e, Bremen, GermanyEpidermal Langerhans cells play an important part in the immune surveillance of the human skin (Schuler et al. 1991;Stingl et al. 1993). So far, beside the fact that multiple studies have focused on the quantification of epidermal Langerhans cells (reviewed inBieber et al. 1988;Breathnach, 1991;Saint-André Marchal et al. 1997), there are varying results concerning the real number and density of Langerhans cells in the human epidermis (Breathnach, 1991). Furthermore, there is only limited information regarding their numerical ratio to other epidermal cells in situ and the percentage of Langerhans cells, which is believed to account for 2–8% of the total epidermal cell population (Holbrook & Wolff, 1993).
Several studies on diseased skin, e.g., in vitiligo (Zelickson & Mottaz, 1968), psoriasis (Bieber et al. 1988;Baadsgaard et al. 1989;Zemelman et al. 1994), contact dermatitis (Kolde & Knop, 1987;Proksch & Brasch, 1997;Teramae et al. 1998;Tsuruta et al. 1999), and atopic dermatitis (Bieber et al. 1988;Bieber & Bruijnzeel, 1990;Yoshida et al. 1997), compared with normal skin, showed different and often contradictory results. A similar situation applies to morphometric time-course studies in vivo and in vitro quantifying the number of Langerhans cells after ultraviolet B radiation (van Praag 1994;Bacci et al. 1998;Okamoto et al. 1999). Several studies have demonstrated the importance of defining Langerhans cells, the selection of the applied enumeration method, and the need for a relatively large sample size in quantifying Langerhans cells (De Jong 1986;Bieber et al. 1988;Emilson & Scheynius, 1995). In a comparative study on six established quantification methods on vertical cryosections of normal and diseased human skin the results differed substantially from one method to another (Bieber et al. 1988).
So far, all morphometric approaches to quantify Langerhans cells were based on standard two-dimensional quantification methods, i.e., counting cells per area unit or per length of basement membrane on two-dimensional vertical sections or optical projections of epidermal sheet preparations using light microscopy (Wolff & Winkelmann, 1967;Bieber et al. 1988;Teramae et al. 1998), electron microscopy (Kolde & Knop, 1986;van Praag et al. 1994), or confocal laser scanning microscopy (Karas et al. 1992;Emilson et al. 1993;Yu et al. 1994;Emilson & Scheynius, 1995;Bacci et al. 1998;Okamoto et al. 1999). For the quantification of cell numbers three-dimensional stereologic quantification methods such as the dissector have been developed, which are not biased by cell size, shape, and distribution (Sterio, 1984). Although the modern dissector method is increasingly used in other fields of bioscience (Mayhew & Gundersen, 1996;Wong et al. 1996) it has so far not been applied on human epidermis. In this study we have combined the disector with the three-dimensional imaging power of the confocal laser scanning microscope (CLSM). Special attention has been paid to the methodologic and technical aspects to adapt the method to the stratified structure of the epidermis taking into account the dendritic shape and the nonrandom distribution of epidermal Langerhans cells.
Materials and methods
Tissue specimens and staining
Skin specimens from breast reduction plastic surgery of eight healthy women (age range 19–37 y) were used after written informed consent was obtained. Three punch biopsies (3 mm) were obtained from each specimen. From each biopsy with a distance of 300 
m eight vertical cryosections (thickness 12 m) were systematically sampled for quantification. All sections were evaluated using a immunofluorescence staining technique. The nuclei were stained with 2 M solution of propidium iodide (Molecular Probes, Leiden, The Netherlands). For the identification of Langerhans cells an indirect immunochemical staining procedure employing anti-CD1a monoclonal staining antibody (isotype IgG1, 
(mouse); Immunotech, Marseille, Cedex9, France) was used.
