CD68 Expression Is Markedly Different in Crohn’s Disease
and the Colitis Associated with Chronic Granulomatous
inherited immunodeficiency disorder characterized by inability of
phagocytes to kill certain bacteria and fungi. Histology from colon
biopsies of CGD patients have shown the presence of inflammation and
granulomas, almost indistinguishable from the findings seen in Crohn’s
disease. We sought to determine if there were any differences in the cell
types that comprise the inflammatory infiltrates in inflamed colon
specimens between the 2 diseases. The objective was to determine
whether the pattern of inflammatory cell composition could be used to
distinguish CGD-associated colitis from Crohn’s colitis.
Chronic granulomatous disease (CGD) is a rare
Methods: Biopsies from 6 patients with Crohn’s disease, 6 pa-
tients with CGD, and 6 control patients were stained with antibodies
to CD3, CD4, CD8, CD68, CD79, CD163, and Foxp3. Positively
staining cells per mm2of lamina propria were calculated for each
antibody, with the exception of CD163, in which a percent area of
lamina propria stained was calculated.
Results: There was a marked difference in the average CD68?
cells per mm2of lamina propria between the 2 groups (Crohn’s:
1104.2 cells/mm2; CGD: 242.3 cells/mm2, P ? 0.0004) and a
significant difference between the CGD group and control group
(Control: 565.4 cells/mm2, CGD: 242.3 cells/mm2, P ? 0.0072).
There were no significant differences between Crohn’s and CGD
biopsies in the other cell types.
Conclusions: This phenomenon provides physicians with a sim-
ple and valuable tool to identify CGD in patients with colonic
(Inflamm Bowel Dis 2009;15:1213–1217)
Crohn’s, colitis, chronic granulomatous disease,
ity of phagocytes to kill certain bacteria and fungi. Patients
with CGD develop recurrent and potentially life-threatening
infections that include pneumonia and abscess formation. Up
to 50% of these patients have gastrointestinal (GI) manifes-
tations of their disease that include colitis and gastric outlet
obstruction.1–3Histology from colon biopsies of CGD pa-
tients have shown the presence of inflammation and granu-
lomas, very similar to the findings seen in Crohn’s disease
(CD).4–6Pigmented histiocytes present in colonic biopsies
have been associated with CGD, and it has been suggested
that granulomas in CGD colitis have more sharply defined
histiocyte aggregates compared to CD granulomas, but these
findings are neither specific nor sensitive.1,3,7Because the
histopathology of colon biopsies of CGD patients can be
almost indistinguishable from those of patients with CD with
standard hematoxylin and eosin (H&E) staining, we sought to
determine if there were any differences in the cell types that
compose the inflammatory infiltrates in inflamed colon spec-
imens between the 2 diseases. The treatment for patients with
CGD and CD is dramatically different and identification of
patients with CGD would improve their prognosis. If a dis-
tinct pattern of inflammatory cell composition could be iden-
tified, it could prove to be a useful tool in distinguishing
CGD-associated colitis from Crohn’s colitis.
hronic granulomatous disease (CGD) is a rare inherited
immunodeficiency disorder characterized by the inabil-
MATERIALS AND METHODS
Approval from our Institutional Review Board (IRB)
was obtained to examine existing colon biopsy specimens for
our study. Colon biopsy samples were studied from 6 patients
with Crohn’s colitis (2 males, 4 females, average age at
biopsy 11.8 years, age range 8.7–14.7 years), 6 patients with
CGD-associated colitis (4 males, 2 females, average age at
biopsy 7.4 years, age range 2.4–19.4 years), and 6 patients
with no history of identifiable disease and histologically
normal colon with no evidence of colitis (5 males, 1 female,
Received for publication November 4, 2008; Accepted December 30,
From the *Division of Gastroenterology, Hepatology, and Nutrition,†De-
partment of Pathology,
Immunology, Children’s Hospital of Philadelphia, Philadelphia, Pennsylva-
Supported in part by the Training Program in Rheumatic Diseases
(5T32AR007442) (to S.L.).
Reprints: Kathleen E. Sullivan, MD, PhD, Division of Allergy and Im-
munology, Children’s Hospital of Philadelphia, 34th St and Civic Ctr. Blvd.,
Philadelphia, PA 19104 (e-mail: email@example.com).
