Investigation of epidermotropism in canine mycosis fungoides ...

Arch Dermatol Res (1995) 287:186-192

9 Springer-Verlag 1995

Thierry Olivry - Peter F. M o o r e , Diane K. Naydan Dimitry M. Danilenko 9 Verena K. Affolter

Investigation of epidermotropism in canine mycosis fungoides: expression of intercellular adhesion molecule-1 (ICAM-1) and beta-2 integrins

Received: 26 October 1993

Abstract In human mycosis fungoides (MF), interactions between LFA-1 (CDIlaJCD18) and ICAM-1 (CD54) are involved in lymphocyte adhesion to keratinocytes. The purpose of this study was to evaluate the expression of ICAM-1, beta-2 integrins and class II major histocompatibility complex molecules (MHC II) on keratinocytes and infiltrating lymphocytes in canine MF. Sections of frozen skin biopsy specimens from normal dogs (n = 3) and dogs with MF (n = 17) were evaluated by immunohistochemistry for expression of ICAM-1, beta-2 integrins, and class II MHC molecules. Our results demonstrated that in canine MF, ICAM-1 was expressed variably on epidermal and follicular keratinocytes. The extent of keratinocyte ICAM-1 expression did not correlate with the degree of lymphocyte epithelial infiltration, nor with lymphocyte LFA-1 expression. This was especially evident in cases of Pagetoid reticulosis-like disease in which prominent lymphocyte epidermotropism was not accompanied by keratinocyte ICAM-1 expression. Keratinocyte class II MHC molecule expression did not correlate with keratinocyte ICAM-1 expression. In conclusion, in canine MF, the lack of statistically significant correlations between epithelial lymphocyte infiltration and keratinocyte ICAM-1 expression, and between keratinocyte ICAM-1 and lymphocyte LFA-I staining, suggests that the LFA-1/ICAM-1 pathway is not the major adhesion mechanism between lymphocytes and keratinocytes. It is suspected that different ligands of the LFA-1 integrin (e.g. ICAM-2) or other adhesion molecules (e.g. CD2/LFA-3, VLA-1) might be involved in the epitheliotropism phenomenon in canine ME

T. Olivry 9P. F. Moore (N~). D. M. Danilenko 9V. K. Affolter Department of Pathology, School of Veterinary Medicine, University of California Davis, Davis, CA 95616, USA D. K. Naydan Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California Davis, Davis, CA 95616, USA

These hypotheses cannot be evaluated in the dog at this time owing to the lack of specific monoclonal antibodies. Key words Dogs 9 Canine 9Mycosis fungoides Cutaneous lymphoma. Epitheliotropism 9 Intercellular adhesion molecule-1 (ICAM-1) - LFA-1 9 Class II major histocompatibility complex 9 Keratinocytes 9Lymphocytes

Introduction The migration of lymphocytes to sites of cutaneous inflammation is dependent upon their adhesion to endothelial cells, extracellular matrix, and epithelial cells. Lymphocyte adhesion is mediated by three families of adhesion molecules: the integrin superfamily (Hynes 1987; Springer 1990), the selectin family (Marx 1989), and the immunoglobulin gene superfamily (Williams and Barclay 1988). Integrins are noncovalently linked heterodimers with a unique subfamily-specific beta-chain and a variable alpha-chain. The beta-2 integrin subfamily (leukocyte integrin subfamily, LeuCAMs) is composed of three related molecules: the lymphocyte function associated antigen-1 (LFA-1, alphaL-beta2, CDlla/CD18), Mac-1 (CR3, alphaM-beta2, CDllb/CD18), and p150,95 (CR4, alphaX-beta2, CDllc/CD18) (Hynes 1987; Springer 1990). Recognized LFA-1 ligands include intercellular adhesion molecule-1 and 2 (ICAM-1, ICAM-2), members of the immunoglobulin superfamily (Rothlein et al. 1986; Marlin and Springer 1987; Staunton et al. 1989). Adhesion of lymphocytes to keratinocytes appears to be mediated principally through the interaction of LFA-1 and ICAM-1 (Dustin et al. 1988). Alternate adhesion pathways may also exist (Mitra et al. 1991). In normal skin, ICAM-1 is expressed on endothelial cells but not on keratinocytes (Dustin et al. 1986; Griffiths et al. 1989; Konter et al. 1989; Vejlsgaard et al. 1989). ICAM-1 expression on keratinocytes can be upregulated by gamma-interferon (Nickoloff et al. 1987; Dustin et al. 1988, Barker et al. 1989; Griffiths et al, 1989), tumor necrosis factor-