Confocal laser scanning microscope
The CLSM allows confocal optical sectioning within thick and, therefore, three-dimensional sections. This feature makes the CLSM a particularly suitable imaging device for the dissector method. A BioRad MRC 1000 attached to a Nikon Diaphot 300 microscope, equipped with a krypton-argon laser (Ion Laser Technology, Salt Lake City, UT) was used. Images were taken using a 60-fold oil immersion objective lens (Nikon, numerical aperture 1.4), a filter combination appropriate for the specific visualization of fluorescein isothiocyanate (488 nm excitation, filter 522 DF 32, displayed green) and propidium iodide (568 nm excitation, filter 605 DF 32, displayed red), and a zoom factor of 1-fold. To ensure identical thickness (about 500 nm z resolution) of single optical sections at both excitation wavelengths applied in double-labeled specimens, pinhole setting was 2.8 for the 488 nm and 2.0 for the 568 nm excitation laser lines. Laser intensity and gain were adapted to yield a distinct image of nuclei and Langerhans cells. Slight variation of these parameters does not influence the quantification using the dissector principle. On average a total of 12 fields of vision were sampled on all sections of each punch biopsy. Within each field of vision a set of six parallel optical sections at a distance of 1.5 m was scanned. Optical sections 3 and 5 were used for quantification, and 1, 2, 4, and 6 for unambiguous identification of Langerhans cell nuclei.
The disector principle
An average of 12 fields of vision were systematically sampled on the eight sections of each punch biopsy, thus yielding 36 fields of vision per individual. For an unequivocal identification of the dendritic-shaped Langerhans cells, nuclei were sampled not cell bodies (Sterio, 1984;Gundersen, 1986). To avoid overestimation by multiple counting, adjacent sections were needed to clarify whether two profiles belong to the same nucleus. Furthermore, using serial optical sections, differentiation between Langerhans cells and other epidermal cells surrounded by CD1a-stained dendritic processes was facilitated.
Applying the dissector method (Sterio, 1984) the number of Langerhans cells and other epidermal cell nuclei per volume unit were quantified. The disector is a probe that samples isolated objects, i.e., nuclei, with a uniform probability in three-dimensional space, irrespective of their size and shape. A first optical section, the so-called ''reference section'', was set into the cryosection. Nuclei were sampled within an unbiased counting frame (Gundersen, 1977) with a known area (140 
200 m2 for Langerhans cell nuclei and 140 20 m2 for other epidermal cell nuclei) (Figure 1 and Figure 2). All nuclei sampled in this way were ''looked up'' for in a second parallel section at a distance h of 3 m and those nuclei not sectioned by this ''look-up'' section were counted. (Gundersen, 1986;Gundersen et al. 1988;Howard & Reed, 1998). The disector test system was calibrated using a scan of a standardized ruler at the same magnification and the same zoom factor. The distance h should not exceed &DF;1 4 to &DF;1 3 of the height of the nuclei (Gundersen et al. 1988) so that no nuclei are left undetected between the sections.
Figure 1.
Reference section: section 3 of the set of six parallel optical sections. Unbiased counting frame superimposed: 140 m 200 m area for Langerhans cells, and 140 m 20 m area for other epidermal cells counting. Fully drawn lines are exclusion lines; inclusion lines are dashed. Point grid for the determination of the area covered by propidium iodide stained epidermis. Note the four Langerhans cells transected. Other epidermal cells surrounded by CD1a-reactive dendrites are characterized by a perinuclear rim of unstained cytoplasm (double arrow). Sections 1, 2, 4, and 6 are used for an easier and unambiguous identification of the Langerhans cells nuclei.
Figure 2.
Lookup section: section 5 a known distance h = 3 m apart from section 3. Only the single Langerhans cell marked with an arrow in Figure 1 is counted, because its nucleus is transected by section 3 and not by section 5. The other epidermal cells are counted the same way. If all Langerhans cells in section 3 were counted, their number would be overestimated by 300%.
As the number of other epidermal cells is about 50 times higher than that of Langerhans cells the width of the counting frame for the Langerhans cells (width 200 m) was chosen 10 times larger than the one for the other epidermal cells (width 20 m) (Figure 1 and Figure 2). In addition, the Langerhans cells were counted using the disector bidirectionally (Howard & Reed, 1998), i.e., the first section as the reference section and the second section as the look up plane for the first disector and vice versa for the second dissector.
The number of nuclei in a volume unit, the so-called numerical density N^v was calculated using the following equations:
Q– was the number of counted cells within the total volume V^ of all evaluated disectors of each patient. As the irregularly shaped epidermis did not cover the full area of the counting frame (Figure 1 and Figure 2), the evaluated area of the disector was estimated in the reference section using a point grid:
P was the total number of all cross-points on the epidermis and a/p was the area associated with each test point (20 40 m2).
