- Email: [email protected]
T-cell subsets and langerhans cells in wart lesions
Immunology Letters. 6 (1983) 191 196 Elsevier BiomedicalPress
T-CELL SUBSETS AND LANGERHANS
CELLS IN WART LESIONS
Yvette C H A R D O N N E T , Paule BEAUVE, Jacqueline VIAC and Daniel S C H M I T T INSERM U 209, CNRS ERA 788, Clinique Dermatologique (Pavilion R), H6pital Edouard Herriot, 69374 Lyon Cedex 2, France
(Received 18 January 1983) (Accepted 8 February 1983)
1. Summary The presence of T-lymphocytes and Langerhans cells was assessed on tissue sections of 28 human warts from various localizations and in normal epidermis by indirect immunofluorescence using specific monoclonal antibodies. In most cases, the density of Langerhans cells was dramatically reduced in wart epidermis as compared to normal epidermis and a mild cellular infiltrate could be noted. In some lesions, OKT4 and OKT8 cell populations were present and the number of Langerhans cells was high both in dermis and epidermis. These data are compatible with the role of Langerhans cells and the existence of a local cellular immune reaction in human papilloma virus-infected tissues. No correlation could yet be established with the virus type on the skin localization.
macrophages and Langerhans cells (LC), indicative of enhanced cellular activity [5,6]. In the present study, we examined wart sections for the presence of infiltrating cells, T-lymphocytes and LC in correlation with viral antigen, in 28 human papillomas of various localizations, with the use of indirect immunofiuorescence and monoclonal antibodies directed against T-cell populations and LC [9-11]. Such an immunohistochemical study has not been done so far.
3. Materials and methods 3.1. Wart specimens
Twenty-eight human warts from various localizations (Table 1) removed surgically were frozen in liquid nitrogen and stored at -20 ° C. They were taken from patients that were 12 to 40 years old. Biopsies of normal skin were used as controls.
2. Introduction Skin tumors are characterized by the presence of mononuclear cell infiltrates indicative of host reaction against the tumour. The grade of infiltrate is related to the tumour regression [1,2]. Cellular immune response is suspected to play an important role during wart evolution [1,3-6]. In regressing plane warts, histological changes consisted in a mild perivascular infiltration by mononuclear cells in the upper dermis and within the epidermis [4,7,8]. Ultrastructural changes were reported in
3.2. Antibodies Monoclonal antibodies directed against various human T-cell antigens were used: OKT3, OKT4 and OKT8 from Ortho Pharmaceuticals, U.S.A. [9] and B16 from Dr. J. Brochier, Lyon, France [10,11], respectively specific for circulating, helper/inducer, suppressor/cytotoxic T-cells and LC. Viral antigen was detected with a human papilloma virus (H PV) antiserum raised in guinea pig [12] against SDS-dissociated purified virus (d-HPV). Reference d-HPV antiserum prepared in rabbit was also used (kindly provided by Dr. G. Orth, Paris, France).
Key words: human papillomavirus(HPV) - warts - monoclonal antibodies T-cells Langerhanscells- immunofluorescence
0165 2478/83/0000-0000/$3.00 © 1983ElsevierScience Publishers B.V.
191
Table 1 Phenotype of T-lymphocytes and LC observed in warts Wart samples Localization
Monoclonal antibodies Case no.
