- Email: [email protected]
Histopathological and Immunohistochemical Study of Trichoblastoma in the Rabbit
J. Comp. Path. 2017, Vol. 157, 126e135
Available online at www.sciencedirect.com
ScienceDirect www.elsevier.com/locate/jcpa
NEOPLASTIC DISEASE
Histopathological and Immunohistochemical Study of Trichoblastoma in the Rabbit M. K. Kok*, J. K. Chambers*, N. Ushio*, A. Watamori†, Y. Miwa†, H. Nakayama* and K. Uchida* * Laboratory of Veterinary Pathology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo and † Miwa Exotic Animal Hospital, 1-25-5 Komagome, Toshima-ku, Tokyo, Japan
Summary Trichoblastoma is the most common skin tumour in the rabbit. The aim of the present study was to characterize the histological and immunohistochemical features of trichoblastoma in 27 rabbits. Common sites of tumour occurrence were the neck (6/30, 20%), head (5/30, 16.7%), flank (4/30, 13.3%) and hindlimb (4/30, 13.3%). Histologically, rabbit trichoblastoma was categorized into ribbon (10/30, 33.3%), trabecular (8/ 30, 26.7%) and mixed types (12/30, 40%). The tumour tissue showed close interaction with the surrounding stroma where prominent fibroblastic aggregation, known as papillary mesenchymal bodies, was frequently observed (24/30; 80%). Peritumoural stroma of all cases was stained by Alcian blue (at pH 2.5 with weaker staining at pH 1.0). Immunohistochemically, the peripheral palisading basal-type cells of the tumour were positive for cytokeratin (CK) 14 while the inner cells were typically positive for CK17, differing from the immunohistochemical profile of the rabbit epidermis and hair follicle. The present study suggests that uncontrolled embryonic trichogenesis is involved in the development of trichoblastoma in the rabbit. Ó 2017 Elsevier Ltd. All rights reserved. Keywords: immunohistochemistry; papillary mesenchymal bodies; rabbit; trichoblastoma
Introduction Rabbits (Oryctolagus cuniculus) have been gaining popularity as a household pet worldwide (von Bomhard et al., 2007; Schepers et al., 2009; Fisher, 2010; Rooney et al., 2014). von Bomhard et al. (2007) provided a comprehensive retrospective study, in which trichoblastoma was identified as the most common skin tumour in rabbits, accounting for 19.3% of all tumours and tumour-like lesions. The incidence of hair follicle tumours (in particular trichoblastoma) is higher in companion animals, such as dogs, cats and rabbits, than in man, where basal cell carcinoma is predominant (Abramo et al., 1999; Rubin et al., 2005; Bath-Hextall et al., 2007; von Bomhard et al., 2007). Correspondence to: J. K. Chambers (e-mail: [email protected]. ac.jp). 0021-9975/$ - see front matter http://dx.doi.org/10.1016/j.jcpa.2017.06.005
Trichoblastoma, previously referred to as ‘basal cell tumour of epidermal origin’, received its current nomenclature in the latest World Health Organization (WHO) Classification of Tumours of Domestic Animals, due to its histological resemblance to the germinative epithelial cells of the hair follicle during hair follicle morphogenesis (Goldschmidt et al., 1998). Microscopically, trichoblastoma had been described as comprising basal-type epithelial cell proliferation in cords, arranged as palisading or lobulated structures separated by collagenous stroma, forming a well-demarcated tumour lesion without significant invasion of the surrounding tissues (Goldschmidt et al., 1998; Gross et al., 2005; De Vico et al., 2011). In domestic animals, trichoblastoma can be classified microscopically as ribbon, medusoid, solid, granular, trabecular or spindle, based on the tumour cell arrangement and cellular Ó 2017 Elsevier Ltd. All rights reserved.
127
Trichoblastoma in Rabbits
morphology (Goldschmidt and Goldschmidt, 2017). In several cases of canine trichoblastoma, the tumour formation was accompanied by abortive hair papilla formation (‘papillary mesenchymal body’) and preservation of the dermal papilla during tumour development (De Vico et al., 2011; Campos et al., 2014; Mineshige et al., 2014). Despite being the most common cutaneous neoplasm in rabbits, the histopathology of rabbit trichoblastoma has not been well described. The aim of the present study was to perform a comprehensive pathological study on rabbit trichoblastoma through histological and immunohistochemical evaluations of components of the normal rabbit skin, as well as those of trichoblastoma.
Materials and Methods Samples
Thirty surgically resected, formalin-fixed skin tumour samples from 27 rabbits were examined. The clinical history of these animals is given in Table 1. Three rabbits developed trichoblastoma on two occasions, each time involving different regions of the body. These six tumours were evaluated independently (Table 1). Based on microscopical examination, 29 samples were diagnosed as trichoblastoma and one as malignant trichoblastoma (Table 1). Radiographical and ultrasonographical examinations were performed and no metastasis of tumour cells was detected in any case of trichoblastoma. Histopathology
Tissue samples were fixed in a 10% phosphatebuffered formalin solution, processed routinely and embedded in paraffin wax. Sections (4 mm) were stained with haematoxylin and eosin (HE), periodic acideSchiff (PAS), Masson’s trichrome and Alcian blue (pH 1.0 and 2.5). Masson’s trichrome and Alcian blue stains were used to evaluate the histochemical properties of the tumour stroma. PAS stain was used to detect cytoplasmic glycogen in the tumour cells. The tumours were classified according to the WHO International Histological Classification of Tumours of Domestic Animals and its recent update (Goldschmidt et al., 1998; Goldschmidt and Goldschmidt, 2017). Immunohistochemistry
Dewaxed sections were subjected to antigen retrieval and non-specific reactions were blocked by immersing the sections in H2O2 3% in methanol at room temperatur-
Table 1 Clinical information and histological classification Case number
Breed
Age
Trichoblastoma 1 HLE 5y 2* HLE 5y 3 HLE 9y 4 HLE 4y 5 MR 6y5m 6 Mix 7y 7 Mix 8y 8 Mix 9y7m 9 No Data 10 y 4 m 10 No Data 3y 11* MR 7y3m 12* MR 8y 13 ND 8y 14* Mix 5y 15 Mix 5y 16 Mix 6y 17 Mix 5y 18 No Data 12 y 19* HLE 8y 20 HLE 11 y 21 HLE 6y 22 ND 7y6m 23 ND 9y 24 ND 4y 25 JW 8y5m 26* Mix 5y 27 Mix 4y 28 No Data 6 y 6 m 29 No Data 12 y Malignant trichoblastoma 30 Mix 7y5m
Sex
Location
Histological subtype
F M F F M F F M M F FN FN M F F M M F M M M F FN F M F M M M
Dorsal trunk Neck Inguinal Neck Neck Shoulder Neck Shoulder Neck Dorsal trunk Axilla Flank Dorsal trunk Abdomen Head Hindlimb Head Forelimb Head Inguinal Neck Hindlimb Flank Base of tail Flank Hindlimb Head Head Hindlimb
Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed
M
Flank
Mixed
HLE, Holland lop ear; JW, Jersey wooly; MR, mini Rex; ND, Netherland dwarf; F, female; M, male; FN, neutered female. * Trichoblastoma occurred twice in three rabbits (cases 2 and 19; 11 and 12; and 14 and 26, each derive from the same animal).
e for 5 min followed by incubation in 8% skimmed milk at 37 C for 30 min. The sections were incubated with one of the primary antibodies (Table 2) at 4 C overnight and then washed three times with Trisbuffered saline (TBS). Sections were then incubated with Dako EnVision+ System horseradish peroxidase (HRP)-labelled anti-mouse polymer (Dako, Tokyo, Japan) at 37 C for 40 min. After washing three times with TBS the sections were subjected to chromogen treatment with 0.05% 3,30 -diaminobenzidine and 0.03% H2O2 in TriseHCl buffer, and counterstained with Mayer’s haematoxylin. For negative controls, sections were incubated with TBS instead of primary antibodies, while the panel of antibodies in the present study were used as irrelevant reagent controls to verify the specificity of each antibody (RamosVara and Miller, 2014).
M.K. Kok et al.
