Keratin expression in equine normal epidermis and cutaneous papillomas using monoclonal antibodies

J. Comp. Path. 1990Vol. 102

Keratin Expression in Equine Normal Epidermis and Cutaneous Papillomas Using Monoclonal Antibodies M. Hamada*,

T. O y a m a d a ~ , H. Y o s h l k a w a ~ , T . Y o s h l k a w a ~ and C. Itakura*

*Department of Comparative Pathology, Faculty of VeterinaryMedicine, Hokkaido University, Sapporo 060, Japan and ~Department of VeterinaryPathology, School of Veterina~),Medicine and Animal Sciences, Kitasato University, Towada 034, Aomori, Japan Summary Keratin expressions in normal equine epidermis and experimentally induced equine papillomas were studied by immunohistochemical methods with three different human cytokeratin monoclonal antibodies, 3413B4 (directed against component 1), 3413E12 (directed against components 1, 5, 10, 11) and 35[3H11 (directed against component 8). Staining patterns with 3413B4 and 3413E12 in the normal equine epidermis did not differ from those in the normal human epidermis. In the early developing papilloma, keratinocytes showed an abnormal suprabasal staining pattern and expressed an additional 56 kD keratin protein detected by 3413E12. In the advanced papilloma, cytolytic cells in the outer spinous and the granular layers did not stain positively with any of the three antibodies used. In both early and advanced papillomas, the expression of high molecular weight keratin proteins, as detected by 3413B4and 34~E12, did not correlate with the degree of keratinization. By electron microscopy, keratinocytes in the advanced papilloma showed a marked decrease of tonofibrils and desmosome-tonofilament complex. These alterations may result t?om an abnormality in both proliferation and functional terminal differentiation of keratinocytes in the papilloma. There were obvious differences in staining patterns with 3513H11 between the normal human and equine epidermis; 54 kD keratin protein was expressed in suprabasal layers of the equine normal and papillomatous epidermis. Thus, this keratin protein may be regarded as a "permanent" marker for the equine epidermis. Introduction Intermediate-sized filaments (ISF) are identified in almost all vertebrate cells and form one of the three classes of cytoskeletal proteins along with microfilaments (MF) and microtubules (MT) (Lazarides, 1980; Hashimoto, Eto, Matsumoto and Hori, 1983). ISF can be divided biochemically and immunologically into five types; vimentin, desmin, keratin, glial filament protein and neurofilament protein (Lazarides, 1982; Hashimoto el al., 1983). Among the five types of ISF, keratin type ISF (KISF) is the most complicated in terms of subunit composition (Lazarides, 1980; Cooper, Schermer and Sun, 1985), and a total of 19 human epithelial KISF has been catalogued both in vivo and in 0021-9975/90/040405+ 16 $03.00/0

© 1990AcademicPressLimited

406

M. Harnada et al.

vitro (Moll, Franke, Volc-Platzer and Krepler, 1982). KISF are a group of water-insoluble proteins (40 to 70 kD) that form 10-nm desmosome-associated tonofilaments in the cytoplasm (Franke, Schmid, Osborn and Weber, 1978; Franke, Appelhans, Schmid, Freudenstein, Osborn and Weber, 1979; Hashimoto et al., 1983). They are actively synthesized in the inner, living cell layers of the mammalian epidermis and represent the major differentiation product (Franke et al., 1978, 1979; Hashimoto et al., 1983). Eventually, the keratin forms the bulk of the outer, fully differentiated, dead, cornified layer (Franke et al., 1978, 1979; Hashimoto el al., 1983). The subunit composition of the keratin filaments varies depending on cell type, period of embryonic development, stage of histological differentiation, cellular growth environment (Franke, Weber, Osborn, Schmid and Freudenstein, 1978; Franke et al., 1979; Franke, Schmid, Grund and Geiger, 1982; Moll el al., 1982) and disease state (Thaler, Fukuyama, Inoue, Coan and Epstein, 1978; Kubilus, Baden and McGilvray, 1980; Loning, Staguet, Thivolet and Seifert, 1980; Staguet, Viac and Thivolet, 1981; Wu and Rheinwald, 1981; Moll et al., 1982; Baden, McGilvray, Cheng, Lee and Kubilus, 1987). Four major keratin proteins (50, 56.5, 58 and 65 to 70 kD) have been identified in normal human epidermis (Fucks and Green, 1980; Bowden and Cunliffe, 1981; Moll et al., 1982; Winter, Schweizer and Goerttler, 1983). Recent studies with the antisera to different molecular weight keratin proteins have supported the concept that 50 kD and 58 kD keratin proteins represent "permanent" markers (Tseng, Jarvinen, Nelson, Huang, Woodcock-Mitchell and Sun, 1982; Sun, Eichner, Nelson, Tseng, Weiss, Jarvinen and Woodcock-Mitchell, 1983; Winter et al., 1983), 56.5 kD and 65 to 70kD keratin proteins "keratinizing" markers (Fucks and Green, 1980; Sun et al., 1983; Winter et al., 1983; Eichner, Bonits and Sun, 1984) and 48 kD and 56 kD keratin proteins "hyperproliferative" markers (Sun et al., 1983; Weiss, Eichner and Sun, 1984) for keratinocytes. KISF are not only a keratogenetic precursor but also play a major role in cytoskeletal organization and directly or indirectly influence maintenance of the cell shape and function of keratinocytes (Loren, Knapp, O'Guin and Sawyer, 1983; Winter et al., 1983). This paper describes KISF in equine normal epidermis and cutaneous papillomas by a combination ofimmunohistochemical staining for cytokeratin with three monoclonal human antikeratin antibodies, 3413B4, 3413E12 and 35~3Hll and electron microscopy. Materials and Methods

