Histological and genotypical characterization of feline

cutaneous mycobacteriosis: a retrospective study of

formalin-fixed paraffin-embedded tissues
Blackwell Publishing Ltd

Jennifer L. Davies*, Jennifer A. Sibley,

Sherry Myers, Edward G. Clark and
Greg D. Appleyard*
*Department of Veterinary Pathology, Western College of Veterinary
Medicine, University of Saskatchewan, Saskatoon, Saskatchewan,
Canada, S7N 5B4
Department of Veterinary Microbiology, Western College of
Veterinary Medicine, University of Saskatchewan, Saskatoon,
Saskatchewan, Canada, S7N 5B4
Prairie Diagnostic Services, Western College of Veterinary Medicine,
University of Saskatchewan, Saskatoon, Saskatchewan, Canada, S7N 5B4
Dr Davies was supported by an Inter-provincial Graduate Student
Fellowship. Funding for this study was obtained from the WCVM
Companion Animal Health Fund.
Previously presented as a poster at American College of Veterinary
Pathologists Conference, Banff, November 2003.
Correspondence: G. D. Appleyard, Provincial Laboratory for
Public Health, Calgary, Alberta, Canada, T2N 4W4.
E-mail: g.appleyard@provlab.ab.ca

Abstract
Twenty-nine cases presumptively diagnosed as feline
cutaneous mycobacteriosis were evaluated microscopically with haematoxylin and eosin and modified
Fites stained sections using archived formalin-fixed
paraffin-embedded tissue specimens. Lesions were
characterized histologically as feline leprosy (7 cases
lepromatous and 16 cases tuberculoid) or atypical
mycobacteriosis (3 cases); three cases did not fit
these criteria and were classified as miscellaneous.
Actinomycetales-specific polymerase chain reaction
(PCR) of variable regions 1, 2 and 3 of the 16S ribosomal
RNA (rRNA) gene and subsequent sequence analysis
of the amplicons were performed to identify the species
of mycobacteria associated with each case. Together,
this study identified 10 different Actinomycetales
organisms with greater than 98% nucleotide sequence
identity to named species, nine were of the genus
Mycobacterium and eight were associated with feline
leprosy (both lepromatous and tuberculoid). Based on
this study, we conclude that feline cutaneous mycobacteriosis should be considered as a syndrome with
varied clinical and histological presentations associated with a variety of different Mycobacterium species, organisms other than Mycobacterium sp. may be
associated with feline cutaneous mycobacteriosis
lesions, and molecular diagnostic techniques can be
an important tool for identifying agents associated
with lesions of feline cutaneous mycobacteriosis.
Accepted 6 March 2006

Introduction
Mycobacteria-associated skin diseases in cats are complex and confusing, primarily because of the number of
species of mycobacteria involved, many of which are difficult or impossible to culture, and because of the variety
of disease manifestations involved. There can be different
histological reaction patterns for the same organism, similar histological reaction patterns for different organisms
and similar clinical presentations for different processes.
Classification schemes for these diseases are difficult to
apply, i.e. whether the disease localizes primarily to the
skin and subcutis or if it is a systemic problem that is only
recognized initially as skin disease. Our ability to obtain
a clear understanding of mycobacteria-associated skin
disease in cats is made difficult by low numbers of case
reports in the literature, and small case series from different geographical locations around the world. Few studies,
however, have been carried out to assess the diversity of
organisms that may be involved.
Three manifestations are recognized including feline
leprosy, cutaneous tuberculosis and atypical mycobacteriosis. Feline leprosy refers to a Mycobacterium sp. infection in which single or multiple granulomas form in the
skin or subcutis with acid-fast bacilli (AFB) that can be
difficult to culture using routine bacteriologic methods.1,2
It occurs in two forms, lepromatous and tuberculoid, most
often in cats under 3 years old with lesions on the head
and limbs and has an accompanying peripheral lymphadenopathy. The condition occurs in temperate, coastal
climates and has been reported in New Zealand, Australia,
western Canada, the Netherlands, Britain and the USA.
The causative agent is presumed to be Mycobacterium
lepraemurium, a fastidious, slow-growing organism,2,3
however, recent studies using the polymerase chain reaction (PCR) have recently demonstrated the presence of
other mycobacteria including novel species.46
Feline cutaneous tuberculosis is most often caused by
infection with M. bovis and rarely with M. tuberculosis.1
The alimentary tract is most commonly affected with
tubercles developing in the intestine and mesenteric
lymph nodes following ingestion of milk or meat from
infected cattle. Cutaneous lesions can develop from bite
wounds with local spread but may also develop as a
consequence of dissemination from the gastrointestinal
tract. Lesions are characterized by dermal nodules, ulcers,
plaques, draining tracts and abscesses that exude a
thick, yellow to green fluid and are commonly associated
with localized or generalized lymphadenopathy.2,3 Face,
neck, shoulders, forepaws and, less commonly, ventral
thorax and tail are expected sites for these cutaneous
lesions.

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology. 17; 155162

155

JL Davies et al.