OKT3
OKT4
OKT8
Antiserum d-HPV
Hand
1 2 3 4 5 6 7
+
+
-
+
(+)
++ (+) +
++ (+) +
(+) + +
I
++ +
(+)
+++ ++ ++
+
1 1 1* 2 2 2*
8
(+)
+
-
2
9 10" 11 12 13
+
+ +
+ +
2 2 2 (2) 2
(+)
(+)
-
-
+ (+)
+ -
+ + + +
-
+
+
(+)
-
+
+ (+) + -
-
+
(+) (+)
+ + (+)
+ +
(+) ++
(+) ++
(+) +
++ +
2* 2
(+) (+)
+ + (+)
+ ++ ++
(2) (2) (2)
15 16 17 18
(+)
Wrist Elbow
19 20
+
Dorsum foot
21 22 23
(+)
(+)
24 25
(+)
+
26 27 28
(+) + (+)
Genitals
+
-
14
Face
E -
-
HPV type (histology)
BI6 D
Deep plantar
Cellular infiltration
Normal epidermis
-
(+)
(+)
+ + +
-
+
2
+ (+) (+)
+ +
+ ++ ++ ++
2 2 2 2
++ +
2 2
+ + +
2 2 2
++
Monoclonal antibodies: OKT3, specific for circulating lymphocytes; OKT4, specific for helper/inducer T-cells; OKT8, specific for suppressor/ cytotoxic T-cells; B16, specific for Langerhans cells (D: dermis; E: epidermis), d-HPV: antiserum raised to SDS dissociated purified HPV. intensity of fluorescence: (+), few cells are slightly positive; +, few cells are intensely positive; ++, many cells are highly fluorescent; , sections are completely negative. Intensity of infiltrate: (+), few mononuclear cells; +, small infiltrate; ++, mild infiltrate; + + + , strong inflammatory reaction. *The type of virus was also determined by DN A cleavage pattern with Hind 11 + Hind I11 restriction enzymes and the Southern hybridization technique under stringe conditions. (2) = not certain.
3.3.
Indirect irnmunofluorescence technique
Five micron frozen sections of warts and normal epidermis were prepared,
a i r - d r i e d a n d f i x e d in c o l d
a c e t o n e f o r 10 r a i n a t 4 ° C, A d j a c e n t 192
sections were
used for precise comparative
analysis. The slides
were incubated
f o r 45 m i n a t r o o m t e m p e r a t u r e
the appropriate
dilution of antibodies: mouse mo-
noclonal antibodies
1:5 a n d d - H P V
antiserum
with (gui-
nea pig 1:10, rabbit 1:100). After washing in phosphate-buffered saline (PBS) pH 7.2, the appropriate dilution of FITC conjugates was added: goat antimouse from Meloy at 1:20, goat anti-guinea pig and goat anti-rabbit from Nordic at 1:30. Following 30 min incubation at room temperature, the slides were washed in PBS, mounted with polyvinyl alcohol medium [13] and examined with a Zeiss microscope. 3.4. Morphological studies Hematoxylin eosin-stained sections were examined to ascertain the histological wart criterias, the presence of infiltrate, the cytopathic effect of virus in upper epidermal cell layers which leads to the determination of HPV type [14,15]. 3.5. Biochemical determination of the virus type in warts In a few cases, viral DNA was extracted by the Hirt technique [16]. DNA was digested for 1 h at 37°C with a mixture of restriction enzymes (Hind II + Hind III; Boehringer, Mannheim, France) under conditions described by Orth et al. [14] and Favre et al. [17]. Fragments of DNA migrated in 1% agar gels. Molecular weights of the fragments were determined on ethidium bromide-stained gels under UV transillumination. Phage DNA digested with Hind III endonuclease was used as marker. HPV types were determined after nitrocellulose transfer of fragments and Southern blot hybridization with [32p]labeled probes of HPV1 and HPV2.
4. Results
Each of the 28 warts under study (Table 1) showed a typical histology. In many cases, epidermis was disorganized as compared to normal skin biopsies. Twenty-two warts out of 28 showed a mononuclear inflammatory infiltrate. In most cases (11/22) the infiltrate was mild. In the 11 others, the strongest reaction was seen as shown in Fig. I A. OKT3-positive cells were found in 9 specimens. OKT4-positive cells were detected in 3 cases (nos. 4, 18, 27). OKT8-positive cells (Fig. 1B) were present in 7 biopsies (nos. 4, 10, 23, 25-28). No infiltration and no OKT-positive cells were de-
tected in normal epidermis. With BI6, 7/26 specimens were negative in epidermis (6 of them showed an infiltrate and 2 contained a number of LC in the dermis). Among 19 other biopsies, the number of positive LC and the fluorescence intensity were highly variable. In most cases (11/14), positive cells were present both in dermis and epidermis (Fig. 1D). Conversely normal skin biopsies contained LC only in epidermis (Fig. 1C). Viral antigen was detected in some keratinocytes of the epidermal upper layer of 21 warts. Most of them (19/28) showed a typical histology of HPV2, and 4 of them an HPVI histology. In 3 cases, the type of virus was confirmed by biochemical technique (no. 4: HPVIa; nos. 7 and 24: HPV2). One hand wart (no. 4) presented an intense inflammatory reaction. It contained viral antigen and was typed as HPVIa. Positive fluorescent cells were found in the dermis with OKT4 and OKT8. The number of LC increased highly in the epidermis as compared to normal epidermis. The dermis also contained positive B16 cells.