Tissues

Five 2-year-old Thoroughbred horses were used for this study. These animals were each injected intradermally or subcutaneously in the skin of the muzzle. The inoculum was obtained from 70 naturally occurril,aog " equine cutaneous papillomas which were stored in 50 per cent glycerin buffer at 4 C for a month. The papilloma tissues were washed with saline and ground to make a 10 per cent suspension in saline with penicillin-streptomycin (10000IU per ml). The ground suspension was then centrifuged at 170 g for 5 rain to separate and get rid of the coarse particles. Under general anaesthesia, 0"2 ml of the bacteria-free inoculum per site was injected

Keratin Expression in Equine Epidermis and Papillomas

407

intradermally in the skin of the muzzle (Hamada, Oyamada, Yoshikawa, Yoshikawa and Itakura, 1990). Sixty skin specimens from experimentally induced papillomas were obtained by biopsy. The first detectable papillomas occurred 18 to 27 days after inoculation, and were about 1 mm in diameter and grey or white in colour. Their size and number increased with time. The largest papillomas reached about 10 mm in both diameter and height 39 to 55 days after inoculation. Their surface was papillary or cauliflower-like in appearance and light-brown, grey or white. In cross section of the biopsied tissues, the epidermis was irregularly thickened but sharply demarcated from the dermis. In addition, 10 biopsies from normal equine muzzle skin from three 2-year-old Thoroughbred horses were also examined.

Histopathology Tissue specimens were fixed in 10 per cent neutral buffered formalin, embedded in paraffin wax and cut serially 4 pm thick. Sections were stained with haematoxylin and eosin (HE) and by immunohistochemical methods.

Immunohistochemistry The presence of papillomavirus group-specific antigen was determined by the avidin-biotin technique in deparaffinized and hydrated slides. Non-specific binding was blocked by incubation in 10 per cent normal goat serum in pho,s~hate-buffered saline (PBS) pH 7"6, for 30 min. Slides were incubated for 12 h at 4 C with rabbit anti-bovine papillomavirus-I (1 in 500) disrupted by sodium dodecyl sulphate (Dako Corp.), then rinsed in PBS. The reaction was completed with the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) and diaminobenzidine as the substrate to detect rabbit IgG. Slides were counterstained with light green. Normal rabbit immunoglobulin (IgG fraction) (Dako Corp.) was substituted as a negative control. A bovine cutaneous papilloma was used as a positive control because of the numerous positive nuclei in the ceils of the outer spinous and granular layers. Three mouse monoclonal anti-cytokeratin antibodies were used and their specificities are listed in Table 1. Sections were deparaffinized and hydrated. In experiments designed to unmask the keratin antigens, sections were preincubated for 60 min at 37°C with 100 mM Tris-HC1 (pH 7"5) and 0'005 per cent trypsin (in 100 mM Tris-HC1, pH 8"0). After preincubation, endogenous peroxidase was blocked by incubation in 0"3 per cent H20 2 in methanol. Then, sections were treated with 4 per cent normal horse serum and stained with each anti-cytokeratin antibody (4°C for 12 h) and the avidin-biotin peroxidase complex (ABC) method with the Vectastain ABC kit (Vector laboratories, Burlingame, CA). Antibody affinity was demonstrated with diaminobenzidine as the peroxidase substrate. Slides were examined under a light microscope for the location and amount of brown reaction product in the section. In each case, small samples of unfixed tissues were frozen in liquid nitrogen and stored at - 8 0 ° C immediately after biopsy. From deep-frozen tissue specimens, 6-gm thick sections were cut with a Lipshaw 50 cryostat. These were air-dried, fixed for 10 min in acetone and hydrated in PBS, after which an indirect immunofluorescence (IF) method was used. Each of the three monoclonal antibodies (Table 1) was used individually (4°C for 12 h). The appropriate fluorescein isothiocyanate-conjugated (FITC) goat anti-mouse IgG I or IgM (Enzo Biochem., Inc. cat. No. MA-904) was then applied (4°C for 12 h).

