Granulomatous

Colitis of Boxer Dogs

Melanie Craven, BVetMed, PhD, MRCVSa,*,
Caroline S. Mansfield, BVMS, MVMb,
Kenneth W. Simpson, BVM&S, PhDa

KEYWORDS
 Histiocytic ulcerative colitis  E coli  Enrofloxacin  IBD  AIEC

Granulomatous colitis (GC) is an uncommon type of inflammatory bowel disease (IBD),

predominant in Boxer dogs younger than 4 years.1–4 There are sporadic reports of GC
in other dog breeds, particularly young French bulldogs,5,6 and in the authors’ obser-
vation. Affected dogs typically present with signs of colitis, hematochezia, and weight
loss, progressing to cachexia in severe cases (Fig. 1).1,7–9 GC was first reported in the
United States in 19651 and later emerged in Australia, Japan, and Europe, becoming
better known as histiocytic ulcerative colitis. However, the authors subscribe to the
original name as described by Van Kruiningen and colleagues1 for several reasons.
First, this name more accurately reflects the histopathologic appearance of the inflam-
matory response, that is, a mix of macrophages, lymphocytes and neutrophils, almost
invariably reported by pathologists as granulomatous inflammation (Fig. 2).2,3,9–13
Second, a histiocyte is a fixed tissue macrophage, whereas the mucosa in GC is tran-
siently packed with recruited macrophages that egress with successful treatment.14
The cytoplasm of macrophages in GC stains positive with periodic acid–Schiff (PAS)
(see Fig. 2, inset), a unique and pathognomonic feature that is strikingly similar to that
of Whipple disease in humans.1 Whipple disease is a rare, systemic bacterial infection
primarily affecting the small intestine. It is caused by Tropheryma whipplei and diag-
nosed by the presence of PAS-positive macrophages in intestinal biopsies.15,16
Because of this similarity and following the occurrence of GC in 9 Boxer dogs from
the same kennel, 6 of which responded to chloramphenicol treatment,1 an infectious
cause has long been suspected in GC. Thus, initial studies focused on searching for
a GC-associated pathogen. Electron microscopic imaging of colon mucosa revealed
occasional bacteria in 4 of 13 affected dogs and abundant coccobacillary structures
resembling Chlamydia within the macrophages of 5 dogs.17 In a later report of GC, the

The authors have nothing to disclose.

a
Department of Veterinary Clinical Sciences, Cornell University, Tower Road, Ithaca, NY
14853–6401, USA
b
University of Melbourne, 250 Princes Highway, Werribee, Victoria 3030, Australia
* Corresponding author. Department of Veterinary Clinical Sciences, VMC 2013, College of
Veterinary Medicine, Cornell University, Tower Road, Ithaca, NY 14853–6401.
E-mail address: [email protected]

Vet Clin Small Anim 41 (2011) 433–445

doi:10.1016/j.cvsm.2011.01.003 vetsmall.theclinics.com
0195-5616/11/$ – see front matter Ó 2011 Elsevier Inc. All rights reserved.
434 Craven et al

Fig. 1. Cachexia in a young Boxer dog with severe GC.

isolation of Mycoplasma spp from the colon of 4 of 11 dogs and the draining lymph
nodes of 3 of 11 dogs raised the possibility of Mycoplasma as a causative agent.
However, experimental inoculation of 8-week-old Boxer puppies with the isolated
Mycoplasma spp did not induce GC.18
With no definitive evidence for a specific pathogen, other investigators suggested
that the scant bacteria visualized within the superficial mucosa were opportune
invaders of an inflamed and ulcerated mucosa.7,19 A primary immune-mediated path-
ogenesis was presumed, and the mucosal immune response in GC was evaluated
using immunohistochemistry.7,19 This evaluation revealed increased numbers of
IgG1 plasma cells, CD31 T cells, L1 cells, and major histocompatibility complex class
II cells, analogous to ulcerative colitis in humans.20 Until 2004, the mainstay of treat-
ment of GC involved immunosuppression with agents such as corticosteroids and
azathioprine in combination with antibiotic therapy and dietary change.3 Responses
to treatment were generally poor, frequently resulting in euthanasia. GC became
considered an incurable immune-mediated disease.4–7,19

Fig. 2. GC-affected colon mucosa showing mucosal ulceration; goblet cell loss; and dense
cellular infiltration with macrophages, lymphocytes, plasma cells, and eosinophils (hematox-
ylin-eosin, original magnification  40). Inset: oamy macrophages positive on periodic acid–
Schiff (PAS) staining, pathognomonic for GC (original magnification  200).
 Granulomatous Colitis of Boxer Dogs 435