5. Discussion
This preliminary study showed changes in LC distribution of wart epidermis as compared to normal skin. In a few specimens, T-lymphocytes were present without significant predominance of a subset. With the classical histological technique, the infiltrate was commonly observed in plane warts [8] and more rarely in common warts [18,19]. In our series, most of warts showed an infiltrate. Surprisingly, in a number of warts, the density of LC in the epidermis was extremely reduced and in some cases, they were not detected, but the viral antigen was often present. These results indicate that HPV infection induces either disappearance of LC or changes of their antigenic membrane properties. It cannot yet be related to a special localization of the lesion or to a type of virus. However, the duration of most lesions in a patient could be an important factor. On the contrary, in a few biopsies, the density of LC was much higher than those in normal epidermis. Moreover, these cells were also present in the 193
A
C Fig. 1. Monocellular infiltrate and Langerhans cells in a biopsy of hand warts. A: histology of monocellular infiltration (Hematoxylin staining; X 1200). B: OKT8-positive cells of infiltrate in immunofluorescence (X 1200). C: LC cells of normal epidermis recognized by B 16 monoclonal antibody in immunofluorescence 0 < 1200). (Dotted white line: dermis-epidermis junction.) D: LC cells in dermis and epidermis of warts (X 1200). 194
upper dermis. Such an increased number of LC could suggest an activation of these cells in some lesions. Ultrastructural changes of LC in the epidermis were reported during spontaneous involution of plane warts [6]. LC showed signs of enhanced cellular activity with an increase number of LC granules. Similarly, a b n o r m a l LC were found by electron microscopy in the epidermis at the sites of vaccinia virus inoculation [20]. These cells contained an unusually increased number of LC granules. In n o n viral skin tumours, the number of L C always increased irrespective of their benign or malignant nature [2,21] as demonstrated by electron microscopy and immunofluorescence. Further studies are needed to clarify the role of the virus in the interactions between LC and keratinocytes in warts and to elucidate changes of antigenic membrane properties. Infection of LC with H P V could explain their lack of reactivity but remains to be demonstrated. Such a study could be investigated by electron microscopy. In a similar study on vaccinia virus, the particles were detected inside LC of epidermis at the inoculation sites [20]. The number of T-lymphocytes is low in wart lesions, as compared to benign and malignant skin tumours. Both T-cell subsets were present with no significant predominance. A high number of OKT8positive cells were found in benign tumours whereas in highly malignant tumours, it was reduced [2]. Our data suggest that either depletion or stimulation of T-cell subsets and of LC in warts might be related to the evolution of viral infection. It is well known that skin-mediated sensitization to different antigens is dependent on the density of LC [22,23]. Depletion of epidermal LC usually leads to host tolerance of these antigens [24-26]. Decreased density of epidermal LC and tolerance to H P V might be responsible for the development of lesions. On the contrary, a strong infiltrate and a high density of LC in the epidermis, but also in dermis, are suggestive of cell-mediated immune reaction directed towards the tumour. This could favour the hypothesis that T-cells and LC participate in local wart rejection [27]. Further studies of these cells in clinically regressing warts from various localizations will give more conclusive information.