Electron Microscopy Freshly obtained tissues were dissected into small blocks (2 to 3 ram) and fixed in 2 per cent glutaraldehyde in 0" 1 i cacodylate buffer, pH 7"2. Blocks were post-fixed in 1 per cent OsO 4 and embedded in Epon 812. Sections were stained with uranyl acetate and lead citrate and examined with a Hitachi electron microscope, model HU-12A.

4:08

M. Ha-mada e t al.

%

o

0

0

a

o

0

o-

0

o~ o=

"~

~ ~

~

o 0 II

o~

o ~

II

E

K e r a t i n E x p r e s s i o n i n E q u i n e E p i d e r m i s and P a p i l l o m a s

409

Results

Microscopic Findings Early papillomas consisted of mild to marked proliferation of basal cells, mild to moderate acanthosis and mild hyper- and parakeratosis (Fig. 1). No PV group-specific antigens were detected in their keratinocytes. Advanced papillomas were composed of marked proliferation of all the epidermal layers associated with pale prickle cells and ballooning degeneration of granular cells (cytolysis) (Fig. 2). The cytolytic cells usually contained large basophilic keratohyalin granules. Intranuclear inclusion bodies (IIB) were detected in cells of the outer spinous and granular layers and parakeratotic cells.

Immunohistochemistry The staining results in the normal epidermis and papillomatous tissues with monoclonal antibodies against three different types ofcytokeratins, designated 3413B4, 3413E12 and 3513Hll, are summarized in Table 2. There were no qualitative differences in the pattern and specificity ofstainings by IF and ABC methods.

Fig. 1. Early papilloma consisting of proliferated basal cells, acanthosls and para- and hyperkeratosls. HE x 140. Fig. 2, Advanced papilloma. Note large pale prickle cells (cytolysis) in the outer layer of acanthosis, ballooning degeneration of granular cells and intranuelear inclusion bodies in their cells (arrow heads). These cells contain large amounts of keratohyallne granules, Toluidine blue x 440.

