Ulcerative Balanitis and Vulvitis of Dorper Sheep in South Africa: a Study on Its Aetiology and Clinical Features

INTRODUCTION Ulcerative balanoposthitis and vulvo-vaginitis of sheep is a venereal disease characterised by erosion and ulceration on the glans penis and vulval labiae. Although it has been recognised in South Africa since 1979


INTRODUCTION
Ulcerative balanoposthitis and vulvovaginitis of sheep is a venereal disease characterised by erosion and ulceration on the glans penis and vulval labiae.
Although it has been recognised in South Africa since 1979 36,37 , publications dealing with it have only appeared during the last decade.Sheep of the Dorper breed appear to be the most frequently affected 4,15,36 .The distribution of the disease in South Africa, in general, corresponds with that of Dorper sheep and therefore tends to be restricted to the dry areas of the Northern Cape, Western Cape, Eastern Cape, Kwa-Zulu-Natal and Free State provinces 15 .
Several authors have intimated that the disease only occurs in ewes as no mention was made in their publications of the simultaneous involvement of associated rams 2,11 .On the other hand, others have described the occurrence of the disease in both ewes and rams running together 4,8,12,36,38 .
Since the first report on the occurrence of the disease in South Africa, issues relating to the epidemiology, aetiology and control have not been adequately addressed.However, several investigators were of the opinion that ulcerative balanoposthitis and vulvovaginitis could be associated with a Mycoplasma sp.infection 4,15,36 .
The principal objectives of this study were to determine whether ulcerative balanoposthitis and vulvovaginitis in South Africa is a multifactorial disease in which the major causative organism is a mollicute, and whether the disease manifests mainly in sheep of both sexes under the age of 36 months.

Study area
The study was conducted between September 2001 and March 2002 in 4 districts, namely Namaqualand, Hay, Barkly West and Hopetown, in the Northern Cape Province, South Africa and Beaufort West in the Western Cape Province (Fig. 1).A total of 15 sheep farms (Table 1) in these 5 districts was visited to study the clinical features of the disease and to collect specimens for microbiological investigation.

Study animals
The study focused exclusively on Dorper sheep, as this breed is the one primarily affected by the disease in South Africa.It is a breed with a white body and a black head (Fig. 2).Fat deposits on the rump, tail and brisket are absent.Under good management conditions mature rams can weigh 82-91 kg and ewes

ABSTRACT
Ovine ulcerative balanitis and vulvitis in sheep of the Dorper breed has been observed in South Africa since 1979.Its aetiology has not been conclusively resolved, and there is some discrepancy in descriptions of its clinical features.In order to identify the pathogenic microorganism/s that contribute to the occurrence of the disease, the microflora in the genital tracts of both clinically healthy and affected sheep were isolated and compared.Bacteriological examination of materials from affected and unaffected sheep resulted in the isolation of Arcanobacterium pyogenes from 44.2 % and 17.2 % of them respectively.This difference is statistically significant (P < 0.01).Seventy-four per cent of the isolates originated from severe clinical cases.Mycoplasmas were isolated from 49.3 % of 116 clinically normal sheep and 78.2 % of 104 affected sheep.There were significant differences in their rates of isolation in clinical groups (P < 0.05).Of all the mycoplasma isolates, Mycoplasma mycoides mycoides large colony variant (MmmLC) was isolated from 61.5 % of clinically diseased sheep while 6.0 % of the isolates were from apparently healthy animals (P < 0.05).The study threw light on the prevalence of mycoplasmas in the genital tract of apparently healthy sheep and, at the same time the identity of the mycoplasma pathogen associated with ulcerative balanitis and vulvitis was revealed.The findings of this investigation therefore confirmed the involvement of mycoplasma, particularly that of MmmLC large colony, in the disease in Dorper sheep in South Africa, and it was concluded that this microorganism is an important pathogen of balanitis and vulvitis in them.The study furthermore demonstrated a probable synergism between A. pyogenes and MmmLC.Finding these 2 organisms together occurred 53.4 times more frequently in the affected sheep than in the unaffected, which emphasises the probable multifactorial nature of the disease.The association between age and the presence of clinical signs was statistically significant.It was found that young sheep were more likely to have lesions than adult sheep.Clinical observations showed that the typical ulceration appears to be confined to the glans penis and lips of the vulva; no ulceration was observed on the shaft of the penis and prepuce or vaginal vestibule.In uncomplicated cases inflammation of the prepuce and vaginal vestibule is not a regular feature of the disease.Therefore the names ulcerative balanitis and vulvitis most accurately describe the nature of the disease in South Africa.
54-63 kg.Lambs can reach a live weight of 34 kg at about 6 months of age 10 .Where possible the number of animals sampled on each farm was at least 5 affected and 5 healthy for each sex group.The age of each animal sampled was estimated according to the number of temporary and/or permanent incisor teeth.Thus sheep with no permanent incisors were judged to be less than 12 months of age, those with 2, 4 or 6 permanent incisors to be 12-14, 18-20 or 26-30 months of age respectively, and the sheep with 8 were considered to be adults.In addition a record was made on the presence or absence of the lesions associated with the disease and of the location and severity of the lesions in those sheep showing them.

