alterations of epidermal proliferation and cytokeratin expression in skin biopsies from heavy...

8/9/2019 Alterations of Epidermal Proliferation and Cytokeratin Expression in Skin Biopsies From Heavy Draught Horses With

1/12

2005 European Society of Veterinary Dermatology 373

Veterinary Dermatology 2005, 16, 373384

BlackwellPublishingLtd

Alterations of epidermal proliferation and cytokeratin expression inskin biopsies from heavy draught horses with chronic pastern

dermatitis

FLORIAN GEBUREK*, BERNHARD OHNESORGE*, ECKEHARD DEEGEN*, RENATEDOELEKE and MARION HEWICKER-TRAUTWEIN

*Klinik fr Pferde, Tierrztliche Hochschule Hannover, Bischofsholer Damm 15, D-30173 Hannover, Germany,

Institut fr Pathologie, Tierrztliche Hochschule Hannover, Bnteweg 17, D-30559 Hannover, Germany

(

Received

16 August

2004; accepted

17 August

2005)

Abstract

We report the historical, clinical and histopathological characteristics of skin lesions in biopsies from

37 heavy draught horses with chronic pastern dermatitis. The skin lesions were divided into four macroscopic

groups: scaling (group I, n

= 5), hyperkeratotic and hyperplastic plaque-like lesions (group II, n

= 14), nodular

skin masses (group III, n

= 16) and verrucous skin lesions (group IV, n

= 2). The principal histological findings

were hyperkeratosis and epidermal hyperplasia. There was a gradual increase in epidermal hyperplasia fromgroups I to IV, suggesting that the lesions represent different stages of disease. In all cases, there was perivascular

dermatitis dominated by T lymphocytes with an increase in MHC class II-positive dendritic-like cells. Immuno-

histochemical labelling for cytokeratins CK5/6(4), CK10 and CK14 indicated a change in their expression

pattern. This correlated with the degree of epidermal hyperplasia, indicating abnormal differentiation of

keratinocytes. There was a statistically significant correlation between the severity of skin lesions and several other

factors including increasing age, increasing cannon circumference, prominence of anatomical structures such as

fetlock tufts of hairs, ergots and chestnuts, and bulges in the fetlock region.

INTRODUCTION

Pastern dermatitis, also known as greasy heel, scratches

or mud fever in the Anglo-American literature, or as

mauke in the German literature, is a progressive

inflammatory skin lesion involving the posterior

pasterns of horses.

13

Pastern dermatitis occurs in all

breeds, but is most common in heavy draught horses

with feathering.

3

Affected horses initially show oedema

and scaling that progresses to exudation and crusting.

Pastern dermatitis is thought to be multifactorial with

many potential causes, e.g. infection with bacteria

(

Staphylococcus

spp., Dermatophilus congolensis

),

dermatophytosis, chorioptic mange, trombiculosis,photosensitization, vasculitis or contact with chemical

irritants.

3

Recently, the occurrence of papillomatous

pastern dermatitis associated with spirochetes and

nematodes identified as Pelodera strongyloides

in a

Tennessee Walking Horse was reported.

4

When the

aetiology is indeterminable, the term idiopathic

pastern dermatitis is used.

13

In these chronic cases,

the skin of the posterior pasterns becomes hyperkera-

totic and lichenified. Finally, hyperplasia is visible as

papillomatous or multifocal circumscribed verrucous

masses referred to as grapes.

13

These chronic, verru-

cous pastern lesions, which almost exclusively occur in

heavy draught horses, are known as warzenmauke

(

warze

= wart in English; muche

= German medievalword for a lameness-inducing disease of horses) in

the German veterinary literature. Although chronic

hyperplastic skin lesions of the pasterns of heavy

draught horses have been described for centuries, detailed

descriptions of the histopathology are not available,

except from older reports or dissertations mainly by

German or French authors.

59

The purpose of this study was to describe the clinical

and histopathological features of chronic pastern

dermatitis in biopsies from 37 heavy draught horses of

different breeds. The proliferative activity of epidermal

basal cells and expression of cytokeratins were charac-terized immunohistochemically with antibodies to the

proliferation marker Ki-67 and cytokeratins CK5/6(4),

CK10 and CK14. The dermal cellular immune response

was assessed with antibodies for leucocyte and MHC

class II antigens.