410

M. H a m a d a e t al.

1. Normal Epidermis of Muzzle. 34flB4: All the cells of the spinous and granular layers and the cornified layer showed a uniform and moderately strong staining reaction, while the basal cells were negative (Fig. 3). 34flE12: All the cells of all the epidermal layers and the cornified layer showed a uniform and moderately strong staining reaction (Fig. 4). 35flHl1: All the cells of the suprabasal layers showed a uniform and moderately strong staining reaction, while the cornified layer and the basal cells were negative (Fig. 5). The hair-forming cells were intensely positive with each of the three monoclonal antibodies. 2. Papilloma. The staining of keratinocytes in papillomas was determined on the basis of a given monoclonal antibody ability to stain the normal epidermis (Table 2). 34~B4: In the early papillomas, cells of the suprabasal layers were stained moderately to intensely, while the basal cells were negative and the cornified layer showing hyper- and parakeratosis reacted poorly (Fig. 6a). In the advanced papillomas, almost all the cells of the outer spinous and granular layers showing hyperplasia and cytolysis were negative (Fig. 6b). Slightly cytolytic cells in these layers showed a highly disorganized staining pattern consisting of intense, moderate and weak stainability (heterogeneous staining) (Fig. 6c and d). The cornified layer was weakly positive or negative (Fig. 6c and d). No positive reactions were seen in keratinocytes of the more advanced papillomas which had a large number of cells with IIB, markedly proliferated basal cells, cytolytic prickle and granular cells, and hyper- and parakeratotic cornified layer (Fig. 6e). 34flE12: In the early papillomas, suprabasal cells were intensely stained, while the basal cells reacted poorly or negatively (suprabasal staining pattern). However, the basal cells gradually showed either a weak or moderate reaction as the keratinocytes proliferated. The hyper- and parakeratotic cornified layer in these lesions stained weakly or negatively (Fig. 7a). In the advanced papillomas, cytolytic cells in the outer spinous and granular layers revealed a disorganized staining pattern, similar to the pattern obtained with 3413B4. The parakeratotic cornified layer was negative, and the proliferated basal cells were moderately stained (Fig. 7b). 35~Hl1: In the early papillomas, all prickle and granular cells were intensely positive, while the proliferated basal cells showed a negative reaction and the hyper- and parakeratotic cornified layer stained weakly (Fig. 8a). In the advanced papillomas, the cytolytic lesions in the outer spinous and granular layers demonstrated a disorganized stain.i.rt~ pattern, and keratinocytes in these layers showed heterogeneous staining, similar to that with 34~B4 and 3413E12. In this lesion, the hyper- and parakeratotic cornified layer also demonstrated the heterogeneous staining, and the basal cells reacted negatively (Fig. 8b).

Electron Microscopic Findings The cornified layer in the papilloma was greatly thickened and lacked the loosely packed keratin filaments which were present in the normal epidermis.

Keratin Expression in Equine Epidermis and Papillomas

411

Fig, 3.

Normal epidermis stained with 3413B4. Note positive reaction in cells of all tile layers except the basal cell layer. Arrows indicate the dermal-epidermal junction. ABC x 125.

Fig. 4.

Normal epidermis stained with 3413E12. Note positive reaction in cells of the entire epidermis. ABC x 125.

Fig. 5,

Normal epidermis stained with 35!3H 11. Note positive reaction in cells of the suprabasal layers. The cornified layer shows nonspecifie reaction (asterisks). Arrows indicate the dermal-epidermal junction. ABC x 190.

412

M. Hamada

et al.

I -o

+

I ~ +

+

Q

I

i

I

i

0

o

$ 0

e~

+ +

+ + +

+ + +

+

I

"o

+ +

+ +

I H + + +

+ +

+ +

+ + o

+

II + +

II + 0

It

I

Keratin Expression in Equine Epidermis and Papillomas

Fig. 6.

413

Early (a) and advanced (b-e) papillomas stained with 3413B4. (a) Hyperplastic epidermis almost showing the normal staining pattern. Note moderate to intense reaction in the suprabasal layers, but negative reaction in basal cells, ABG x 90. (b) Hyperplastic and cytolytic lesions revealing heterogeneous staining in cells of the outer spinous and granular layers, Note weak reaction in the cornified layer showing parakeratosis. ABC x 90, (c) Note markedly heterogeneous staining keratinocytes of the outer spinous and granular layers. The cornified layer was weakly positive. ABC x 180. (d) Same lesion as (c). Indirect IF × 180. (e) Note almost negative reaction in keratinocytes of the entire layers with hyperplastic and cyto[ytic lesions in the more advanced papilloma comparing with positive reaction in keratinocytes of the suprabasal layers and in the eornified layer in the adjacent normal epidermis. ABC x 90.

414

M. H a m a d a e t a l .

i i~

i

i

ii iI!i!i~!i li!!i!~!i!ifi!!~ili i!ill .....

~i~!~ii'~i!i!,~/~i~i~i@i

i!ii~

!/i!iii~i!:il;il/iiiii! ili;;i!~!i~ilii:ii!~;!ii~ii!i!i;iiiili~i ;iliiiii!iilI

Fig. 7. Early (a) and advanced (b) papillomas stained with 34~E12, (a) Hyperplastic epidermis. Note the intense reaction of cells of the suprabasal layers and weak to moderate reaction of the proliferated basal cells. ABC x 140. (b) Hyperplastic and cytolytie lesions, demonstrating a disorganized staining pattern. Note heterogeneous staining in keratinocytes of the outer spinous and granular layers (arrows), moderate reaction of proliferated basal cells and negative reaction in the cornified layer showing parakeratosis. ABC x 175. Fig. 8. Early (a) and advanced (b) papillomas stained with 3513H11. Note moderate to intense staining of cells of the suprabasal layers. (a) Hyperplastic epidermal lesion. ABC x 210. (b) Hyperplastic and cytolytic lesions. Indirect IF x 210.