Cavalieri's principle
To minimize the bias caused by shrinkage, on the same cryosections the reference volume was estimated applying Cavalieri's principle. The epidermal volume of a 3 mm punch was used as a sampling unit, which is defined by its surface and is comparable for its special location on the body surface. Total quantities of Langerhans cells and other epidermal cells within this sampling unit were evaluated. The volume of the epidermis in a 3 mm punch biopsy can be determined if the sum of the areas of equidistant parallel sections is multiplied by the distance T (300 m) between the sections, provided that the first section is placed at random between 0 and 1 T.
Pictures of the propidium iodide and CD1a-stained sections were taken on ordinary slide film. The slides were projected at a final magnification M of 125-fold on to a point grid with an area associated with each test point a/p of 157.5 mm2. The area of the epidermis on each section was estimated by counting the cross-points on the epidermis (Gundersen, 1986;Bauer et al. 1995;Howard & Reed, 1998). Summing up all cross-points 
P the volume of the epidermis in each punch biopsy can be calculated using the following formula:
This principle is unbiased and not influenced by the shape of the epidermis or the direction of the cutting plane. Formulae for estimating CE for the area and volume determination are given elsewhere (Gundersen, 1986;Bauer et al. 1995;Howard & Reed, 1998).
For each patient the total number of Langerhans cells and other epidermal cells N^(cells)epidermis in the average epidermal volume V^¯ (epidermis) of the three biopsies was calculated:
As the surface of a 3 mm punch biopsy is 
r2 before processing the number of cells given is unbiased as cells per mm2 skin surface and the average thickness of the nucleated part of the epidermis can be calculated as the volume divided by the surface of the punch.
Results
Localization of Langerhans cells in the epidermis
One section of each punch biopsy was stained with hematoxylin–eosin to rule out inflammatory reactions or exocytosis. Langerhans cells were usually located in the suprabasal layers, but sometimes they were also found in other nucleated layers of the epidermis. Although their distribution pattern was not quantified, it was obvious that Langerhans cells were not distributed equally throughout the suprabasal layer. Sometimes they were found clustering and sometimes there were apparent Langerhans cell-free areas.
Numerical densities of Langerhans cells and other epidermal cells, epidermal volume, and thickness
The dissector provides an estimate of the numerical density N^v of cells within the volume V of the evaluated disector probes. The number of Langerhans cells was found to be 18,217–27 955 per mm3 (mean 
SD 23,409 3076) and the number of other epidermal cells 1,032,687–1,514,185 per mm3 (mean SD 1232 998 153 573) (Table 1). As numerical densities are biased to an unknown extent by shrinkage of the reference volume V, these values have to be multiplied by the epidermal reference volume to achieve unbiased results.
Table 1 - Average volumes V(epidermal) of the epidermis in a 3 mm punch of the eight patients investigated.
![Table 1 - Average volumes V(epidermal) of the epidermis in a 3[thinsp]mm punch of the eight patients investigateda - Unfortunately we are unable to provide accessible alternative text for this. If you require assistance to access this image, please contact help@nature.com or the author](/common/images/table_thumb.gif)
Full table The average volume of epidermis V^¯ (epidermis) after preparation for the three punch biopsies of each patient ranged from 0.269 to 0.550 mm3 (mean SD 0.427 0.107) (Table 1). The coefficient of error for area determination CE(A) using a point grid was between 1.80 and 4.87% (mean 2.55%) and the coefficient of error for the volume determination CE(V) from a set of systematic sampled sections was between 3.85 and 10.77% (mean 5.97%). These results with a total CE for the volume estimation of less than 10% in most cases show the high accuracy of the method. The thickness of the nucleated part of the epidermis ranged from 38.03 to 77.78 m (mean SD 60.34 15.06).
Total number of epidermal Langerhans cells and other epidermal cells per punch biopsy
By multiplying the numerical density of Langerhans cells and other epidermal cells with the average volume of the epidermis the number of Langerhans cells and other epidermal cells per 3 mm punch biopsy can be calculated without the influence of shrinkage (Eqn 4). The number of Langerhans cells for each patient was calculated between 6447 and 12,765 (mean SD 9855 2268) for a 3 mm punch of mammary skin and the number of other epidermal cells was between 334,454 and 739,289 (mean SD 522,733 137,314) (Table 1).