Feline atypical mycobacteriosis is most often characterized by panniculitis and chronic or recurrent fistulous
tracts and nodules associated with a previous history of
trauma and typically affects the inguinal fat pad and ventral
abdomen. Lesions are usually associated with saprophytic
mycobacteria ubiquitous in soil and water. Potentially
pathogenic species for cats include M. fortuitum, M. phlei,
M. smegmatis, M. chelonae and M. thermoresistible.3,7
A novel tuberculosis syndrome was reported by GunnMoore and colleagues8 is a syndrome that primarily
affects young male cats considered to be avid hunters.
The syndrome is characterized by submandibular lymphadenopathy and cutaneous lesions, which primarily
involve the face, extremities, tail base and perineum. The
agent involved was found to be M. microti.
Mycobacteriosis in cats has also been associated with
slow-growing mycobacteria including M. xenopi,9 M.
terrae10 and members of the M. avium complex (MAC).3
While members of MAC can cause localized disease, they
are more commonly associated with disseminated disease (i.e. not limited to the skin).
In recent years, several cases of feline cutaneous
mycobacteriosis were described and could not be classified into the recognized manifestations of this disease.5,11
Malik et al.5 have suggested a new classification scheme
wherein feline leprosy comprises two distinct clinical syndromes. One syndrome occurs in young cats (< 4 years)
and is associated with tuberculoid histology, scant to moderate AFB, ulceration, necrosis and a rapidly progressive
clinical course attributable to infection with M. lepraemurium. The other syndrome occurs in older cats (> 9 years)
and is associated with lepromatous histology, numerous
intracellular AFB that stain with haematoxylin, an absence
of necrosis and ulceration, a slowly progressive clinical
course and is associated with a novel Mycobacterium species with high nucleotide sequence identity to M. leprae,
M. haemophilum and M. malmoense. Coincident with
Malik et al.s work, a generalized granulomatous dermatitis, fasciitis and multisystemic granulomatous disease
associated with another novel Mycobacterium, M. visible,
has been reported by the authors of this monograph.11,12
These reports indicate that feline cutaneous mycobacteriosis does not always conform to the three accepted
clinical manifestations of disease and is associated with
more Mycobacterium species than previously recognized.
Molecular DNA techniques, such as PCR and 16S ribosomal RNA (rRNA) gene sequencing, have advanced our
understanding of Mycobacterium taxonomy and the
microbial diversity associated with feline cutaneous mycobacteriosis. Molecular genetic identification schemes are
an alternative to culture, for rapid and accurate diagnosis,
and enable recognition of previously undescribed taxa,
particularly those that have not yet been cultured.13,14 In
this monograph, we describe the results of a retrospective
histological and genotypical analysis of 29 presumptive
cases of feline cutaneous mycobacteriosis using formalinfixed, paraffin-embedded tissues collected between
1971 and 2001. The objectives were to histologically evaluate archived cases of feline cutaneous mycobacteriosis
and to correlate histology with results of DNA sequencebased identification of mycobacteria from the same
samples.
156

Materials and methods

Histopathology
Clinical data and tissue specimens from 29 cases of presumptive
feline cutaneous mycobacteriosis submitted between 1971 and 2001
were obtained from Prairie Diagnostic Services (Saskatoon, Saskatchewan, Canada, n = 21), Abbey Veterinary Services (Devon, UK, n = 1)
and Alpha Scientific Limited (Hamilton, New Zealand, n = 8). On three
occasions (cases 1, 7, and 8), lesions redeveloped after the initial
investigation and tissue had been resubmitted at a different time, up
to 1 year later from the same animal. When this occurred, the earlier
sample is labelled A and the latter B. Historical case data included geographic origin, breed, age and sex of the cats.
Tissues had been fixed in 10% neutral buffered formalin during the
period of transport to the diagnostic laboratory and embedded in paraffin
blocks upon arrival. Sections (4.5 m thick) were stained with haemotoxylin and eosin (H&E) and modified Fites (acid-fast) stains. Tissue sections were examined microscopically to obtain qualitative information
regarding mycobacteria (numbers, location, visibility using H&E stain)
and the accompanying host response (type, location of inflammation,
presence or absence of caseous necrosis, epidermal ulceration and
peripheral nerve involvement). Based on histological features of the
lesions, cases were categorized as one of the following: tuberculoid form
of feline leprosy (pyogranulomatous dermatitis and panniculitis, multifocal
to coalescing areas of necrosis, moderate to scant intrahistiocytic AFB,
but best observed in areas of necrosis), lepromatous form of feline
leprosy (pyogranulomatous dermatitis and panniculitis, multinucleate
giant cells, absence of necrosis, abundant intrahistiocytic AFB), atypical
mycobacteriosis (pyogranulomatous panniculitis, centred around extracellular fat vacuoles containing AFB) or other (pyogranulomatous panniculitis
or nodular granulomatous dermatitis, numerous extracellular fat vacuoles,
multifocal areas of necrosis, moderate to no AFB). Tissue sections were
examined without prior knowledge of the results of the molecular
genetic analysis of the Mycobacterium species identified in each case.