The entire cornified layer, except the most superficial layer, was composed of an extremely dense and amorphous substance. In the granular and parakeratotic layers, nuclei were often filled with viral particles which had a hexagonal profile, ranged from 40 to 45 nm in diameter and frequently showed crystalline arrays. In the granular cells, there were swollen and vacuolated mitochondria and rough endoplasmic reticulum (RER), tonofibrils were absent in their paranuclear areas and tonofilaments were aggregated in the periphery of the

Keratin Expression in Equine Epidermis and Papillomas

415

cytoplasm. In addition, membrane coating granules (MCG) were not observed and large, irregular-shaped, electron-dense keratohyaline granules were evident in the cytoplasm (Fig. 9). In the outer spinous layer, the cytoplasm of the cells with nuclear viral particles contained few tonofibrils, ribosomes, mitochondria and RER. As was observed in the granular cells, various-sized, round electron-dense bodies were scattered in their cytoplasm. Sometimes, round electron-dense bodies containing aggregates of tonofilaments and small vacuoles were observed in the cytoplasm (Fig. 10). In these prickle cells, there were few desmosomes in the cell membrane and few tonofibrils seen to converge into desmosomes at the alSices of the cell processes (Fig. 11). In addition, intracytoplasmic desmosomes were observed in these cells. Very few, small MCG (about 100 to 300 nm) appeared in the periphery of the cytoplasm. The intercellular spaces were relatively widened. Although no virus particles were seen in basal cells, they were markedly enlarged with involuted nuclear membranes and swollen nucleoli and showed evidence of increased mitosis. In the cytoplasm, tonofilaments were moderately decreased in number and were sometimes present as bundles. Discussion

Recent studies in man have shown that 50 kD and 58 kD keratin proteins are present in cells of all the epidermal layers including those of the relatively undifferentiated basal layer, supporting the concept that they represent "permanent" markers for keratinocytes (Fucks and Green, 1980; Winter et al., 1983; Eichner el al., 1984). On the other hand, 56"5 kD and 65 to 68 kD keratin proteins are associated with more differentiated cells above the basal layer (Tseng el al., 1982). Thus, these latter keratin proteins are regarded as markers for keratinization (Tseng et al., 1982~"~Vinter et al., 1983). In our studies with the 3413B4 and 34~E 12 antibodies, there was no difference between the normal equine and human epidermis. These findings clearly establish that there are similarities in "permanent" and "keratinizating" markers of keratin proteins between the normal human epidermis and normal equine epidermis. Although keratinocytes in the entire epidermis showed a positive reaction with 3413E12 antibody in the normal equine epidermis, keratinocytes in the early papillomas showed an abnormal suprabasal staining pattern with the same antibody recognizing 58 kD keratin protein (Gown and Vogel, 1982, 1984, 1985; Shah, Tabibzadeh and Gerber, 1987) which is present predominantly in the basal cells in the normal human epidermis (Sun and Green, 1978). At least two keratin proteins, 50 kD and 58 kD, are required for keratin filament formation in basal cells of the human epidermis (Lee and Baden, 1976; Steinert, Idler and Zimmerman, 1976; Milstone, 1981). Weiss et al. (1983) developed a monoclonal antibody, designated AE1, which selectively recognizes 50 kD keratin protein and is specific for basal cells of the normal human epidermis and ichthyosis vulgaris epidermis. With this antibody, the suprabasal staining pattern has been observed in various epidermal disease conditions including psoriasis, verruca, seborrhoeic keratosis and actinic

416

Fig. 9.

M. H a m a d a e t al.

A granular cell with prominent nucleoli and numerous viral particles (inset) in the nucleus. Note extensive vacuolation of mitochondria and rough endoplasmic reticulum and the absence of tonofibrils in the paranuclear areas. Arrows indicate an aggregation of tonofilaments in the periphery of the cytoplasm. × 4400, inset x 11 550.

Fig. 10. A prickle cell. Note a mass of electron-dense material (arrow) having fine tonofilaments and small vacuoles in the cytoplasm. There are bundles of disorientated tonofibrils in the peripheral cytoplasm, × 3300. Fig. 11. A prickle cell. Note disappearance of desmosome-tonofilaments complex and aggregation of tonofilaments in perinuclear area, × 5500.