MATERIALS AND METHODS

Historical and clinical details

Data from 37 heavy draught horses of different breeds

with chronic pastern dermatitis were collected (Table 1).

For each horse, this included details of housing, feedingregimes, amount of work, use for breeding, condition

of the hair/hooves, duration of the skin lesions, presence

or absence of white markings or pruritus in the foot

Correspondence: M. Hewicker-Trautwein, Institut fr Pathologie,

Tierrztliche Hochschule Hannover, Bnteweg 17, D-30559 Hannover,

Germany. E-mail: [email protected]


2/12

374 F Geburek et al

.

2005 European Society of Veterinary Dermatology, Veterinary Dermatology

, 16

, 373384

region and presence or absence of other skin lesions.

The cleanliness of stable facilities and quality of the

air and light were graded subjectively as good = 1,moderate = 2 or bad = 3. The ground where horses

were kept outdoors was recorded as pasture/soil, sand,

or rubber meadows. Coat condition was graded subjec-

tively as good = 1 (clean, smooth, shiny), moderate = 2

(partly dirty, partly rough, partly dull) or bad = 3 (dirty,

rough, dull). The condition of the hooves was graded

as good = 1 (clean, no trimming necessary, little or no

thrush) or bad = 2 (dirty, trimming necessary, noticeable

thrush).

A scoring system was developed for evaluating the

pastern skin lesions according to their type, extent and

severity. The clinical extent and macroscopic appear-

ance of skin lesions were recorded and documented bypreparing schematic drawings and by colour photography.

Parameters recorded were erythema, exudation, crust

formation, scaling, hyperkeratotic and hyperplastic

plaque-like lesions, erosions/excoriations, greasiness and

malodorous skin surface, nodular masses, or verrucous

masses with rugged surfaces. Each of these nine features(if observed) was multiplied by 13, depending on the

severity of the lesion (1 = mild, 2 = moderate, 3 = severe).

These were added together to give a total severity score

for each horse. The cannon circumference of all four

feet was measured in centimetres with a tape measure

at the smallest circumference of the metacarpal and

metatarsal bone immediately beneath the carpal or

tarsal joint. The average cannon circumference was

then calculated. Normal anatomical structures, i.e.

fetlock tufts of hairs (feathering), ergots, chestnuts

and bulges in the fetlock region were recorded. Skin

scrapings were examined for Chorioptes

sp.

Biopsy specimens

From each horse, at least one punch biopsy of 6 mm in

diameter was taken from the diseased skin. The location

Table 1. Clinical data, sites and numbers of biopsy specimens and macroscopic lesions on the feet of 37 heavy draught horses with chronic

pastern dermatitis

Case no. Cold blood breed Age (years) Sex

Cannon circumference

(in cm)