Molecular analysis
From each paraffin block, ten 10 m tissue sections were obtained
using a microtome mounted with a sterile disposable blade that was
discarded after each case. Sections were deparaffinized using two 1mL xylene and 80% ethanol washes and dried in a rotary evaporator.
DNA was extracted from the tissue samples using 600 L of lysis buffer
[100 mM NaCl, 500 mM Tris (pH 8.0), 10% sodium dodecylsulfate],
vortexing with 200 L of 0.1 mm diameter zirconia beads (Biospec
Products Inc., Bartlesville, OK, USA) and digesting for 3 h with proteinase K (0.2 mg mL1). Two solvent extractions with phenolchloroform
(1:1) and one extraction with chloroform alone were performed and
nucleic acids were concentrated by precipitation in cold 95% ethanol.
The following primers were used to amplify a portion of the
Mycobacterium 16S rRNA gene. Two sets of primers were used:
17F 5 -CATGCAAGTCGAACGGAAAG-3 and 525R 5 -TTTCACGAACAACGCGACAA-3, and 53F 5-GAGTGGCGAACGGGTGAGTAA3 and 485R 5-TTACGCCCAGTAATTCCGGACAA-3.15 Primers 17F
and 525R amplify approximately 500 bp of the 16S rRNA gene in the
V2 and V3 region.4,16 Primers 53F and 485R amplify a smaller portion
of the same region and were used in a nested reaction in combination
with 17F and 525R primers in cases where the primary reaction failed
to generate sufficient product to be observed on a gel. This occurred
in those cases where scant numbers of AFB were observed histologically. Both regular and nested assays were performed using the same
amplification conditions: a 2-min initial denaturation phase at 94 C;
denaturation, annealing and elongation steps were performed for
1 min at 94 C, 2 min at 60 C and 2 min at 72 C, respectively. A final
elongation step was performed at 72 C for 10 min. PCR products
were visualized on 1.5% agarose gels with ethidium bromide staining
and purified using the QIAquick PCR Purification Kit (QIAGEN Inc.,
Mississauga, Ontario, Canada). These were sequenced directly, in
both directions, using the same primers (DNA Services Laboratory,
Plant Biotechnology Institute, National Research Council, Saskatoon,
Saskatchewan, Canada). Negative controls included an extraction blank
(no sample in the extraction) and reagent blanks (no DNA in PCR).
DNA sequence analysis was performed using the ClustalV algorithm

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

Feline cutaneous mycobacteriosis

(Megalign, DNA Star, Madison, WI, USA) and searches of both GenBank16
and RIDOM17 databases.
Sequence identity among isolates of the same Mycobacterium
species using the equivalent partial 16S rRNA loci found in GenBank
generally ranged from 98% to 100% but as low as 96%, while
sequence identity between differing species was generally less than
96% (data not shown). Since nucleotide sequence identity may vary
between isolates depending on which gene is selected for analysis,
it may not be possible, or appropriate, to definitively assign a minimum nucleotide identity value to characterize isolates of a given species. We have therefore only reported a tentative assignment of
species relatedness for organisms with a nucleotide sequence identity greater than 98%. Since sequences similarities between 98% and
100% are only tentative associations with the species they most
resemble, a relatedness-group designation was also applied.

Results
Twenty-nine cases of presumptive cutaneous mycobacteriosis were examined; 20 from British Columbia, Canada,

8 from New Zealand and 1 from the UK. Cats ranged in

age from 10 months to 14 years and included 17 males
(7 intact, 10 castrated), 11 females (1 intact, 10 spayed) and
one animal of unknown sex. The majority of cats were
domestic breeds (19 DSH, 3 DLH) while only one was
pure-bred (Siamese). The breed was unknown to us for six
cases. Our case series did not include any cases of presumptive cutaneous tuberculosis. Classification of lesions
and subsequent molecular genetic analysis of associated
mycobacteria are summarized in Table 1. None of the
cases had routine mycobacterial culture performed.
Tuberculoid form of feline leprosy
There were 16 cases with tuberculoid histology from the
coastal areas of western Canada, New Zealand and the
UK. Cats ranged from 10 months to 14 years with an
average of 3.5 years. Five were sexually intact (4 male
and 1 female) and 11 were neutered (7 castrated males

Table 1. Cases of feline cutaneous mycobacteriosis, geographical origin and associated infectious agents, listed by histological type

Case Breed

Age
(years)

Sex

Origin Lesion history

14
16
12

DSH
2.5
Siamese
2
DSH
N/A

M(N) BC
M(N) BC
M
NZ

11

DLH

M(N) BC

15
4
5
6
2
10

DSH
DSH
DSH
N/A
N/A
DSH

14
0.8
4
2
13
1.5

DSH

13
9

DSH
DSH

1.5
10

1A
1B
7A

DSH
DSH
N/A

1.5
2
1

7B
8A
8B
18
20
23

N/A
DSH
DSH
DLH
N/A
N/A

2
9
10
7
3
3

F(S)
M(N)
M(N)
M(N)
N/A
F(S)

BC
BC
BC
BC
BC
BC

17
19
21
22
24

DSH
DSH
DSH
DSH
DSH

2
1.5
9
14
2

M(N)
F(S)
M
F(S)
F(S)

BC
BC
NZ
BC
NZ

25
26

DSH
DLH

3.5
4

M
M

NZ
NZ

27
28
29

N/A
DSH
DSH

1.5
3
2.5

F(S) BC
M(N) NZ
F(S) BC

F(S)
M
M
M(N)
F(S)
M

BC
BC
BC
BC
BC
NZ

M(N) BC
F
F(S)

NZ
UK

M(N) BC
M(N) BC
F(S) BC

ulcerated mass on scrotum

two nodules in skin
three masses: flanks,
hind limb, chest
2-cm mass on right leg,
two smaller masses
mass on right hind limb
3-cm mass on upper lip
mass on hock
sore near tail
three masses: flank, tail, elbow
three ulcerated lumps under
chin, one mass on ear
ulcerated mass on distal limb,
two smaller masses
two masses near base of ear
two masses
right carpus
unknown
chronic, ulcerated mass on leg,
second small mass
unknown
mammary mass
unknown
lower eyelid
two masses
large mass on forelimb,
several other nodules
mass on left tibia
right cheek
right tarsus
ulcerated sore on head
chronic draining wound in inguinal
region, previous history of cat bite
mass on thorax
recurrent pustules and
abscesses on chest
growth on flank
nonhealing abscess
regressing lesion previously
diagnosed as feline leprosy