187 alpha (Griffiths et al. 1989), and interleukin-1 (Groves et al. 1992). N e o e x p r e s s i o n of I C A M - 1 on keratinocytes has been observed in various h u m a n i n f l a m m a t o r y and neoplastic skin diseases i n c l u d i n g mycosis fungoides (MF), and corresponds to intraepidermal infiltration of l y m p h o c y t e s (Griffiths et al. 1989; Vejlsgaard et al. 1989; B o e h n c k e et al. 1992; Sterry et al. 1992). Similarly, class II m a j o r histocompatibility c o m p l e x molecules ( H L A - D R ) are not expressed on n o r m a l h u m a n keratinocytes but can be upregulated by g a m m a - i n t e r feron (Griffiths et al. 1989; Vejlsgaard et al. 1989). Neoexpression of H L A - D R can be detected on keratinocytes in M F and various i n f l a m m a t o r y dermatoses (Griffiths et al. 1989; Vejlsgaard et al. 1989; W i l l e m z e et al. 1985; D r i j k o n i n g e n et al. 1986; Pincelli et al. 1986), but the significance of this f i n d i n g is uncertain. Since c a n i n e M F differs m a r k e d l y from M F in h u m a n s in respect of the i m m u n o p h e n o t y p e of the neoplasti cells (CD8 + versus C D 4 +, respectively), as well as the persistence of e p i d e r m o t r o p i s m in a d v a n c e d stages of the disease in dogs (Moore et al. 1993), we decided to study some molecular aspects of epidermotropism in canine M E Since interactions b e t w e e n l y m p h o c y t e LFA-1 and keratinocyte I C A M - 1 have b e e n implicated in l y m p h o c y t e - k e r a t i n o cyte adhesion in h u m a n cases of MF, the expression of these adhesion molecules was investigated in the c a n i n e counterpart. The correlation of the expression of M H C II molecules and I C A M - 1 on keratinocytes was additionally determined, since both molecules can be upregulated b y g a m m a - i n t e r f e r o n (Griffiths et al. 1989; Vejlsgaard et al. 1989). In this study we demonstrated that in canine MF, the expression of I C A M - 1 on keratinocytes was not correlated with the degree of l y m p h o c y t e epitheliotropism nor with l y m p h o c y t e LFA-1 expression.

Material and methods

Table 1 Monoclonal antibody panel

Antibody

Epitope

Dilution~ Source

CL18 CAll.4D3 CA16.1Bll 904 b CA11.6A1 CA1.4E9 CA2.1C12 2D16 c

ICAM-1 CDlla CDIla CD 1lb CDllc CD18 MHC II MHC II

50 10 50 Undiluted 10 50 50 10

Dr. Dr. Dr. Dr. Dr. Dr. Dr. Dr.

C.W. Smith P.F. Moore P.F. Moore C.W. Smith P.F. Moore P.F. Moore P.F. Moore P.F. Moore

Houston, Tex. Davis, Calif. Davis, Calif. Houston, Tex. Davis, Calif. Davis, Calif. Davis, Calif. Davis, Calif.

Dilutions from hybridoma tissue culture supernatants b Antihuman monoclonal antibody cross-reactive with the dog c Antirhesus macaque monoclonal antibody cross-reactive with the dog

Martel Laboratory of Leukocyte Biology, Houston, Tex. (Smith et al. 1991). The 904 anti-human CDI lb antibody has been shown to crossreact with the canine species (Giger et al. 1987; Danilenko et al. 1992). Other mAbs have previously been developed in our laboratory, and have been proven to specifically identify various leukocyte populations in the dog (Moore et al. 1986; Moore et al. 1990; Danilenko et al. 1992).

