ARTICLE
Basal cell carcinoma (BCC) of the skin, the most common human cancer,
is a locally invasive tumor that grows slowly and rarely metastasizes.
BCCs are frequently associated with a peritumoral mononuclear cell infiltrate
that may represent a specific antitumor response [1, 2]. The infiltrate
surrounding BCC tumor islands largely consists of T cells with a predominance
of the CD4+/T-helper over the CD8+/T-cytotoxic subset,
similar to that found in typical delayed hypersensitivity reactions [3,
4]. Natural killer cells and B cells are seen in much smaller numbers.
To enable infiltrating leukocytes to influence tumor growth, these cells
first need to be recruited from the circulation. The entry of lymphocytes
from the circulation into an organ is mediated by interactions between
adhesion molecules on vascular endothelial cells (so called vascular adressins)
and circulating leukocytes (termed homing receptors) [5]. Recent interest
has focused upon three cytokine-inducible leukocyte adhesion molecules,
designated intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion
molecule-1 (VCAM-1) and E-selectin [6-8]. Another area of interest involving
adhesion molecules concerns leukocyte-tumor cell interaction. In order
to control the proliferation of the tumor by the surrounding cellular
infiltrate, expression of adhesion molecules on tumor cells is thought
to be required, and it has been suggested that failure to express adhesion
molecules would enable malignant cells to escape immunosurveillance [9].
In the present study, the distribution of a selected panel of adhesion
molecules and their T cell ligands is tested in a group of thirty-three
basal cell carcinomas of either the nodular- and morpheaform type.
Materials and methods
Tissue specimens
Fresh-frozen skin samples were obtained from patients undergoing Mohs'
micrographic surgery (MMS) for BCC, all present on sunexposed areas of
the head. The BCCs consisted of 23 nodular types and 10 with morpheaform
type growth, according to histopathological analysis. Samples were immediately
frozen in liquid nitrogen and stored at 70° C until required.
Normal skin specimens (n = 5) from various body sites were used as a control.
Immunohistochemistry
Frozen, 5 µm sections, were cut, air dried, and fixed for 10 min
in acetone. Primary antibodies were applied at the appropriate dilution
for 30 min at room temperature. Depending on the species of the primary
antibodies, the secondary antibody was either a biotinylated goat anti-mouse
immunoglobulin or a biotinylated goat anti-rat immunoglobulin (Dakopatts,
Copenhagen). In the third step, streptavidin-peroxidase was applied. HRP-enzyme
activity was visualized using 3-amino-ethylcarbizole as a chromogen. Sections
were counterstained with hematoxylin. For positive controls the following
specimens were used: positive patch test reaction biopsy sites (ICAM-1,
E-selectin) and in vitro TNF-alpha-stimulated skin (E-selectin,
VCAM-1).
Negative controls included omission of the primary antibody and substitution
of the primary antibody for an irrelevant one. Primary antibodies used
are summarized in Table I.
Quantification
Expression of endothelial ICAM-1, VCAM-1, and E-selectin was quantified
according to the following criteria: 0, no staining; ±, weak focal
staining; +, moderate staining of whole vessels, ++, strong staining of
whole vessels. The extent of MoAb staining of leukocytic infiltrates was
quantified as follows: 0, no staining; ±, 0-25% positive stained
cells; +, 25-50% positive stained cells, ++, 50-75% positive stained cells,
and +++, > 75% positive stained cells.
Results
In 15 cases (solid type n = 6, morpheaform type n = 9) of the 33 BCCs
studied, only a discrete inflammatory infiltrate was present around the
tumor islands. The remaining cases (solid type n = 17, morpheaform n =
1) displayed a more prominent lymphocytic infiltrate especially perivascular
and at the base of the tumor. In none of the cases were lymphocytes found
infiltrating the BCC tumor nests.
In normal skin, moderate ICAM-1 expression (+) of whole vessels was
found, whereas endothelial E-selectin and VCAM-1 stained negative (0).
When compared to normal skin, increased endothelial ICAM-1 expression
(+/++) in BCC was mainly restricted to those peritumoral areas, where
a prominent lymphocytic infiltrate was present (Fig.
1a).
In BCC tumor specimens, endothelial VCAM-1 and E-selectin expression
was either undetectable or weakly positive (±) with granular staining
of individual endothelial cells (Fig.
1b and c).
No endothelial staining differences between solid and morpheaform BCC
were observed. The peritumoral infiltrate contained mostly LFA-1-expressing
cells (++), with fewer lymphocytes variably staining for VLA-4 (±)
and CLA (±). The results are summarized in Table
II.
In none of the cases examined could ICAM-1, VCAM-1 or E-selectin expression
be identified on tumor cells or overlying benign epidermal keratinocytes.
Discussion
In the last decade, several reports concerning the local immune response
in BCC have attempted to elucidate the function of the cellular inflammatory
infiltrate surrounding these skin cancers and their possible role in controlling
tumor growth [1-3]. A tumor-specific, cellular immune response is considered
to be associated with adhesion molecule expression on endothelial cells
for recruitment of T cells into the tissue, and on tumor cells for lymphocyte-target
cell interaction [9].