128
Table 2 Primary antibodies and protocols for immunohistochemistry Antibody specificity
Type
Dilution
CK AE1/AE3
mAb (AE1/AE3)
RTU
CK5/6
mAb (D5/16 B4)
1 in 100
CK14
mAb (NCL-LL002)
1 in 50
CK17
mAb (E3)
1 in 40
CK18
mAb (Ks18.04)
RTU
CK19
mAb (b170)
RTU
mAb (V9)
1 in 200
Vimentin
Antigen retrieval
Source
HIER (Citrate buffer, pH 6.0), 121 C, 10 min HIER (Dako Target Retrieval Solution, pH 9.0), 121 C, 10 min HIER (Citrate buffer, pH 6.0), 121 C, 10 min HIER (Dako Target Retrieval Solution, pH 9.0), 121 C, 10 min PIER (Proteinase K), room temperature, 30 min PIER (Proteinase K), room temperature, 30 min HIER (Citrate buffer, pH 6.0), 121 C, 10 min
Dako, Tokyo, Japan Dako Leica Biosystems, UK Dako Progen Biotechnik, Frankfurt, Germany Leica Biosystems Dako
CK, cytokeratin; HIER, heat-induced epitope retrieval; mAb, monoclonal antibody; PIER, proteolytic-induced epitope retrieval; RTU, readyto-use.
Scoring
The intensity of Alcian blue (pH 2.5) staining in the tumour stroma and of vimentin expression in the papillary mesenchymal bodies (PMBs) was graded as: , negative; +, weakly positive; ++, moderately positive; and +++, strongly positive. The tumour stroma in all cases showed faint positive staining with Alcian blue stain at pH 1.0 and scoring was not applicable. The intensity of PAS staining and immunohistochemical labelling other than for vimentin were graded as: , 0%; +, 1e10%; ++, 11e50%; +++, 51e75%; and ++++,
76e100% positive tumour cells. The immunohistochemical scoring was performed by consensus of two Japanese College of Veterinary Pathologists boardcertified pathologists (KU, JKC) and MKK. Any disagreements were reviewed together to reach a consensus score.
Results Tumour Occurrence
Clinical information is shown in Table 1. The 27 rabbits included 14 males and 13 females aged from 3 to
Table 3 Histochemical and immunohistochemical characteristics of normal rabbit skin Tissue
Stain AB
Interfollicular epidermis Basal layer Suprabasal layer Stratum corneum Hair follicle Infundibulum Isthmus * Suprabulbar * Bulb Matrix cell Dermal papilla + Sebaceous gland Clear cells Basal cells
Immunohistochemistry PAS
CK AE1/AE3
CK5/6
CK14
CK17
CK18
CK19
Vimentin
+ + +
+ +
+ +†
+
+ +
+ + + +
+ + + +
+ + + +†
+ + +
+‡ +‡
+
+ +
+
+ +
+
AB, Alcian blue (pH 2.5); CK, cytokeratin; PAS, periodic acideSchiff. * Dermal lining of the suprabulbar and isthmus regions was moderately positive for Alcian blue at pH 2.5 and weakly positive at pH 1.0. † Positive in the basal layer of epidermis and outer root sheath of the hair follicle. In addition, expression was also seen in the suprabasal layer of the hyperplastic epidermis and cortical cells above the anagen hair bulb. ‡ Positive in few cells in the outer root sheath of the isthmus and suprabulbar region.
Trichoblastoma in Rabbits
129
Fig. 1. Histological and immunohistochemical (IHC) features of rabbit epidermis and hair follicle. (A) Outer root sheath is PAS positive. (B) The dermal lining of the hair follicles is Alcian blue (pH 2.5) positive. (C) The basal and suprabasal layers of the epidermis are strongly positive for CK5/6. IHC. (D) The basal and suprabasal layers of the hyperplastic epidermis are positive for CK14. IHC. (E) The basal layer of the epidermis is positive for CK17. IHC. (F) Some cells in the dermal papilla are positive for vimentin. IHC. (G) The outer root sheath of the hair follicle is positive for CK5/6. IHC. (H) The cortical cells arising from matrix cells of the anagen bulb are positive for CK14. IHC. (I) The outer root sheath of the hair follicle is positive for CK17. IHC. (J) The outer root sheath of the hair follicle is partly positive for CK19. IHC. (K) The dermal papilla is positive for vimentin. IHC. Bar, 40 mm.
12 years (median 7 years). Multiple occurrence of trichoblastoma was observed in three rabbits at different times and involving different areas of the body. The breeds included Holland lop ear (6/27; 22.2%), Netherland dwarf (4/27; 14.8%), mini Rex (2/27; 7.4%) and Jersey wooly (1/27; 3.7%). More than half were either mixed breed or the breed was not known (14/ 27; 51.9%). The tumours arose on the neck (6/30, 20%), head (5/30, 16.7%), flank (4/30, 13.3%), hindlimb (4/30, 13.3%), dorsal trunk (3/30, 10%), shoulder (2/30, 6.7%), inguinal region (2/30, 6.7%), abdomen (1/30, 3.3%), forelimb (1/30, 3.3%), axilla (1/30, 3.3%) or base of the tail (1/30, 3.3%). Histochemical and Immunohistochemical Profile of Normal Rabbit Skin
The histochemical and immunohistochemical features of normal rabbit skin are summarized in Table 3. PAS staining was observed of the outer root sheath of the suprabulbar and isthmus segments of the hair follicle in a coarsely, granular pattern (Fig. 1A), while the dermal lining of these two segments was moderately positive for Alcian blue at pH 2.5 (Fig. 1B) and weakly positive at pH 1.0. The dermal papilla of the hair follicle was strongly positive for Alcian blue at pH 1.0 and 2.5. In the epidermis, cytokeratin (CK) AE1/AE3 was expressed in all layers. CK5/6 was expressed in the basal and suprabasal layers (Fig. 1C). CK14 was expressed in the basal layer as well as the suprabasal layer of the hyperplastic epidermis (Fig. 1D). CK17 was expressed in the basal layer (Fig. 1E). Vimentin
was expressed by some cells in the underlying dermal papilla (Fig. 1F). In the hair follicles, CK AE1/AE3 was expressed in all layers, except for the hair matrix and dermal papilla cells. CK5/6 was expressed in the outer root sheath of the hair follicle (Fig. 1G). CK14 was expressed in the outer root sheath and occasionally in the cortical cells arising from the hair matrix of the anagen hair bulb (Fig. 1H). CK17 was expressed in the outer root sheath of the hair follicle (Fig. 1I). CK19 was infrequently expressed in the outer root sheath of the hair follicle (Fig. 1J). Vimentin was expressed by the dermal papilla cells within the hair bulb (Fig. 1K). In the sebaceous glands, the basal cells expressed CK AE1/AE3, CK5/6, CK14 and CK17, while the differentiated clear cells expressed CK AE1/AE3 and CK14. In the present study, the expression of CK18 was not detected in the epidermis and its adnexa. Histological Features of Trichoblastoma
The histological features of the present cases are summarized in Table 4. Almost all tumours were diagnosed as trichoblastoma, except for one case, which was diagnosed as malignant trichoblastoma. Tumour types were ribbon (10/30, 33.3%), trabecular (8/30, 26.7%) or mixed (12/30, 40.0%). In general, the tumour masses were well circumscribed, but some dermal masses occasionally extended into the subcutaneous tissue. In the ribbon type (cases 1e10), tumour cells were arranged in thin, branching cords of two to four cell layers (Fig. 2A). In the trabecular
M.K. Kok et al.
130
Table 4 Histological and immunohistochemical features of rabbit trichoblastomas Case number
Histological subtype
Trichoblastoma 1 Ribbon 2* Ribbon 3 Ribbon 4 Ribbon 5 Ribbon 6 Ribbon 7 Ribbon 8 Ribbon 9 Ribbon 10 Ribbon 11* Trabecular Trabecular 12* 13 Trabecular 14* Trabecular 15 Trabecular 16 Trabecular 17 Trabecular 18 Trabecular 19* Mixed 20 Mixed 21 Mixed 22 Mixed 23 Mixed 24 Mixed 25 Mixed 26* Mixed 27 Mixed 28 Mixed 29 Mixed Malignant trichoblastoma 30 Mixed
Presence of PMBs†
AB‡
PASx
Immunohistochemistry CK AE1/AE3x
CK5/6x
CK14x
CK17x
CK18x
CK19x
Vimentink
+ + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + ++ + ++ + ++ + + + + ++ + + + +
+ ++ ++ ++ + + + ++ + ++ ++ + ++ + ++ ++ ++
++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++
++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++
++ + + + +++ + + + ++ + ++ + ++ ++ +++ ++ ++ + + +++ + + ++ + ++ ++ ++ ++ ++
++ + ++ + ++ ++ ++ ++ + ++++ +++ ++ +++ ++ ++ +++ ++++ ++ ++++ + ++ ++ ++ ++++ ++++ ++ ++ ++ +++
+ + + +
+++ +++ +++ +++ +++ +++ +++ +++ ND +++ +++ +++ ND +++ +++ +++ +++ +++ ND +++ ND +++ ND +++ ND +++ +++ +++ +++
+
++
++
++++
++++
+
++++
+++
AB, Alcian blue pH 2.5; CK, cytokeratin; ND, not done (PMBs were absent in these cases); PAS, periodic acideSchiff; PMBs, papillary mesenchymal bodies. * Trichoblastoma occurred twice in three rabbits (cases 2 and 19; 11 and 12 and 14 and 26). † Presence of PMBs in the peritumoural stroma: , absent; +, present. ‡ Tumour cells were negative for Alcian blue (pH 1.0 and 2.5) while the tumour stroma showed positive staining. Qualitative scoring of tumour stroma for Alcian blue at pH 2.5: , negative; +, weak positive; ++, moderate positive; +++, strong positive. At pH 1.0, tumour stroma of all cases showed faint positive staining. x Percentage of positive tumour cells: , negative; +, 1e10%; ++, 11e50%; +++, 51e75%; ++++, 76e100%. k Tumour cells were negative for vimentin expression while the tumour stroma and PMBs showed positive expression. The positive labelling of vimentin in PMBs was evaluated qualitatively and categorized as: , negative; +, weak positive; ++, moderate positive; +++, strong positive.