Copyright © 2009 Crohn’s & Colitis Foundation of America, Inc.
‡Kathleen E. Sullivan: Division of Allergy and
Inflamm Bowel Dis ● Volume 15, Number 8, August 2009
average age at biopsy 10.1 years, age range 6.3–16.0 years)
(Table 1). The 6 patients with CGD were comprised of 3 with
mutations in gp91, 2 with mutations in gp47, and 1 with a
mutation in gp22. All biopsies were obtained prior to therapy.
Four of the patients with CGD had significant infections
leading to their diagnosis while 2 patients presented with
colitis and were subsequently found to have CGD.8
Samples and Staining
Immunohistochemical analyses were performed on for-
malin-fixed paraffin-embedded tissue from colon biopsies cut
at 4–5 ?m. Antibodies were applied as per the manufactur-
er’s instructions with appropriate positive and negative con-
trols. The antibodies used were CD3 (Dako, Denmark), CD4
(Zymed, San Francisco, CA), CD8 (Vector Labs, Burlin-
game, CA), CD68 (Dako), CD79 (Dako), CD163 (Vector
Labs), Foxp3 (Abcam, Cambridge, MA). These antibodies
were used for the identification of T-cells (CD3), T-helper
cells (CD4), cytotoxic T-cells (CD8), macrophages (CD68,
CD163), B-cells (CD79), and T-regulatory cells (Foxp3).
Intracellular staining for CD68 was performed using
PE-CD68 and FITC-CD14 (BD Biosciences, San Jose, CA).
Fixation and permeabilization was achieved using Cytofix/
Cytoperm according to the manufacturer’s instructions (BD
Two regions of inflammation were identified by a pa-
thologist for each inflamed biopsy sample stained with H&E.
For histologically normal biopsies, 2 random regions were
identified. The corresponding locations on the biopsies were
identified in the sections stained with different antibodies to
detect cell surface molecules characteristic of specific cell
types. Digital pictures of these identified areas of interest
were taken under 20? magnification using a Leica DM4000B
microscope, SPOT RT-SE microscope digital camera, and
SPOT Software v. 4.6 (Diagnostic Instruments, Sterling
Heights, MI). Cells per square millimeter of lamina propria
for each region were calculated using Image-Pro Plus v. 6.1
software (Media Cybernetics, Bethesda, MD) for slides
stained for CD3, CD4, CD8, CD68, CD79, and Foxp3. The 2
values for each biopsy were averaged to calculate an average
cells per square millimeter. For slides stained for CD163, an
average percent area was calculated using the same software.
P-values were calculated using a 2-tailed Student’s t-test.
Positively stained cells for every antibody were found
diffusely throughout the inflamed lamina propria (Figs. 1, 2).
Granulomata were seen in 3 of 6 biopsies from patients with
CD and in all 6 biopsies from patients with CGD. There were
no significant differences in the average number of cells per
square millimeter of lamina propria for the CD3, CD4, CD8,
CD79, and Foxp3 stains when comparing CGD biopsies to
CD biopsies (Fig. 3). Surprisingly, B cells were nearly as
common as T cells in the inflamed tissues. However, there
was a marked difference in CD68? cells per square milli-
meter of lamina propria between the 2 groups (CD: 1104.2
cells/mm2; CGD: 242.3 cells/mm2, P ? 0.0004). In fact, the
number of CD68? cells counted in the CGD biopsies was
significantly fewer than the number of cells counted in nor-
mal biopsies (Normal: 565.4 cells/mm2, CGD: 242.3 cells/
mm2, P ? 0.0072). Only 1 slide per patient underwent
quantitative analysis but CD68 staining was uniformly de-
creased in lamina propria from CGD patients. CD68 expres-
sion in other tissues was variable in intensity.
To elucidate whether the decrease in CD68 staining in
CGD biopsies was a result of fewer macrophages in inflamed
CGD colon biopsies or impaired expression of CD68, the
colon biopsies were also stained for CD163, another marker
of cells of monocyte/macrophage lineage.9Because of the
diffuse staining pattern of CD163, it was not possible to
accurately count the number of cells per square millimeter,
and an alternative calculation, the percentage of lamina pro-
pria area stained for CD163, was determined. The CGD and
CD biopsies showed a similar pattern of staining, which was
significantly more than the staining in normal colon biopsies.