Total number of epidermal Langerhans cells and other epidermal cells per mm2 of skin surface
As the surface of a 3 mm punch is 1.5 mm2, the number of cells can also be given as cells per mm2 of skin surface. The number of Langerhans cells per mm2 was between 912 and 1806 (mean SD 1394 321) The number of other epidermal cells per mm2 showed the same high variation as well with values between 47,315 and 104,588 (mean SD 73,952 19,426) (Table 1). The average area covered by one Langerhans cells ranged from 554 to 1096 m2.
Ratio of other epidermal cells to Langerhans cells in human skin
In spite of the high variation of cells per mm2 their ratio appears to be rather constant for mammary skin with 47.92–58.97 other epidermal cells per one Langerhans cell (mean SD 52.88 4.32). The coefficient of correlation (0.952) for this linear relation was highly significant (p < 0.0001, 95% confidence interval 0.721–0.990) (Table 1, Figure 3).
Figure 3.
Number of other epidermal cells and corresponding number of Langerhans cells per mm2 of skin surface for the eight investigated patients with a coefficient of correlation of 0.952. Linear regression showing a fixed ratio of 1–52.88.
Full figure and legend (12K)Discussion
This study investigated the number of epidermal Langerhans cells and their numerical ratio to other epidermal cells using a three-dimensional stereologic method (the CLSM disector method). It is widely accepted that Langerhans cells form a reticular trap within the epidermis where they take up and process antigens (Schuler et al. 1991;Stingl et al. 1993). Therefore, it is of interest to specify their number and distribution in the epidermis. Previous studies, however, have shown a high interindividual and intraindividual variation of the skin surface density of Langerhans cells ranging from 460 to 1170 per mm2 (Berman et al. 1983;Chen et al. 1985;Breathnach, 1991). This previously found high interindividual variation was confirmed by our study, showing a variation of Langerhans cells per mm2 of breast skin from 912 to 1805. This high variability seems to arise by the true biologic interindividual variation. Obviously, this true biologic variability cannot be influenced methodologically or by larger sample sizes. Yet, we found the numerical ratio of Langerhans cells to other epidermal cells to be rather constant: one Langerhans cell per 53 other epidermal cells, i.e., Langerhans cells represent 1.86% of all epidermal cells. As the percentage of Langerhans cells in human epidermis seems to be fixed, the high variation of their surface density can be explained by the varying thickness of the epidermis (Table 1). Previous studies have only investigated the number of Langerhans cells per skin surface area overlooking this biologic decisive relationship. In our study the number of other epidermal cells showed a high variability (47,000–105,000 per mm2), which is in the same range as an earlier paper reporting the surface density of epidermal cells to be approximately 50,000 per mm2 (Bergstresser et al. 1978). As the keratinocyte is the numerically dominating cell type within the epidermis, this ratio most likely represents the ratio between Langerhans cells and keratinocytes. This numerical ratio proved to be strikingly constant between individuals, in spite of the varying epidermal thickness. It could be speculated that this ratio is regulated to a rated value.
Previously, quantification of epidermal Langerhans cells has been performed applying two-dimensional methods either on epidermal sheet preparations (Wolff & Winkelmann, 1967;Berman et al. 1983;Emilson & Scheynius, 1995;Saint-André Marchal et al. 1997;Bacci et al. 1998) or on vertical sections (De Jong et al. 1986;Bieber et al. 1988;Emilson et al. 1993;Van Praag et al. 1994;Zemelman et al. 1994;Emilson & Scheynius, 1995). Methods using epidermal sheets for Langerhans cell counting offer a better identification of the completely embedded and untransected Langerhans cells. Yet, by using epidermal sheets the results cannot be related to morphologic parameters such as papillomatosis, the number of keratinocytes, and especially the epidermal thickness; however, the epidermal thickness seems to be a very important factor as indicated by the strong correlation of Langerhans cells to other epidermal cells in our study. Methods applied in previous studies using vertical sections were based on two-dimensional quantification. Bias can be introduced in several ways: (i) by the fact that bigger cells have a higher probability of being sectioned than smaller ones, i.e., the probability being sectioned is proportional to the cell size (Gundersen et al. 1988); (ii) by the fact that a varying section thickness influences the number of visible cells, the so-called Holmes effect (Gundersen, 1986); and (iii) by the so-called ''edge-effect'', i.e., by including all ''edging'' cell profiles in a two-dimensional counting frame, so that an overestimation of their number might occur (Gundersen, 1986). In addition, the two-dimensional quantification of Langerhans cells on vertical sections is complicated by the ambiguous assignment of the sectioned dendrites to individual Langerhans cells (De Jong et al. 1986;Bieber et al. 1988). Both methods have in common that they are influenced to an unknown extent by the shrinkage of cells and reference volume (Braendgaard & Gundersen, 1986;Howard & Reed, 1998).