Keratin Expression in Equine Epidermis and Papillomas

417

keratosis (Weiss, Guillet, Freedberg, Far.mer, Small and Weiss, 1983; Winter et al., 1983). As a result, it was pointed out that this staining pattern was not disease-specific, but was related to a hyperproliferative state of keratinocytes (Weiss et al., 1983). The possible mechanisms for such an abnormal suprabasal staining pattern include the loss of keratin proteins or its masking by synthesis of new keratin species in the basal layer and modification ofkeratins leading to the exposure or unmasking of additional sites in cells above the basal layer (Woodcock-Mitchell, Eichner, Nelson aiad Sun, 1982). These reports support our present findings on the suprabasal.staining pattern in the early papillomas, suggesting that the above mechanisms gause some changes in 58 kD keratin protein during the terminal differentiation of keratinocytes in a hyperproliferative state of the equine epidermis. In the present observations with 3413E12 antibody, there was a close relationship between the degree of'proliferative changes of the basal and suprabasal cells in the papilloma and:their positive staining. This antibody recognizes a 56 kD keratin protein, which is not detectable in the normal human epidermis (Gown and Vogel, :1982, 1984, 1985; Weiss et al., 1984; Shah et al., 1987). The 56 kD and 48 kD keratin proteins are also identified in all keratinocytes present in cases of physiological or pathological hyperproliferation (Woodcock-Mitchell et al., ~1982; Eichner et al., 1984). For example, the suprabasal staining pattern was recognized in all benign epidermal diseases positive for 48 kD keratin protein (Weiss et al., 1983). Therefore, the 48 kD and 56 kD keratin proteins are regarded as markers for hyperproliferative human keratinocytes. The present immunoiiistochemical expressions with 3413E12 antibody were almost the same as those seen in the human epidermis (Weiss et al., 1983). It is probable that 56 kD keratin protein is synthesized in keratinocytes of the early papilloma, and thus it may be available as one of molecular markers for a hyperproliferative state in the equine epidermis. Our findings with 3413E12 antibody suggest that keratin expression can be influenced by hyperproliferative changes of keratinocytes in equine papilloma. Acantholytic cells and ballooning degenerative granular ceils in the advanced papilloma showed negative reactions with each of the three antibodies used. This is considered to result from the loss or decrease of high molecular weight (HMW) keratin proteins in such degenerative keratinocytes. The electron microscopy of these cytolytic cells and basal cells, which showed a marked decrease of tonofibrils and disappearance of a desmosome-tonofilaments complex, supports the immunohistochemical findings. In both initial and advanced papillomas, the expressions of HMW keratin proteins with 34~B4 and 34~E12 antibodies did not correlate with the degree ofkeratinization. The same abnormal staining patterns have been observed in different types of human papillomas, resulting from a marked decrease of 67 kD keratin protein (Shah et al., 1987). These fndings suggest that the synthesis of H M W keratin proteins has declined in the more differentiated keratinocytes in the equine papilloma and that a non-species-specific decrease of these proteins occurs in PV infection. Electron microscopically loose fibrous materials were lost in the cornified layer of the present papilloma. This finding suggests a reduction of KISF in the outer spinous and granular layers of the papilloma. In addition, the cornified layer contained electron-dense,

418

Keratin Expression in Equine Epidermis and Papillomas

amorphous materials and keratinocytes of the outer spinous and granular layers also contained large, electron-dense, often irregular-shaped keratohyaline granules and sometimes tonofilaments, These are related to the excessive production or retention of keratohyaline granules (Fulton, Doane and Macpherson, 1970). Thus, the cornified layer in papillomas might be formed by substances derived from keratohyaline granules without KISF. Fulton and co-workers (1970) observed an aberrant form of keratohyaline granules and the loss of tonofilaments in equine papillomas and considered that these alterations resulted from an interference of normal keratin synthesis by equine PV. Furthermore, the present electron microscope findings revealed a marked decrease of MCG in cytolytic keratinocytes. Wilgram and Weistock (1966) found a large number of MCG in a markedly desquamated skin disorder in man and suggested that they were related to a desquamation or an exfoliation of keratin in the cornified layer. Hence, the decrease of MCG in the equine papilloma may be closely related to the retention of keratin. The suprabasal layers in the present study of normal and papillomatous epidermis reacted with 3513Hll antibody. This antibody recognizes only 54 kD keratin protein, which is present in normal human simple epithelia but absent in the normal human epidermis (Gown and Vogel, 1982, 1984, 1985; Shah et al., 1987). Therefore, the 54kD keratin protein may be absent from human epidermis and present in equine epidermis, but it may also be present or absent in many other species which were not examined. It is suggested that this protein, as defined by this monoclonal antibody, be regarded as a "permanent" marker for the equine epidermis, having been observed in both normal and papillomatous epidermis. Cytoskeletal alterations in keratinocytes of papillomas examined here indicate disturbances in their keratin synthesis. Abnormalities of morphological as well as functional differentiation would have occurred in the lesions. References