Immunoenzymatic staining procedures Frozen sections were fixed in acetone and stained with a three-step labelled streptavidin method (Cartun and Pedersen 1989) with the mAbs listed in Table 1. The biotinylated horse anti-mouse secondary antiserum (Vector Laboratories, Burlingame, Calif.) was applied at a dilution of 1 : 400 for 30 min. Streptavidin-horseradish peroxidase (Zymed Laboratories, South San Francisco, Calif.) was applied for 30 min at a dilution of 1 : 400. The sections were rinsed in phosphate-buffered saline (PBS) between each step. Diaminobenzidine hydrochloride (Sigma Chemical, Saint Louis, Mo.) was used as chromogen. Normal canine spleen was used as a positive control. Negative controls were obtained by using PBS in place of the primary mAb. To minimize variation due to technical factors, all cases were stained with the same primary antibody on the same day.

Biopsy specimens Quantitation of results Control skin biopsy specimens were obtained from three sites (face, gingiva, front leg) from normal canine skin (n = 3). MF biopsy specimens were obtained from lesional canine skin (n = 17). The T-cell nature of epidermotropic lymphocytes was confirmed by immunohistochemistry (P. F. Moore; data not shown) (Moore et al. 1993). Canine CD3 epsilon was detected by a polyclonal rabbit antiserum (Dako, Carpinteria, Calif.) which is specific for a 13-mer peptide sequence from the cytoplasmic domain of human CD3 epsilon (Mason et al. 1989). This sequence is highly conserved in several mammalian species including the dog (Nash et al. 1991). Fresh biopsy specimens were bisected, embedded in OCT compound (Tissue Tek, Baxter Diagnostics, McGaw Park, Ill.), snap-frozen in Freon 22 (dichlorodifluoromethane) cooled to freezing in liquid nitrogen, and stored at -70 ~C. Cryostat sections (5 gm) were air dried and stored at -70 ~C for subsequent staining. Immunohistochemistry

Monoclonal antibody panel (Table 1) The monoclonal antibodies (mAb) against canine ICAM-1 and human CD1 lb were obtained from Dr. C, W. Smith of the Speros P.

The immunostained sections were examined with a light microscope at X200 magnification and assessed semiquantitatively.

Lymphocyte epitheliotropism Areas of most severe lymphocytic infiltration in the epidermis and hair follicle root sheaths were identified. The degree of lymphocyte infiltration was graded as absent (0), mild (l), moderate (2) or marked (3). Lymphocyte infiltration in the superficial and deep dermis was recorded as present (+) or absent (-).

Staining The extent and intensity of antibody staining of keratinocytes and neoplastic lymphocytes were recorded at each selected site of most severe epitheliotropism. The extent of staining in the selected area was graded on a scale of 0 to 4 depending upon the percentage of cells stained: 0-5% = (0), 5-25% = (1), 25-50% = (2), 50-75% = (3), > 75% = (4). The intensity of the staining was graded on a scale of 0 to 2: no staining = (0), faint staining = (1), intense staining = (2).

188 Table 2 Epithelial lymphocyte infiltration, keratinocyte ICAM-1 expression, and lymphocyte LFA-1 expression Case

217747 219796 221498 225341 227075 229559 231443 236516 253584 6253 251328 256585 259461 6151227 a 260441 a 232326 a 239158 a

Epidermis

Follicles

Lymphocytes

Degree of ICAM-1 infiltration extent

ICAM-1 intensity

Degree of infiltration

ICAM-1 extent

ICAM-1 intensity

C D l l a (4D3) CDlla (1Bll) CD18 extent extent extent

2 2 2 2 2 2 3 2 1 2 2 0 2 2 3 3 2

1 1 2 1 2 1 2 2 2 2 2 0 1 0 1 0 0

2 2 2 2 3 NA 3 NA 2 2 2 3 2 3 1 2 NA

0 1 2 0 2 NA 3 NA 3 2 1 0 2 1 0 0 NA

0 1 2 0 1 NA 2 NA 2 2 1 0 1 1 0 0 NA

1 0 1 0 0 0 1 0 3 1 1 1 1 2 1 0 0

1 1 2 1 4 1 3 2 3 2 4 0 2 0 1 0 0

Woringer-Kolopp-like disease NA: Not Assessable Statistical analysis The extent of keratinocyte ICAM-1 staining was compared with the degree of epidermal and follicular lymphocyte infiltration. It was also compared with the extent of keratinocyte MHC II molecule (1C12 mAb) expression and with the extent of lymphocyte CD 11a staining (4D3 and 1B 11 mAbs). The Pearson product-moment correlation coefficients (r) were determined using B.M.D.P 386 software (B.M.D.P. Statistical Software, Los Angeles, Calif.) (Dixon 1990). A Pearson product-moment correlation coefficient was considered significant if greater than 1.75.