Acknowledgements This work was done with grants from Fbd~ration Nationale des Centres de Lutte contre le Cancer 1981 1982. The authors are indebted to Prof. J. Thivolet and the clinicians of Clinique Dermatologique, Pav. R, H6pital E. Herriot, Lyon, who collected the biopsies, and to G. Orth, M. Favre and O. Croissant (Institut Pasteur, Paris) for determination of H P V type.
References [1] Viac, J., Bustamante, R. and Thivolet, J. (1977) Br. J. Dermatol. 97, 1 10. [2] Fernandez-Bussy,R., Cambazard, F., Mauduit, G., Schmitt, D. and Thivolet, J. (1983) Europ. J. Cancer, in press. [3] Morison,W. L. (1974) Br. J. Dermatol. 90, 531 534. [4] Tagami, H., Takigawa, M., Ogino, A., Imamura, S. and Ofuji, S. (1977) Arch. Dermatol. 113, 1209-1213. [5] Oguchi, M., Komura, J., Tagami, H. and Ofuji, S. (1981) Arch. Dermatol. Res. 270, 403-411. [6] Oguchi, M., Komura, J., Tagami, H. and Ofuji, S. (1981) Arch. Dermatol. Res. 271, 55-61. [7] Berman, A. and Winkelmann, R. K. (1977) Arch. Dermatol. 113, 1219-1221. [8] Tagami, H., Ogino, A., Takigawa, M., Imamura, S. and Ofuji, S. (1974) Br. J. Dermatol. 90, 147 154. [9] Kung, P. C., Goldstein, C., Reinherz, E. L. and Schlossman, S. F. (1979) Science206, 337-340. [I0] Yonish-Rouach, E,, Viac, J., Schmitt, D., Cordier, G., Thivolet, 3. and Brochier,J. (1982) in: Current Concepts in Human Immunology and Cancer Immunomodulation (Serrou, B., Rosenfeld, C. and Daniels, J., eds.) pp. 55-62, Elsevier/ North-Holland, Amsterdam. [I 1] Schmitt, D., Souteyrand, P., Brochier,J., Czernielewski,J. and Thivolet, J. (1982) Acta Dermatovener. (Stockholm) 62, 193-199. [12] Jenson, A. B., Rosenthal, J. R., Olson, C., Pass, F., Lancaster, W. D. and Shah, K. (1980) J. Nat. Cancer Inst. 64, 495 500. [13] Rodriguez, J., Deinhardt, F. (1960) Virology 12, 316-317. [14] Orth, G., Jablonska, S., Favre, M., Croissant, O., Obalek, S., Jarzabek-Chorzelska, M. and Jibard, N. (1981) J. Invest. Dermatol. 76, 97-102. [15] Gross,G., Pfister, H., Hagedorn, M. and Gissman, L. (1982) J. Invest. Dermatol. 78, 160-164. [16] Hirt, B. (1967) J. Mol. Biol. 26, 365 369. [17] Favre, M., Jibard, N. and Orth, G. (1982) Virology 119, 298-309. [18] Matthews, R. S. and Shirodaria, P. V. (1973) Lancet 1, 689 691. [19] Brodersen, 1. and Genner, J. (1973) Acta Dermatovener. (Stockholm) 53, 461-464. 195
[20] Nagao, S., lnaba, S. and lijima, S. (1976) Arch. Dermatol. Res. 256, 23-31. [21] Lisi, P. 0973) Acta Dermatovener. (Stockholm) 53, 425-428. [22] Horio, T. and Okamoto, H. (1982) J. Invest. Dermatol. 78, 402-405. [23] Miyachi, Y. and Takigawa, M. 0982) J. Invest. Dermatol. 78,363-365.