Study design
The study was designed as a case-control study in which diseased animals (cases) and non-diseased animals (controls) were selected and compared in terms of the presence of the risk factor 34 .The data were recorded on a standard data sheet before they were entered into a computer database.
The variables analysed included temporal and spatial data, population variables such as age, sex and disease status and microbiological data.A dichotomous variable with either yes (Y) or no (N) as an entry was used to denote the disease status of each animal sampled and to describe if certain microorganisms were isolated from the genital swabs examined.

Specimen collection
Genital swab specimens were collected from both clinically normal and diseased sheep using commercial sterile cotton swabs.A total of 220 vulval and penile swabs consisting of 116 and 104 swabs from healthy and clinically affected sheep, respectively, were examined.The methods used for taking swabs, and procedures for the detection and identification of microorganisms, have been described in detail by Kidanemariam  (2003)  22 .
Specimens were obtained from the vulvae by spreading the vulval lips, inserting a swab and rubbing it back and forth gently before removing it.Swab samples from rams were obtained by gentle rubbing of the glans penis and preputial mucosa after manual extrusion of the penis from the prepuce.
Swabs were placed into cryovials containing transport medium appropriate for the different selected microorganisms.The transport media included Hayflick's broth for mycoplasma isolation, brain heart infusion broth for isolation of bacteria, and phosphate buffered saline for viral isolation.Transport medium was not used for swabs to be used for Chlamydophila antigen detection.The cryovials were placed in liquid nitrogen and transported to the laboratory where they were stored at -85 °C until processed.

Isolation and identification of bacteria
Swabs were streaked onto plates containing Columbia blood agar (Difco) enriched with 5-6 % horse blood which were then incubated in an atmosphere of 5 % CO2 at 37 °C for 24 hours, with further re-incubation for 36-72 hours if no growth was observed.Single colonies of the different colony types recognised on each plate were selected and re-inoculated onto plates containing blood agar and MacConkey (Difco) agar.This was repeated until pure growth was obtained.Smears of the bacterial cultures were stained by Gram's method for light microscopy and the cultures were subjected to the catalase and oxidase tests and oxidation-fermentation (O-F) reactions.Cultures that conformed to certain criteria were subjected to additional biochemical tests.
Biochemical characterisation of isolates was done using conventional methods that included a panel of sugars, and 3

Isolation and identification of mycoplasmas
The standard growth media used for the propagation of field isolates and reference strains of Mycoplasma were Hayflick's agar and broth, Chalquest agar and broth, ureaplasma agar and broth, and rabbit meat infusion broth.The latter was also used to grow Mycoplasma reference strains that were used as antigens to inoculate adult white New Zealand rabbits for the production by standard means of hyperimmune sera for the indirect immunofluorescent antibody test.
Thawed swab specimens were both streaked on plates containing mycoplasma and ureaplasma media and used to inoculate the broth media.These were incubated in a moist chamber at 37 °C in an atmosphere containing 5 % CO2.Evidence of growth in broth medium was monitored by turbidity of the medium.When satisfactory growth had been obtained, aliquots of broth culture were inoculated onto mycoplasma agar plates.All plates were examined daily or on alternate days, using a dissection microscope at ×20-40 magnification.The colonies were subsequently identified by the indirect immunofluorescent antibody test using the antisera produced in the rabbits.
The reference Mycoplasma spp.used as antigens for the production of positive serological controls were 8 reference For the indirect staining procedure employed in this study, agar blocks containing colonies in situ were used.Rabbit antiserum was used as the primary antibody, followed by addition of fluoresceinconjugated goat anti-rabbit IgG (The Binding Site, Birmingham, UK) as secondary antibody.Phosphate buffered saline (PBS) at pH 7.2 was used as a diluent for the conjugate, antisera and Evans' blue, and PBS with 0.5 % Tween twenty was used for washing.
Fluorescence microscopy was achieved with the use of a binocular microscope equipped with an epi-fluorescent attachment and Osram HBO 100 W/2 highpressure mercury vapour lamp (Nikon corporation, Japan).A green excitatory filter (G 2A) and yellow barrier filter were used.The microscope was fitted with a photomicrographic apparatus (Model UFX-DX).