At least five molecular pathways operate to recruit T cells to the peripheral
tissue through endothelial cell/T cell interactions that include ICAM-1
and ICAM-2/LFA-1, VCAM-1/VLA-4, E-selectin/CLA and hyaluronic acid/CD44
pathways [10].
ICAM-1, which is inducible by IFN-gamma and TNF-alpha, is expressed
by a wide variety of cells, including resting endothelial cells, and binds
virtually all circulating white cells [6, 7]. E-selectin and VCAM-1 expression
correlates with endothelial cell activation and are both inducible by
TNF-alpha and IL-1 [11]. VCAM-1 contributes to selective recruitment of
T cells and eosinophils but not neutrophils [12], whereas E-selectin functions
as a tissue-selective endothelial cell-adhesion molecule at sites of chronic
inflammation and is thought to be important for recruitment of skin-homing
T lymphocytes expressing the cutaneous lymphocyte-associated antigen (CLA)
[13, 14].
Firstly, recent literature concerning adhesion molecules in BCC will
be summarized. Taylor et al., who studied thirty BCCs, demonstrated
negative ICAM-1 staining of tumor cells [15]. This was confirmed by Barker
et al. who examined six cases of BCC [16]. However, they demonstrated
in areas of inflammatory infiltrate, a more intense ICAM-1 expression
by dermal endothelial cells as compared to normal skin. Groves et al.
showed variable staining patterns of endothelial E-selectin expression
in a group of eleven BCCs [17, 18]. Finally, Pentel et al. in a
group of 16 BCCs and Viac et al. in a group of five BCCs, described
comparable results [19, 20]. BCC tumor cells were all ICAM-1, E-selectin
and VCAM-1 negative. Endothelial cells within the surrounding areas of
the tumors weakly expressed ICAM-1 and rarely expressed E-selectin and
VCAM-1.
In the present study, we investigated the expression and distribution
of the adhesion molecules ICAM-1, E-selectin and VCAM-1, together with
their leukocyte receptors LFA-1, CLA and VLA-4 respectively, in thirty-three
BCCs of different histological subtypes. We demonstrated moderate ICAM-1
and minimal VCAM-1 and E-selectin expression by endothelial cells of blood
vessels surrounding BCC tumor lobules. This is in accordance with the
staining results of their counter-molecules, i.e. the demonstrated
positivity for LFA-1 and the minimal staining for VLA-4 and CLA.
In neither the nodular nor the morpheaform types of BCCs, was ICAM-1,
E-selectin or VCAM-1 expression by tumor cells or overlying benign epidermal
keratinocytes identified.
Adhesion of lymphocytes to target cells via surface molecules
is thought to be required for T lymphocyte-mediated immune reactions.
It has been postulated that BCC can escape immunosurveillance because
of the absence of adhesion molecules on tumor cells, preventing binding
of T lymphocytes [9, 16]. This could explain the histological observation
that T cells do not infiltrate BCC tumor lobules.
The inability of BCC tumor cells to express adhesion molecules is not
absolute. Taylor et al. presented a study where they demonstrated
induction of ICAM-1 expression on tumor cells after BCC tissue had been
incubated in vitro with IFN-gamma [15]. They put forward three
possible explanations for the discrepancy between the in vivo and
in vitro observations: a) insufficient in vivo cytokine
levels, b) a barrier preventing cytokines from reaching and interacting
with tumor cells, and c) a lower level of sensitivity.
In our opinion, the absence of adhesion molecules on both BCC tumor
cells and the overlying epidermal keratinocytes as demonstrated in this
study, together with the moderate ICAM-1 and minimal VCAM-1 or E-selectin
expression on the surrounding peritumoral endothelial cells could probably
best be explained by insufficient in vivo cytokine levels. Insufficient
in vivo cytokine levels can be caused by an insufficient number
of activated T lymphocytes responding to antigen stimulation, or, as Yamamura
et al. demonstrated, an accumulation of type-2 instead of type-1
T helper cells, resulting in the production of IL-4, IL-5, and IL-10 cytokines
instead of the cytokines IL-2, IFN-gamma and TNF-alpha [21-23].
In summary, endothelial cells are critical elements in the evolution
of cellular infiltrates. Our results indicate resting endothelial cells
around BCCs, with moderate ICAM-1 expression and minimal E-selectin/VCAM-1
expression, recruiting particularly LFA-1 positive lymphocytes in both
nodular and morpheaform type BCC. Histologically, these LFA-1 positive
lymphocytes do not infiltrate the BCC tumor lobules. This is possibly
due to the demonstrated absence of adhesion molecule expression on BCC
tumor cells, interfering with the essential cell-cell contact between
the effector immune cell and target cell.
REFERENCES
1. Hunt MJ, Halliday GM, Weedon D, Cooke BE, StC. Barnetson R. Regression
in basal cell carcinoma: an immunohistochemical analysis. Br J Dermatol
1994; 130: 1-8.