This reflects macrophage infiltration into inflamed tissues
To ensure that comparable types of inflammation were
TABLE 1. Demographic Data for Patients with Crohn’s Disease and CGD
8.7-14.7 yrs (mean 11.8)
2 males, 4 females
5-ASA (5 patients), immunomodulator
(3), infliximab (2), metronidazole
(3), steroid (1)
2.4 years-19.4 years (mean 7.4)
4 males, 2 females
5-ASA (2 patients), immunomodulator (1),
infliximab (1), metronidazole (2), steroid
(1), interferon gamma (2)
gp91 (3 patients), p47 (2), p22 (1)Mutations
Immunomodulators include 6-mercaptopurine, azathioprine, or methotrexate.
Liu et al
Inflamm Bowel Dis ● Volume 15, Number 8, August 2009
being examined, biopsies with granulomas were compared
(Fig. 2B). Even within granulomas, CGD patients had de-
creased expression of CD68 compared to patients with CD.
Biopsies from CGD patients from regions other than the
colon also demonstrated decreased expression of CD68 com-
pared to controls or biopsies from patients with CD (data not
shown). Peripheral blood monocytes were examined for
CD68 expression by intracellular cytokine staining. Although
there was a trend toward decreased CD68 expression, it did
not reach statistical significance (data not shown).
The nitroblue tetrazolium (NBT) slide test and the
dihydrorhodamine 123 (DHR 123) flow cytometric assay are
commonly used to test for chronic granulomatous dis-
ease.10,11However, they are not widely available and there
have been no reports of using immunohistochemical tech-
niques to screen for CGD. Although the presence of granu-
lomas is an expected finding in biopsy specimens from pa-
tients with CGD, it is also a common finding in intestinal
biopsy specimens from patients with CD.
Because the histology of colonic biopsies in the 2
diseases can be so similar, it is possible that there are patients
diagnosed with CD who actually suffer from CGD. An ex-
ample that CGD can be mistaken for CD is the case report by
Ramanuja et al,12who described a woman diagnosed with
CGD at the age of 53, but had carried the diagnosis of CD for
inflammatory cell markers. Biopsies were stained as described
in Materials and Methods.
CD68. A: Inflamed colon. B: Granulomas. There was a clear de-
crease in staining for CD68 in all 6 CGD biopsies when com-
pared with CD.
FIGURE 3. Comparison of various inflammatory cell types per
square millimeter of lamina propria in CD colon biopsies, CGD
colon biopsies, and histologically normal colon biopsies. There
was a significant difference in CD68-positive cells per mm2of
lamina propria when comparing CD biopsies (1104.2 cells/
mm2) and CGD biopsies (242.3 cells/ mm2) (P ? 0.0004) as well
as normal biopsies (565.4 cells/ mm2) and CGD biopsies (P
Inflamm Bowel Dis ● Volume 15, Number 8, August 2009 CD68 Expression
almost 30 years prior. We felt that by using immunohisto-
chemical markers to determine the concentrations of cell
types composing the inflammatory infiltrates of CD and CGD
colon biopsies, we may be able to identify a pattern histo-
logically distinguishing the 2 diseases.
Our results demonstrated no clear differences in any of
the cell markers, except for a dramatic difference in CD68
staining. The staining for CD68 in CGD colon biopsies was
not only significantly less than the staining in CD biopsies,
but it was statistically significantly less than the staining in
normal colon biopsies. Given that the staining for CD163,
another macrophage marker, was similar in CD and CGD
(and both were significantly more than the staining in normal
colon), we felt that the lack of CD68 staining did not indicate
a dearth of macrophages in the inflamed colon of CGD
patients, but rather a defect specific to CD68 expression.