RECENT DISCOVERIES
GC and Invasive Escherichia coli
The search for an infectious cause of GC was reignited by reports of long-term remis-
sion in dogs treated with enrofloxacin.3,21–23 The application of culture-independent
molecular methods, namely, immunohistochemistry and fluorescence in situ hybrid-
ization (FISH), enabled the identification of mucosally invasive E coli. Using a polyclonal
E coli antibody, immunoreactivity was documented in the lamina propria macrophages
and the regional lymph nodes of 10 affected dogs.22 Also, immunostaining of colonic
mucosa gave positive results with antibodies against Salmonella, Campylobacter, and
Lawsonia intracellularis. Concurrent work using advanced molecular methods demon-
strated the presence of metabolically active invasive E coli packed within colonic
macrophages.23 This finding was accomplished using FISH, a technique that uses
fluorescent molecules attached to oligonucleotide probes that hybridize to bacterial
16S ribosomal DNA (rDNA). Fluorescent labeling enables clear visualization of bacte-
rial morphology and spatial localization, even against a busy background of severe
inflammation. In this study, FISH analysis was done in 13 dogs with GC with a eubac-
terial 16S rDNA library construction generated from GC mucosa. In all dogs evaluated,
the authors discovered intramucosal and macrophage invasion exclusively by E coli
(Fig. 3).23 GC-associated E coli were shown to lack genes associated with virulence
present in diarrheagenic E coli and were able to invade epithelial cells and persist
within macrophages.23 This pathogen-like behavior is similar to that of a newly identi-
fied E coli pathotype, the adherent and invasive E coli (AIEC) that is increasingly asso-
ciated with Crohn’s disease (CD) in humans.24,25
A direct causal role for E coli in GC pathogenesis is supported by the correlation
between clinical remission of the disease and eradication of invasive E coli using
enrofloxacin.14,23 A series of 7 dogs with histologically confirmed GC and intramucosal
E coli invasion confirmed by FISH were treated with enrofloxacin (7  3 mg/kg/d for
9.5  4 weeks) and reevaluated by repeat histology and FISH (Fig. 4).14 Long-term
clinical remission coincided with the eradication of invasive E coli in 4 dogs. In
a relapsing case, the E coli were enrofloxacin resistant and the animal was euthanized
because of refractory disease. The result of PAS staining in this study remained posi-
tive for more than 6 months despite remission of clinical signs and eradication of the

Fig. 3. FISH image (original magnification  40) of GC colon mucosa showing typical clusters
of E coli within the mucosa (red arrow) and intracellularly with macrophages (yellow
arrows). Inset: invasive E coli within a macrophage. E coli-Cy3 probe (red) with non-
EUB3386FAM (green) and 4’,6-diamidino-2-phenylindole (4’-6-diamidino-2-phenylindole
[DAPI]) (nuclei in blue).
436 Craven et al

Fig. 4. Colon mucosa from 2 dogs with GC before and after enrofloxacin treatment.
Pretreatment sections: histologically, (A, C) there is severe loss of glandular structure and
cellular infiltration in both cases. Mucosal infiltration with macrophages that show positive
with PAS (E, G) is a dominant feature. FISH (I, K) shows invasive E coli (Insets I and K, magni-
fied E coli 100). Posttreatment sections: 10 weeks after initial diagnosis in dog 1, inflamma-
tion is resolving, but mild PAS staining persists (B, F) and the result of FISH is negative for
bacterial invasion (J). In dog 2, enrofloxacin resistance developed, and after 3 months of en-
rofloxacin treatment, severe inflammation and positive staining with PAS persist (D, H).
Result of FISH remains positive for E coli invasion (L). (A–D) Hematoxylin-eosin, original
magnification  60; (E–H) PAS, original magnification  60; and (I–L) FISH, original magni-
fication  60.

invasive E coli. The reasons for this positivity are not clear, but it is important to note
that the complete histologic remission of disease seems to lag behind clinical
improvement, a feature also reported in Whipple disease (see Fig. 4).26
The importance of appropriate antimicrobial selection in the treatment of GC was
recently demonstrated in a prospective study of 14 GC cases.27 In this study, the
E coli isolates from 6 of 6 complete responders were enrofloxacin sensitive, whereas
those from 4 of 4 nonresponders and 2 of 4 partial responders were enrofloxacin resis-
tant. Clinical response was directly influenced by susceptibility of E coli to enrofloxacin
(P<.01).
Taken as a whole, this evidence indicates a 1:1 correlation between GC and invasive
E coli in 32 cases collectively evaluated by FISH to date.14,23,27 This discovery has
transformed the diagnostic approach, therapy, and prognosis of GC.

Genetics
Because GC is breed specific and rare, it is suspected to be an autosomal recessive
genetic defect involving the immune system that confers susceptibility to E coli inva-
sion. Research is currently being undertaken to identify the genetic basis of GC, and
a genome-wide association scan (GWAS) is underway.28,29 The principle of a GWAS is
to observe the frequency with which certain alleles are present in affected and control
groups in order to identify disease associations in candidate genes. The Broad Insti-
tute Dog Genome Project identified more than 2.5 million single nucleotide polymor-
phisms (SNPs) in the Boxer dog and 10 additional dog breeds and developed
a custom canine SNP array the GeneChip Canine Genome 2.0 Array in collaboration
with Affymetrix Inc, Santa Clara, CA, USA.30,31 This array relies on the hybridization of
 Granulomatous Colitis of Boxer Dogs 437