Histology
type

Species that
the sequence
most resembles*

% sequence
identity (matches/
fragment length)

Relatednessgroup

T
T
T

M. lepraemurium
M. lepraemurium
M. lepraemurium

98.1 (460/469)
99.1 (465/469)
99.4 (466/469)

A
A
A

97.4 (457/469)

T
T
T
T
T
T

M. intracellulare
AF144747
AF144747
AF144747
M. szulgai
*

98.9 (465/470)
98.2 (449/457)
100 (457/457)
98.9 (452/457)
98.9 (464/469)
97.8 (447/457)

B
C
C
C
E
*

M. mucogenicum

98.0 (448/457)

T
T

98.6 (445/451)
100 (471/471)

D
F

T
T
T

M. mucogenicum
M. tuberculosis
complex
M. lepraemurium
M. intracellulare
M. lepraemurium

99.8 (468/469)
99.8 (477/478)
100 (473/473)

A
B
A

T
T
T
L
L
L

M. intracellulare
*
M. septicum
M. lepraemurium
M. lepraemurium
M. lepraemurium

98.3 (461/469)
95.7 (445/465)
99.1 (453/457)
99.8 (468/469)
99.4 (466/469)
99.8 (468/469)

B
*
G
A
A
A

L
L
L
L
A

*
M. intracellulare
*
AJ294746
*

97.9 (458/468)
99.1 (465/469)
95.2 (437/459)
98.3 (461/469)
96.1 (441/457)

*
B
*
H
*

A
A

M. smegmatis
*

99.6 (495/497)
96.3 (490/509)

I
*

M
M
M

*
*
Rhodococcus
erythropolis

93.2 (428/459)
97.1 (444/457)
99.1 (454/458)

*
*
J

* < 98% nucleotide sequence identity not considered a match to previously characterized sequence.
Histology type: T, tuberculoid; L, lepromatous; A, atypical; M, miscellaneous.

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

157

JL Davies et al.

Figure 1. Photomicrograph: Feline skin, case 14. Pyogranulomatous

dermatitis with large central focus of caseous necrosis typical of the
tuberculoid form of feline leprosy. H&E. Bar = 100 m.

Figure 2. Photomicrograph: Feline skin, case 14. Acid-fast bacilli in

necrotic foci, typical of the tuberculoid form of feline leprosy. Modified
Fites acid-fast stain. Bar = 20 m.

and 4 spayed females). Lesions generally consisted of

single or multiple cutaneous nodules on the head and limbs,
with fewer occurrences on the scrotum, tail and ventral
abdomen.
Microscopically, lesions consisted of multifocal to coalescing pyogranulomas with central caseous necrosis
(Fig. 1) and moderate to many peripheral lymphocytes and
plasma cells in the dermis and panniculus. Multinucleate
giant cells were scant to moderate and only present in
some cases. Bacteria were not visible with H&E stain, but
modified Fites stain revealed variable numbers of AFB,
mainly in the areas of necrosis (Fig. 2). Peripheral nerves
were not involved.
In the tuberculoid group (except cases 1, 7 and 8), partial
16S rRNA gene sequences from three cases were identified as 100% match or were similar to M. lepraemurium
(Tables 1 and 2). One case (case 15) had sequence that
was similar to M. intracellulare. Two cases (cases 10 and
11) had sequences that were too variable to be matched.
Three cases had sequences that were identified as, or
were similar to, a novel species (AF144747). One case
had sequence similar to M. szulgai and two cases had
sequence similar to M. mucogenicum. Another in this
group (case 9) yielded partial 16S rRNA gene sequence
with 100% nucleotide identity over 471 bp with a member
of the M. tuberculosis complex.
For three cats (cases 1, 7 and 8) biopsies were submitted for histology on two different occasions. A single

organism was identified by molecular genetic analysis

in each biopsy; however, the organisms were not the
same for each of the paired samples. Two cases had
sequences that were identical or similar to M. lepraemurium and M. intracellulare, one case had sequences
similar to M. septicum and one case revealed sequences
too variable to match to a previously characterized
sequence.
Lepromatous form of feline leprosy
There were seven cases with lepromatous histology from
the coastal areas of western Canada and New Zealand.
Age of cats in this group ranged from 1.5 years to 14 years
with an average of 5.7 years. One cat was sexually intact
(male), five were neutered (two castrated males and three
spayed females) and the sex of one cat was not known.
In general, lesions included single or multiple cutaneous
nodules on either the head (lip, cheek, eyelid) or limbs.
Microscopically, there was pyogranulomatous dermatitis
and panniculitis characterized by diffuse sheets of epithelioid macrophages and neutrophils with moderate to abundant multinucleate giant cells (Fig. 3). Caseous necrosis
was absent. Inflammation of the peripheral nerves was
absent. On H&E-stained sections, the cytoplasm of the
giant cells and macrophages was granular, suggesting the
presence of moderate to abundant intracellular bacteria
and this was confirmed with modified Fites staining of
intracellular AFB (Fig. 4).