type (cases 11e18), tumour cells proliferated in a lobular or fascicular pattern with minimal surrounding connective tissue stroma (Fig. 2B). In the mixed type (cases 19e29), both trabecular and ribbon patterns of tumour proliferation were observed. Multifocal clusters of small-sized stromal cells with dense basophilic nuclei and indistinct cytoplasm, termed as ‘papillary mesenchymal bodies’ (PMBs) (Brooke et al., 1989), were observed within the peritumoural stroma in the majority of cases (24/30, 80%; cases 1e8, 10e12, 14e18, 20, 22, 24 and 26e30) (Fig. 2AeC). The PMBs were stained blue with Mas-
son’s trichrome stain. In case 30, the trichoblastoma consisted of poorly demarcated, multiple expansive lobules of basal-type cells, proliferating in a mixture of ribbon, trabecular and solid patterns. In the solid areas, the nests of densely packed tumour cells were in association with the overlying epidermis and accompanied by abundant dermal papillae beneath the tumour lobules (Fig. 2D). The tumour cells in solid nests exhibited a higher degree of cellular atypia and continually expanded along the long axis of epidermis with frequent infiltration into the deeper dermis. Tumour cells were observed at the lateral and deep
Trichoblastoma in Rabbits
131
Fig. 2. Histological features of rabbit trichoblastoma. (A) Ribbon type. Tumour cells are arranged in thin branching cords with the presence of PMBs (arrows) in the stroma. HE. Bar, 100 mm. (B) Trabecular type. Tumour cells are arranged in a lobular pattern and prominent PMBs are observed (arrows) at the periphery of the lobules. HE. Bar, 100 mm. (C) Trabecular type. Higher magnification of a PMB consisting of cells with sparse cytoplasm and prominent oval-shaped nucleus (arrow) at the periphery of the tumour lobule. HE. Bar, 25 mm. (D) Malignant trichoblastoma. The tumour tissue (below) is contiguous with the overlying epidermis. HE. Bar, 100 mm. (E) Mixed type. Trabecular (left) and ribbon (right) types. Basal-type cells of the ribbon type exhibit PAS-positive granules. Bar, 50 mm. (F) Trabecular type. Tumour parenchyma is surrounded by Alcian blue (pH 2.5)-positive stroma. Bar, 100 mm.
margins. Based on the cellular pleomorphism and invasive growth, case 30 was diagnosed as a malignant trichoblastoma. Weak to moderate PAS-positive coarse granules were observed in the cytoplasm of the tumour cells of ribbon (4/10, 40%; cases 1, 2, 6 and 10), trabecular (8/8, 100%; cases 11e18) and mixed types (6/12, 50%; cases 19, 20, 22, 23, 26 and 30), respectively (Fig. 2E). All tumours in the present study possessed peritumoural stroma, which was weakly to moderately positive for Alcian blue at pH 2.5 (Fig. 2F), but faintly positive at pH 1.0. Immunohistochemical Features of Trichoblastoma
The immunohistochemical features of rabbit trichoblastoma are summarized in Table 4. All tumours
in the present study showed uniform and intense cytoplasmic expression of CK AE1/AE3 and CK5/6 (Fig. 3A). Palisading basal cells displayed moderate to strong cytoplasmic expression of CK14 in ribbon, trabecular and mixed types (Fig. 3B). In contrast, the inner cells of these tumours were positive for CK17 (Fig. 3C). The difference in CK17 expression between tumour types was attributed to the proportion of inner cells within the lesions examined. Only four cases (13, 16, 17 and 23) showed few scattered immunopositive areas for CK19, which were localized to the tumour parenchyma with differentiation towards luminal structures (Fig. 3D). CK18 was not expressed in any case. Intense immunoreactivity to vimentin was observed in the PMBs, while the tumour
M.K. Kok et al.
132
Fig. 3. Immunohistochemical features of rabbit trichoblastoma. (A) Ribbon type. Strong CK5/6 expression in the tumour, surrounded by CK5/6-negative dense PMBs. Bar, 100 mm. (B) Trabecular type. Strong CK14 expression in palisading basal-type tumour cells at the periphery of the tumour. Bar, 100 mm. (C) Mixed type. CK17 expression in the inner cells of the tumour. Bar, 100 mm. (D) Trabecular type. CK19 expression in a few parts of the tumour. Bar, 100 mm. (E) Trabecular type. Vimentin is expressed strongly in the PMBs (arrows) and moderately in the tumour stroma (arrowhead). Bar, 50 mm. (F) Malignant trichoblastoma. Inward invagination of CK17-positive tumour cells from the epidermis to form tumour lobules. Bar, 100 mm.
stroma showed a weak to moderate reaction (Fig. 3E). In the malignant trichoblastoma, a similar immunohistochemical pattern for all markers was observed, except for some areas of dense and solid tumour cell nests. In these areas, basal-type cells at the periphery of the lobules were weakly positive for CK14. In contrast, strong CK17 expression was observed, extending from the overlying epidermis and branching deeply towards the central part of the tumour lobules (Fig. 3F).
Discussion Rabbit trichoblastomas in the present study were located mainly on the neck, head, flank and dorsal
trunk. No breed or sex predilection was observed. Multiple occurrence of trichoblastoma was observed in three rabbits, suggesting the predisposition of rabbits towards the development of this tumour. Histopathologically, the trichoblastomas were of ribbon, trabecular or mixed type, as observed in equivalent canine and feline tumours. These patterns may be accounted for by the intensity of collagenous stromal proliferation. For instance, increased stromal elements in the ribbon type may limit the proliferation of tumour cells, resulting in their cord-like arrangement, while less stromal tissue in the trabecular type allows ample space for the tumour cells to proliferate, resulting in lobular, fascicular or nodular structures (De Vico et al., 2011).
Trichoblastoma in Rabbits
In the present study, weak to moderate PAS staining of the tumour cells was found in 18 (60%) of 30 trichoblastomas, which suggests occasional differentiation of the tumour cells to those of the outer root sheath. In addition, both the peritumoural stroma and the dermal lining of the normal hair follicle segments exhibited stronger Alcian blue intensity at pH 2.5 as compared with pH 1.0, suggesting that hyaluronic acid may be the main acid mucin in these structures. The histochemical findings highlight a possible origin of trichoblastoma from the suprabulbar and isthmus regions of the hair follicle in rabbits. PMBs are a histopathological feature characterized by distinct clusters of specialized fibroblasts in the peritumoural stroma, and these structures closely resemble the dermal papilla (Brooke et al., 1989; Campos et al., 2014). In man, PMBs have been identified in the lesions of trichoepithelioma and trichoblastic fibroma (Brooke et al., 1989; Altman et al., 1995; du Toit et al., 2016), but in animals PMBs have been described only in a few cases of canine trichoblastoma (Campos et al., 2014; Mineshige et al., 2014). In the present study, we observed abundant PMBs in the majority of the tumours, implying that aberrant dermal papilla formation is more prominent in rabbit trichoblastomas than in those arising in other species. Although immunohistochemistry (IHC) is rarely required to reach a definitive diagnosis of trichoblastoma, CK profiling is useful to determine the cellular origin and the differentiation of trichoepithelial tumours (Demirkesen et al., 1995; Ohnishi and Watanabe, 1999; Walter, 2000; Kato et al., 2007; Campos et al., 2014). CK14 is a basal cell marker of epidermis and hair follicle, while CK18 and CK19 are expressed in the luminal cells of apocrine glands (Walter, 2000; Kato et al., 2007; Bragulla and Homberger, 2009). CK17 is expressed in the outer root sheath of hair follicles (Chu and Weiss, 2002; Bragulla and Homberger, 2009). In man, CK17 and CK19 are also known as epidermal stem cell markers and, in particular, the expression of CK17 has been associated with ectodermal placode formation during the early stage of hair follicle morphogenesis (McGowan and Coulombe, 1998; Bragulla and Homberger, 2009). As epithelial tumours are known to conserve the CK associated with their cell of origin (Moll et al., 2008), CK markers were used to investigate the nature of rabbit trichoblastoma. In the present study, CK AE1/AE3 and CK5/6, which are general markers for epithelial tissues, were expressed diffusely in the trichoblastomas as well as in the normal epidermis and hair follicles.