Results Quantitation of epidermal and follicular infiltration In most cases, c o m b i n e d epidermal, follicular, and dermal l y m p h o c y t e infiltration was observed. Epidermal l y m p h o cyte infiltration varied from absent (grade 0) to m a r k e d (grade 3) (Table 2). Follicular infiltration, when assessable (n = 14), varied from m i l d (grade 1) to m a r k e d (grade 3). In cases where epidermal infiltration was absent, folliculotropism was still detectable. L y m p h o c y t e infiltration was present in the superficial dermis in 14 cases and in the deep dermis in 10 cases. In 4 cases, l y m p h o c y t e infiltration was restricted to the epidermis and hair follicles, as in Pagetoid reticulosis ( W o r i n g e r - K o l o p p disease) in huFig.lA, B Immunohistochemical staining of ICAM-1 in canine skin. A In normal canine skin, ICAM-1 expression is absent from epidermal keratinocytes. ICAM-1 can be detected on the endothelium of the dermal vasculature (arrow) (x 140). B Strong expression of ICAM-1 on keratinocytes and infiltrating lymphocytes in a case of canine mycosis fungoides (case no. 227075) (• 175)

1 0 1 NA 0 0 3 0 3 3 2 2 3 NA 0 0 0

3 4 4 4 3 4 4 4 4 2 4 4 4 4 4 4 4

189

Fig.2A-D Immunohistochemical staining of ICAM-1 and LFA-1 (mAb CD1 la/4D3) in canine mycosis fungoides. A LFA-1 is not expressed on epidermotropic and dermal lymphocytes (case no. 227075) (x 125). B same biopsy as A. ICAM-1 is markedly expressed on epidermal keratinocytes (case no. 227075) (x 125). C LFA-1 can be detected with high intensity on both epidermotropic and dermal lymphocytes (case no. 6151227) (x 125). D Same area as C. ICAM-1 is not expressed on keratinocytes (case no. 6151227) (x 125)

marts (Lever and Schaumburg-Lever 1990; Walder and Gross 1992).

Expression of ICAM-1 in normal skin and in mycosis fungoides In normal skin ICAM-1 was expressed on the endothelium of the superficial and deep dermal vascular plexus but was not detected on keratinocytes (Fig. 1A). In mycosis fungoides keratinocyte ICAM-1 expression was present in the epidermis in 13 of 17 (76%) cases (Fig. 1B) and in hair follicles in 9 of 14 (64%) cases (Table 2). The intensity of the staining varied from faint to intense, and usually exhibited greater intensity on epidermal cells than on follicular keratinocytes (Table 2). In cases with deep dermal lymphocyte infiltration, a feature suggestive of advanced tumor-stage disease, keratinocyte ICAM-1 staining was no different from cases with minimal superficial dermal infiltration (patch-plaque stage) (data not shown). In the four cases of Woringer-Kolopp-like disease in which epidermotropism was moderate to marked and der-

mal infiltration was poor, keratinocyte ICAM-1 expression was absent or minimal (Table 2). The correlation coefficients between the extent of ICAM-1 expression on epidermal keratinocytes and the degree of epidermal lymphocyte infiltration and between the extent of ICAM-1 expression on follicular keratinocytes and the degree of follicular lymphocyte infiltration were not statistically significant (r = 0.07 and 0.28, respectively).