Isolation of viruses
Primary and secondary lamb foetal kidney (LFK) cell cultures were used in an attempt to isolate a virus or viruses.The LFK cells were grown in Eagles' Minimal Essential Medium (MEM) with Earle's salts supplemented with 5 % heatinactivated foetal bovine serum (FBS).Gentamicin was added to the medium at a final concentration of 50 µg per m .

Chlamydophila antigen detection
The Clearview ® Chlamydia MF (Clearview ® , Unipath Limited, Bedford, UK) test kit was used in an attempt to detect Chlamydophila antigen from genital swabs collected for this purpose from the sheep.
Although, the test was developed to detect Chlamydia trachomatis antigen from human genital swabs, it has been shown that it also detects Chlamydophila psittaci subspecies ovis antigen.It provides a simple direct detection assay which is highly sensitive, specific and rapid.

Data analysis
All relevant data generated in the study were recorded in a data-capturing format and entered into a computer database for subsequent analysis.The statistical package used to store and analyse the data was EpiInfo 2000 version 1.0 (Centers for Disease Prevention and Control, Department of Health and Human Services, USA).The variables to be assessed were included in an EpiInfo questionnaire and were analysed using the ANALYSIS and STATCALC facilities of the EpiInfo software.Descriptive statistics were employed to designate the different types of microorganisms encountered in both the healthy and clinically infected sheep.The association of each isolate (risk factor) with the disease process was assessed using an odds ratio.The differences were analysed using the chi-square test, and the levels of significance were taken as P < 0.05 and confidence intervals (CI) were set at 95 %.

Clinical observations
Early signs of vulvitis were characterised by swollen and reddened vulvae visible at a distance in short-docked Dorper ewes (Figs 3, 4).In some ewes the swelling was accompanied by the development of discrete mucosal ulcers at the mucocutaneous junction of the vulval labia.
On closer examination, shallow, blisterlike wounds covered with scabs could be found.No lesions were observed in the vestibule of the vulva/vagina.
On manual withdrawal of the penis of affected rams, the characteristic lesions could readily be seen involving the soft glans of the penis (Figs 5, 6).This comprised the presence of 1 of more discrete ulcers on the glans penis, each of which had a sharply defined edge.Hyperaemia of the glans penis was invariably present.The rest of the penile tissue and preputial mucosa appeared unaffected.

Bacteria isolated from healthy and affected sheep
The microbiological flora of 104 clinically affected genitals of both ewes and rams from sheep in 15 flocks was compared with that of 116 clinically unaffected genitals.A total of 20 species of bacteria was identified from the 220 specimens.Only 6 (2.7 %) of the 220 samples showed no growth of bacterial organisms.The isolates were usually mixed populations and only 8 specimens yielded pure cultures of Arcanobacterium pyogenes.A summary of the bacteria isolated from the genital swabs of the sheep is shown in Table 2.
Arcanobacterium pyogenes was isolated from sheep affected with balanitis or vulvitis significantly more often than the other species of bacteria (P < 0.001).Of the isolated strains of A. pyogenes, 73.9 % (34/46) were from sheep showing severe balanitis or vulvitis whilst 21.6 % (12/46) were isolated from sheep that showed mild lesions.This difference is statistically significant (P < 0.01) and the odds ratio analysis showed an association between A. pyogenes isolation and the degree of severity of the lesion (OR = 3.36; 95 % CI = 1.75, OR < 6.47; chi-square = 15.97;P = 0.000).