133
In contrast, CK14 and 17 were localized within different tumour compartments, while their coexpression was observed in the normal rabbit epidermis and outer root sheath of the hair follicle. In the malignant trichoblastoma, the inward invagination of CK17-positive tumour cells from the epidermis was observed (Fig. 2F). This lesion was morphologically similar to the appearance of normal hair follicle formation in fetal human and murine skin, where there is downward expansion of CK17positive cells from the placode that eventually forms the primary hair germ (McGowan and Coulombe, 1998; Coolen et al., 2010). Differential labelling of CK14 and CK17, including the observed invagination of CK17positive tumour cells from the epidermis (case 30) and the presence of peritumoural PMBs, which resembled dermal condensation in hair follicle morphogenesis, suggests that the development of rabbit trichoblastoma may mimic the intermediate stage of embryonic hair follicle formation. A close interaction between tumour cells of trichoblastoma and adjacent dermal components occurred, leading to the aggregation of dermal mesenchymal cells and increased mucinous substance around the tumour lobules. Increased dermal cellular densities, PMB formation and acidic mucin production simulate the embryonic dermal layer, and may provide an optimal microenvironment for continuous development of an epithelial tumour (Jahoda, 1992; Yang and Cotsarelis, 2010). At the later stage, the abundant dermal fibroblasts produce sufficient collagen fibres, which eventually limit the space available for tumour cell differentiation, and, as mentioned earlier, this may contribute to the different proliferative pattern and well-demarcated nature of trichoblastoma. The significance of CK19 expression in the rabbit hair follicle and in trichoblastomas is not clear, as the expression was infrequently observed and, if any, the reactivity was weak. Consistent with a previous study of a canine trichoblastoma (Campos et al., 2014), expression of vimentin in PMBs, as well as in the dermal papillae of the hair follicle, was observed in the present study. In addition, PMBs were frequently localized within the peritumoural stroma and were stained blue with Masson’s trichrome stain. Taken together, these findings suggest that the PMBs may be derived from the germ of the dermal papilla. In conclusion, the present study has described the histopathological features and immunohistochemical profile of rabbit trichoblastoma as well as normal rabbit skin. Similar to findings in other
M.K. Kok et al.
134
species, rabbit trichoblastoma has an immunohistochemical resemblance to the embryonic hair follicle. In addition, PMBs observed within the peritumoural stroma are an important histological feature of rabbit trichoblastoma and these structures resemble embryonic dermal condensation during hair follicle development.
Acknowledgments The authors are grateful to Ms. S. Kato for invaluable technical assistance. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of Interest Statement The authors declare that there are no potential conflicts of interest related to publication of this research.
References Abramo F, Pratesi F, Cantile C, Sozzi S, Poli A (1999) Survey of canine and feline follicular tumours and tumourlike lesions in central Italy. Journal of Small Animal Practice, 40, 479e481. Altman DA, Mikhail GR, Johnson TM, Lowe L (1995) Trichoblastic fibroma: a series of 10 cases with report of a new plaque variant. Archives of Dermatology, 131, 198e201. Bath-Hextall F, Leonardi-Bee J, Smith C, Meal A, Hubbard R (2007) Trends in incidence of skin basal cell carcinoma. Additional evidence from a UK primary care database study. International Journal of Cancer, 121, 2105e2108. Bragulla HH, Homberger DG (2009) Structure and functions of keratin proteins in simple, stratified, keratinized and cornified epithelia. Journal of Anatomy, 214, 516e559. Brooke JD, Fitzpatrick JE, Golitz LE (1989) Papillary mesenchymal bodies: a histologic finding useful in differentiating trichoepithehomas from basal cell carcinomas. Journal of the American Academy of Dermatology, 21, 523e528. Campos AG, Cogliati B, Guerra JM, Matera JM (2014) Multiple trichoblastomas in a dog. Veterinary Dermatology, 25, 48e49. Chu PG, Weiss LM (2002) Keratin expression in human tissues and neoplasms. Histopathology, 40, 403e439. Coolen NA, Schouten KC, Middelkoop E, Ulrich MM (2010) Comparison between human fetal and adult skin. Archives of Dermatological Research, 302, 47e55. De Vico G, Cataldi M, Maiolino P, Carella F, Beltraminelli S et al. (2011) Fractal pattern of canine trichoblastoma. Analytical and Quantitative Cytology and Histology, 33, 151e157.
du Toit JP, Schneider JW, Visser WI, Jordaan HF (2016) The clinicopathological spectrum of trichoepitheliomas: a retrospective descriptive study. International Journal of Dermatology, 55, 270e277. Demirkesen C, Hoede N, Moll R (1995) Epithelial markers and differentiation in adnexal neoplasms of the skin: an immunohistochemical study including individual cytokeratins. Journal of Cutaneous Pathology, 22, 518e535. Fisher PG (2010) Standards of care in the 21st century: the rabbit. Journal of Exotic Pet Medicine, 19, 22e35. Goldschmidt MH, Dunstan RW, Stannard AA, von Tscharner C, Walder EJ et al. (1998) Tumor with adnexal differentiationHistological Classification of Epithelial and Melanocytic Tumors of the Skin of Domestic Animals (WHO), 2nd Series, Armed Forces Institute of Pathology, Washington DC, pp. 21e32. Goldschmidt MH, Goldschmidt KH (2017) Epithelial and melanocytic tumors of the skin. In: Tumors in Domestic Animals, 5th Edit, DJ Meuten, Ed., Wiley, Ames, pp. 88e141. Gross TL, Ihrke PJ, Walder EJ, Affolter VK (2005) Follicular tumors. In: Skin Diseases of the Dog and Cat: Clinical and Histopathological Diagnosis, 2nd Edit, Blackwell Science, Oxford, pp. 604e640. Jahoda CA (1992) Induction of follicle formation and hair growth by vibrissa dermal papillae implanted into rat ear wounds: vibrissa-type fibres are specified. Development, 115, 1103e1109. Kato K, Uchida K, Nibe K, Tateyama S (2007) Immunohistochemical studies on cytokeratin 8 and 18 expressions in canine cutaneous adnexus and their tumors. Journal of Veterinary Medical Science, 69, 233e239. McGowan KM, Coulombe PA (1998) Onset of keratin 17 expression coincides with the definition of major epithelial lineages during skin development. Journal of Cell Biology, 143, 469e486. Mineshige T, Yasuno K, Sugahara G, Tomishita Y, Shimokawa N et al. (2014) Trichoblastoma with abundant plump stromal cells in a dog. Journal of Veterinary Medical Science, 76, 735e739. Moll R, Divo M, Langbein L (2008) The human keratins: biology and pathology. Histochemistry and Cell Biology, 129, 705e733. Ohnishi T, Watanabe S (1999) Immunohistochemical analysis of cytokeratin expression in various trichogenic tumors. American Journal of Dermatopathology, 21, 337e343. Ramos-Vara JA, Miller MA (2014) When tissue antigens and antibodies get along: revisiting the technical aspects of immunohistochemistry e the red, brown, and blue technique. Veterinary Pathology, 51, 42e87. Rooney NJ, Blackwell EJ, Mullan SM, Saunders R, Baker PE et al. (2014) The current state of welfare, housing and husbandry of the English pet rabbit population. BMC Research Notes, 7, 942. Rubin AI, Chen EH, Ratner D (2005) Basal-cell carcinoma. New England Journal of Medicine, 353, 2262e2269.
135
Trichoblastoma in Rabbits
Schepers F, Koene P, Beerda B (2009) Welfare assessment in pet rabbits. Animal Welfare, 18, 477e485. von Bomhard W, Goldschmidt MH, Shofer FS, Perl L, Rosenthal KL et al. (2007) Cutaneous neoplasms in pet rabbits: a retrospective study. Veterinary Pathology, 44, 579e588. Walter J (2000) A cytokeratin profile of canine epithelial skin tumours. Journal of Comparative Pathology, 122, 278e287.
Yang CC, Cotsarelis G (2010) Review of hair follicle dermal cells. Journal of Dermatological Science, 57, 2e11.