The CLSM is known to be a powerful tool for three-dimensional quantitative analyses. It has been used to describe the three-dimensional morphologic characteristics of single Langerhans cells (Karas et al. 1992). Recently, it has been suggested that the problems in Langerhans cell quantification can be overcome using the CLSM (Emilson & Scheynius, 1995). As in their study the cells were counted on two-dimensional optical sections of epidermal sheets the same problems as mentioned above occur (Yu et al. 1994;Bacci et al. 1998). In other studies cell volume densities were quantified performing an extended focus projection of three-dimensional data obtained with a CLSM (Emilson et al. 1993;Emilson & Scheynius, 1995). An extended focus projection is the sum of several parallel optical sections resulting in a two-dimensional image where all optical sections are in focus. On these two-dimensional images cell volume densities and not cell numbers were quantified; however, the cell volume density does not give information about the real cell number, as it is influenced by cell volume, shrinkage, or swelling.
The problem of shrinkage affects many quantification methods, especially those that enumerate cell densities, i.e., cells per volume unit, per surface area, or per length of basement membrane. It is due to different preparation, varying morphologic techniques, and uneven hydration of tissue resulting in a varying amount of shrinkage of cells and their reference volume (Braendgaard & Gundersen, 1986;Howard & Reed, 1998). In our study this problem was solved by estimating the reference volume using Cavalieri's principle, where the volume of the dissector probe and the total epidermal volume of a 3 mm punch were both measured after undergoing the same shrinkage (see Eqn 4). The epidermal volume within a 3 mm punch seems to be a suitable sampling unit, as it is clearly defined by the diameter of a circular area in vivo.
Another method used to quantify Langerhans cells is flow cytometry. Flow cytometric investigations found epidermal Langerhans cells to constitute 2.5% of all epidermal cells in forearm skin (Ashworth et al. 1989); however, the results of flow cytometry are biased by the proteolysis of target phenotypes during tissue disaggregation (Barrett et al. 1995). Other sources of error are the tendency of Langerhans cells to adhere to keratinocytes (Ashworth et al. 1989), the occurrence of cell debris due to dermoepidermal separation procedures, and the formation of clumps (Glade et al. 1996). Furthermore, there is a complete loss of tissue morphology. Thus, findings cannot be related to particular layers of the epidermis or to regional morphologic changes. Compared with flow cytometry, the major advantage of the CLSM dissector is that cells can be quantified and characterized within their morphologic and functional environment with high efficiency.
In conclusion, by applying the three-dimensional dissector method we could clearly determine that a constant ratio of epidermal Langerhans cells to other epidermal cells in healthy skin is present for a given region of the body, i.e., for mammary skin. The findings corroborate the hypothesis that a regulative interaction and a spatial relationship between Langerhans cells and other epidermal cells exists (Potten & Allen, 1976;Tang et al. 1993). This network of Langerhans cells seems to be designed to keep an economic balance to the other constituents of the epidermis. Our results further suggest that one epidermal Langerhans cell might claim a certain territory of the epidermis that is patrolled; penetrating chemical or biologic substances in this defined area are trapped and processed by the assigned Langerhans cells. Furthermore, parallel to the epidermal melanin unit (Fitzpatrick & Breathnach, 1963), our results support the existence of an ''epidermal Langerhans cell unit'' (Wolff, 1972), where one Langerhans cell keeps a territory with 53 epidermal cells under immune surveillance, and possibly also interacts with keratinocytes in an as yet unknown fashion. In future studies the combination of the three-dimensional dissector method with modern immunologic techniques might provide a new tool to gain further insight into regulatory mechanisms and new information regarding different physiologic and pathophysiologic conditions known to affect Langerhans cells. For example, recent studies suggest that in tumor tissue and in regional lymph nodes the presence and density of antigen-presenting cells such as dendritic cells (Toriyama et al. 1993;Lotze & Jaffe, 1999) and Langerhans cells play an important part in host defense mechanisms against tumor growth, extension, and the establishment of metastasis. A statistically significant improved prognosis has been reported for higher numbers of tumor-associated dendritic cells for several tumor types (Lotze & Jaffe, 1999). An unbiased and reproducible morphometric method such as the CLSM dissector therefore opens new possibilities in the interpretation of their special functional impact on the prognosis of malignant diseases.