fluorescently labeled fragments of SNP-containing DNA to complementary DNA olig-

omers that are tiled on a silicon wafer. The SNP genotype calls are made using the
integration of fluorescent signal intensities at each location.
The GWAS of GC has revealed GC-associated SNPs in the gene encoding neutrophil
cytosolic factor (NCF) 2.28,29 This gene encodes a cytosolic subunit, p67phox, of the
multiprotein complex NADPH oxidase.32,33 Within phagocytes, NADPH oxidase plays
a crucial role in innate immunity by reducing molecular oxygen to superoxide, gener-
ating numerous toxic reactive oxygen species (ROS). ROS are used as microbicidal
agents against pathogens in the respiratory burst generated by phagocytic cells.34,35
An ineffective respiratory burst results in a compromised ability to eliminate intracellular
pathogens, particularly catalase-producing bacteria and fungi.36 Mutations in NCF2 in
humans are known to cause chronic granulomatous disease (CGD), a disease complex
comprising immunodeficiency disorders and predisposition to chronic infections.36,37
Patients with CGD can develop colitis with striking histologic similarities to GC,
including macrophages that stain positive with PAS.37–39 The initial screening test for
CGD in humans is the evaluation of the neutrophil respiratory burst in peripheral blood.
The authors have recently evaluated a flow cytometric method of assessing the neutro-
phil respiratory burst in dogs and have demonstrated marked reductions in the neutro-
phil oxidative burst in 2 dogs with GC compared with healthy controls.29 It is notable
that a recent GWAS in human IBD has identified CD-associated SNPs in the gene
encoding another NADPH oxidase complex subunit, NCF4.40
To summarize, a GWAS of GC has identified NCF2 (a gene associated with CGD in
humans) as a candidate gene, and further disease mapping and phenotypic charac-
terization by neutrophil function testing are ongoing. The bacteria involved in the
gastrointestinal manifestations of CGD are poorly characterized, but their striking simi-
larities with those involved in GC suggest a potential role for E coli. The identification of
NCF2 in GC suggests that the Boxer dog may prove to be a useful model for CGD, but
further work is required to confirm NCF2 gene involvement.

AIEC in Crohn’s Disease

The association of AIEC with GC is similar to findings in humans with IBD, especially
ileal CD, one of the most prevalent forms of IBD occurring in humans. CD is a hetero-
geneous group of disorders resulting from the convergence of multiple factors such as
genetically determined susceptibility, altered immune tolerance of the enteric bacteria,
and environmental triggers. The role of the resident microflora in CD pathogenesis was
initially thought to arise from a lack of immunologic tolerance and an overly aggressive
T-cell response to microbial components in individuals with genetic susceptibility (eg,
nucleotide-binding oligomerization domain containing 2 [NOD2] mutation).41 More
recent work shows that CD is in fact associated with specific alterations in the status
quo of the intestinal microbial ecosystem that can develop independent of genetic
susceptibility.42–44 This phenomenon termed dysbiosis refers to an altered balance
of “aggressive” species (eg, Bacteroidetes, Proteobacteria) versus “protective”
species (eg, Firmicutes). Specific pathogens, such as Mycobacterium tuberculosis,
Salmonella, Helicobacter, and Listeria, are frequently cited in CD pathogenesis but
cause and effect have never been convincingly demonstrated. New evidence for
a specific pathogen in CD lies in the ability of AIEC to invade and persist intracellularly
in intestinal epithelial cells and macrophages and to induce granulomatous lesions in
vitro.24,25,41,45–50
This unique group of E coli was first associated with IBD when recovered from 100%
of the biopsy specimens of early ileal CD lesions and 65% of chronically inflamed ileal
resections, compared with 3.7% of colonic biopsies from the same patients, and 6%
438 Craven et al