April 26th, 2017 ½ Received, Accepted, June 13th, 2017
Available online at www.sciencedirect.com
ScienceDirect www.elsevier.com/locate/jcpa
NEOPLASTIC DISEASE
Histopathological and Immunohistochemical Study of Trichoblastoma in the Rabbit M. K. Kok*, J. K. Chambers*, N. Ushio*, A. Watamori†, Y. Miwa†, H. Nakayama* and K. Uchida* * Laboratory of Veterinary Pathology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo and † Miwa Exotic Animal Hospital, 1-25-5 Komagome, Toshima-ku, Tokyo, Japan
Summary Trichoblastoma is the most common skin tumour in the rabbit. The aim of the present study was to characterize the histological and immunohistochemical features of trichoblastoma in 27 rabbits. Common sites of tumour occurrence were the neck (6/30, 20%), head (5/30, 16.7%), flank (4/30, 13.3%) and hindlimb (4/30, 13.3%). Histologically, rabbit trichoblastoma was categorized into ribbon (10/30, 33.3%), trabecular (8/ 30, 26.7%) and mixed types (12/30, 40%). The tumour tissue showed close interaction with the surrounding stroma where prominent fibroblastic aggregation, known as papillary mesenchymal bodies, was frequently observed (24/30; 80%). Peritumoural stroma of all cases was stained by Alcian blue (at pH 2.5 with weaker staining at pH 1.0). Immunohistochemically, the peripheral palisading basal-type cells of the tumour were positive for cytokeratin (CK) 14 while the inner cells were typically positive for CK17, differing from the immunohistochemical profile of the rabbit epidermis and hair follicle. The present study suggests that uncontrolled embryonic trichogenesis is involved in the development of trichoblastoma in the rabbit. Ó 2017 Elsevier Ltd. All rights reserved. Keywords: immunohistochemistry; papillary mesenchymal bodies; rabbit; trichoblastoma
Introduction Rabbits (Oryctolagus cuniculus) have been gaining popularity as a household pet worldwide (von Bomhard et al., 2007; Schepers et al., 2009; Fisher, 2010; Rooney et al., 2014). von Bomhard et al. (2007) provided a comprehensive retrospective study, in which trichoblastoma was identified as the most common skin tumour in rabbits, accounting for 19.3% of all tumours and tumour-like lesions. The incidence of hair follicle tumours (in particular trichoblastoma) is higher in companion animals, such as dogs, cats and rabbits, than in man, where basal cell carcinoma is predominant (Abramo et al., 1999; Rubin et al., 2005; Bath-Hextall et al., 2007; von Bomhard et al., 2007). Correspondence to: J. K. Chambers (e-mail: [email protected]. ac.jp). 0021-9975/$ - see front matter http://dx.doi.org/10.1016/j.jcpa.2017.06.005
Trichoblastoma, previously referred to as ‘basal cell tumour of epidermal origin’, received its current nomenclature in the latest World Health Organization (WHO) Classification of Tumours of Domestic Animals, due to its histological resemblance to the germinative epithelial cells of the hair follicle during hair follicle morphogenesis (Goldschmidt et al., 1998). Microscopically, trichoblastoma had been described as comprising basal-type epithelial cell proliferation in cords, arranged as palisading or lobulated structures separated by collagenous stroma, forming a well-demarcated tumour lesion without significant invasion of the surrounding tissues (Goldschmidt et al., 1998; Gross et al., 2005; De Vico et al., 2011). In domestic animals, trichoblastoma can be classified microscopically as ribbon, medusoid, solid, granular, trabecular or spindle, based on the tumour cell arrangement and cellular Ó 2017 Elsevier Ltd. All rights reserved.
127
Trichoblastoma in Rabbits
morphology (Goldschmidt and Goldschmidt, 2017). In several cases of canine trichoblastoma, the tumour formation was accompanied by abortive hair papilla formation (‘papillary mesenchymal body’) and preservation of the dermal papilla during tumour development (De Vico et al., 2011; Campos et al., 2014; Mineshige et al., 2014). Despite being the most common cutaneous neoplasm in rabbits, the histopathology of rabbit trichoblastoma has not been well described. The aim of the present study was to perform a comprehensive pathological study on rabbit trichoblastoma through histological and immunohistochemical evaluations of components of the normal rabbit skin, as well as those of trichoblastoma.
Materials and Methods Samples
Thirty surgically resected, formalin-fixed skin tumour samples from 27 rabbits were examined. The clinical history of these animals is given in Table 1. Three rabbits developed trichoblastoma on two occasions, each time involving different regions of the body. These six tumours were evaluated independently (Table 1). Based on microscopical examination, 29 samples were diagnosed as trichoblastoma and one as malignant trichoblastoma (Table 1). Radiographical and ultrasonographical examinations were performed and no metastasis of tumour cells was detected in any case of trichoblastoma. Histopathology
Tissue samples were fixed in a 10% phosphatebuffered formalin solution, processed routinely and embedded in paraffin wax. Sections (4 mm) were stained with haematoxylin and eosin (HE), periodic acideSchiff (PAS), Masson’s trichrome and Alcian blue (pH 1.0 and 2.5). Masson’s trichrome and Alcian blue stains were used to evaluate the histochemical properties of the tumour stroma. PAS stain was used to detect cytoplasmic glycogen in the tumour cells. The tumours were classified according to the WHO International Histological Classification of Tumours of Domestic Animals and its recent update (Goldschmidt et al., 1998; Goldschmidt and Goldschmidt, 2017). Immunohistochemistry
Dewaxed sections were subjected to antigen retrieval and non-specific reactions were blocked by immersing the sections in H2O2 3% in methanol at room temperatur-
Table 1 Clinical information and histological classification Case number
Breed
Age
Trichoblastoma 1 HLE 5y 2* HLE 5y 3 HLE 9y 4 HLE 4y 5 MR 6y5m 6 Mix 7y 7 Mix 8y 8 Mix 9y7m 9 No Data 10 y 4 m 10 No Data 3y 11* MR 7y3m 12* MR 8y 13 ND 8y 14* Mix 5y 15 Mix 5y 16 Mix 6y 17 Mix 5y 18 No Data 12 y 19* HLE 8y 20 HLE 11 y 21 HLE 6y 22 ND 7y6m 23 ND 9y 24 ND 4y 25 JW 8y5m 26* Mix 5y 27 Mix 4y 28 No Data 6 y 6 m 29 No Data 12 y Malignant trichoblastoma 30 Mix 7y5m
Sex
Location
Histological subtype
F M F F M F F M M F FN FN M F F M M F M M M F FN F M F M M M
Dorsal trunk Neck Inguinal Neck Neck Shoulder Neck Shoulder Neck Dorsal trunk Axilla Flank Dorsal trunk Abdomen Head Hindlimb Head Forelimb Head Inguinal Neck Hindlimb Flank Base of tail Flank Hindlimb Head Head Hindlimb
Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Ribbon Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Trabecular Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed Mixed
M
Flank
Mixed
HLE, Holland lop ear; JW, Jersey wooly; MR, mini Rex; ND, Netherland dwarf; F, female; M, male; FN, neutered female. * Trichoblastoma occurred twice in three rabbits (cases 2 and 19; 11 and 12; and 14 and 26, each derive from the same animal).
e for 5 min followed by incubation in 8% skimmed milk at 37 C for 30 min. The sections were incubated with one of the primary antibodies (Table 2) at 4 C overnight and then washed three times with Trisbuffered saline (TBS). Sections were then incubated with Dako EnVision+ System horseradish peroxidase (HRP)-labelled anti-mouse polymer (Dako, Tokyo, Japan) at 37 C for 40 min. After washing three times with TBS the sections were subjected to chromogen treatment with 0.05% 3,30 -diaminobenzidine and 0.03% H2O2 in TriseHCl buffer, and counterstained with Mayer’s haematoxylin. For negative controls, sections were incubated with TBS instead of primary antibodies, while the panel of antibodies in the present study were used as irrelevant reagent controls to verify the specificity of each antibody (RamosVara and Miller, 2014).
M.K. Kok et al.
128
Table 2 Primary antibodies and protocols for immunohistochemistry Antibody specificity
Type
Dilution
CK AE1/AE3
mAb (AE1/AE3)
RTU
CK5/6
mAb (D5/16 B4)
1 in 100
CK14
mAb (NCL-LL002)
1 in 50
CK17
mAb (E3)
1 in 40
CK18
mAb (Ks18.04)
RTU
CK19
mAb (b170)
RTU
mAb (V9)
1 in 200
Vimentin
Antigen retrieval
Source
HIER (Citrate buffer, pH 6.0), 121 C, 10 min HIER (Dako Target Retrieval Solution, pH 9.0), 121 C, 10 min HIER (Citrate buffer, pH 6.0), 121 C, 10 min HIER (Dako Target Retrieval Solution, pH 9.0), 121 C, 10 min PIER (Proteinase K), room temperature, 30 min PIER (Proteinase K), room temperature, 30 min HIER (Citrate buffer, pH 6.0), 121 C, 10 min
Dako, Tokyo, Japan Dako Leica Biosystems, UK Dako Progen Biotechnik, Frankfurt, Germany Leica Biosystems Dako
CK, cytokeratin; HIER, heat-induced epitope retrieval; mAb, monoclonal antibody; PIER, proteolytic-induced epitope retrieval; RTU, readyto-use.