References
| 1. | Ashworth J, Kahan MC & Breathnach SM. Flow cytometric analysis and sorting of HLA-DR+ and CD I+ Langerhans cells. Br J Dermatol (1989) 121: 11–18. | PubMed | ISI | ChemPort | |
| 2. | Baadsgaard O, Gupta AK, Taylor RS, Ellis CN, Voorhees JJ & Cooper KD. Psoriatic epidermal cells demonstrate increased numbers and function of non-Langerhans antigen-presenting cells. J Invest Dermatol (1989) 92: 190–195. | Article | PubMed | ISI | ChemPort | |
| 3. | Bacci S, Romagnoli P & Streilein JW. Reduction in number and morphologic alteration of Langerhans cells after UVB radiation in vivo are accompanied by an influx of monocytoid cells into the epidermis. J Invest Dermatol (1998) 111: 1134–1138 10.1046/j.1523-1747.1998.00406.x. | Article | PubMed | ISI | ChemPort | |
| 4. | Barrett AW, Ross DA & Goodacre JA. Limitaions of flow cytometry in the analysis of CD Ia/HLA DR+ human oral mucosal Langerhans cells. Oral Dis (1995) 1: 49–53. | PubMed | ChemPort | |
| 5. | Bauer J, Gries W & Bahmer FA. Volume estimation of multicellular colon carcinoma spheroids using Cavalieri's principle. Pathol Res Pract (1995) 191: 1192–1197. | PubMed | ISI | ChemPort | |
| 6. | Bergstresser PR, Pariser RJ & Taylor JR. Counting and sizing of epidermal cells in normal human skin. J Invest Dermatol (1978) 70: 280–284. | Article | PubMed | ISI | ChemPort | |
| 7. | Berman B, Chen VL, France DS, Dotz WI & Petroni G. Anatomical mapping of epidermal Langerhans cell densities in adults. Br J Dermatol (1983) 109: 553–558. | PubMed | ISI | ChemPort | |
| 8. | Bieber T & Bruijnzeel C. Langerhans cells in the physiopathology of atopic dermatitis. Ann Dermatol Venereol (1990) 117: 1985–1193. |
| 9. | Bieber T, Ring J & Braun-Falco O. Comparison of different methods for enumeration of Langerhans cells in vertical cryosections of human skin. Br J Dermatol (1988) 118: 385–392. | PubMed | ISI | ChemPort | |
| 10. | Braendgaard H & Gundersen HJG. The impact of recent stereological advances on quantitative studies of the nervous system. J Neurosci Methods (1986) 18: 39–78. | PubMed | ISI | ChemPort | |
| 11. | Breathnach SM. Origin, cell lineage, ontogeny, tissue distribution, and kinetics of langerhans cells. In: Schuler G (ed)Epidermal Langerhans Cells (1991) Boca Raton: CRC Press pp 23–48. |
| 12. | Chen H, Yuan J, Wang Y & Silvers WK. Distribution of ATPase-positive Langerhans cells in normal adult human skin. Br J Dermatol (1985) 113: 707–711. | PubMed | ISI | ChemPort | |
| 13. | De Jong MCJM. Blanken R, Nanninga J, van Voorst Vader PC, Popperna S. Defined in situ enumeration of T6 and HLA-DR expressing epidermal Langerhans cells: Morphologic and methodologic aspects. J Invest Dermatol (1986) 87: 698–702. | Article | PubMed | ChemPort | |
| 14. | Emilson A, Lindberg M, Forslind B & Scheynius A. Quantitative and 3-dimensional analysis of Langerhans'cells following occlusion with patch test using confocal laser scanning microscopy. Acta Derm Venereol (1993) 73: 323–329. | PubMed | ISI | ChemPort | |
| 15. | Emilson A & Scheynius A. Quantitative and three-dimensional analysis of human Langerhans cells in epidermal sheets and vertical skin sections. J Histochem Cytochem (1995) 43: 993–998. | PubMed | ISI | ChemPort | |
| 16. | Fitzpatrick TB & Breathnach AS. Das epidermale Melanin-Einheit-System. Dtsch Med Wochenschr (1963) 147: 481. | ChemPort | |