of healthy control ileal biopsies.25 Numerous subsequent studies have confirmed

these observations,45,46,51 but the precise role of AIEC in CD, that is, whether it is
a secondary invader or a primary pathogen, remains the subject of much debate.
These commensal flora are, however, increasingly cited as emerging pathogens in
CD because the number of E coli in CD have been shown to be strongly correlated
with the severity of disease (P<.001) and in a FISH-based study, invasive E coli
were found only in inflamed mucosa.45
AIEC are unique in that they do not possess any of the known virulence genes for
invasion used by enteroinvasive or enteropathogenic E coli, or Shigella strains.45,46,52
They are similar to extraintestinal avian and uropathic E coli strains in phylogeny and
virulence gene profile.45,53 A functional change in the resident E coli, characterized by
proliferation and upregulation of virulence genes, has been suggested to account for
the ability of AIEC to invade and persist in the epithelial cells and macrophages of
patients with CD.24,25,45 There is emerging evidence that AIEC use specific mecha-
nisms to facilitate cellular invasion and survival, such as flagellin,48 type I pili,54 cell
adhesion molecule CEACAM6, which acts as a receptor for AIEC,51 the stress
response protein Gp96,55 and long polar fimbriae.25,42,45
A role for AIEC in CGD has not yet been appreciated but is implied by our under-
standing not only of the GC pathogenesis but also of the pathophysiologic similarities
between CD and CGD that culminate in defective bacterial killing, which includes
phagocyte dysfunction because of defective NADPH oxidase36,40 and polymorphisms
of genes regulating clearance of intracellular pathogens.56–58 The most well-
recognized CD-associated polymorphism involves NOD2, which encodes an intracel-
lular sensor for a bacterial wall peptidoglycan, and facilitates a nuclear factor-k
B–mediated proinflammatory and antibacterial response.57 More recently, GWAS in
patients with CD has shown disease-associated polymorphisms in autophagy
pathway components, which also play important roles in eliminating intracellular
microbes, autophagy-related 16-like protein 1, and immunity-related GTPase family,
M.40,59 Evidently, an increasingly emerging theme in the pathophysiology of IBD
involves the abnormal interfacing of host immunity with resident intestinal microbes,
which has major implications for disease management.
DIAGNOSIS
Clinical Features
GC typically affects Boxer dogs younger than 4 years with no sex predilection, and
some reports describe clinical signs in animals as young as 6 weeks.2,7,14 Clinical
signs are typical of colitis, that is, frequent small-volume diarrhea, hematochezia,
mucoid feces, and tenesmus. The degree of hematochezia is often significantly
greater than for other types of colitis, and affected dogs may fail to thrive or may
lose weight. Affected dogs are usually clinically well and afebrile but may be lethargic
with severe disease. Differential diagnoses aside from GC include idiopathic IBD,
enteric parasites, and infectious agents (whipworms, hookworms, Giardia, Cryptospo-
ridium, Salmonella, Campylobacter, fungal infections, pathogen causing protothe-
cosis), neoplasia, rectoanal polyps, chronic intussusceptions, and rectal stricture. It
is not uncommon in the authors’ experience for there to have been a recent episode
of suspected infectious gastroenteritis (eg, caused by Salmonella or Campylobacter)
that may act as a trigger for GC.

Traditional Diagnostics
Routine clinicopathologic testing is usually unremarkable but may detect mild to
moderate anemia. This diagnosis could reflect anemia of chronic disease, or
 Granulomatous Colitis of Boxer Dogs 439

hemorrhage if hematochezia is severe. The degree and chronicity of blood loss can be
sufficient in rare cases to cause iron deficiency anemia, characterized by red cell
microcytosis and hypochromia. Hypoalbuminemia may also occur in some affected
dogs because of hemorrhage, protein exudation via diffusely ulcerated mucosa,
anorexia, and inflammation (albumin is a negative acute phase protein). Parasitologic
analysis of feces is usually unrewarding but is required to exclude other causes of clin-
ical signs. Imaging studies (radiographs, ultrasound) are largely unremarkable but may
be useful to detect other causes of large-bowel signs (eg, partial intestinal obstruction,
abdominal masses, chronic intussusceptions, prostatomegaly). Definitive diagnosis is
achieved by ruling out other causes of clinical signs and histologic confirmation on
colonic mucosal biopsies. Mucosal pinch biopsies obtained via endoscopy are
adequate for diagnosis, but a patchy distribution of disease and PAS staining is not
uncommon; hence the authors suggest obtaining a minimum of 10 endoscopic biop-
sies. Grossly, the colonic mucosa may be reddened, cobblestoned, and ulcerated.
The histologic appearance of GC is unique relative to other types of colitis in dogs
because of the severe mucosal ulceration and infiltration of the submucosa and lamina
propria with macrophages that stain positive with PAS (see Fig. 2).7,9,14,60 Additional
histologic features include mucosal ulceration, loss of goblet cells, and cellular infiltra-
tion with granulocytes and lymphocytes.1,17 Enlargement of draining lymph nodes, or
more rarely, generalized lymphadenopathy, can develop as a result of lymphoid
hyperplasia and macrophage infiltration.17,18

FISH Analysis
Demonstration of invasive E coli in GC is now integral to disease diagnosis and
management and is best accomplished using FISH (see Fig. 3). The advantage of
FISH over other methods such as Gram staining is that it uses fluorescent probes
that bind with high specificity to bacterial rDNA and increase the likelihood of visual-
izing bacteria in tissues with a busy inflammatory background. The degree of cellular
infiltration and the foamy appearance of macrophages in GC make it difficult to differ-
entiate bacteria from cytoplasmic contents, granules, and inflammatory debris using
routine stains. Poor visualization is likely the reason why the association with E coli
was not uncovered in earlier studies because cultures positive for E coli were actually
obtained from the colic lymph nodes of 7 affected Boxer dogs in 1966, but the E coli
were considered to be secondary invaders.60
FISH is performed on formalin-fixed paraffin-embedded colonic mucosal biopsy
specimens. An E coli–specific probe is colocalized with a eubacterial probe and slides
spotted with other bacteria are used to control probe specificity, such as Salmonella,
Proteus, Klebsiella, Enterococcus, Staphylococcus, Streptococcus, and genera of
Clostridiales. A negative FISH result does not completely exclude E coli invasion
because a patchy distribution of invasion can occur. Thus, a minimum of 10 mucosal
biopsies is recommended. Other reasons for false-negative results include the pres-
ence of dead or dying bacteria during biopsy sampling, low bacterial numbers, over-
fixation, and sulfasalazine treatment. False-positive results on FISH are also possible
but unlikely, given the additional probes used as positive and negative controls. FISH
for GC is currently performed by the Simpson Laboratory at Cornell University, and
additional information is available online at www.vet.cornell.edu/labs/simpson.