Scoring
The intensity of Alcian blue (pH 2.5) staining in the tumour stroma and of vimentin expression in the papillary mesenchymal bodies (PMBs) was graded as: , negative; +, weakly positive; ++, moderately positive; and +++, strongly positive. The tumour stroma in all cases showed faint positive staining with Alcian blue stain at pH 1.0 and scoring was not applicable. The intensity of PAS staining and immunohistochemical labelling other than for vimentin were graded as: , 0%; +, 1e10%; ++, 11e50%; +++, 51e75%; and ++++,
76e100% positive tumour cells. The immunohistochemical scoring was performed by consensus of two Japanese College of Veterinary Pathologists boardcertified pathologists (KU, JKC) and MKK. Any disagreements were reviewed together to reach a consensus score.
Results Tumour Occurrence
Clinical information is shown in Table 1. The 27 rabbits included 14 males and 13 females aged from 3 to
Table 3 Histochemical and immunohistochemical characteristics of normal rabbit skin Tissue
Stain AB
Interfollicular epidermis Basal layer Suprabasal layer Stratum corneum Hair follicle Infundibulum Isthmus * Suprabulbar * Bulb Matrix cell Dermal papilla + Sebaceous gland Clear cells Basal cells
Immunohistochemistry PAS
CK AE1/AE3
CK5/6
CK14
CK17
CK18
CK19
Vimentin
+ + +
+ +
+ +†
+
+ +
+ + + +
+ + + +
+ + + +†
+ + +
+‡ +‡
+
+ +
+
+ +
+
AB, Alcian blue (pH 2.5); CK, cytokeratin; PAS, periodic acideSchiff. * Dermal lining of the suprabulbar and isthmus regions was moderately positive for Alcian blue at pH 2.5 and weakly positive at pH 1.0. † Positive in the basal layer of epidermis and outer root sheath of the hair follicle. In addition, expression was also seen in the suprabasal layer of the hyperplastic epidermis and cortical cells above the anagen hair bulb. ‡ Positive in few cells in the outer root sheath of the isthmus and suprabulbar region.
Trichoblastoma in Rabbits
129
Fig. 1. Histological and immunohistochemical (IHC) features of rabbit epidermis and hair follicle. (A) Outer root sheath is PAS positive. (B) The dermal lining of the hair follicles is Alcian blue (pH 2.5) positive. (C) The basal and suprabasal layers of the epidermis are strongly positive for CK5/6. IHC. (D) The basal and suprabasal layers of the hyperplastic epidermis are positive for CK14. IHC. (E) The basal layer of the epidermis is positive for CK17. IHC. (F) Some cells in the dermal papilla are positive for vimentin. IHC. (G) The outer root sheath of the hair follicle is positive for CK5/6. IHC. (H) The cortical cells arising from matrix cells of the anagen bulb are positive for CK14. IHC. (I) The outer root sheath of the hair follicle is positive for CK17. IHC. (J) The outer root sheath of the hair follicle is partly positive for CK19. IHC. (K) The dermal papilla is positive for vimentin. IHC. Bar, 40 mm.
12 years (median 7 years). Multiple occurrence of trichoblastoma was observed in three rabbits at different times and involving different areas of the body. The breeds included Holland lop ear (6/27; 22.2%), Netherland dwarf (4/27; 14.8%), mini Rex (2/27; 7.4%) and Jersey wooly (1/27; 3.7%). More than half were either mixed breed or the breed was not known (14/ 27; 51.9%). The tumours arose on the neck (6/30, 20%), head (5/30, 16.7%), flank (4/30, 13.3%), hindlimb (4/30, 13.3%), dorsal trunk (3/30, 10%), shoulder (2/30, 6.7%), inguinal region (2/30, 6.7%), abdomen (1/30, 3.3%), forelimb (1/30, 3.3%), axilla (1/30, 3.3%) or base of the tail (1/30, 3.3%). Histochemical and Immunohistochemical Profile of Normal Rabbit Skin
The histochemical and immunohistochemical features of normal rabbit skin are summarized in Table 3. PAS staining was observed of the outer root sheath of the suprabulbar and isthmus segments of the hair follicle in a coarsely, granular pattern (Fig. 1A), while the dermal lining of these two segments was moderately positive for Alcian blue at pH 2.5 (Fig. 1B) and weakly positive at pH 1.0. The dermal papilla of the hair follicle was strongly positive for Alcian blue at pH 1.0 and 2.5. In the epidermis, cytokeratin (CK) AE1/AE3 was expressed in all layers. CK5/6 was expressed in the basal and suprabasal layers (Fig. 1C). CK14 was expressed in the basal layer as well as the suprabasal layer of the hyperplastic epidermis (Fig. 1D). CK17 was expressed in the basal layer (Fig. 1E). Vimentin
was expressed by some cells in the underlying dermal papilla (Fig. 1F). In the hair follicles, CK AE1/AE3 was expressed in all layers, except for the hair matrix and dermal papilla cells. CK5/6 was expressed in the outer root sheath of the hair follicle (Fig. 1G). CK14 was expressed in the outer root sheath and occasionally in the cortical cells arising from the hair matrix of the anagen hair bulb (Fig. 1H). CK17 was expressed in the outer root sheath of the hair follicle (Fig. 1I). CK19 was infrequently expressed in the outer root sheath of the hair follicle (Fig. 1J). Vimentin was expressed by the dermal papilla cells within the hair bulb (Fig. 1K). In the sebaceous glands, the basal cells expressed CK AE1/AE3, CK5/6, CK14 and CK17, while the differentiated clear cells expressed CK AE1/AE3 and CK14. In the present study, the expression of CK18 was not detected in the epidermis and its adnexa. Histological Features of Trichoblastoma
The histological features of the present cases are summarized in Table 4. Almost all tumours were diagnosed as trichoblastoma, except for one case, which was diagnosed as malignant trichoblastoma. Tumour types were ribbon (10/30, 33.3%), trabecular (8/30, 26.7%) or mixed (12/30, 40.0%). In general, the tumour masses were well circumscribed, but some dermal masses occasionally extended into the subcutaneous tissue. In the ribbon type (cases 1e10), tumour cells were arranged in thin, branching cords of two to four cell layers (Fig. 2A). In the trabecular
M.K. Kok et al.
130
Table 4 Histological and immunohistochemical features of rabbit trichoblastomas Case number
Histological subtype
Trichoblastoma 1 Ribbon 2* Ribbon 3 Ribbon 4 Ribbon 5 Ribbon 6 Ribbon 7 Ribbon 8 Ribbon 9 Ribbon 10 Ribbon 11* Trabecular Trabecular 12* 13 Trabecular 14* Trabecular 15 Trabecular 16 Trabecular 17 Trabecular 18 Trabecular 19* Mixed 20 Mixed 21 Mixed 22 Mixed 23 Mixed 24 Mixed 25 Mixed 26* Mixed 27 Mixed 28 Mixed 29 Mixed Malignant trichoblastoma 30 Mixed
Presence of PMBs†
AB‡
PASx
Immunohistochemistry CK AE1/AE3x
CK5/6x
CK14x
CK17x
CK18x
CK19x
Vimentink
+ + + + + + + + + + + + + + + + + + + + + + +
+ + + + + + + + + + + + + + + ++ + ++ + ++ + + + + ++ + + + +
+ ++ ++ ++ + + + ++ + ++ ++ + ++ + ++ ++ ++
++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++
++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++ ++++
++ + + + +++ + + + ++ + ++ + ++ ++ +++ ++ ++ + + +++ + + ++ + ++ ++ ++ ++ ++
++ + ++ + ++ ++ ++ ++ + ++++ +++ ++ +++ ++ ++ +++ ++++ ++ ++++ + ++ ++ ++ ++++ ++++ ++ ++ ++ +++
+ + + +
+++ +++ +++ +++ +++ +++ +++ +++ ND +++ +++ +++ ND +++ +++ +++ +++ +++ ND +++ ND +++ ND +++ ND +++ +++ +++ +++
+
++
++
++++
++++
+
++++
+++
AB, Alcian blue pH 2.5; CK, cytokeratin; ND, not done (PMBs were absent in these cases); PAS, periodic acideSchiff; PMBs, papillary mesenchymal bodies. * Trichoblastoma occurred twice in three rabbits (cases 2 and 19; 11 and 12 and 14 and 26). † Presence of PMBs in the peritumoural stroma: , absent; +, present. ‡ Tumour cells were negative for Alcian blue (pH 1.0 and 2.5) while the tumour stroma showed positive staining. Qualitative scoring of tumour stroma for Alcian blue at pH 2.5: , negative; +, weak positive; ++, moderate positive; +++, strong positive. At pH 1.0, tumour stroma of all cases showed faint positive staining. x Percentage of positive tumour cells: , negative; +, 1e10%; ++, 11e50%; +++, 51e75%; ++++, 76e100%. k Tumour cells were negative for vimentin expression while the tumour stroma and PMBs showed positive expression. The positive labelling of vimentin in PMBs was evaluated qualitatively and categorized as: , negative; +, weak positive; ++, moderate positive; +++, strong positive.