Expression of beta-2 integrins in normal skin and in mycosis fungoides In normal skin epidermal or follicular staining with the C D l l a , C D l l b , C D l l c , and CD18 mAbs was not observed. In mycosis fungoides C D l l a , b , c were not appreciably expressed by epidermal or follicular epithelium in any of the cases, although CD18 was detectable on epidermal and follicular epithelium in three cases. Variable lymphocyte C D l l a expression was observed in 10 of 17 (59%) cases with the 4D3 mAb (Fig. 2A-D) and in 8 of 15 (53%) cases with the 1 B l l mAb (Table 2). Minimal lymphocyte C D l l b expression was present in 2 of 16 (12.5%) cases. Lymphocyte C D l l c was detected in one case (6%). Intense lymphocyte CD18 staining was detected in all cases. The correlation coefficients between the extent of keratinocyte ICAM-1 and the extent of lymphocyte LFA-I (CD1 la) staining (4D3 and 1B 11 mAbs) were not statistically significant (r = 0.22 and 0.39, respectively).

190 Expression of class II MHC molecules in normal skin and in mycosis fungoides In normal skin epidermal or follicular staining with the 1C12 and 2D16 mAbs were not observed. In mycosis fungoides epidermal 1C12 staining was present to a variable extent in 8 of 16 (50%) cases. Follicular staining was detected in 6 of 16 (37.5%) cases. In all cases, the intensity of the staining varied from faint to intense. Keratinocyte 2D16 staining was not detected. The correlation coefficient between the extent of epidermal 1C12 and ICAM-1 staining was not statistically significant (r = 0.55). Lymphocyte 1C12 staining was detected in 14 of 16 (87.5%) cases. Faint lymphocyte 2D16 staining was present in 2 of 17 (12%) cases.

Discussion This study demonstrated that in canine MF, keratinocytes may express ICAM-1 and various other surface glycoproteins. However, there was no statistically significant correlation between keratinocyte ICAM-1 staining and the degree of lymphocyte epitheliotropism, or between the expression of lymphocyte LFA-1 and keratinocyte ICAM1. In canine MF, epidermal and follicular keratinocytes did not appreciably express beta-2 integrins. Similarly, the expression of MHC II molecules by keratinocytes was uncommon and unpredictable, and was less frequently observed than the expression of ICAM-1 by keratinocytes. The fact that gamma-interferon is able to upregulate both ICAM-1 and MHC II molecules on keratinocytes, implies that either the kinetics of ICAM-1 and MHC II molecule upregulation by gamma-interferon are different (Griffiths et al. 1989; Vejlsgaard et al. 1989), of that mechanisms other than neoplastic lymphocyte gamma-interferon production are perhaps involved in the induction of these glycoproteins. In humans, adhesion of lymphocytes to keratinocytes appears to be mediated principally by the binding of LFA1 and ICAM-1 surface glycoproteins (Dustin et al. 1988). In normal human skin, ICAM-1 is expressed only on endothelial cells but not on keratinocytes (Dustin et al. 1986; Griffiths et al. 1989; Konter et al. 1989; Vejlsgaard et al. 1989). ICAM-1 can be upregulated on keratinocytes by various cytokines including gamma-interferon (Nickoloff et al. 1987; Dustin et al. 1988, Barker et al. 1989; Griffiths et al. 1989), and this upregulation depends upon keratinocyte differentiation and appears more prominent on basal cells (Kashihara-Sawami and Norris 1992). In humans, ICAM-1 neoexpression on keratinocytes has been observed with variable intensity in MF and various inflammatory dermatoses (Griffiths et al. 1989; Vejlsgaard et al. 1989; Boehncke et al. 1992; Sterry et al. 1992). In human MF, the expression of ICAM-1 on keratinocytes correlates with the intraepithelial infiltration of lymphocytes. With progression of MF to advanced rumor stages, lymphocyte epidermotropism decreases and is accompanied by a marked diminution of keratinocyte