2. Curson C, Weedon D. Spontaneous regression in basal cell carcinomas.
J Cutan Pathol 1979; 6: 432-7.
3. Habets JM, Tank B, Vuzevski VD, van Reede EC, Stolz E, van Joost
T. Characterization of the mononuclear infiltrate in basal cell carcinoma:
a predominantly T cell-mediated immune response with minor participation
of Leu-7+ (natural killer) cells and Leu-14+ (B)
cells. J Invest Dermatol 1988; 90: 289-92.
4. Claudy AL, Viac J, Schmitt D, Alario A, Thivolet J. Identification
of mononuclear cells infiltrating basal cell carcinomas. Acta Derm
Venereol 1976; 56: 361-5.
5. Shimizu Y, Newman W, Tanaka Y, Shaw S. Lymphocyte interactions with
endothelial cells. Immunol Today 1992; 13: 106-12.
6. Springer TA. Adhesion receptors of the immune system. Nature
1990; 346: 425-34.
7. Barker JNWN, MacDonald DM. Cutaneous lymphocyte trafficking in the
inflammatory dermatoses. Br J Dermatol 1992; 126: 211-5.
8. Groves RW, Ross EL, Barker JNWN, MacDonald DM. Vascular cell adhesion
molecule-1: expression in normal and diseased skin and regulation in
vivo by interferon gamma. J Am Acad Dermatol 1993; 29: 67-72.
9. Nickoloff BJ, Griffiths CEM, Barker JNWN. The role of adhesion molecules,
chemotactic factors, and cytokines in inflammatory and neoplastic skin
disease-1990 update. J Invest Dermatol 1990; 94: 151S-7S.
10. Wakita H, Takigawa. E-selectin and vascular cell adhesion molecule-1
are critical for initiating trafficking of helper-inducer/memory T cells
in psoriatic plaques. Arch Dermatol 1994; 130: 457-63.
11. Pober JS, Bevilacqua MP, Mendrick DL, Lapierre LA, Fiers W, Gimbrone
Jr MA. Two distinct monokines, interleukin-1 and tumor necrosis factor,
each independently induce the biosynthesis and transient expression of
the same antigen on the surface of cultured human vascular endothelial
cells. J Immunol 1986; 136: 1680-7.
12. Petzelbauer P, Pober JS, Keh A, Braverman IM. Inducibility and expression
of microvascular endothelial adhesion molecules in lesional, perilesional,
and uninvolved skin of psoriatic patients. J Invest Dermatol 1991;
103: 300-5.
13. Picker LJ, Kishimoto TK, Smith CW, Warnock TK, Butcher EC. ELAM-1
is an adhesion molecule for skin-homing T cells. Nature 1991; 349:
796-9.
14. Bos JD, de Boer OJ, Tibosch E, Das PK, Pals ST. Skin-homing T lymphocytes:
detection of cutaneous lymphocyte-associated antigen (CLA) by HECA-452
in normal human skin. Arch Dermatol Res 1993; 285: 179-83.
15. Taylor RS, Griffiths CEM, Brown MD, Swanson NA, Nickoloff BJ. Constitutive
absence and interferon-gamma-induced expression of adhesion molecules
in basal cell carcinoma. J Am Acad Dermatol 1990; 22: 721-6.
16. Barker JNWN, Allen MH, MacDonald DM. Distribution of intercellular-adhesion
molecule-1 (ICAM-1) and lymphocyte-function-associated antigen-1 (LFA-1)
in epidermal tumours. Clin Exp Dermatol 1990; 15: 331-4.
17. Groves RW, Allen MH, Barker JNWN, Haskard DO, MacDonald DM. Endothelial
leucocyte adhesion molecule-1 (ELAM-1) expression in cutaneous inflammation.
Br J Dermatol 1991; 124: 117-23.
18. Groves RW, Allen MH, Ross EL, Ahsan G, Barker JNWN, MacDonald DM.
Expression of selectin ligands by cutaneous squamous cell carcinoma. Am
J Pathol 1993; 143: 1220-5.
19. Pentel M, Helm KF, Malony MM. Cell surface molecules in basal cell
carcinomas. Dermatol Surg 1995; 21: 858-61.
20. Expression of adhesion receptors in epidermal tumors: correlation
with TNF-alpha expressing cells. Anticancer Res 1995; 15: 551-6.
21. Guillén FJ, Day CL, Murphy GF. Expression of activation antigens
by T cells infiltrating basal cell carcinomas. J Invest Dermatol
1985; 85: 203-6.
22. Markey AC, Churchill LJ, Allen MH, MacDonald DM. Activation and
inducer subset phenotype of the lymphocytic infiltrate around epidermally
derived tumors. J Am Acad Dermatol 1990; 23: 214-20.
23. Yamamura M, Modlin RL, Ohmen JD, Moy RL. Local expression of antiinflammatory
cytokines in cancer. J Clin Invest 1993; 91: 1005-10.
|
Version imprimable