Isolation and identification of mollicutes
A total of 222 mollicutes were isolated from genital swabs of 104 clinically affected and 116 unaffected sheep.Of the 222 isolates, Mycoplasma represented 153 (69.0 %), Ureaplasma 54 (24.3 %) and Acholeplasma 15 (6.7 %).The isolation rates of mycoplasmas from healthy and infected sheep varied significantly (P < 0.05).The likelihood that mycoplasmas would be isolated from affected sheep was 3.86 times greater than it was from apparently healthy sheep.Mycoplasma mycoides mycoides large colony variant (MmmLC) was the most frequently isolated mycoplasma (71 of 153; 46.4 %).Thirteen specimens (18.3 %) yielded pure cultures of  MmmLC while the remaining 58 (81.7 %) were found in mixed cultures with 1 or more species of mycoplasma.Of the 54 ureaplasma strains isolated, 24 (45 %) were from affected and 30 (54 %) were from healthy sheep.
When the isolates from individual clinically affected sheep were compared, 64 (61.5 %) sheep had MmmLC.The difference in the isolation of this species between healthy and affected groups is statistically significant (P < 0.01; chisquare = 7.6; P = 0.000).Sheep with clinical lesions were 24.46 times more likely to have been infected with MmmLC than healthy sheep (95 % CI = 9.76-64.01).
Of the 64 isolates of MmmLC from the 104 clinically affected sheep, 46 (71.9 %) were from sheep manifesting severe clinical signs with varying degrees of ulceration, and 18 (28.1 %) were from mildly affected sheep (Table 3).
When the association between the isolate and degree of disease severity was evaluated, a significant association (P < 0.05) was evident between the degree of the disease severity and the rate of isola-tion of MmmLC (odds ratio = 2.89; 95 % CI = 1.54 < OR < 5.43; chi-square = 12.9; P = 0.000).In other words, sheep from which MmmLC was isolated were 2.89 times more likely to have manifested severe lesions than mild lesions.
The association between age and the presence of clinical signs was statistically significant (P < 0.05).It was found that young sheep (0-6 teeth) were more likely to have lesions than adult sheep.
As A. pyogenes and MmmLC were the 2 most common isolates in sheep with clinical signs of the disease, their possible synergistic role in the disease process was assessed.They were simultaneously isolated from 33 (31.7 %) sheep out of 104 clinically affected sheep and from 1 (0.9 %) sheep out of 116 healthy sheep (Table 4).The odds ratio analysis showed that when A. pyogenes and MmmLC were present together, they were 53.5 times more likely to occur in clinically affected sheep than in unaffected sheep (Table 5).

Examination for Chlamydophila and viruses
No Chlamydophila antigens were detected in 20 selected swabs each from affected and unaffected sheep, using the Clearview ® chlamydia antigen detection ELISA kit.
No viruses were isolated in the cell cultures from the 160 specimens examined from affected sheep.

DISCUSSION
Although, Trichard et al. 36 and Trichard and Van Tonder 37 described posthitis and vaginitis as part of the disease syndrome, the involvement of the prepuce, vaginal and vulval vestibule was essentially absent in the affected sheep examined in this study (Figs 3-6).This observation is consistent with the description of the disease given by Dent 9 , Webb and Chick 38 and Deas 8 .In almost all the cases examined the tissues affected were the soft tissue of the glans penis of rams and the muco-cutaneous junction of the vulval lips of ewes.According to Deas 8 the fibro-elastic tissues of the shaft of the penis are less likely to be affected by such lesions than the soft tissues of the glans penis.The name ulcerative balanitis and vulvitis as suggested by Deas 8 , Linklater and Smith 24 , Dunn 12 and Greig 13 is therefore a more appropriate description of the disease as seen in South Africa.
In the present case-control study, mycoplasmas were isolated from 59.1 % (130/220) of the sheep, with MmmLC being the predominant species (61.5 %).There was a difference in the isolation of mycoplasma organisms between healthy and affected sheep.isms were isolated 3.86 times more often from the affected sheep than from the unaffected sheep.This result points to an association of these mollicutes with ulcerative genital disease of sheep in South Africa.It confirms the suppositions expressed in earlier publications that mycoplasma organisms are important contributing pathogenic agents in ulcerative balanoposthitis and vulvovaginitis in sheep 4,15,36 .The results of this study provide strong evidence for the association of MmmLC with a clinical genital tract infection in Dorper sheep in South Africa.This postulation is supported by the fact that there was a higher rate of isolation of MmmLC from clinically affected sheep compared with the apparently normal flock mates (Table 3).Furthermore, the absence of viruses, Chlamydophila species and inconsistent isolation of different species of bacteria from clinically infected sheep indicates that these organisms are probably not responsible for the clinical signs observed.However, it seems probable that bacteria may contribute by initiating or aggravating the condition when present in association with mycoplasmas.This is substantiated by the combined isolation of A. pyogenes and MmmLC from 31.7 % of sheep with clinical signs of the disease (Tables 4, 5).
The study has shown, for the first time, an association of MmmLC and A. pyogenes with genital tract infection of sheep in South Africa.Mycoplasmas are welldefined pathogens acting singly or in consort with other agents.Although it cannot be assumed with certainty, it is possible that on the basis of the results obtained in this investigation, the concurrent isolation of A. pyogenes and MmmLC supports the concept that the bacterial pathogen is the primary factor that 'opens the door ' for mycoplasmal invasion with subsequent pathological consequences.Arcanobacterium pyogenes is considered as the most pathogenic bacterium residing on mucosal surfaces as it can cause tissue damage by means of its virulence factor, the exotoxin pyolysin 5 .The organism has for example been isolated from 3 of 10 pus specimens taken from goats suffering from vulvitis in Nigeria 17 .Furthermore, MmmLC has an endotoxic activity which is able to induce damage to epithelial cells of the genital tract 30 .It is therefore probable that these endotoxins can be held responsible for aggravating the cellular damage of the genital system of sheep colonised by this Mycoplasma species.
It seems likely that A. pyogenes and MmmLC act synergistically to induce the characteristic lesions of the disease.A synergistic interaction of A. pyogenes with other organisms in a variety of other clinical conditions has been described 27,29,32 .
In this study, Mycoplasma capricolum was isolated from 7 sheep with lesions of ulcerative balanitis and vulvitis (Table 3).This organism is usually associated with polyarthritis in sheep and goats 19 , but has also been isolated from the genital tract of sheep in England in an outbreak of vulvovaginitis and balanoposthitis 20 .However, from the low isolation rate detected in the investigation reported here, it appears unlikely that M. capricolum played a primary pathogenic role.
Although Mycoplasma species group 7 was isolated from 23 % of affected sheep with an odds ratio of 5.19 (Table 3), it is difficult to establish whether or not it has any significant role in the clinical signs of ulcerative balanitis and vulvitis.Moreover, there are no reports in the literature which incriminate this organism as a causal agent of any disease.
Nine of the 15 strains of A. laidlawii isolated were from animals showing balanitis or vulvitis (data not shown).Although, A. laidlawii is usually regarded