196
[24] Silberberg, I. (1973) Acta Dermatovener. (Stockholm) 53, 1-12. [25] Silberherg-Sinakin, I. and Thorbecke, G. J. (1980) J. Invest. DermatoL 75, 61 67. [26] Shelley, W. B. and Juhlin, L. (1976) Nature (London) 261, 46-47. [27] Thivolet, J., Viac, J. and Staquet, M. J. 0982) Int. J. Dermatol. 21, 94 98
T-CELL SUBSETS AND LANGERHANS
CELLS IN WART LESIONS
Yvette C H A R D O N N E T , Paule BEAUVE, Jacqueline VIAC and Daniel S C H M I T T INSERM U 209, CNRS ERA 788, Clinique Dermatologique (Pavilion R), H6pital Edouard Herriot, 69374 Lyon Cedex 2, France
(Received 18 January 1983) (Accepted 8 February 1983)
1. Summary The presence of T-lymphocytes and Langerhans cells was assessed on tissue sections of 28 human warts from various localizations and in normal epidermis by indirect immunofluorescence using specific monoclonal antibodies. In most cases, the density of Langerhans cells was dramatically reduced in wart epidermis as compared to normal epidermis and a mild cellular infiltrate could be noted. In some lesions, OKT4 and OKT8 cell populations were present and the number of Langerhans cells was high both in dermis and epidermis. These data are compatible with the role of Langerhans cells and the existence of a local cellular immune reaction in human papilloma virus-infected tissues. No correlation could yet be established with the virus type on the skin localization.
macrophages and Langerhans cells (LC), indicative of enhanced cellular activity [5,6]. In the present study, we examined wart sections for the presence of infiltrating cells, T-lymphocytes and LC in correlation with viral antigen, in 28 human papillomas of various localizations, with the use of indirect immunofiuorescence and monoclonal antibodies directed against T-cell populations and LC [9-11]. Such an immunohistochemical study has not been done so far.
3. Materials and methods 3.1. Wart specimens
Twenty-eight human warts from various localizations (Table 1) removed surgically were frozen in liquid nitrogen and stored at -20 ° C. They were taken from patients that were 12 to 40 years old. Biopsies of normal skin were used as controls.
2. Introduction Skin tumors are characterized by the presence of mononuclear cell infiltrates indicative of host reaction against the tumour. The grade of infiltrate is related to the tumour regression [1,2]. Cellular immune response is suspected to play an important role during wart evolution [1,3-6]. In regressing plane warts, histological changes consisted in a mild perivascular infiltration by mononuclear cells in the upper dermis and within the epidermis [4,7,8]. Ultrastructural changes were reported in
3.2. Antibodies Monoclonal antibodies directed against various human T-cell antigens were used: OKT3, OKT4 and OKT8 from Ortho Pharmaceuticals, U.S.A. [9] and B16 from Dr. J. Brochier, Lyon, France [10,11], respectively specific for circulating, helper/inducer, suppressor/cytotoxic T-cells and LC. Viral antigen was detected with a human papilloma virus (H PV) antiserum raised in guinea pig [12] against SDS-dissociated purified virus (d-HPV). Reference d-HPV antiserum prepared in rabbit was also used (kindly provided by Dr. G. Orth, Paris, France).
Key words: human papillomavirus(HPV) - warts - monoclonal antibodies T-cells Langerhanscells- immunofluorescence
0165 2478/83/0000-0000/$3.00 © 1983ElsevierScience Publishers B.V.
191
Table 1 Phenotype of T-lymphocytes and LC observed in warts Wart samples Localization
Monoclonal antibodies Case no.