Antimicrobial Susceptibility Testing

Although FISH analysis may identify the presence, and sometimes specific species of
bacteria, it is only moderately helpful for antimicrobial selection because it does not
reveal antimicrobial resistance genes or intracellular versus extracellular bacterial
440 Craven et al

location. It is also necessary to culture colonic mucosa, particularly (but not only) when
invasive E coli are documented, in order to determine antimicrobial susceptibility. It is
of course impossible to be certain that the E coli strain isolated is in fact the invasive
strain and not just a surface colonizer. However, in the authors’ experience (M. C. and
K. W. S.), only 1 or 2 E coli strains are usually cultivable from 1 to 2 colon biopsies
because the invading pathotype is likely to predominate, having outcompeted other
strains. Collection of 2 to 3 mucosal biopsies into Luria-Bertani broth for gram-
negative enrichment and antimicrobial sensitivities of all isolated E coli strains is rec-
ommended (further information and sampling kits for FISH and culture are available at
www.vet.cornell.edu/labs/simpson).
Future Directions
Pending further evaluation of the genetic basis of GC, it is possible that a genetic
screening test may become available in the near future. Initial screening of patients
with CGD is accomplished by simple tests of neutrophil function, and this may also
become a useful diagnostic tool in GC, if NCF2 gene involvement is confirmed.

TREATMENT

Before recent developments, treatment with standard recommended protocols for

idiopathic colitis, namely dietary modification and therapy with metronidazole/tylosin,
sulfasalazine, prednisolone, and azathioprine, failed to produce satisfactory clinical
results.5,7,10,11 The administration of enrofloxacin alone, 5 mg/kg once daily for a total
of 6 to 8 weeks, has been associated with long-term remission.3,14,23,27 It is now
apparent that a successful response to treatment hinges on the successful eradication
of invasive E coli. Thus the antimicrobial used must not only be able to kill E coli but
also achieve an adequate intracellular concentration.
In order to optimize antimicrobial selection against GC-associated E coli, a recent
study analyzed antimicrobial sensitivity profiles in 14 GC cases and discovered enro-
floxacin resistance in 43% (Table 1). The resistant E coli strains were uniformly
resistant to all fluoroquinolones tested (ciprofloxacin, marbofloxacin, enrofloxacin, da-
nofloxacin) and tended to harbor resistance to other macrophage-penetrating antimi-
crobials, such as chloramphenicol, florfenicol, rifampin, trimethoprim-sulfa (TMPS),

Table 1
Prevalence of antimicrobial resistance in E coli strains isolated from 14 GC-affected
Boxer dogs versus 17 healthy dogs

Resistant Strains (%) Resistant Dogs (%)

Antimicrobial GC Healthy GC Healthy
Amoxicillin-clavulanate 35* 8 57* 12
Ampicillin 49** 15 64** 18
Cefoxitin 30*** 0 50*** 0
Tetracycline 48* 18 64* 24
Trimethoprim-sulfa 44*** 6 57** 7
Ciprofloxacin 35*** 0 43** 0
Gentamicin 13 18 14 36
Chloramphenicol 17* 0 21 0

GC significantly different from healthy using Fisher exact test: *P<.05, **P<.01,***P<.001.
Data from Craven M, Dogan B, Schukken A, et al. Antimicrobial resistance impacts clinical
outcome of granulomatous colitis in boxer dogs. J Vet Intern Med 2010;24(4):819–24.
 Granulomatous Colitis of Boxer Dogs
Fig. 5. Summary of the approach to diagnosis and treatment of GC. CBC, complete blood cell count; H&E, hematoxylin-eosin; IFA, immunofluorescence
assay; NSAID, nonsteroidal antiinflammatory drug.

441
442 Craven et al

tetracyclines, and clarithromycin. Empirical treatment with enrofloxacin before per-