Table 2. Number of samples with histological features and Mycobacterium sp. sequence relatedness groups identified by molecular techniques
Feline leprosy
Relatedness group

Species that the sequence most resembles*

Tuberculoid

Lepromatous

Atypical

Miscellaneous

A
B
C
D
E
F
G
H
I
J
*

M. lepraemurium
M. intracellulare
Mycobacterium sp. (AF144747)
M. mucogenicum
M. szulgai
M. tuberculosis complex
M. septicum
Mycobacterium sp. (AJ294746)
M. smegmatis
Rhodococcus erythropolis
Insufficient match to previously characterized sequence

5
3
3
2
1
1
1
0
0
0
3

3
1
0
0
0
0
0
1
0
0
2

0
0
0
0
0
0
0
0
1
0
2

0
0
0
0
0
0
0
0
1
2

*< 98% nucleotide sequence identity not considered a match to previously characterized sequence.

158

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

Feline cutaneous mycobacteriosis

Figure 3. Photomicrograph: Feline skin, case 18. Numerous

multinucleate giant cells in the lepromatous form of feline leprosy.
H&E. Bar = 50 m.

Figure 4. Photomicrograph: Feline skin, case 18. Numerous

intracellular acid-fast bacilli characteristic of the lepromatous form
of feline leprosy. Modified Fites acid-fast stain. Bar = 20 m.

Figure 6. Photomicrograph: Feline skin, case 22. Numerous

intracellular acid-fast bacilli showing parallel stacking. Modified
Fites acid-fast stain. Bar = 25 m.

Figure 7. Photomicrograph: Feline skin, case 24. Atypical

mycobacteriosis demonstrating pyogranulomas centred on
extracellular fat vacuoles with the organisms (arrowheads). H&E.
Bar = 50 m.

sections that were acid fast with modified Fites acid-fast

stain (Figs 5 and 6).

Figure 5. Photomicrograph: Feline skin, case 22. Numerous

filaments and bacilli typical (arrowheads) of some novel
mycobacteria. H&E. Bar = 20 m.

Partial 16S rRNA gene sequences from three cases in

the lepromatous group revealed sequences similar to M.
lepraemurium (Tables 1 and 2). One case (case 19) had
sequences similar to M. intracellulare. Two cases (cases
17 and 21) had sequences too variable to match. One case
(case 22), with sequence similar to unnamed species of
mycobacteria (AJ294746),5 had lesions with numerous
intracellular, basophilic filaments and rods on H&E-stained

Feline atypical mycobacteriosis

These three cases were obtained from New Zealand. Cats
were young, ranging from 2 years to 4 years in age. Two
were male (both intact) and one was a spayed female.
Cutaneous lesions consisted of draining wounds and
abscesses on the thorax and the inguinal area. Microscopically, lesions were characterized by pyogranulomas centred on clear extracellular fat vacuoles and in one case
organisms were visible with H&E stain (Fig. 7). Caseous
necrosis was marked in two of the cases. In all three cases
AFB in extracellular fat vacuoles were scant (Fig. 8). One
case revealed sequence similar to M. smegmatis and one
case had sequences too variable to match but could be
characterized based on 16S rRNA gene variable region 3
sequence motifs to be members of the rapidly growing
group of mycobacteria.
Miscellaneous cases
This group consisted of two cases from western Canada
and one from New Zealand. Cats in this category were
young, ranging from 1.5 years to 3 years in age. Two were

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

159

JL Davies et al.

Figure 8. Photomicrograph: Feline skin, case 24. Acid-fast bacilli

within extracellular fat vacuoles characteristic of atypical
mycobacteriosis. Modified Fites acid-fast stain. Bar = 50 m.

spayed females and one was a castrated male. Two cats

had marked pyogranulomatous panniculitis suggestive of
atypical mycobacteriosis; however, AFB were scant to
moderate in areas of necrosis and were not located within
extracellular fat vacuoles, characteristic of atypical mycobacteriosis cases.18 Partial 16S rRNA sequencing revealed
the lesions contained organisms whose sequence was
too variable to match previously characterized sequence.
The third miscellaneous case (case 29) showed multifocal,
nodular, pyogranulomatous perifollicular inflammation,
and modified Fites stained tissue sections were negative
for AFB, a finding which would be unusual for feline cutaneous mycobacteriosis, and partial 16S rRNA gene
sequence was similar to Rhodococcus erythropolis.

Discussion
Historically, the term feline leprosy refers to mycobacterial
infections in which single, or multiple, granulomas form
in the skin, or subcutis, in association with AFB that cannot be cultured in vitro using routine mycobacteriological
methods.1,2 Mycobacterium lepraemurium is the putative
pathogen based on in vivo transmission studies19 and
delayed-type hypersensitivity reactions in cats20 and,
more recently, other species of mycobacteria have been
associated with this disease including M. avium, M.
intracellulare, an unnamed Mycobacterium species
(AJ294746) with highest nucleotide identity to M.
malmoense, M. leprae, and M. haemophilum4,5 and a
slow-growing species of the M. terrae complex.9
Partial or whole sequencing of the 16S rRNA gene is a
well-established method used to identify bacteria, particularly mycobacteria,14,15,21 but is incapable of differentiating some mycobacterial species, particularly isolates of
the M. avium and M. tuberculosis complexes.22 Caution
needs to be exercised in the interpretation of such data
and species relatedness should be viewed as tentative.
Sources of sequence variation encountered in this study
likely included damage to DNA archived in formalin-fixed
paraffin-embedded tissues, microheterogeneity between
isolates within species and microheterogeneity between
copies of the 16S rRNA gene within individual bacterial
cells.22
160