OKT3
OKT4
OKT8
Antiserum d-HPV
Hand
1 2 3 4 5 6 7
+
+
-
+
(+)
++ (+) +
++ (+) +
(+) + +
I
++ +
(+)
+++ ++ ++
+
1 1 1* 2 2 2*
8
(+)
+
-
2
9 10" 11 12 13
+
+ +
+ +
2 2 2 (2) 2
(+)
(+)
-
-
+ (+)
+ -
+ + + +
-
+
+
(+)
-
+
+ (+) + -
-
+
(+) (+)
+ + (+)
+ +
(+) ++
(+) ++
(+) +
++ +
2* 2
(+) (+)
+ + (+)
+ ++ ++
(2) (2) (2)
15 16 17 18
(+)
Wrist Elbow
19 20
+
Dorsum foot
21 22 23
(+)
(+)
24 25
(+)
+
26 27 28
(+) + (+)
Genitals
+
-
14
Face
E -
-
HPV type (histology)
BI6 D
Deep plantar
Cellular infiltration
Normal epidermis
-
(+)
(+)
+ + +
-
+
2
+ (+) (+)
+ +
+ ++ ++ ++
2 2 2 2
++ +
2 2
+ + +
2 2 2
++
Monoclonal antibodies: OKT3, specific for circulating lymphocytes; OKT4, specific for helper/inducer T-cells; OKT8, specific for suppressor/ cytotoxic T-cells; B16, specific for Langerhans cells (D: dermis; E: epidermis), d-HPV: antiserum raised to SDS dissociated purified HPV. intensity of fluorescence: (+), few cells are slightly positive; +, few cells are intensely positive; ++, many cells are highly fluorescent; , sections are completely negative. Intensity of infiltrate: (+), few mononuclear cells; +, small infiltrate; ++, mild infiltrate; + + + , strong inflammatory reaction. *The type of virus was also determined by DN A cleavage pattern with Hind 11 + Hind I11 restriction enzymes and the Southern hybridization technique under stringe conditions. (2) = not certain.
3.3.
Indirect irnmunofluorescence technique
Five micron frozen sections of warts and normal epidermis were prepared,
a i r - d r i e d a n d f i x e d in c o l d
a c e t o n e f o r 10 r a i n a t 4 ° C, A d j a c e n t 192
sections were
used for precise comparative
analysis. The slides
were incubated
f o r 45 m i n a t r o o m t e m p e r a t u r e
the appropriate
dilution of antibodies: mouse mo-
noclonal antibodies
1:5 a n d d - H P V
antiserum
with (gui-
nea pig 1:10, rabbit 1:100). After washing in phosphate-buffered saline (PBS) pH 7.2, the appropriate dilution of FITC conjugates was added: goat antimouse from Meloy at 1:20, goat anti-guinea pig and goat anti-rabbit from Nordic at 1:30. Following 30 min incubation at room temperature, the slides were washed in PBS, mounted with polyvinyl alcohol medium [13] and examined with a Zeiss microscope. 3.4. Morphological studies Hematoxylin eosin-stained sections were examined to ascertain the histological wart criterias, the presence of infiltrate, the cytopathic effect of virus in upper epidermal cell layers which leads to the determination of HPV type [14,15]. 3.5. Biochemical determination of the virus type in warts In a few cases, viral DNA was extracted by the Hirt technique [16]. DNA was digested for 1 h at 37°C with a mixture of restriction enzymes (Hind II + Hind III; Boehringer, Mannheim, France) under conditions described by Orth et al. [14] and Favre et al. [17]. Fragments of DNA migrated in 1% agar gels. Molecular weights of the fragments were determined on ethidium bromide-stained gels under UV transillumination. Phage DNA digested with Hind III endonuclease was used as marker. HPV types were determined after nitrocellulose transfer of fragments and Southern blot hybridization with [32p]labeled probes of HPV1 and HPV2.
4. Results
Each of the 28 warts under study (Table 1) showed a typical histology. In many cases, epidermis was disorganized as compared to normal skin biopsies. Twenty-two warts out of 28 showed a mononuclear inflammatory infiltrate. In most cases (11/22) the infiltrate was mild. In the 11 others, the strongest reaction was seen as shown in Fig. I A. OKT3-positive cells were found in 9 specimens. OKT4-positive cells were detected in 3 cases (nos. 4, 18, 27). OKT8-positive cells (Fig. 1B) were present in 7 biopsies (nos. 4, 10, 23, 25-28). No infiltration and no OKT-positive cells were de-
tected in normal epidermis. With BI6, 7/26 specimens were negative in epidermis (6 of them showed an infiltrate and 2 contained a number of LC in the dermis). Among 19 other biopsies, the number of positive LC and the fluorescence intensity were highly variable. In most cases (11/14), positive cells were present both in dermis and epidermis (Fig. 1D). Conversely normal skin biopsies contained LC only in epidermis (Fig. 1C). Viral antigen was detected in some keratinocytes of the epidermal upper layer of 21 warts. Most of them (19/28) showed a typical histology of HPV2, and 4 of them an HPVI histology. In 3 cases, the type of virus was confirmed by biochemical technique (no. 4: HPVIa; nos. 7 and 24: HPV2). One hand wart (no. 4) presented an intense inflammatory reaction. It contained viral antigen and was typed as HPVIa. Positive fluorescent cells were found in the dermis with OKT4 and OKT8. The number of LC increased highly in the epidermis as compared to normal epidermis. The dermis also contained positive B16 cells.