Baden, H. P., McGilvray, N., Cheng, C. K., Lee, L. D. and Kubilus, J. (1987). The keratin polypeptides of psoriatic epidermis. Journal of Investigative Dermatology, 70, 294-297. Bowden, P. E. and Cunliffe, W.J. (1981). Modification of human prekeratin during epidermal differentiation. Journal of Biochemistry, 199, 145-154. Cooper, D., Schermer, A. and Sun, T-T. (1985). Classification of human epithelia and their neoplasms using monoclonal antibodies to keratins: Strategies, applications, and limitation. Laboratory Investigation, 52, 243-256. Eichner, R., Bonits, P. and Sun, T-T. (1984). Classification of epidermal keratins according to their immunoreactivity, isoelectric point, and mode of expression. Journal of Cell Biology, 98, 1388-1394. Franke, W. W., Appelhans, O., Schmid, E., Freudenstein, C., Osborn, M. and Weber, K. (1979). Identification and characterization of epithelial cells in mammalian tissues by immunofluorescence microscopy using antibodies to prekeratin. Differentiation, 15, 7-25. Franke, W. W., Schmid, E., Grund, C. and Geiger, B. (1982). Intermediate filament proteins in nonfilamentous structures: Transient disintegration and inclusion of subunit protein in granular aggregates. Cell, 30, 103-113. Franke, W. W., Schmid, E., Osborn, M. and Weber, K. (1978). Different intermediate-sized filaments distinguished by immunofluorescence microscopy. Proceedings of the National Academy of Sciences of the United States of America, 75, 5034-5038.