This study revealed the presence of 16S rRNA genes

associated with skin lesions from 29 cases of presumptive
feline cutaneous mycobacteriosis, which shared high
sequence similarity (> 98%) with nine different mycobacteria [M. lepraemurium, M. intracellulare, M. septicum, M.
smegmatis, M. mucogenicum, M. szulgai, Mycobacterium
sp. (AF144747), M. tuberculosis complex and Mycobacterium
sp. (AJ294746)]. Nine cases had sequence that was too
variable to be matched to previously characterized sequence
with certainty. Mycobacterium szulgai is a slow-growing
species and, to the best of our knowledge, has not been
reported in association with lesions from cats. Mycobacterium mucogenicum, a rapidly growing species found
in drinking water has been associated with outbreaks of
nosocomial disease such as post-traumatic skin infections and catheter sepsis23 and M. septicum is a newly
identified species also associated with catheter-related
bacteremia.24
In this study, lesions from three cats were sampled at
two different times. Unfortunately, the case histories did
not reveal whether the lesions were both present at the
initial sampling so we were unable to determine whether
there was evidence of a common inciting agent or pathogenesis. Despite two different sequences being associated with the samples from each cat, histology was the
same (tuberculoid).
Three of 16 tuberculoid form feline leprosy cases in this
study were associated with sequences similar to Mycobacterium sp. Murphy (AF144747). This organism was
described as the causative agent of canine leproid granuloma syndrome (LGS) in Australia and, more recently, has
been recognized in similarly affected dogs from the
USA.25,26 In dogs, LGS is considered to be self-limiting and
regresses with or without surgical intervention. An unusual histological finding in our feline cases was the presence of caseous necrosis, a feature not previously
described in canine LGS,27 and the distribution of the
lesions in cats differed from that of dogs. In this study, the
lesions were not limited to the head but also occurred on
the hind limbs and tail. The anatomic location of lesions
from cats in this study suggests inoculation of the organism through cat-fight wounds, which was further supported as two of three of these cases were from sexually
intact males. All three of our feline cases were from the
coastal regions of British Columbia, which has a cool,
moist climate similar to the climate in places where canine
LGS occurs.
The most commonly reported causes of atypical mycobacteriosis include M. fortuitum, M. smegmatis, M.
chelonae and M. phlei.2 In this study, we observed panniculitis associated with an organism similar to M. smegmatis.
Three cases in this study could not be classified within
the three traditional forms of cutaneous mycobacteriosis.
Rhodococcus erythropolis was identified from one case
that had previously been diagnosed as consistent with
feline leprosy; however, AFB were not present in the
lesion and so it did not fit the definition of feline leprosy.
Rhodococcus erythropolis infections have not been
described in the cat to date, but R. equi has been
described as the cause of pyogranulomatous skin disease
and cellulitis in one cat.28 In humans, R. erythropolis is
generally considered to be nonpathogenic but has been

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

Feline cutaneous mycobacteriosis

reported in one immunosuppressed patient with multiple cutaneous nodules.29 The etiological relevance of the
R. erythropolis finding is uncertain as it may be a skin
contaminant.
In this study, feline leprosy cases could not be segregated into the same two syndromes observed by Malik
et al.5 In our study, both lepromatous and tuberculoid histology was noted in all ages of cats, whereas Maliks group
found the lepromatous form limited to older cats (> 9 years
old). Mycobacterium lepraemurium was associated with
both the lepromatous and tuberculoid forms of disease in
our study, whereas Maliks group found this organism was
limited to the tuberculoid form. Malik also found a novel
species of Mycobacterium associated with the lepromatous form. In agreement with Maliks study, we identified
one older cat with lesions identical to those described for
the lepromatous form and the associated organism had
16S rRNA gene sequence closely resembling the organism (AJ294746) with highest nucleotide identity to M.
leprae, M. haemophilum and M. malmoense. These findings
suggest that both forms (lepromatous and tuberculoid)
of feline leprosy occur in both old and young cats, that M.
lepraemurium is not limited to tuberculoid lesions and
that this novel species may cause disease more often in
Australia compared to North America.
Our case series did not include any cases of presumptive feline cutaneous tuberculosis. It was therefore a surprise to identify a sequence (case 9) with 100% identity to
members of the M. tuberculosis complex. Histologically,
this case is consistent with the tuberculoid form of feline
leprosy.
Accepted definitions of feline cutaneous mycobacterioses appear to be oversimplified as these diseases have
overlapping histological features and associated mycobacteria making an accurate and consistent diagnosis difficult.
Failure to culture AFB should not be considered specific
evidence of M. lepraemurium infection as many other fastidious, slow-growing mycobacteria are unlikely to grow
in vitro.13 Many of these mycobacteria can be opportunistic
or novel organisms. Molecular diagnostic techniques such
as genotyping maybe required to identify the causative
agents of feline cutaneous mycobacteriosis to gain further
understanding of this syndrome and assess zoonotic risk.
Lastly, a variety of Mycobacterium spp. may result in similar clinical and histological picture.

3.
4.

5.

6.

7.

8.

9.
10.

11.

12.
13.

14.

15.

16.

17.

18.

19.