5. Discussion
This preliminary study showed changes in LC distribution of wart epidermis as compared to normal skin. In a few specimens, T-lymphocytes were present without significant predominance of a subset. With the classical histological technique, the infiltrate was commonly observed in plane warts [8] and more rarely in common warts [18,19]. In our series, most of warts showed an infiltrate. Surprisingly, in a number of warts, the density of LC in the epidermis was extremely reduced and in some cases, they were not detected, but the viral antigen was often present. These results indicate that HPV infection induces either disappearance of LC or changes of their antigenic membrane properties. It cannot yet be related to a special localization of the lesion or to a type of virus. However, the duration of most lesions in a patient could be an important factor. On the contrary, in a few biopsies, the density of LC was much higher than those in normal epidermis. Moreover, these cells were also present in the 193
A
C Fig. 1. Monocellular infiltrate and Langerhans cells in a biopsy of hand warts. A: histology of monocellular infiltration (Hematoxylin staining; X 1200). B: OKT8-positive cells of infiltrate in immunofluorescence (X 1200). C: LC cells of normal epidermis recognized by B 16 monoclonal antibody in immunofluorescence 0 < 1200). (Dotted white line: dermis-epidermis junction.) D: LC cells in dermis and epidermis of warts (X 1200). 194
upper dermis. Such an increased number of LC could suggest an activation of these cells in some lesions. Ultrastructural changes of LC in the epidermis were reported during spontaneous involution of plane warts [6]. LC showed signs of enhanced cellular activity with an increase number of LC granules. Similarly, a b n o r m a l LC were found by electron microscopy in the epidermis at the sites of vaccinia virus inoculation [20]. These cells contained an unusually increased number of LC granules. In n o n viral skin tumours, the number of L C always increased irrespective of their benign or malignant nature [2,21] as demonstrated by electron microscopy and immunofluorescence. Further studies are needed to clarify the role of the virus in the interactions between LC and keratinocytes in warts and to elucidate changes of antigenic membrane properties. Infection of LC with H P V could explain their lack of reactivity but remains to be demonstrated. Such a study could be investigated by electron microscopy. In a similar study on vaccinia virus, the particles were detected inside LC of epidermis at the inoculation sites [20]. The number of T-lymphocytes is low in wart lesions, as compared to benign and malignant skin tumours. Both T-cell subsets were present with no significant predominance. A high number of OKT8positive cells were found in benign tumours whereas in highly malignant tumours, it was reduced [2]. Our data suggest that either depletion or stimulation of T-cell subsets and of LC in warts might be related to the evolution of viral infection. It is well known that skin-mediated sensitization to different antigens is dependent on the density of LC [22,23]. Depletion of epidermal LC usually leads to host tolerance of these antigens [24-26]. Decreased density of epidermal LC and tolerance to H P V might be responsible for the development of lesions. On the contrary, a strong infiltrate and a high density of LC in the epidermis, but also in dermis, are suggestive of cell-mediated immune reaction directed towards the tumour. This could favour the hypothesis that T-cells and LC participate in local wart rejection [27]. Further studies of these cells in clinically regressing warts from various localizations will give more conclusive information.