Fig. 1 :
Fig. 1: Map of the Northern Cape Province indicating the study sites.

Fig. 6 :
Fig. 6: Ulcer that damaged a large part of the glans penis.

Table 5 : Odds ratio analysis to determine the association between the mutual isolation of Arcanobacterium pyogenes and MmmLC and clinical disease status.
Arcanobacterium pyogenes isolated from clinical cases of ulcerative balanitis and vulvitis in Dorper sheep in South Africa.Journal of the South African Veterinary Association 76(4): 204-208 24.Linklater K A, Smith M C 1993 Diseases of the reproductive system and udder.In Color atlas of diseases of the sheep and goat.Wolfe, London: 123-135 25.Livingstone C W, Gauer B B, Shelton M 1978 A specific ureaplasmal serotype associated with ovine uterine infections.American Journal of Veterinary Research 39: 1699-1701 26.Livingstone C W, Gauer B B 1983 Occurrence of Mycoplasma spp.(2D) in a Texas sheep flock.American Journal of Veterinary Research 44: 868-869 27.McCaughey W J, Ball H J 1981 Distribution of ureaplasmas in the urogenital tract of ewes.Veterinary Record 109: 472 28.McCaughey W J, Ball H J 1983 Ureaplasmas in ewes.Veterinary Record 12: 441 29.Roberts D S 1967 The pathogenic synergy of Fusiformis necrophorus and Corynebacterium pyogenes.II.The response of F. necrophorus to a filterable product of C. pyogenes.British Journal of Experimental Pathology 48: 66-73 30.Rosendal S 1986 Mycoplasma.In Gyles C L, Thoen C O (eds) Pathogenesis of bacterial infections in animals (1st edn).Iowa State University Press, Ames: 205-214 31.Singh N, Rajyan B S, Mohanty G C 1974 Granular vulvovaginitis in goats associated with Mycoplasma agalactiae.Cornell Veterinarian 64: 435-442 32.Takeuchi S, Nakajima Y, Hashimoto K 1983 Pathogenic synergism of Fusobacterium necrophorum and other bacteria in formation of liver abscesses in BALB/c mice.Japanese Journal of Veterinary Science 45: 775-781 33.Tarigan S, Webb R F, Kirkland D 1987 Caprine herpesvirus from balanoposthitis.Australian Veterinary Journal 64: 321 34.Thrusfield M 1995 Observational studies.In Veterinary Epidemiology (2nd edn).Blackwell Science, The University Press, Cambridge: 221-223 35.Tiwana J S, Singh N 1982 Isolation of Acholeplasma oculi from genital lesions in sheep.Veterinary Record 111: 417 36.Trichard C J V, Jordan P, Prozesky L, Jacobsz E P, Henton M M 1993 The identification of Mycoplasma mycoides mycoides Large Colony as the aetiological agent of balanoposthitis and vulvovaginitis in sheep in South Africa.