CD68 is glycoprotein that is recognized as a monocyte/
macrophage marker and is a member of the lysosomal-asso-
ciated membrane protein (LAMP) family.13It is primarily
found intracellularly, mainly located in endosomal and lyso-
somal structures, and though its precise function is unclear, it
is thought to play a role in endocytosis or lysosomal traffick-
ing. LAMP family proteins may also offer protection of the
lysosomal membrane from hydrolytic enzymes, and they may
serve as extracellular presenters of carbohydrate ligands to
selectins, allowing homing of cells to certain organ sites.14
CD68 is not a completely specific marker for monocytes and
macrophages. It has also been found in the primary granules
of neutrophils, as well as in certain epithelial cells.9,15Cyto-
toxic necrotizing factor type 1 (CNF1), a Rho-activating
protein toxin from pathogenic strains of Escherichia coli,
induces macropinocytosis in epithelial cells and increases
expression of CD68 in these cells.15These epithelial cells
behave like phagocytes, capable of digesting bacteria and
The decreased CD68 expression in CGD patients is
currently without explanation since there does not appear to
be a direct relationship between CD68 and the NADPH
oxidase defect in CGD. The answer may lie in the generation
of reactive oxygen species (ROS). In addition to their micro-
bicidal role, ROS such as superoxide and hydrogen peroxide
can regulate other cellular processes such as transcription
factor function, proliferation, apoptosis, and cytokine produc-
tion.16–19Thus, CGD patients’ cells have compromised gen-
eration of ROS, this could theoretically lead to a dysregula-
tion of cellular and immunologic processes. CNF1 has not
only been shown to increase expression of CD68 in epithelial
cells, but it has also been reported to increase the production
of ROS as well. This would be compatible with the theory
that CD68 expression is dependent on the production of ROS,
although further studies in other cell lines would be necessary
to substantiate this claim. Alternatively, the macrophage in-
filtration in CGD patients may be fundamentally different
cells and may have a more DC-like phenotype with low
CD68 expression. This could reflect a different cytokine
milieu or different antigenic stimulation.20
There is one previous immunohistochemical study that
evaluated staining for CD68 in a CGD colon specimen. This
was reported by Mitomi et al6in 1999, who reported the case
of an 18-year-old male with CGD who suffered from intrac-
table diarrhea and a colitis histologically resembling CD. The
patient underwent a left-sided colectomy, and the surgical
specimen was stained for CD68 and compared to 13 surgical
specimens from patients with Crohn’s colitis. They found that
the CGD specimen had an average of 112 CD68? cells/mm2,
the Crohn’s specimens with granulomas had an average of
102 ? 75 CD68? cells/mm2, and the Crohn’s specimens
without granulomas had an average of 79 ? 59 CD68?
cells/mm2. No statistical analysis was performed since there
was only 1 CGD specimen, but there was likely no statistical
difference. The CD68? cell concentrations were markedly
less than the concentrations we calculated, and there could be
multiple explanations for this. Instead of using an ocular
micrometer, we used computer software that enabled us to
exclude colonic epithelium, crypts, and crypt epithelium in
our calculated area of interest. Another reason for the lower
cell counts in the Mitomi et al study may have been due to
their selection of fields. Twenty random fields were selected
in their samples, and due to the patchy nature of CD and CGD
colitis, these probably included fields of uninflamed tissue.
The regions we selected in our biopsies all consisted of
inflamed tissue. Finally, it is possible that the fixation tech-
nique employed by the Mitomi et al study on their surgical
specimens influenced the amount of CD68 cell staining com-
pared to our techniques.
Regardless of the mechanism by which CD68 expres-
FIGURE 4. Comparison of percent of lamina propria stained
with CD163 in CD colon biopsies, CGD colon biopsies, and his-
tologically normal colon biopsies. Both CD (27.5%) and CGD
to normal biopsies (13.6%) (P ? 0.007, CD versus normal; P
? 0.004, CGD versus normal), and there was no significant dif-
ference when comparing CD to CGD staining (P ? 0.55).
Liu et al
Inflamm Bowel Dis ● Volume 15, Number 8, August 2009
sion is decreased in CGD, it may prove to be a valuable
phenomenon that would allow for a simple method to screen
for CGD in patients with GI disease. GI manifestations are
common in CGD, and there are cases of patients diagnosed
with CD who were later identified as suffering from CGD.4,12
CD68 staining of colon biopsies may be an effective initial
screen for CGD. CD68 staining would be a simple technique
that should be readily accessible to pathologists in most
clinical settings. Patients with findings suspicious for CGD
could then have definitive testing with nitroblue tetrazolium,
or preferably, the more sensitive fluorescent assay based on
We thank Dan Martinez (CHOP Pathology Core Lab)
for help with the SPOT and Image-Pro Plus software.