Acknowledgements This work was done with grants from Fbd~ration Nationale des Centres de Lutte contre le Cancer 1981 1982. The authors are indebted to Prof. J. Thivolet and the clinicians of Clinique Dermatologique, Pav. R, H6pital E. Herriot, Lyon, who collected the biopsies, and to G. Orth, M. Favre and O. Croissant (Institut Pasteur, Paris) for determination of H P V type.
References [1] Viac, J., Bustamante, R. and Thivolet, J. (1977) Br. J. Dermatol. 97, 1 10. [2] Fernandez-Bussy,R., Cambazard, F., Mauduit, G., Schmitt, D. and Thivolet, J. (1983) Europ. J. Cancer, in press. [3] Morison,W. L. (1974) Br. J. Dermatol. 90, 531 534. [4] Tagami, H., Takigawa, M., Ogino, A., Imamura, S. and Ofuji, S. (1977) Arch. Dermatol. 113, 1209-1213. [5] Oguchi, M., Komura, J., Tagami, H. and Ofuji, S. (1981) Arch. Dermatol. Res. 270, 403-411. [6] Oguchi, M., Komura, J., Tagami, H. and Ofuji, S. (1981) Arch. Dermatol. Res. 271, 55-61. [7] Berman, A. and Winkelmann, R. K. (1977) Arch. Dermatol. 113, 1219-1221. [8] Tagami, H., Ogino, A., Takigawa, M., Imamura, S. and Ofuji, S. (1974) Br. J. Dermatol. 90, 147 154. [9] Kung, P. C., Goldstein, C., Reinherz, E. L. and Schlossman, S. F. (1979) Science206, 337-340. [I0] Yonish-Rouach, E,, Viac, J., Schmitt, D., Cordier, G., Thivolet, 3. and Brochier,J. (1982) in: Current Concepts in Human Immunology and Cancer Immunomodulation (Serrou, B., Rosenfeld, C. and Daniels, J., eds.) pp. 55-62, Elsevier/ North-Holland, Amsterdam. [I 1] Schmitt, D., Souteyrand, P., Brochier,J., Czernielewski,J. and Thivolet, J. (1982) Acta Dermatovener. (Stockholm) 62, 193-199. [12] Jenson, A. B., Rosenthal, J. R., Olson, C., Pass, F., Lancaster, W. D. and Shah, K. (1980) J. Nat. Cancer Inst. 64, 495 500. [13] Rodriguez, J., Deinhardt, F. (1960) Virology 12, 316-317. [14] Orth, G., Jablonska, S., Favre, M., Croissant, O., Obalek, S., Jarzabek-Chorzelska, M. and Jibard, N. (1981) J. Invest. Dermatol. 76, 97-102. [15] Gross,G., Pfister, H., Hagedorn, M. and Gissman, L. (1982) J. Invest. Dermatol. 78, 160-164. [16] Hirt, B. (1967) J. Mol. Biol. 26, 365 369. [17] Favre, M., Jibard, N. and Orth, G. (1982) Virology 119, 298-309. [18] Matthews, R. S. and Shirodaria, P. V. (1973) Lancet 1, 689 691. [19] Brodersen, 1. and Genner, J. (1973) Acta Dermatovener. (Stockholm) 53, 461-464. 195
[20] Nagao, S., lnaba, S. and lijima, S. (1976) Arch. Dermatol. Res. 256, 23-31. [21] Lisi, P. 0973) Acta Dermatovener. (Stockholm) 53, 425-428. [22] Horio, T. and Okamoto, H. (1982) J. Invest. Dermatol. 78, 402-405. [23] Miyachi, Y. and Takigawa, M. 0982) J. Invest. Dermatol. 78,363-365.
196
[24] Silberberg, I. (1973) Acta Dermatovener. (Stockholm) 53, 1-12. [25] Silberherg-Sinakin, I. and Thorbecke, G. J. (1980) J. Invest. DermatoL 75, 61 67. [26] Shelley, W. B. and Juhlin, L. (1976) Nature (London) 261, 46-47. [27] Thivolet, J., Viac, J. and Staquet, M. J. 0982) Int. J. Dermatol. 21, 94 98