Acknowledgements
The authors express their sincere gratitude to Prairie Diagnostics Services, Emily Walder, Thelma Lee Gross,
Malcolm Silkstone and Angus Black for providing case
material. We thank M. Siobhan Hughes for technical
advice and the WCVM Companion Animal Health Fund for
financial sponsorship of this study. We thank Dr Joyce
Wolfe (Public Health Agency of Canada, Mycobacteriology) for critical analysis and advice on the manuscript.

References
1. Gunn-Moore D, Shaw S. Mycobacterial disease in the cat. In Practice 1997; 19: 493501.
2. Pederson NC. Atypical mycobacteriosis. In: Pederson NC, ed.

20.
21.

22.

23.

24.

Feline Infectious Diseases. California: American Veterinary Publications, 1988; 194197.

Lemarie SL. Mycobacterial dermatitis. Veterinary Clinics of North
America: Small Animal Practice 1999; 29: 1291301.
Hughes MS, Ball NW, Beck L-A et al. Determination of the etiology of presumptive feline leprosy by 16S rRNA gene analysis.
Journal of Clinical Microbiology 1997; 35: 246471.
Malik R, Hughes MS, James G et al. Feline leprosy: two different
clinical syndromes. Journal of Feline Medicine and Surgery 2002;
4: 4359.
Hughes MS, James MJ, McCarroll J et al. PCR studies of feline
leprosy cases. Journal of Feline Medicine and Surgery 2004; 6:
23543.
Malik R, Wigney DI, Dawson D et al. Infection of the subcutis and
skin of cats with rapidly growing mycobacteria: a review of microbiological and clinical findings. Journal of Feline Medicine and Surgery 2000; 2: 3548.
Gunn-Moore DA, Jekins PA, Lucle VM. Feline tuberculosis: a
literature review and discussion of 19 cases caused by an
unusual mycobacterial variant. Veterinary Record 1996; 138:
538.
Tomasovic AA, Rac R, Purcell DA. Mycobacterium xenopi in a skin
lesion of a cat. Australian Veterinary Journal 1976; 52: 103.
Henderson SM, Baker J, Williams R et al. Opportunistic mycobacterial granuloma in a cat associated with a member of the Mycobacterium terrae complex. Journal of Feline Medicine and
Surgery 2003; 5: 3741.
Appleyard GD, Clark EG. Histologic and genotypic characterization of a novel Mycobacterium species found in three cats. Journal of Clinical Microbiology 2002; 40: 242530.
Tortoli E. Latin grammar headaches. Journal of Clinical Microbiology 2003; 41: 5838.
Springer B, Stockman L, Teschner K et al. Two-laboratory collaborative study on the identification of mycobacteria: molecular versus phenotypic methods. Journal of Clinical Microbiology 1996;
34: 296303.
Cai H, Archambault M, Prescott JF. 16S ribosomal RNA
sequence-based identification of veterinary clinical bacteria. Journal of Veterinary Diagnostic Investigation 2003; 15: 4659.
Skuce RA, Hughes MS, Taylor MJ et al. Detection of pathogenic
mycobacteria of veterinary importance. In: Sachse K, Frey J, eds.
Methods of Molecular Biology, Vol. 216. Totowa, NJ: Humana
Press Inc., 2002: 20120.
Altschul SF, Madden TL, Schffer AA et al. Gapped BLAST and
PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 1997; 25: 3389402.
Harmsen D, Rothgnger J, Frosch M et al. RIDOM: ribosomal differentiation of medical micro-organisms database. Nucleic Acids
Research 2002; 30: 4167.
Gross TL, Ihrke PJ, Walder EJ (eds). In: Veterinary Dermatopathology: Macroscopic and Microscopic Evaluation of Canine and Feline
Skin Disease. Mosby Yearbook Inc., Saint Louis, MO, 1992: 170.
Schiefer HB, Middleton DM. Experimental transmission of a
feline mycobacterial skin disease (feline leprosy). Veterinary
Pathology 1983; 20: 46071.
Leiker DA, Poelma FG. On the etiology of cat leprosy. International Journal of Leprosy 1974; 42: 3125.
Rogall T, Wolters J, Flohr T et al. Towards a phylogeny and definition of species at the molecular level within the genus Mycobacterium. International Journal of Systematic Bacteriology 1990;
40: 32330.
Clarridge JE. Impact of 16S rRNA gene sequence analysis for
identification of bacteria on clinical microbiology and infectious
diseases. Clinical Microbiological Reviews 2004; 17: 84062.
Covert TC, Rodgers MR, Reyes AL et al. Occurrence of nontuberculous mycobacteria in environmental samples. Applied and Environmental Microbiology 1999; 65: 24926.
Schinsky MF, McNeil MM, Whitney AM et al. Mycobacterium
septicum sp. nov., a new rapidly growing species associated with
catheter-related bacteraemia. International Journal of Systematic
and Evolutionary Microbiology 2000; 50: 57581.

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.

161

JL Davies et al.