Biopsied

foot/feet

Total number

of biopsies

Macroscopic

lesion

1 Mecklenburgisch* 8 MC 27.0 LH, RH 3 Sc

2 Rhenisch German 6 MC 32.0 LH 2 Sc

3 Rhenisch German 9 F 27.5 RH 2 Sc4 Westphalian 5 F 28.25 RH 2 Sc

5 Lower Saxon 2 F 26.0 LH 2 Sc

6 Unknown 5 MC 26.75 LF, RH 4 Hhp

7 Westphalian 3 M 28.0 RH 2 Hhp

8 Altmrkisch* 17 F 31.0 RH 2 Hhp

9 Rhenisch German 4 M 29.0 LH 1 Hhp

10 Westphalian 3 M 27.5 RH 1 Hhp



13 Westphalian 4 M 28.0 LH 2 Hhp

14 Westphalian 4 M 27.5 LH 2 Hhp

15 Belgian Draught 10 M 31.0 LH 2 Hhp

16 Mecklenburgisch 7 F 29.0 RH 2 Hhp

17 Rhenisch German 5 M 32.0 RH 2 Hhp

18 Mecklenburgisch 7 MC 28.75 RH 2 Hhp19 Lower Saxon 3 F 26.0 LH 2 Hhp

20 Belgian Draught 6 M 32.0 RF 3 Nm

21 Rhenisch German 9 M 32.5 RH 2 Nm

22 Percheron 15 M 34.5 LH 2 Nm

23 Percheron 13 M 33.5 LH 2 Nm

24 Percheron 9 MC 34.0 RH 2 Nm

25 Rhenisch German 19 MC 31.25 RH 2 Nm

26 Westphalian 7 M 33.0 LH 2 Nm

27 Westphalian 10 M 29.5 RH 1 Nm

28 Saxon 13 M 37.5 RH 2 Nm

29 Westphalian 8 M 31.5 LH 2 Nm

30 Rhenisch German 8 M 29.5 LH 2 Nm

31 Rhenisch German 7 M 31.5 LH 2 Nm

32 Polish 14 F 24.5 LH 2 Nm

33 Saxon 7 M 33.5 LH 1 Nm34 Altmrkisch 8 M 31.25 RH 2 Nm

35 Lower Saxon 9 F 27.0 RH 2 Nm

36 Westphalian 10 M 32.0 LH 3 Vm

37 Rhenisch German 12 F 37.0 RH 1 Vm

C1 Altmrkisch 3 M 27,5 LH 1 None

C2 Rhenisch German 3 M 28.5 RH 1 None

M, male; MC, male castrated; F, female; LH, left hind; RH, right hind; LF, left front; RF, right front; C, control horse; *East German

Coldblood; Sc, scaling; Hhp, hyperkeratotic/hyperplastic plaques; Nm, nodular masses; Vm, verruccous masses.


3/12


, 16

, 373384

Pastern dermatitis in heavy draught horses 375

and total number of biopsies taken from the 37 cases

are given in Table 1. Tissue specimens were fixed in 4%

neutral buffered formaldehyde, embedded in paraffin

wax, cut at 4

m, mounted on coated slides (SuperFrost

Plus, Menzel-Glser, Wiesbaden, Germany) and stained

with haematoxylin and eosin or cresyl echt violet for

demonstration of mast cells. Selected sections were

stained with by the Warthin Starry silver staining

method. For comparison, from all 37 horses one 6-mm

punch biopsy was taken from the macroscopically

normal skin of the lateral neck. Control 6-mm skin

biopsies were included from two heavy draught horses

without clinical dermatological abnormality. These

were taken from the pasterns (C 1 and C 2 in Table 1)

and from the neck of horse C2.

Antibodies and immunohistochemical labellingprocedures

Antibodies used for immunohistochemical labellingare shown in Table 2.

Paraffin sections were dewaxed and rehydrated. Endo-

genous peroxidase activity was blocked by incubation

for 30 min in a solution of 0.5% H

2

O

2

in 85% ethanol.

For antigen retrieval, sections stained with antibodies

to Ki-67 or cytokeratins were pretreated by heating in

a microwave oven for 2

5 min in citric acid buffer

(0.01 mol L

1

, pH 6.0). For detection of CD3 and CD79a

antigens, sections were pretreated with unmasking

fluid G (Biologo, Kronshagen, Germany) for 10 min at

95

C. Sections stained with antibodies to MHC class

II antigens were incubated for 15 min at room temper-ature (rt) in a 0.25% solution of Triton X-100, rinsed in

phosphate-buffered saline (PBS), and pretreated with

a 0.05% pronase E (Merck, Darmstadt, Germany)

solution for 20 min at 37

C. Following incubation with

inactivated normal goat serum for 20 min at rt and

incubation with the primary antibodies (1820 h at

4

C) biotin-conjugated goat-antimouse IgG (H + L)

or goat-antirabbit IgG (H + L) (both from Vector

Laboratories, Burlingame, USA), diluted 1 : 200 in

PBS containing 10% inactivated normal horse serum

were used as second antibodies for 30 min at rt. After

incubation with the ABC solution (Vectastain Elite

ABC Kit, Vector Laboratories, Burlin-game, CA,USA) for 30 min at rt, peroxidase activity was detected

by incubation in 3-amino-9-ethylcarbazol solution

(AEC-Plus, Dako, Hamburg, Germany). Sections were

counterstained with Mayers haematoxylin. Negative

control sections, in which the primary antibody was

replaced by appropriately diluted normal mouse or

rabbit serum, were included in all staining runs.

Positive control tissues were equine lymph node for

CD3, CD79a, MHC II and Ki-67 antigens and normal

human and equine skin for CK5/6(4), CK10 and

CK14.

Statistical analysis

SAS software (version 6.04, Statistical Analysis Insti-

tute, Cary, NC, USA) was used for statistical analysis.