type (cases 11e18), tumour cells proliferated in a lobular or fascicular pattern with minimal surrounding connective tissue stroma (Fig. 2B). In the mixed type (cases 19e29), both trabecular and ribbon patterns of tumour proliferation were observed. Multifocal clusters of small-sized stromal cells with dense basophilic nuclei and indistinct cytoplasm, termed as ‘papillary mesenchymal bodies’ (PMBs) (Brooke et al., 1989), were observed within the peritumoural stroma in the majority of cases (24/30, 80%; cases 1e8, 10e12, 14e18, 20, 22, 24 and 26e30) (Fig. 2AeC). The PMBs were stained blue with Mas-
son’s trichrome stain. In case 30, the trichoblastoma consisted of poorly demarcated, multiple expansive lobules of basal-type cells, proliferating in a mixture of ribbon, trabecular and solid patterns. In the solid areas, the nests of densely packed tumour cells were in association with the overlying epidermis and accompanied by abundant dermal papillae beneath the tumour lobules (Fig. 2D). The tumour cells in solid nests exhibited a higher degree of cellular atypia and continually expanded along the long axis of epidermis with frequent infiltration into the deeper dermis. Tumour cells were observed at the lateral and deep
Trichoblastoma in Rabbits
131
Fig. 2. Histological features of rabbit trichoblastoma. (A) Ribbon type. Tumour cells are arranged in thin branching cords with the presence of PMBs (arrows) in the stroma. HE. Bar, 100 mm. (B) Trabecular type. Tumour cells are arranged in a lobular pattern and prominent PMBs are observed (arrows) at the periphery of the lobules. HE. Bar, 100 mm. (C) Trabecular type. Higher magnification of a PMB consisting of cells with sparse cytoplasm and prominent oval-shaped nucleus (arrow) at the periphery of the tumour lobule. HE. Bar, 25 mm. (D) Malignant trichoblastoma. The tumour tissue (below) is contiguous with the overlying epidermis. HE. Bar, 100 mm. (E) Mixed type. Trabecular (left) and ribbon (right) types. Basal-type cells of the ribbon type exhibit PAS-positive granules. Bar, 50 mm. (F) Trabecular type. Tumour parenchyma is surrounded by Alcian blue (pH 2.5)-positive stroma. Bar, 100 mm.
margins. Based on the cellular pleomorphism and invasive growth, case 30 was diagnosed as a malignant trichoblastoma. Weak to moderate PAS-positive coarse granules were observed in the cytoplasm of the tumour cells of ribbon (4/10, 40%; cases 1, 2, 6 and 10), trabecular (8/8, 100%; cases 11e18) and mixed types (6/12, 50%; cases 19, 20, 22, 23, 26 and 30), respectively (Fig. 2E). All tumours in the present study possessed peritumoural stroma, which was weakly to moderately positive for Alcian blue at pH 2.5 (Fig. 2F), but faintly positive at pH 1.0. Immunohistochemical Features of Trichoblastoma
The immunohistochemical features of rabbit trichoblastoma are summarized in Table 4. All tumours
in the present study showed uniform and intense cytoplasmic expression of CK AE1/AE3 and CK5/6 (Fig. 3A). Palisading basal cells displayed moderate to strong cytoplasmic expression of CK14 in ribbon, trabecular and mixed types (Fig. 3B). In contrast, the inner cells of these tumours were positive for CK17 (Fig. 3C). The difference in CK17 expression between tumour types was attributed to the proportion of inner cells within the lesions examined. Only four cases (13, 16, 17 and 23) showed few scattered immunopositive areas for CK19, which were localized to the tumour parenchyma with differentiation towards luminal structures (Fig. 3D). CK18 was not expressed in any case. Intense immunoreactivity to vimentin was observed in the PMBs, while the tumour
M.K. Kok et al.
132
Fig. 3. Immunohistochemical features of rabbit trichoblastoma. (A) Ribbon type. Strong CK5/6 expression in the tumour, surrounded by CK5/6-negative dense PMBs. Bar, 100 mm. (B) Trabecular type. Strong CK14 expression in palisading basal-type tumour cells at the periphery of the tumour. Bar, 100 mm. (C) Mixed type. CK17 expression in the inner cells of the tumour. Bar, 100 mm. (D) Trabecular type. CK19 expression in a few parts of the tumour. Bar, 100 mm. (E) Trabecular type. Vimentin is expressed strongly in the PMBs (arrows) and moderately in the tumour stroma (arrowhead). Bar, 50 mm. (F) Malignant trichoblastoma. Inward invagination of CK17-positive tumour cells from the epidermis to form tumour lobules. Bar, 100 mm.
stroma showed a weak to moderate reaction (Fig. 3E). In the malignant trichoblastoma, a similar immunohistochemical pattern for all markers was observed, except for some areas of dense and solid tumour cell nests. In these areas, basal-type cells at the periphery of the lobules were weakly positive for CK14. In contrast, strong CK17 expression was observed, extending from the overlying epidermis and branching deeply towards the central part of the tumour lobules (Fig. 3F).
Discussion Rabbit trichoblastomas in the present study were located mainly on the neck, head, flank and dorsal
trunk. No breed or sex predilection was observed. Multiple occurrence of trichoblastoma was observed in three rabbits, suggesting the predisposition of rabbits towards the development of this tumour. Histopathologically, the trichoblastomas were of ribbon, trabecular or mixed type, as observed in equivalent canine and feline tumours. These patterns may be accounted for by the intensity of collagenous stromal proliferation. For instance, increased stromal elements in the ribbon type may limit the proliferation of tumour cells, resulting in their cord-like arrangement, while less stromal tissue in the trabecular type allows ample space for the tumour cells to proliferate, resulting in lobular, fascicular or nodular structures (De Vico et al., 2011).
Trichoblastoma in Rabbits
In the present study, weak to moderate PAS staining of the tumour cells was found in 18 (60%) of 30 trichoblastomas, which suggests occasional differentiation of the tumour cells to those of the outer root sheath. In addition, both the peritumoural stroma and the dermal lining of the normal hair follicle segments exhibited stronger Alcian blue intensity at pH 2.5 as compared with pH 1.0, suggesting that hyaluronic acid may be the main acid mucin in these structures. The histochemical findings highlight a possible origin of trichoblastoma from the suprabulbar and isthmus regions of the hair follicle in rabbits. PMBs are a histopathological feature characterized by distinct clusters of specialized fibroblasts in the peritumoural stroma, and these structures closely resemble the dermal papilla (Brooke et al., 1989; Campos et al., 2014). In man, PMBs have been identified in the lesions of trichoepithelioma and trichoblastic fibroma (Brooke et al., 1989; Altman et al., 1995; du Toit et al., 2016), but in animals PMBs have been described only in a few cases of canine trichoblastoma (Campos et al., 2014; Mineshige et al., 2014). In the present study, we observed abundant PMBs in the majority of the tumours, implying that aberrant dermal papilla formation is more prominent in rabbit trichoblastomas than in those arising in other species. Although immunohistochemistry (IHC) is rarely required to reach a definitive diagnosis of trichoblastoma, CK profiling is useful to determine the cellular origin and the differentiation of trichoepithelial tumours (Demirkesen et al., 1995; Ohnishi and Watanabe, 1999; Walter, 2000; Kato et al., 2007; Campos et al., 2014). CK14 is a basal cell marker of epidermis and hair follicle, while CK18 and CK19 are expressed in the luminal cells of apocrine glands (Walter, 2000; Kato et al., 2007; Bragulla and Homberger, 2009). CK17 is expressed in the outer root sheath of hair follicles (Chu and Weiss, 2002; Bragulla and Homberger, 2009). In man, CK17 and CK19 are also known as epidermal stem cell markers and, in particular, the expression of CK17 has been associated with ectodermal placode formation during the early stage of hair follicle morphogenesis (McGowan and Coulombe, 1998; Bragulla and Homberger, 2009). As epithelial tumours are known to conserve the CK associated with their cell of origin (Moll et al., 2008), CK markers were used to investigate the nature of rabbit trichoblastoma. In the present study, CK AE1/AE3 and CK5/6, which are general markers for epithelial tissues, were expressed diffusely in the trichoblastomas as well as in the normal epidermis and hair follicles.