forming colon biopsy was associated with the isolation of resistant E coli (P<.01),
perhaps associated with inadequate duration of treatment.27 In skeletally immature
dogs, damage to developing cartilage is a potential adverse effect of enrofloxacin
treatment and practitioners may be understandably reluctant to prescribe the drug
for extended periods. However, it is important in GC that a sufficient duration of enro-
floxacin is given, even if significant clinical improvement is noted within 1 to 2 weeks.
Cartilage defects are rarely appreciated and are perhaps the lesser of the 2 evils
because resistance to enrofloxacin in GC is significantly associated with a poor
outcome.27 Currently, the suggested treatment regimen for cases with enrofloxacin-
sensitive E coli is 5 to 10 mg/kg every 24 hours for a minimum of 6 weeks. The authors
suggest enrofloxacin specifically because it has been proved to induce remission.
Other fluoroquinolones such as ciprofloxacin and marbofloxacin may also be effective
but have not been evaluated. Posttreatment colonoscopy and biopsy are advisable to
demonstrate remission of disease and successful eradication of E coli invasion. The
role of other medications such as mesalamine or sulfasalazine as adjunctive treatment
in GC is unknown. Mesalamine has been shown to downregulate cytokine production
in response to AIEC in vitro61 and may have synergistic effects alongside antimicro-
bials in the treatment of GC, but this drug has yet to be critically evaluated.
A poor response to treatment is usually associated with development of enrofloxa-
cin resistance. Repeat colonoscopic biopsy for FISH and culture is required to guide
further treatment. Potential reasons for development of enrofloxacin resistance in GC
include treatment with an inadequate dosage and duration of enrofloxacin and the
acquisition of resistance plasmids.62,63 In enrofloxacin-resistant cases, the antimicro-
bial selection should be determined by susceptibility testing. Aside from the spectrum
of activity, it is of critical importance that the antimicrobial used is capable of pene-
trating macrophages. Agents likely to do so include chloramphenicol, florfenicol,
TMPS, tetracyclines, clarithromycin, and rifampicin.63 When a multidrug-resistant
strain of E coli is present, the authors recommend considering the use of a combination
antimicrobial protocol, to include a fluoroquinolone and several other of these
macrophage-penetrating agents (eg, chloramphenicol, TMPS). Even though the
E coli may be resistant to these agents individually, they may have synergistic effects
when the drugs are administered together over an extended period (the authors
suggest at least 1 month beyond the resolution of clinical signs). This perhaps
heavy-handed approach seems to be justified when considering that refractory cases
are usually euthanized. No obvious universal alternative was identified as the next-
best antimicrobial to enrofloxacin in the aforementioned case series, but of note is
that 100% of the GC-associated E coli strains were sensitive to the aminoglycoside
amikacin. However, the molecular properties of amikacin result in a poor ability to
penetrate mammalian cells, precluding its clinical application for the treatment of intra-
cellular infections. Additional therapies for GC are clearly needed, and future direc-
tions include intracellular targeting of amikacin, E coli vaccination, and gene transfer
therapy. Fig. 5 provides a summary of the diagnosis and management of GC.

REFERENCES

1. Van Kruiningen HJ, Montali RJ, Strandberg JD, et al. A granulomatous colitis of
dogs with histologic resemblance to Whipple’s disease. Pathol Vet 1965;2(6):
521–44.
2. Churcher RK, Watson AD. Canine histiocytic ulcerative colitis. Aust Vet J 1997;
75(10):710–3.
 Granulomatous Colitis of Boxer Dogs 443

3. Hostutler RA, Luria BJ, Johnson SE, et al. Antibiotic-responsive histiocytic ulcer-
ative colitis in 9 dogs. J Vet Intern Med 2004;18(4):499–504.
4. Van Kruiningen HJ. Granulomatous colitis of boxer dogs: comparative aspects.
Gastroenterology 1967;53(1):114–22.
5. Stokes JE, Kruger JM, Mullaney T, et al. Histiocytic ulcerative colitis in three non-
boxer dogs. J Am Anim Hosp Assoc 2001;37(5):461–5.
6. Tanaka H, Nakayama M, Takase K. Histiocytic ulcerative colitis in a French
bulldog. J Vet Med Sci 2003;65(3):431–3.
7. Hall EJ, Tennant BJ, Payne-Johnson CE, et al. Boxer colitis. Vet Rec 1992;
130(7):148.
8. Fletcher WD. Advanced granulomatous colitis in a Siberian Husky (a case
report). Vet Med Small Anim Clin 1978;73(11):1409.
9. Kennedy PC, Cello RM. Colitis of boxer dogs. Gastroenterology 1966;51(5):
926–31.
10. Hill FW, Sullivan ND. Histiocytic ulcerative colitis in a Boxer dog. Aust Vet J 1978;
54(9):447–9.
11. Lindberg R, Segall T. Histiocytic ulcerative colitis in a boxer. A case report. Nord
Vet Med 1977;29(12):552–5.
12. Arai K, Sudo M, Saeki Y, et al. Granulomatous colitis in a Boxer dog. Journal of the
Japan Veterinary Medical Association 1983;36:29–33.
13. van der Gaag I, Happe RP, Wolvekamp WT. A boxer dog with chronic hypertrophic
gastritis resembling Menetrier’s disease in man. Vet Pathol 1976;13(3):172–85.
14. Mansfield CS, James FE, Craven M, et al. Remission of histiocytic ulcerative
colitis in Boxer dogs correlates with eradication of invasive intramucosal Escher-
ichia coli. J Vet Intern Med 2009;23(5):964–9.
15. Afshar P, Redfield DC, Higginbottom PA. Whipple’s disease: a rare disease revis-
ited. Curr Gastroenterol Rep 2010;12(4):263–9.
16. Desnues B, Al Moussawi K, Fenollar F. New insights into Whipple’s disease and
Tropheryma whipplei infections. Microbes Infect 2010;12(14–15):1102–10.
17. Van Kruiningen HJ. The ultrastructure of macrophages in granulomatous colitis of
Boxer dogs. Vet Pathol 1975;12(5–6):446–59.
18. Bowe PS, Van Kruiningen HJ, Rosendal S. Attempts to produce granuloma-
tous colitis in Boxer dogs with a mycoplasma. Can J Comp Med 1982;
46(4):430–3.
19. German AJ, Hall EJ, Kelly DF, et al. An immunohistochemical study of histiocytic
ulcerative colitis in boxer dogs. J Comp Pathol 2000;122(2–3):163–75.
20. Hanauer SB. Inflammatory bowel disease: epidemiology, pathogenesis, and ther-
apeutic opportunities. Inflamm Bowel Dis 2006;12(Suppl 1):S3–9.
21. Davies DR, O’Hara AJ, Irwin PJ, et al. Successful management of histiocytic ulcer-
ative colitis with enrofloxacin in two Boxer dogs. Aust Vet J 2004;82(1–2):58–61.
22. Van Kruiningen HJ, Civco IC, Cartun RW. The comparative importance of E. coli
antigen in granulomatous colitis of Boxer dogs. APMIS 2005;113(6):420–5.
23. Simpson KW, Dogan B, Rishniw M, et al. Adherent and invasive Escherichia coli
is associated with granulomatous colitis in boxer dogs. Infect Immun 2006;74(8):
4778–92.
24. Darfeuille-Michaud A, Neut C, Barnich N, et al. Presence of adherent Escherichia
coli strains in ileal mucosa of patients with Crohn’s disease. Gastroenterology
1998;115(6):1405–13.
25. Darfeuille-Michaud A, Boudeau J, Bulois P, et al. High prevalence of adherent-
invasive Escherichia coli associated with ileal mucosa in Crohn’s disease. Gastro-
enterology 2004;127(2):412–21.
444 Craven et al