1. Ament ME, Ochs HD. Gastrointestinal manifestations of chronic gran-
ulomatous disease. N Engl J Med. 1973;288:382–387.
2. Schappi MG, Smith VV, Goldblatt D, et al. Colitis in chronic granulo-
matous disease. Arch Dis Child. 2001;84:147–151.
3. Marciano BE, Rosenzweig SD, Kleiner DE, et al. Gastrointestinal in-
volvement in chronic granulomatous disease. Pediatrics. 2004;114:462–
4. Isaacs D, Wright VM, Shaw DG, et al. Chronic granulomatous disease
mimicking Crohn’s disease. J Pediatr Gastroenterol Nutr. 1985;4:498–
5. Huang JS, Noack D, Rae J, et al. Chronic granulomatous disease caused
by a deficiency in p47(phox) mimicking Crohn’s disease. Clin Gastro-
enterol Hepatol. 2004;2:690–695.
6. Mitomi H, Mikami T, Takahashi H, et al. Colitis in chronic granuloma-
tous disease resembling Crohn’s disease: comparative analysis of CD68-
positive cells between two disease entities. Dig Dis Sci. 1999;44:452–
7. Schappi MG, Klein NJ, Lindley KJ, et al. The nature of colitis in chronic
granulomatous disease. J Pediatr Gastroenterol Nutr. 2003;36:623–631.
8. Liu S, Abrams D, Baldassano RN, et al. Prevalence of chronic granu-
lomatous disease in pediatric patients diagnosed with Crohn’s disease.
Inflamm Bowel Dis. 2008;14:727–728.
9. Lau SK, Chu PG, Weiss LM. CD163: a specific marker of macrophages
in paraffin-embedded tissue samples. Am J Clin Pathol. 2004;122:794–
10. O’Gorman MR, Corrochano V. Rapid whole-blood flow cytometry
assay for diagnosis of chronic granulomatous disease. Clin Diagn Lab
11. Baehner RL, Nathan DG. Quantitative nitroblue tetrazolium test in
chronic granulomatous disease. N Engl J Med. 1968;278:971–976.
12. Ramanuja S, Wolf KM, Sadat MA, et al. Newly diagnosed chronic
granulomatous disease in a 53-year-old woman with Crohn disease. Ann
Allergy Asthma Immunol. 2005;95:204–209.
13. Holness CL, Simmons DL. Molecular cloning of CD68, a human mac-
rophage marker related to lysosomal glycoproteins. Blood. 1993;81:
14. Holness CL, da Silva RP, Fawcett J, et al. Macrosialin, a mouse
macrophage-restricted glycoprotein, is a member of the lamp/lgp family.
J Biol Chem. 1993;268:9661–9666.
15. Travaglione S, Falzano L, Fabbri A, et al. Epithelial cells and expression
of the phagocytic marker CD68: scavenging of apoptotic bodies follow-
ing Rho activation. Toxicol In Vitro. 2002;16:405–411.
16. Nakamura H, Nakamura K, Yodoi J. Redox regulation of cellular acti-
vation. Annu Rev Immunol. 1997;15:351–369.
17. Hehner SP, Breitkreutz R, Shubinsky G, et al. Enhancement of T cell
receptor signaling by a mild oxidative shift in the intracellular thiol pool.
J Immunol. 2000;165:4319–4328.
18. Hildeman DA, Mitchell T, Teague TK, et al. Reactive oxygen species
regulate activation-induced T cell apoptosis. Immunity. 1999;10:735–
19. Heltzer M, Jawad A, Rae J, et al. Diminished T cell numbers in patients
with chronic granulomatous disease. Clin Immunol. 2002;105:273–278.
20. Krutzik SR, Tan B, Li H, et al. TLR activation triggers the rapid
differentiation of monocytes into macrophages and dendritic cells. Nat
Inflamm Bowel Dis ● Volume 15, Number 8, August 2009CD68 Expression