133
In contrast, CK14 and 17 were localized within different tumour compartments, while their coexpression was observed in the normal rabbit epidermis and outer root sheath of the hair follicle. In the malignant trichoblastoma, the inward invagination of CK17-positive tumour cells from the epidermis was observed (Fig. 2F). This lesion was morphologically similar to the appearance of normal hair follicle formation in fetal human and murine skin, where there is downward expansion of CK17positive cells from the placode that eventually forms the primary hair germ (McGowan and Coulombe, 1998; Coolen et al., 2010). Differential labelling of CK14 and CK17, including the observed invagination of CK17positive tumour cells from the epidermis (case 30) and the presence of peritumoural PMBs, which resembled dermal condensation in hair follicle morphogenesis, suggests that the development of rabbit trichoblastoma may mimic the intermediate stage of embryonic hair follicle formation. A close interaction between tumour cells of trichoblastoma and adjacent dermal components occurred, leading to the aggregation of dermal mesenchymal cells and increased mucinous substance around the tumour lobules. Increased dermal cellular densities, PMB formation and acidic mucin production simulate the embryonic dermal layer, and may provide an optimal microenvironment for continuous development of an epithelial tumour (Jahoda, 1992; Yang and Cotsarelis, 2010). At the later stage, the abundant dermal fibroblasts produce sufficient collagen fibres, which eventually limit the space available for tumour cell differentiation, and, as mentioned earlier, this may contribute to the different proliferative pattern and well-demarcated nature of trichoblastoma. The significance of CK19 expression in the rabbit hair follicle and in trichoblastomas is not clear, as the expression was infrequently observed and, if any, the reactivity was weak. Consistent with a previous study of a canine trichoblastoma (Campos et al., 2014), expression of vimentin in PMBs, as well as in the dermal papillae of the hair follicle, was observed in the present study. In addition, PMBs were frequently localized within the peritumoural stroma and were stained blue with Masson’s trichrome stain. Taken together, these findings suggest that the PMBs may be derived from the germ of the dermal papilla. In conclusion, the present study has described the histopathological features and immunohistochemical profile of rabbit trichoblastoma as well as normal rabbit skin. Similar to findings in other
M.K. Kok et al.
134
species, rabbit trichoblastoma has an immunohistochemical resemblance to the embryonic hair follicle. In addition, PMBs observed within the peritumoural stroma are an important histological feature of rabbit trichoblastoma and these structures resemble embryonic dermal condensation during hair follicle development.
Acknowledgments The authors are grateful to Ms. S. Kato for invaluable technical assistance. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Conflict of Interest Statement The authors declare that there are no potential conflicts of interest related to publication of this research.
References Abramo F, Pratesi F, Cantile C, Sozzi S, Poli A (1999) Survey of canine and feline follicular tumours and tumourlike lesions in central Italy. Journal of Small Animal Practice, 40, 479e481. Altman DA, Mikhail GR, Johnson TM, Lowe L (1995) Trichoblastic fibroma: a series of 10 cases with report of a new plaque variant. Archives of Dermatology, 131, 198e201. Bath-Hextall F, Leonardi-Bee J, Smith C, Meal A, Hubbard R (2007) Trends in incidence of skin basal cell carcinoma. Additional evidence from a UK primary care database study. International Journal of Cancer, 121, 2105e2108. Bragulla HH, Homberger DG (2009) Structure and functions of keratin proteins in simple, stratified, keratinized and cornified epithelia. Journal of Anatomy, 214, 516e559. Brooke JD, Fitzpatrick JE, Golitz LE (1989) Papillary mesenchymal bodies: a histologic finding useful in differentiating trichoepithehomas from basal cell carcinomas. Journal of the American Academy of Dermatology, 21, 523e528. Campos AG, Cogliati B, Guerra JM, Matera JM (2014) Multiple trichoblastomas in a dog. Veterinary Dermatology, 25, 48e49. Chu PG, Weiss LM (2002) Keratin expression in human tissues and neoplasms. Histopathology, 40, 403e439. Coolen NA, Schouten KC, Middelkoop E, Ulrich MM (2010) Comparison between human fetal and adult skin. Archives of Dermatological Research, 302, 47e55. De Vico G, Cataldi M, Maiolino P, Carella F, Beltraminelli S et al. (2011) Fractal pattern of canine trichoblastoma. Analytical and Quantitative Cytology and Histology, 33, 151e157.
du Toit JP, Schneider JW, Visser WI, Jordaan HF (2016) The clinicopathological spectrum of trichoepitheliomas: a retrospective descriptive study. International Journal of Dermatology, 55, 270e277. Demirkesen C, Hoede N, Moll R (1995) Epithelial markers and differentiation in adnexal neoplasms of the skin: an immunohistochemical study including individual cytokeratins. Journal of Cutaneous Pathology, 22, 518e535. Fisher PG (2010) Standards of care in the 21st century: the rabbit. Journal of Exotic Pet Medicine, 19, 22e35. Goldschmidt MH, Dunstan RW, Stannard AA, von Tscharner C, Walder EJ et al. (1998) Tumor with adnexal differentiationHistological Classification of Epithelial and Melanocytic Tumors of the Skin of Domestic Animals (WHO), 2nd Series, Armed Forces Institute of Pathology, Washington DC, pp. 21e32. Goldschmidt MH, Goldschmidt KH (2017) Epithelial and melanocytic tumors of the skin. In: Tumors in Domestic Animals, 5th Edit, DJ Meuten, Ed., Wiley, Ames, pp. 88e141. Gross TL, Ihrke PJ, Walder EJ, Affolter VK (2005) Follicular tumors. In: Skin Diseases of the Dog and Cat: Clinical and Histopathological Diagnosis, 2nd Edit, Blackwell Science, Oxford, pp. 604e640. Jahoda CA (1992) Induction of follicle formation and hair growth by vibrissa dermal papillae implanted into rat ear wounds: vibrissa-type fibres are specified. Development, 115, 1103e1109. Kato K, Uchida K, Nibe K, Tateyama S (2007) Immunohistochemical studies on cytokeratin 8 and 18 expressions in canine cutaneous adnexus and their tumors. Journal of Veterinary Medical Science, 69, 233e239. McGowan KM, Coulombe PA (1998) Onset of keratin 17 expression coincides with the definition of major epithelial lineages during skin development. Journal of Cell Biology, 143, 469e486. Mineshige T, Yasuno K, Sugahara G, Tomishita Y, Shimokawa N et al. (2014) Trichoblastoma with abundant plump stromal cells in a dog. Journal of Veterinary Medical Science, 76, 735e739. Moll R, Divo M, Langbein L (2008) The human keratins: biology and pathology. Histochemistry and Cell Biology, 129, 705e733. Ohnishi T, Watanabe S (1999) Immunohistochemical analysis of cytokeratin expression in various trichogenic tumors. American Journal of Dermatopathology, 21, 337e343. Ramos-Vara JA, Miller MA (2014) When tissue antigens and antibodies get along: revisiting the technical aspects of immunohistochemistry e the red, brown, and blue technique. Veterinary Pathology, 51, 42e87. Rooney NJ, Blackwell EJ, Mullan SM, Saunders R, Baker PE et al. (2014) The current state of welfare, housing and husbandry of the English pet rabbit population. BMC Research Notes, 7, 942. Rubin AI, Chen EH, Ratner D (2005) Basal-cell carcinoma. New England Journal of Medicine, 353, 2262e2269.
135
Trichoblastoma in Rabbits
Schepers F, Koene P, Beerda B (2009) Welfare assessment in pet rabbits. Animal Welfare, 18, 477e485. von Bomhard W, Goldschmidt MH, Shofer FS, Perl L, Rosenthal KL et al. (2007) Cutaneous neoplasms in pet rabbits: a retrospective study. Veterinary Pathology, 44, 579e588. Walter J (2000) A cytokeratin profile of canine epithelial skin tumours. Journal of Comparative Pathology, 122, 278e287.
Yang CC, Cotsarelis G (2010) Review of hair follicle dermal cells. Journal of Dermatological Science, 57, 2e11.
April 26th, 2017 ½ Received, Accepted, June 13th, 2017