ICAM-1 expression (Vejlsgaard et al. 1989). In two cases of S6zary syndrome, the leukemic form of MF, expression of ICAM-1 on keratinocytes was markedly decreased in association with the inability of malignant T cells to produce gamma-interferon (Nickoloff et al. 1989). ICAM-1 can also be expressed on malignant lymphocytes in human MF (Sterry et al. 1992). In canine MF, neoexpression of ICAM-1 was variably detected on epidermal and follicular keratinocytes as well as infiltrating lymphocytes. ICAM-1 staining appeared less prominent on follicular keratinocytes than on epidermal keratinocytes, even when lymphocytic folliculotropism was marked. In canine MR as opposed to the human disease, keratinocyte ICAM-1 expression could still be observed in cases associated with deep dermal lymphocyte infiltration, a feature of advanced tumor-stage neoplasia. In the dog, the expression of ICAM-1 on keratinocytes did not correlate significantly with the degree of epidermal or follicular neoplastic lymphocytic infiltration. This finding markedly differs from those of similar studies in human MF (Griffiths et al. 1989; Vejlsgaard et al. 1989; Sterry et al. 1992). Neoplastic lymphocytes variably expressed beta-2 integrin molecules, which is a similar result to those of previous studies in humans and dogs (Nickoloff et al. 1989; Moore et al. 1990; Sterry et al. 1992). However, in canine MF, there was no statistically significant correlation between keratinocyte ICAM-1 and its ligand LFA-1 on lymphocytes. In four of our cases of canine MF, neoplastic lymphocyte infiltration was restricted to the epidermis and hair follicles as in human Pagetoid reticulosis (WoringerKolopp disease). In these canine cases, the expression of ICAM-1 on keratinocytes and LFA-l on lymphocytes was minimal or absent despite lymphocyte epidermotropism. In humans, recent data suggest that some cases of Pagetoid reticulosis represent a cutaneous malignancy that originates from gamma/delta T lymphocytes (Berti et al. 1991; Heald et al. 1992). In these reports, there were no data concerning the expression of ICAM-1 or LFA-1. The results obtained in this study suggest that adhesion of lymphocytes to keratinocytes in canine Pagetoid reticulosis may not involve that ICAM-1/LFA-1 pathway. This adhesion pathway has yet to be evaluated in human Pagetoid reticulosis. An apparent discrepancy between CDlla,b,c and CD18 expressions on infiltrating lymphocytes in our study of canine MF was observed. The expression of CD18 was apparent in cases in which expression of CDll,a,b,c was not detectable. A possible explanation for this phenomenon may be that the anti-CD18 mAb CA1.4E9 has a higher affinity than do the various CDll mAbs. Although this explanation is plausible, our two C D l l a mAbs had different affinities yet yielded similar results. An alternative explanation is that discordant CD11 a/CD18 expression may occur in neoplastic T lymphocytes in canine ME This has been observed in human B-cell lymphoma (Picker et al. 1987). Presumably, in these instances, the surface expression of a CD 11a/CD 18 heterodimer is lacking. However, this could not be evalu-

191

ated in tissue sections owing to the inability to readily distinguish surface and cytoplasmic antigen expression in lymphoid cells. In canine MF, the lack of statistically significant correlations b e t w e e n epithelial l y m p h o c y t e infiltration and kera t i n o c y t e 1CAM-1 expression, and b e t w e e n k e r a t i n o c y t e I C A M - 1 and l y m p h o c y t e L F A - 1 staining, suggests that the L F A - 1 / I C A M - 1 p a t h w a y is p r o b a b l y not the m a j o r a d h e s i o n m e c h a n i s m b e t w e e n l y m p h o c y t e s and keratinocytes in canine e p i d e r m o t r o p i c T-cell l y m p h o m a . It is susp e c t e d that different ligands o f the L F A - 1 integrin (e.g. I C A M - 2 ) (Staunton et al. 1989) or other a d h e s i o n m o l ecules (e.g. C D 2 / L F A - 3 , beta-1 integrins) (Dustin et al. 1987; Konter et al. 1989; DePanfilis et al. 1991; B o e h n c k e et al. 1992; Sterry et al. 1992) m i g h t b e i n v o l v e d in the e p i t h e l i o t r o p i s m p h e n o m e n o n in canine cutaneous T-cell l y m p h o m a . T h e s e a d h e s i o n p a t h w a y s cannot be evaluated in the dog at this time o w i n g to the l a c k o f specific mAbs. Acknowledgements The authors are grateful to Mr. J. Holmes, from the Department of Veterinary Epidemiology, University of California at Davis, for the statistical analysis. This study was funded in part in the Resident Training Fund from the Veterinary Medical Teaching Hospital, and by a grant from the Companion Animal Laboratory, School of Veterinary Medicine, University of California Davis

