweat apparatus. In psoriasis, epithelium-lining
macrophages have been found only in the region of hyperplastic rete ridges,
and not in the adjacent uninvolved skin, and they may therefore be involved
in the regulation of epidermal growth [4]. This process is controlled
by various cytokines such as TNF-alpha, IL-1 and IL-6 produced by fibroblasts
and mononuclear inflammatory cells [9]. Dermal dendrocytes in psoriatic
skin have been shown to produce TNF-alpha, which would play a role in
epidermal growth, the trafficking of T cells into the dermis, and their
activation [10]. The apices of rete ridges in normal human skin from the
palm and the sole are well-developed and wider than those in other locations.
Although the histogenesis of hyperplastic epidermis differs between psoriatic
skin and normal sole skin [11], the greater number of spindle-shaped cells
beneath the dermo-epidermal junction in the normal skin of palms and soles
may correlate with their wider rete ridges.
In the present study we used the term "interstitial cells lining the
epidermis", because our quantitative analysis was done using only routine
histology. Based on the small number of samples for immunohistochemistry
and electron microscopy, most of the spindle-shaped cells and oval cells
lining the epidermis shown by H&E staining must be subepidermal dendrocytes
and macrophages.
Van Den Oord et al. [4] concluded that epithelium-lining macrophages
differed from dermal dendrocytes in that the majority of cells lacked
factor XIIIa-antigen. However, the expression of factor XIIIa in subepidermal
dendrocytes varies depending on the technique, and it could be enhanced
by the addition of TNF-alpha or mast cell secretagogues [2]. Although
subepidermal dendrocytes appear oval or spindle-shaped under light microscopy,
ultrastructurally they display multiple (more than two) thin, elongated
cytoplasmic processes, which are found to be thin, membrane-bound flaps
aligned parallel to the dermo-epidermal junction in a three-dimensional
reconstruction [12]. Another ultrastructural marker of subepidermal dendrocytes
is the fibronexus, which appears as an electron-dense, plasma-membrane-associated
plaque ranging from 200 to 500 nm in diameter [2, 13]. The fibronexus
is typically expressed by cells that require binding interactions between
cytoskeletal actin and extracellular fibronectin to maintain constant
spatial relationships or to promote contractile functions [13]. The fibronexus
serves to promote a fixed relationship between subepidermal dendrocytes
and the dermo-epidermal junction that may represent their functional target
[2]. In contrast, the ultrastructure of epithelium-lining macrophages
has been poorly characterized. In an ultrastructural study of psoriasis
vulgaris, epithelium-lining macrophages consisted of cells showing the
ultrastructural characteristics of subepidermal dendrocytes and cells
showing those of macrophages [14]. These two types of cells are intimately
associated and distributed in a line. Such an intimate spatial relationship
between subepidermal dendrocytes and macrophages has been also documented
in normal human skin [2, 15]. Considered together, under light microscopy,
epithelial-lining macrophages or lining cells represent a mixture of subepidermal
dendrocytes and elongated macrophages. We propose that use of the term
"epithelial-lining macrophages" and "lining cells" should be restricted
to psoriasis where subepidermal dendrocytes and elongated macrophages
form an almost continuous single-cell row beneath the dermo-epidermal
junction.
The functional role of subepidermal dendrocytes and elongated macrophages
beneath the epidermis remains largely unknown. The potential function
of factor XIIIa-producing subepidermal dendrocytes is that of fibrin and
fibronectin stabilization along the base of newly formed cutaneous blisters,
sealing off the eroded skin surface [2]. Subepidermal dendrocytes and
macrophages could be involved in antigen presentation and the activation
of T lymphocytes in immunological skin diseases, such as graft-versus-host
disease [16], psoriasis [10] and atopic eczema [6]. Follicular infundibula
contain a great number of Langerhans cells and indeterminate cells, which
could be sequestered from harmful environmental stresses such as ultraviolet
light, and serve as reservoirs of antigen-presenting cells [17]. Moreover,
a number of dermal dendrocytes expressing human progenitor cell-antigen
CD34 are embedded in the fibrotic follicular sheath [18]. When profound
abrogation of Langerhans cell number and function occurs, the main source
of replenishment is likely to be follicular infundibulum and the papillary
dermis [17, 19]. The large number of epithelium-lining macrophages in
palm and sole skin could be related to the lack of hair follicles in these
locations.
CONCLUSION
In summary, the present study has demonstrated that the number of spindle-shaped
cells, quantified by H& E staining, is significantly greater beneath
the apices of rete ridges than in dermal papillae. The number is greater
in glabrous skin, such as palm and sole, compared with samples of normal
human skin from other locations.
Acknowledgments
The authors thank Dr. Toshifumi Nakagawa, M.D., Kagawa, Japan, for his
helpful comments on the ultrastructure of epithelium-lining macrophages
in psoriasis.