26. Schneider T, Moos V, Loddenkemper C, et al. Whipple’s disease: new aspects of

pathogenesis and treatment. Lancet Infect Dis 2008;8(3):179–90.
27. Craven M, Dogan B, Schukken A, et al. Antimicrobial resistance impacts clinical
outcome of granulomatous colitis in Boxer dogs. J Vet Intern Med 2010;24(4):
819–24.
28. Craven M, Acland G, Mezey J, et al. Genome-wide association scan reveals poly-
morphisms in the P67phox subunit (NCF2) In Boxer dogs with E.coli-associated
granulomatous colitis: a potential model of chronic granulomatous disease
[abstract]. In: Digestive Disease Week. New Orleans (Louisiana), 2010.
29. Craven M, Gao C, Acland G, et al. Genome-wide analysis of granuomatous colitis
in the Boxer dog. Research report. Forum of the ACVIM 2010.
30. Karlsson EK, Baranowska I, Wade CM, et al. Efficient mapping of mendelian traits
in dogs through genome-wide association. Nat Genet 2007;39(11):1321–8.
31. Salmon Hillbertz NH, Isaksson M, Karlsson EK, et al. Duplication of FGF3, FGF4,
FGF19 and ORAOV1 causes hair ridge and predisposition to dermoid sinus in
Ridgeback dogs. Nat Genet 2007;39(11):1318–20.
32. von Rosenvinge EC, O’Donnell TG, Holland SM, et al. Chronic granulomatous
disease. Inflamm Bowel Dis 2010;16(1):9.
33. Holland SM. Update on phagocytic defects. Pediatr Infect Dis J 2003;22(1):87–8.
34. Mizuki K, Kadomatsu K, Hata K, et al. Functional modules and expression of
mouse p40(phox) and p67(phox), SH3-domain-containing proteins involved in
the phagocyte NADPH oxidase complex. Eur J Biochem 1998;251(3):573–82.
35. Han CH, Freeman JL, Lee T, et al. Regulation of the neutrophil respiratory burst
oxidase. Identification of an activation domain in p67(phox). J Biol Chem 1998;
273(27):16663–8.
36. Holland SM. Chronic granulomatous disease. Clin Rev Allergy Immunol 2010;
38(1):3–10.
37. Marciano BE, Rosenzweig SD, Kleiner DE, et al. Gastrointestinal involvement in
chronic granulomatous disease. Pediatrics 2004;114(2):462–8.
38. Schappi MG, Smith VV, Goldblatt D, et al. Colitis in chronic granulomatous
disease. Arch Dis Child 2001;84(2):147–51.
39. Schappi MG, Klein NJ, Lindley KJ, et al. The nature of colitis in chronic granulo-
matous disease. J Pediatr Gastroenterol Nutr 2003;36(5):623–31.
40. Roberts RL, Hollis-Moffatt JE, Gearry RB, et al. Confirmation of association of
IRGM and NCF4 with ileal Crohn’s disease in a population-based cohort. Genes
Immun 2008;9(6):561–5.
41. Sartor RB. Microbial influences in inflammatory bowel diseases. Gastroenterology
2008;134(2):577–94.
42. Sartor RB. Targeting enteric bacteria in treatment of inflammatory bowel
diseases: why, how, and when. Curr Opin Gastroenterol 2003;19(4):358–65.
43. Sartor RB, Muehlbauer M. Microbial host interactions in IBD: implications for
pathogenesis and therapy. Curr Gastroenterol Rep 2007;9(6):497–507.
44. Willing B, Halfvarson J, Dicksved J, et al. Twin studies reveal specific imbalances
in the mucosa-associated microbiota of patients with ileal Crohn’s disease.
Inflamm Bowel Dis 2009;15(5):653–60.
45. Baumgart M, Dogan B, Rishniw M, et al. Culture independent analysis of ileal
mucosa reveals a selective increase in invasive Escherichia coli of novel
phylogeny relative to depletion of Clostridiales in Crohn’s disease involving the
ileum. ISME J 2007;1(5):403–18.
46. Barnich N, Darfeuille-Michaud A. Adherent-invasive Escherichia coli and Crohn’s
disease. Curr Opin Gastroenterol 2007;23(1):16–20.
 Granulomatous Colitis of Boxer Dogs 445