25. Hughes MS, James G, Ball N et al. Identification by 16S rRNA

gene analyses of a potential novel mycobacterial species as an
etiological agent of canine leproid granuloma syndrome. Journal
of Clinical Microbiology 2000; 38: 9539.
26. Foley JE, Borjesson D, Gross TL et al. Clinical, microscopic, and
molecular aspects of canine leproid granuloma in the United
States. Veterinary Pathology 2002; 39: 2349.
27. Malik R, Love DN, Wigney DI et al. Mycobacterial nodular

granulomas affecting the subcutis and skin of dogs (canine leproid

granuloma syndrome). Australian Veterinary Journal 1998; 76: 4037.
28. Patel A. Pyogranulomatous skin disease and cellulitis in a cat
caused by Rhodococcus equi. Journal of Small Animal Practice
2002; 43: 12932.
29. Vernazza PL, Bodmer T, Galeazzi RL. Rhodococcus erythropolis
infection in HIV-associated immunosuppression. Schweizerische
Medizinische Wochenschrift 1991; 121: 10958.

Rsum Cette tude a valu les aspects microscopiques (hematoxyline-osine et coloration modifie
de Fite) de vingt neuf cas de suspicion de mycobactriose fline cutane. Les cas ont t diagnostiqus
comme des lpres flines (7 cas lpromateux et 16 cas tuberculoides), ou mycobactriose atypique (3 cas);
3 cas ne rpondaient pas aux critres et ont t classs en divers. Des ractions par PCR spcifique
dactinomycetales-specific PCR au niveau de rgions variables 1, 2 et 3 de lARN 16 S ont t ralises pour
identifier les espces de mycobactries associes chaque cas. Cette tude a permis didentifier 10
espces diffrentes dActinomycetales avec plus de 98% didentit de structure avec des espces connues, 9 du genre Mycobacterium et 8 associes la lpre fline (lpromateuse et tuberculode). En se basant
sur ces rsultats, nous concluons que les mycobactrioses flines devraient tre considres comme un
syndrome avec des prsentations cliniques et histologiques varies, associes diffrentes espces de
Mycobacterium. Dautres organismes peuvent tre associs aux lsions de mycobactriose fline, et les
techniques de diagnostic molculaire peuvent tre importantes pour identiier les agents responsables des
mycobactrioses flines.
Resumen Veintinueve casos con un presunto diagnosis de micobacterisos cutnea felina fueron evaluados al microscopia utilizando hematoxilina-eosina y una tcnica modificada de Fite. Las muestras utilizadas
procedan de archivos de bloques en parafina fijados con formalina. Histologicamente las muestras correspondieron con lepra felina (7 casos lepromatosos y 16 casos tuberculoides) o micobacterisosis atpica
(3 casos); 3 casos no correspondieron a ninguno de los criterios anteriores y se clasificaron como miscelaneos. Se utilizo una reaccin de polimerasa en cadena (PCR) especifica para las regiones variables 1, 2 y
3 del 16S-rRNA de Actinomicetales, seguida de la secuenciacin de los amplicones con el fin de identificar
las diferentes especies de micobacteria. En total se identificaron 10 especies de Actinomicetales con ms
de un 98% de equivalencia en la secuencia de nucletidos de especies reconocidas, 9 en el gnero Mycobacterium y 8 asociados con lepra felina (ambas formas lepromatosa y tuberculoide). En base a estos resultados concluimos que la lepra felina debe ser considerada un sndrome con una variedad de presentaciones
clnicas e histolgicas, y asociada con diferentes especies del gnero Mycobacterium. Otros organismos
de diferente gnero tambin pueden estar asociados con la enfermedad y tcnicas de diagnstico molecular
pueden ser una herramienta importante en la identificacin de especies asociadas con lesions de micobacteriosis cutnea felina.
Zusammenfassung Von neunundzwanzig Fllen mit der Verdachtsdiagnose einer felinen kutanen
Mykobakteriose wurden Gewebsschnitte von archivierten, Formalin fixierten, in Paraffin eingebetteten
Gewebeproben mittels Hmtoxylin und Eosin sowie mit einer modifizierten Fites Frbung mikroskopisch
untersucht. Die Vernderungen wurden histologisch als feline Lepra charakterisiert (7 Flle waren lepromats und 16 Flle waren tuberkulr) oder als atypische Mykobakteriose (3 Flle); 3 Flle passten nicht zu
diesen Merkmalen und wurden unter Verschiedenes eingeordnet. Ein fr Actinomycetales (Strahlenpilze)
spezifischer PCR der variablen Regionen 1, 2 und 3 des 16S rRNS Gens und nachfolgende Sequenzanalyse
der Amplikons wurden durchgefhrt, um die Spezies der Mykobakterien zu identifizieren, die mit dem jeweiligen Fall assoziiert waren. Insgesamt wurden in dieser Studie 10 verschiedene Actinomycetales Organismen mit einer mehr als 98%igen Identitt der Nukleotidsequenz der genannten Spezies identifiziert,
9 gehrten zum Genus Mykobakterium und 8 standen im Zusammenhang mit feliner Lepra (lepromats
sowohl als auch tuberkulr). Basierend auf den Ergebnissen dieser Studie schlieen wir, dass die feline kutane
Mykobakteriose als ein Syndrom mit variabler klinischer und histologischer Prsentation betrachtet werden
sollte, welches mit einer Vielfalt von verschiedenen Spezies von Mykobakterium assoziiert sein kann.
Ausserdem knnen andere Organismen als Mykobakterium sp. mit Vernderungen der felinen kutanen
Mykobakteriose im Zusammenhang stehen. Molekulardiagnostische Techniken knnen ein wichtiges Mittel
sein, um die Erreger zu identifizieren, die mit den Lsionen der felinen kutanen Mykobakteriose assoziiert
sind.

162

2006 The Authors. Journal compilation 2006 European Society of Veterinary Dermatology.