Article accepted on 12/11 /01
REFERENCES {
1. Weber-Matthiesen K, Sterry W. Organization of the monocyte/macrophage
system of normal human skin. J Invest Dermatol 1990; 95: 83-9.
2. Sueki, H, Whitaker D, Buchsbaum M, Murphy GF. Novel interactions
between dermal dendrocytes and mast cells in human skin. Implications
for matrix repair. Lab Invest 1993; 69: 160-72.
3. Boehncke, WH, Wortmann S, Kaufmann R, Mielke V, Sterry W.
A subset of macrophages located along the basement membrane ("lining cells")
is a characteristic histological feature of psoriasis. Am J Dermatopathol
1995; 17: 139-44.
4. Van den Oord JJ, De Wolf-Peeters C. Epithelium-lining macrophages
in psoriasis. Br J Dermatol 1994; 130: 589-94.
5. Headington JT. The dermal dendrocyte. In: Callen JP, Dahl
MV, Goltz LE, Rasmussen JE, Stegman SJ, eds. Advances in dermatology
vol.1. Illinois: Year Book Medical Publishers, 1986: 159-71.
6. Cerio R, Griffiths CEM, Cooper KD, Nickoloff BJ, Headington
JT. Characterization of factor XIIIa positive dermal dendritic cells in
normal and inflamed skin. Br J Dermatol 1989; 121: 421-31.
7. Sueki H, El Gammal C, Kudoh K, Kligman AM. Hairless guinea
pig skin: anatomical basis for studies of cutaneous biology. Eur J
Dermatol 2000; 10: 357-64.
8. Monteiro MR, Murphy EE, Galaria NA, Whitaker-Menezes D, Murphy
GF. Cytological alterations in dermal dendrocytes in vitro: evidence
for transformation to a non-dendritic phenotype. Br J Dermatol
2000; 143: 84-90.
9. Castells-Rodellas A, Castell JV, Ramirez-Bosca A, Nicolas
JF, Valcuende-Cavero F, Thivolet J. Interleukin-6 in normal skin and psoriasis.
Acta Derm Venereol (Stockh) 1992; 72: 165-8.
10. Nickoloff BJ, Karabin GD, Barker JNWN, Griffiths CEM, Sarma
V, Mitra RS, Elder JT, Kunkel SL, Dixit VM. Cellular localization of interleukin-8
and its inducer, tumor necrosis factor-alpha in psoriasis. Am J Pathol
1991; 138: 129-40.
11. Iizuka H, Ishida-Yamamoto A, Honda H. Epidermal remodelling
in psoriasis. Br J Dermatol 1996; 135: 433-8.
12. Sueki H, Telegan B, Murphy GF. Computer-assisted three-dimensional
reconstruction of human dermal dendrocytes. J Invest Dermatol 1995;
105: 704-8.
13. Singer II, Kawka DW, Kazazis DM, Clark RAF. In vivo
co-distribution of fibronectin and actin fibers in granulation tissue:
immunofluorescence and electron microscope studies of the fibronexus at
the myofibroblast surface. J Cell Biol 1984; 98: 2091-106.
14. Nakashima K, Masada M, Shibazaki Y, Fukuda J, Nakagawa T,
Takaiwa T. Light and electron microscopic observations of lining macrophages
in psoriasis. Proceedings of the 12th Annual Meeting of the Japanese Society
for Psoriasis Research 1997; 156-7.
15. Sueki H, Iijima M. Dermal dendritic cell: its past, present,
and future. Dendritic Cells 1999; 9: 1-6.
16. Yoo YH, Park BS, Whitaker-Menezes D, Korngold R, Murphy GF.
Dermal dendrocytes participate in the cellular pathology of experimental
acute graft-versus-host disease. J Cutan Pathol 1998; 25:
426-34.
17. Gilliam AC, Kremer IB, Yoshida Y, Stevens SR, Tootell E,
Teunissen MBM, Hammerberg C, Cooper KD. The human hair follicle: a reservoir
of CD40+ B7-deficient Langerhans cells that repopulate epidermis
after UVB exposure. J Invest Dermatol 1998; 110: 422-7.
18. Nickoloff BJ. The human progenitor cell antigen (CD34) is
localized on endothelial cells, dermal dendritic cells and perifollicular
cells in formalin-fixed normal skin and on proliferating endothelial cells
and stromal spindle-shaped cells in Kaposi's sarcoma. Arch Dermatol
1991; 127: 523-9.
19. Murphy GF, Katz S, Kligman AM. Topical tretinoin replenishes
CD1a-positive epidermal Langerhans cells in chronically photodamaged human
skin. J Cutan Pathol 1998; 25: 30-4.
|
Version imprimable