47. Bringer MA, Glasser AL, Tung CH, et al. The Crohn’s disease-associated
adherent-invasive Escherichia coli strain LF82 replicates in mature phagolyso-
somes within J774 macrophages. Cell Microbiol 2006;8(3):471–84.
48. Eaves-Pyles T, Allen CA, Taormina J, et al. Escherichia coli isolated from a Crohn’s
disease patient adheres, invades, and induces inflammatory responses in polar-
ized intestinal epithelial cells. Int J Med Microbiol 2008;298(5–6):397–409.
49. Glasser AL, Boudeau J, Barnich N, et al. Adherent invasive Escherichia coli
strains from patients with Crohn’s disease survive and replicate within
macrophages without inducing host cell death. Infect Immun 2001;69(9):
5529–37.
50. Meconi S, Vercellone A, Levillain F, et al. Adherent-invasive Escherichia coli iso-
lated from Crohn’s disease patients induce granulomas in vitro. Cell Microbiol
2007;9(5):1252–61.
51. Barnich N, Carvalho FA, Glasser AL, et al. CEACAM6 acts as a receptor for
adherent-invasive E. coli, supporting ileal mucosa colonization in Crohn disease.
J Clin Invest 2007;117(6):1566–74.
52. Packey CD, Sartor RB. Interplay of commensal and pathogenic bacteria, genetic
mutations, and immunoregulatory defects in the pathogenesis of inflammatory
bowel diseases. J Intern Med 2008;263(6):597–606.
53. Martinez-Medina M, Aldeguer X, Lopez-Siles M, et al. Molecular diversity of
Escherichia coli in the human gut: new ecological evidence supporting the role
of adherent-invasive E. coli (AIEC) in Crohn’s disease. Inflamm Bowel Dis 2009;
15(6):872–82.
54. Ryan P, Kelly RG, Lee G, et al. Bacterial DNA within granulomas of patients with
Crohn’s disease–detection by laser capture microdissection and PCR. Am J Gas-
troenterol 2004;99(8):1539–43.
55. Rolhion N, Barnich N, Bringer MA, et al. Abnormally expressed ER stress
response chaperone Gp96 in CD favours adherent-invasive Escherichia coli inva-
sion. Gut 2010;59(10):1355–62.
56. Barrett JC, Hansoul S, Nicolae DL, et al. Genome-wide association defines more
than 30 distinct susceptibility loci for Crohn’s disease. Nat Genet 2008;40(8):
955–62.
57. Eckmann L, Karin M. NOD2 and Crohn’s disease: loss or gain of function? Immu-
nity 2005;22(6):661–7.
58. Homer CR, Richmond AL, Rebert NA, et al. ATG16L1 and NOD2 interact in an
autophagy-dependent, anti-bacterial pathway implicated in Crohn’s disease
pathogenesis. Gastroenterology 2010;139(5):1630–41.
59. Massey DC, Parkes M. Genome-wide association scanning highlights two
autophagy genes, ATG16L1 and IRGM, as being significantly associated with
Crohn’s disease. Autophagy 2007;3(6):649–51.
60. van Kruiningen HJ. Granulomatous colitis of Boxer dogs [PhD thesis]. New York:
State Veterinary College, Cornell University; 1966.
61. Subramanian S, Rhodes JM, Hart CA, et al. Characterization of epithelial IL-8
response to inflammatory bowel disease mucosal E. coli and its inhibition by
mesalamine. Inflamm Bowel Dis 2008;14(2):162–75.
62. Warren A, Townsend K, King T, et al. Multi-drug resistant escherichia coli with
extended-spectrum beta-lactamase activity and fluoroquinolone resistance iso-
lated from clinical infections in dogs. Aust Vet J 2001;79(9):621–3.
63. Subramanian S, Roberts CL, Hart CA, et al. Replication of colonic Crohn’s
disease mucosal Escherichia coli isolates within macrophages and their suscep-
tibility to antibiotics. Antimicrob Agents Chemother 2008;52(2):427–34.