| 17. | Glade CP, Seegers B, Meulen EFJ, Van Hooijdonk CAEM, Van Erp PEJ & van de Kerkhof PCM. Multiparameter flow cytometric characterization of epidermal cell suspension prepared from normal and hyperproliferative human skin using an optimized thermolysin-trypsin protocol. Arch Dermatol Res (1996) 288: 203–210 10.1007/s004030050047. | Article | PubMed | ISI | ChemPort | |
| 18. | Gundersen HJG. Notes on the estimation of the numerical density of arbitrary profiles: The edge effect. J Microsc (1977) 111: 219–223. | ISI | |
| 19. | Gundersen HJG. Stereology of arbitrary particles. A review if unbiased number and size estimators and the presentation of some new ones, in memory of William R. Thompson. J Microsc (1986) 143: 3–45. | PubMed | ISI | |
| 20. | Gundersen HJG, Bagger P & Bendtsen TF et al. The new stereological tools: Dissector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis. APMIS (1988) 96: 857–881. | PubMed | ISI | ChemPort | |
| 21. | Holbrook KA & Wolff K. The structure and development of skin. In: Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF, (eds)Dermatology in General Medicine (1993) 4th edn New York: McGraw-Hill pp 97–145. |
| 22. | Howard CV & Reed MG. Unbiased stereology. Three-Dimensional Measurement in Microscopy (1998) Oxford. BIOS Scientific Publishers Inc.. |
| 23. | Karas Z, Warchol JB & Jaroszewski J. Three-dimensional reconstruction and stereometric analysis of langerhans cells in mouse epidermis. J Invest Dermatol (1992) 99: 774–778. | Article | PubMed | ISI | ChemPort | |
| 24. | Kolde G & Knop J. Different cellular reaction patterns of epidermal Langerhans cells after application of contact sensitizing, toxic and tolerogenic compounds. A comparative ultrastructural and morphometric time-course analysis. J Invest Dermatol (1987) 89: 19–23. | Article | PubMed | ISI | ChemPort | |
| 25. | Kolde G & Knop J. Ultrastructural morphometry of epidermal Langerhans cells: introduction of a simple method for a comprehensive quantitative analysis of the cells. Arch Dermatol Res (1986) 278: 298–301. | Article | PubMed | ISI | ChemPort | |
| 26. | Lotze MT & Jaffe R. Cancer. Lotze MT, Thomson AW (eds)Dendritic Cells, Biology and Clinical Applications (1999) San Diego: Academic Press pp 325–338. |
| 27. | Mayhew TM & Gundersen HJG. If you assume, you can make an ass out of u and me: a decade of the disector for stereological counting of particles in 3D space. J Anat (1996) 188: 1–15. | PubMed | ISI | |
| 28. | Okamoto H, Mizuno K, Itoh T, Tanaka K & Horio T. Evaluation of apoptotic cells induced by ultraviolet light B radiation in epidermal sheets stained by TUNEL technique. J Invest Dermatol (1999) 113: 802–807 10.1046/j.1523-1747.1999.00757.x. | Article | PubMed | ISI | ChemPort | |
| 29. | Potten CS & Allen TD A. model implicating the Langerhans cell in keratinocyte proliferation control. Differentiation (1976) 5: 43–47. | PubMed | ISI | ChemPort | |
| 30. | van Praag MC. Mulder AA, Claas FH, Vermeer BJ, Mommaas AM. Long-term ultraviolet B-induced impairment of Langerhans cell function: an immunelectron microscopic study. Clin Exp Immunol (1994) 95: 73–77. | PubMed | ChemPort | |
| 31. | Proksch E & Brasch J. Influence of epidermal permeability barrier disruption and Langerhans' cell density on allergic contact dermatitis. Acta Derm Venereol (1997) 77: 102–104. | PubMed | ISI | ChemPort | |