References Barker JNWN, Allen MH, MacDonald DM (1989) The effect of in vivo interferon-gamma on the distribution of LFA-1 and ICAM-1 in normal human skin. J Invest Dermatol 93:439442 Berti E, Cerri A, Cavicchini S, Delia D. Soligo D, Alessi E, Caputo R (1991) Primary cutaneous gamma-delta lymphoma presenting as disseminated pagetoid reticulosis. J Invest Dermatol 96 : 718-722 Boehncke WH, Kellner I, Konter U, Sterry W (1992) Differential expression of adhesion molecules on infiltrating cells in inflammatory dermatoses. J Am Acad Dermatol 26:907-913 Cartun RW, Pedersen C (1989) An imnmnocytochemical technique offering increased sensitivity and lowered cost with a streptavidin-horseradish peroxidase conjugate. J Histotechnol 12 : 273-277 Danilenko DM, Moore PF, Rossitto PV (1992) Canine leukocyte cell adhesion molecules (LeuCAMs): characterization of the CD 11/CD 18 family. Tissue Antigens 40 : 13-21 DePanfilis G, Manara GC, Ferrari C, Torresani C (1991) Adhesion molecules on the plasma membrane of epidermal cells. III. Keratinocytes and Langerhans cells constitutively express the lymphocyte function-associated antigen 3. J Invest Dermatol 96:512-517 Dixon WJ (1990) B.M.D.P. statistical software manual, vol 1. UC Press, Berkeley, pp 267-268 Drijkoningen M, De Wolf-Peeters C, Desmet V (1986) Expression of T6 antigen on the epidermal keratinocytes in various dermatoses (letter). Br J Dermatol 115:511-512 Dustin ML, Rothlein R, Bhan AK, Dinarello CA, Springer TA (1986) Induction by IL-1 and interferon, tissue distribution, biochemistry and function of a natural adherence molecule (ICAM-1). J Immunol 137 : 245-254 Dustin ML, Sanders M, Shaw S, Springer TA (1987) Purified lymphocyte function-associated antigen 3 binds to CD2 and mediates T lymphocytes adhesion. J Exp Med 165 : 677-692

Dustin ML, Singer KH, Tuck DT, Springer TA (1988) Adhesion of T lymphoblasts to epidermal keratinocytes is regulated by interferon gamma and is mediated by intercellular adhesion molecule-1 (ICAM-1). J Exp Med 167:1323-1340 Giger U, Boxer LA, Simpson PJ, Lucchesi BR, Todd RF (1987) Deficiency of leukocyte surface glycoproteins Mol, LFA-1, and Leu M5 in a dog with recurrent bacterial infections: an animal model. Blood 69:1622-1630 Griffiths CEM, Voorhees JJ, Nickoloff BJ (1989) Characterization of intercellular adhesion molecule-1 and HLA-DR expression in normal and inflamed skin: modulation by recombinant gamma interferon and tumor necrosis factor. J Am Acad Dermatol 20:617-629 Groves RW, Ross E, Barker JNWN, Ross JS, Camp RDR, MacDonald DM (1992) Effect on in vivo interleukin-1 on adhesion molecule expression in normal human skin. J Invest Dermatol 98 : 384-387 Heald P, Buckley P, Gilliam A, Perez M, Knobler R, Kacinski B, Edelson R (1992) Correlations of unique clinical, immunotypic, and histologic findings in cutaneous gamma/delta T-cell lymphoma. J Am Acad Dermatol 26 : 865-870 Hynes RO (1987) Integrins: a family of cell surface receptors. Cell 48 : 549-554 Kashihara-Sawami M, Norris DA (1992) The state of differentiation of cultured human keratinocytes determines the level of intercellular adhesion molecule- 1 (ICAM- 1) expresseion induced by gamma interferon. J Invest Dermatol 98:741-747 Konter U, Kellner I, Klein E, Kaufmann R, Mielke V, Sterry W (1989) Adhesion molecule mapping in normal human skin. Arch Dermatol Res 281:454-462 Lever WF, Schaumburg-Lever G (1990) Mycosis fungoides. Histopathology of the skin, 7th edn. JB Lippincott, Philadelphia, pp 819-830 Marlin SD, Springer TA (1987) Purified intercellular adhesion molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen-1 (LFA-1). Cell 51:813-819 Marx JL (1989) New family of adhesion proteins discovered. Science 243 : 1144 Mason DY, Cordell J, Brown M, Palleson G, Ralfldaer E, Rothbard J, Crumpton M, Gatter KC (1989) Detection of T cells in paraffin embedded tissue using antibodies against a sequence from the CD3 antigen. J Clin Pathol 42:1194-1200 Mitra R, Barker J, Stoof T, Stoolman L, Nickoloff BJ (1991) Comparison of lymphocyte adhesion to keratinocytes (KC), fibroblasts (FB), melanocytes (MC): CD-18 (LFA-1) dependent and independent pathways. Clin Res 39 : 234A Moore PF, Ellingsworth LR, Toedter GP (1986) Development and characterization of a monoclonal antibody to class II MHC antigens in rhesus macaques. Tissue Antigens 28 : 88-89 Moore PF, Rossitto PV, Danilenko DM (1990) Canine leukocyte integrins: characterization of a CD 18 homologue. Tissue Antigens 36:211-220 Moore PF, Olivry T, Naydan DK (1994) Canine cutaneous epitheliotropic lymphoma (mycosis fungoides) is a proliferative disorder of CD8+ T-cells. Am J Pathol 144:421-429 Nash RA, Scherf U, Storb R (1991) Molecular cloning of the CD3 epsilon subunit of the T-cell receptor/CD3 complex in the dog. Immunogenetics 33:396-401 Nickoloff B J, Lewinsohn DM, Butcher EC (1987) Allogeneic peripheral blood mononuclear leukocytes prominently adhere to gamma interferon treated cultured keratinocytes. Am J Pathol 9:413-418 Nickoloff B J, Griffiths CEM, Baadsgard O, Voorhees JJ, Hanson CA, Cooper KD (1989) Markedly diminished epidermal keratinocyte expression of intercellular adhesion molecule-1 (ICAM-1) in Sezary syndrome. JAMA 261:2217-2221 Picker LJ, Weiss LM, Medeiros LJ, Wood GS, Warnke RA (1987) Immunophenotypic criteria for the diagnosis of non-Hodgkin's lymphoma. Am J Pathol 128 : 181-201 Pincelli C, Girolomoni G, Benassi L (1986) Expression of T6 antigen on the epidermal keratinocytes in various dermatoses (letter). Br J Dermatol 115 : 512-513