M. Hamada e t al.

419

Franke, W.W., Weber, K., Osborn, M., Schmid, E. and Freudenstein, C. (1978). Antibody to prekeratin: Decoration of tonofilament-like arrays in various cells of epithelial character. Experimental Cell Research, 116, 429-445. Fucks, E. and Green, H. (1980). Changes in keratin gene expression during terminal differentiation of the keratinocyte. Cell, 19, 1033-1042. Fulton, R.E., Doane, F.W. and Macpherson, L.W. (1970). The fine structure of equine papillomas and the equine papillomavirus. Journal of UltrastructureResearch, 30, 328-343. Gown, A.M. and Vogel, A.M. (1982). Monoclonal antibodies to ifitermediate filament proteins of human cells: Unique and cross-reacting antibodies. Journal of Cell Biology, 95, 414--424. Gown, A. M. and Vogel, A. M. (1984). Monoclonal antibodies to human intermediate filament proteins. II. Distribution of filament proteins in normal human tissues. American Journal of Pathology, 114, 309-321. Gown, A. M. and Vogel, A. M. (1985). Monoclonal antibodies to human intermediate filament proteins. III. Analysis of tumors. AmericanJournal of ClinicalPathology, 11, 175-189. Hashimoto, K., Eto, H., Matsumoto, M. and Hori, K. (1983). Anti-keratin monoclonal antibodies: Production, specificities and applications. Journal of Cutaneous Pathology, 10, 529-539. Hamada, M., Oyamad-a, T., Yoshikawa, H., Yoshikawa, T. and Itakura, C. (1990). Histopathological development of equine cutaneous papillomas. Journal of Comparative Pathology 102, 393-403. Kubilus, J., Baden, H. P. and McGilvray, N. (1980). Filamentous protein of basal cell epithelioma: Characteristics in vivo and in vitro. Journal of the National Cancer Institute, 65, 869-875. Lazarides, E. (1980). Intermediate filaments as mechanical integrators of cellular space. Nature, 283, 249-256. Lazarides, E. (1982). Intermediate filaments: A chemically heterogeneous, developmentally regulated class of proteins. Annual Review of Biochemistry, 51, 219-250. Lee, L.D. and Baden, H.P. (1976). Organization of the polypeptide chains in mammalian keratin. Nature (London), 26't, 377-379. Loning, T., Staguet, M.J., Thivolet, J. and Seifert, G. (1980). Keratin polypeptide distribution in normal and diseased human epidermis and oral mucosa. Virchows Archly A PathologicalAnatomy and Histopi~thology,388, 273-288. Loren, W., Knapp, W., O'Guin, M. and Sawyer, R. H. (1983). Rearrangement of the keratin cytoskeleton after combined treatment with microtubule and microfilament inhibitors. Journal of Cell Biolo'gy, 97, 1788-1794. Milstone, L. M. (1981), Isolation and characterization of two polypeptides that form intermediate filaments in bovine esophageal epithelium. Journal of Cell Biology, 88, 317-322. Moll, R., Franke, W. W., Volc-Platzer, B. and Krepler, R. (1982). Different keratin polypeptides in epidermis and other epithelia of human skin: A specific cytokeratin of molecular weight 46 000 in epithelia of the pilosebaceous tract and basal cell epitheliomas. Journal of Cell Biology, 95, 285-295. Shah, K. D., Tabibzadeh, S. S. and Gerber, M. A. (1987). Comparison ofcytokeratin expression in primary and metastatic carcinomas. American Journal of Clinical Pathology, 87, 708-715. Staguet, M.J., Viac, J. and Thivolet, J. (1981). Keratin polypeptide modifications induced by human papillomavirus. Archivesof DermatologicalResearch, 271, 83-90. Steinert, P. M., Idler, W. W. and Zimmerman, S.B. (1976). Self-assembly of bovine epidermal keratin filaments in vitro. Journal of MolecularBiology, 108, 547-567. Sun, T-T., Eichner, R., Nelson, W. G., Tseng, S. C. G., Weiss, R. A., Jarvinen, M. and Woodcock-Mitchell, J. (1983). Keratin classes: Molecular markers for different types of epithelial differentiation. Journal of InvestigativeDermatology, 81, 109-115. Sun, T-T. and Green, H. (1978). The keratin filaments of cultured human epidermal cells. Formation ofintermolecular disulfide bonds during terminal differentiation. Journal of Biological Chemistry, 293, 2053-2060.

490

M. Hamada et aL

Thaler, M. P., Fukuyama, K., Inoue, N., Coan, D. L. and Epstein, W. L. (1978). Two Tris-urea-mercaptoethanol extractable polypeptides found uniquely in scales of patients with psoriasis. Journal of Investigative Dermatology, 70, 38-41. Tseng, S. C. G., Jarvinen, M.J., Nelson, W. G., Huang, J-W., Woodcock-Mitchell, J. and Sun, T-T. (1982). Correlation of specific keratins with different types of epithelial differentiation: Monoclonal antibody studies. Cell, 30, 361-372. Weiss, R.A., Eichner, R. and Sun, T-T. (1984). Monoclonal antibody analysis of keratin expression in epidermal diseases: A 48- and 56-kDalton keratin as molecular markers for hyperproliferative keratinocytes. Journal of Cell Biology, 98, 1397-1406. Weiss, R.A., Guillet, G. Y.A., Freedberg, I.M., Farmer, E.R., Small, E.A. and Weiss, M . M . (1983). The use of monoclonal antibody to keratin in human epidermal disease: Alterations in immunohistochemical staining pattern, Journal of Investigative Dermatology, 81, 224-230. Wilgram, G. F. and Weistock, A. (1966). Advances in genetic dermatology. Archivesof Dermatology, 94, 456-479. Winter, H., Schweizer, J. and Goerttler, K. (1983). Keratin polypeptide composition as a biochemical tool for the discrimination of benign and malignant epithelial lesions in man. Archives of Dermatological Research, 275, 27-34. Woodcock-Mitchell, J , Eichner, R., Nelson, W. G. and Sun, T-T. (1982). Immunolocalization of keratin polypeptides in human epidermis using monoclonal antibodies. Journal of Cell Biology, 95, 580-588. Wu, Y-J. and Rheinwald, J. G. (1981). A new small (40 kD) keratin filament protein made by some cultured human squamous cell carcinomas. Cell, 25, 627-635. [Receivedfor publication, June 23rd, 1989]