The clinical and morphological data were either direct

and quantitative (age, cannon circumference) or qual-

itative, grouped or graded (hygienic condition of the

stable, character of the ground where the horses were

kept on outdoors, feeding, amount of work, use of

stallions for breeding, grooming condition of the hair/

hooves, white markings in the foot region, fetlock tuftsof hairs, ergots, chestnuts, anatomically normal bulges

in the fetlock region, pruritus). The previously mentioned

parameters were correlated using Spearmans rank

correlation.

Differences in clinical severity for paired data were

compared with Wilcoxons signed-rank test, i.e. use of

stallions for breeding, grooming condition of hooves,

anatomically normal bulges in the fetlock region, white

markings in the foot region, pruritus and frequency of

cleaning stables. KruskalWallis test was used when

more than two groups had to be compared, i.e. sex,

amount of work, grooming condition, type of groundhorses were kept on. Means and standard deviations

were calculated. Differences were considered signifi-

cant at P

< 0.05.

Immunolabelled cells were counted in pastern skin

sections and compared to the number in skin sections

from the neck. Immunolabelling for Ki-67 was quan-

tified by counting basal epidermal cells with Ki-67-

positive nuclei in interfollicular epidermal areas in five

high-power fields (HPF) at 400

magnification. The

field of view was 600

m. Mast cells, CD3-, CD79a- and

MHC class II-positive cells were counted in five HPF

(400

) in the upper dermis. The fields were selected to

contain perivascular inflammatory foci. The meannumbers of positively labelled nuclei or cells, respec-

tively, were documented, standard deviations were

calculated and correlation was tested by Spearmans

Table 2. Antibodies used for immunohistochemical staining

Antibody Clone or identification Specificity Dilution Source

Anti-Ki-67 MIB-1 Human Ki-67 1 : 75 Dako

Anti-CK5/6(4) D5/16B4 Human CK 5/6(4) 1 : 3000 Dako

Anti-CK10 DE-K10 Human CK10 1 : 80 Dako

Anti-CK14 LL002 Human CK14 1 : 1000/1 : 50* Novocastra, Newcastle upon Tyne, UK

Anti-CD3 A0452 Human CD3 1 : 300 DakoAnti-CD79a HM57 Human CD79a 1 : 60 Dako

Anti-MHC II H42A Equine MHC II 1 : 120 VMRD, Pullman, Washington, USA

*In biopsies with moderate and severe epidermal hyperplasia the 1 : 1000 dilution resulted in a very weak staining intensity. Therefore, the

sections were also stained with a lower antibody dilution of 1 : 50.


4/12

376 F Geburek et al

.


, 16

, 373384

rank correlation. To compare differences between

numbers of cells, Wilcoxons signed-rank test was used.

RESULTS

Clinical and macroscopic findings

The data obtained from the owners or animal attend-

ants indicated that in the majority of horses (

n

= 17),

the skin lesions had been present for at least 2 years. In

10 cases, the lesions had not been noted or recognized

by the owners or animal attendants as pastern dermatitis.Some horses had additional dermatological diseases

such as facial alopecia (case no. 21), sweet itch (case

nos 3, 23, 25, 35) or inflammation of elbows and knees

(case no. 15). Hoof disease (case no. 6) and ulcerating

subsolar abscesses (case no. 1) were each found in one

horse. All other horses were clinically healthy apart

from the pastern dermatitis.

Based on macroscopic findings, the pastern skin

lesions were divided into four different groups (Table 1).

In horses of group I (

n

= 5), scaling was the predominant

sign (Fig. 1). Lesions in horses of group II (

n

= 14)

were characterized by hyperkeratotic and hyperplastic

plaque-like lesions (Fig. 2) as well as areas with scaling.In horses of group III (

n

= 16), nodular skin masses

were present (Figs 3 and 4), and horses of group IV

(

n

= 2) were affected by verrucous skin lesions with

rugged surfaces (Figs 5 and 6). Horses of groups III

and IV also had scaling and plaque-like lesions and

greasy, malodorous skin with focal excoriations/erosions

(Fig. 6).

Statistical analysis revealed a significant correlation

between the severity of skin lesions and age (

P =

0.0005),

increasing mean cannon circumference (

P

alterations of epidermal proliferation and cytokeratin expression in skin biopsies from heavy...

Documents