192 Rothlein R, Dustin ML, Marlin SD, Springer TA (1986) A human intercellular adhesion molecule (ICAM-1) distinct from LFA1. J Immunol 137 : 1270-1274 Smith CW, Entman ML, Lane CL, Beaudet AL, Ty TI, Youker K, Hawkins HK, Anderson DC (1991) Adherence of neutrophils to canine cardiac myocytes in vitro is dependent on intercellular adhesion molecule-1. J Clin Invest 88 : 1216-1223 Springer TA (1990) Adhesion receptors of the immune system. Nature 346 : 425-434 Staunton DE, Dustin ML, Springer TA (1989) Functional cloning of ICAM-2, a cell adhesion ligand for LFA-1 homologous to ICAM-1. Nature 339 : 61-63 Sterry W, Mielke V, Konter U, Kellner I, Boehncke WH (1992) Role of beta-1 integrins in epidermotropism of malignant T cells. Am J Pathol 141 : 855-860

Vejlsgaard GL, Ralfkiaer E, Aunstrop C, Czajkowski M, Marlin SD, Rothlein R (1989) Kinetics and characterization of intercellular adhesion molecule-1 (ICAM-1) expression on keratinocytes in various inflammatory skin lesions and malignant cutaneous lymphoma. J Am Acad Dermatol 20 : 782-790 Walder EJ, Gross TL (1992) Cutaneous lymphoma. In: Gross TL, Ihrke PJ, Walder EJ (eds) Veterinary dermatopathology. A macroscopic and microscopic evaluation of canine and feline skin diseases. Mosby Year Book, St Louis, pp 476-482 Willemze R, Dijkstra A, Nugteren WM (1985) Intercellular distribution of HLA-DR and T6 antigens in the epidermis of various dermatoses (letter). Br J Dermatol 113 : 802-804 Williams F, Barclay AN (1988) The immunoglobulin superfamily domains for cell surface recognition. Annu Rev Immunol 6 : 381-405 -