| 32. | Saint-André Marchal I, Dezutter-Dambuyant C & Martin J et al. Quantitative assessment of feline epidermal Langerhans cells. Br J Dermatol (1997) 136: 961–965. | PubMed | |
| 33. | Schuler G, Romani N, Stössel H & Wolff K. Structural organization and biological properties of Langerhans cells. In: Schuler G (ed)Epidermal Langerhans Cells (1991) Boca Raton: CRC Press pp 87–135. |
| 34. | Sterio DC. The unbiased estimation of number and sizes of arbitrary particles using the disector. J Microsc (1984) 134: 127–136. | PubMed | ISI | |
| 35. | Stingl G, HauSeries C & Wolff K. The epidermis: An immunologic microenvironment. In: Fitzpatrick TB, Eisen AZ, Wolff K, Freedberg IM, Austen KF, edsDermatology in General Medicine (1993) 4th edn New York: McGraw-Hill pp 172–197. |
| 36. | Tang A, Amagai M, Granger LG, Stanley JR & Udey MC. Adhesion of epidermal Langerhans cells to keratinocytes mediated by E-cadherin. Nature (1993) 361: 82–85. | Article | PubMed | ISI | ChemPort | |
| 37. | Teramae H, Kaneda K, Tsuruta D, Ishii M & Hamada T. Morphometric and ultrastructural analyses of in vivo-activated murine Langerhans cells induced by administration of a streptococcal preparation (OK-432). Arch Dermatol Res (1998) 290: 533–539 10.1007/s004030050348. | Article | PubMed | ISI | ChemPort | |
| 38. | Toriyama K, Wen DR, Paul E & Cochran AJ. Variations in the distribution, frequency, and phenotype of Langerhans cells during the evolution of malignant melanoma of the skin. J Invest Dermatol (1993) 100: 269S–273S. | Article | PubMed | ChemPort | |
| 39. | Tsuruta D, Kaneda K, Teramae H & Ishii M. In vivo activation of Langerhans cells and dendritic epidermal cells in the elicitation phase of murine contact hypersensitivity. Br J Dermatol (1999) 140: 392–399 10.1046/j.1365-2133.1999.02698.x. | Article | PubMed | ISI | ChemPort | |
| 40. | Wolff K. The Langerhans cell. Curr Probl Dermatol (1972) 4: 79. |
| 41. | Wolff K & Winkelmann RK. Quantitative studies on the Langerhans cell population of guinea pig epidermis. J Invest Dermatol (1967) 48: 504–513. | PubMed | ISI | ChemPort | |
| 42. | Wong M, Wuethrich P, Eggli P & Hunziker E. Zone-specific cell biosynthetic activity in mature bovine articular cartilage: a new method using confocal microscopic stereology and quantitative autoradiography. J Orthop Res (1996) 14: 424–432. | Article | PubMed | ISI | ChemPort | |
| 43. | Yoshida A, Imayama S, Sugai S, Kawano Y & Ishibashi T. Increased number of IgE positive Langerhans cells in the conjunctiva of patients with atopic dermatitis. Br J Ophthalmol (1997) 81: 402–406. | PubMed | ISI | ChemPort | |
| 44. | Yu RC, Abrams DC, Alaibac M & Chu AC. Morphological and quantitative analyses of normal epidermal Langerhans cells using confocal scanning laser microscopy. Br J Dermatol (1994) 131: 843–848. | PubMed | ISI | ChemPort | |
| 45. | Zelickson AS & Mottaz JH. Epidermal dendritic cells. Arch Dermatol (1968) 98: 652–659. | Article | PubMed | ISI | ChemPort | |
| 46. | Zemelman V, Van Neer F, Roberts N, Patel P, Langtry J & Staughton RC. Epidermal Langerhans cells, HIV-1 infection and psoriasis. Br J Dermatol (1994) 130: 307–311. | PubMed | ISI | ChemPort | |
Acknowledgments
We thank Dr. Martin, Department of Plastic Surgery, Kliniken Dr. Erler, Nuremberg, Germany and PD Dr. Siebzehnrübel of the Department of Gynecology, University of Erlangen for their collaboration.
