Basic Research—Biology
Co-Expression of Cyclooxygenase-2 and Vascular
Endothelial Growth Factor in Inflamed Human
Pulp: An Immunohistochemical Study
Günseli Güven, DDS, PhD,* Ceyhan Altun, DDS, PhD,* Ömer Günhan, DDS, PhD,†
Taskin Gurbuz, DDS, PhD,‡ Feridun Basak, DDS, PhD,* Erman Akbulut, DDS, PhD,*
and Zafer C. Cehreli, DDS, PhD§
Abstract
Recent data from the medical literature indicates
that cyclooxygenase-2 (COX-2) plays a key role in the
production of vascular endothelial growth factor
(VEGF), a glycoprotein that has the ability to increase
the permeability of blood vessels and to induce
angiogenesis. This study was undertaken to investi-
gate the immunohistological co-expression of COX-2
and VEGF in inflamed human pulp, in conjunction with
the expression of CD34, a transmembrane glycoprotein
expressed in endothelial cells. Pulp tissue of extracted
carious human third molars with a recent history of spon-
taneous pain were collected and processed for immuno-
staining of COX-2, VEGF, and CD34 using the biotin-
streptoavidin method. Healthy pulp samples served as
controls. COX-2 expression was not observed in healthy
pulps, whereas all inflamed pulps demonstrated COX-2-
expressing cells. Similarly, VEGF was not expressed in
normal pulp tissue, but was strongly positive in inflamed
pulps. CD34 was expressed in the endothelium of both
normal and inflamed pulp tissues. Co-expression of COX-2
and VEGF in all consecutive sections of inflamed pulps
could be suggestive of a possible release of VEGF via a
COX-2-dependent pathway. (J Endod 2007;33:18–20)
Key Words
Angiogenesis, CD34, COX-2, inflammation, pulp dental,
vascular endothelial growth factor
From the *Department of Pediatric Dentistry, Center of
Dental Sciences, †Department of Pathology, Faculty of Medi-
cine, Gulhane Medical Academy, Ankara, Turkey; ‡Department
of Pediatric Dentistry, Faculty of Dentistry, Ataturk University,
Erzurum, Turkey; and the §Department of Pediatric Dentistry,
Faculty of Dentistry, Hacettepe University, Ankara, Turkey.
Address requests for reprints to Dr. Günseli Güven, Gül-
hane Askeri Tıp Akademisi, Dişhekimliği Bilimleri Merkezi,
Pedodonti Anabilim Dalı, Etlik, Ankara, Turkey. E-mail address:
guvengunseli@yahoo.com.
0099-2399/$0 - see front matter
Copyright © 2007 by the American Association of
Endodontists.
doi:10.1016/j.joen.2006.07.014
18 Güven et al.
H uman dental pulp, surrounded by the low-compliance structure of dentin and
enamel, is susceptible to tissue damage when there is an increase in interstitial fluid
pressure in an inflammatory state (1, 2). Vascular endothelial growth factor (VEGF),
also known as the vascular permeability factor, is a glycoprotein that has the ability to
increase the permeability of blood vessels (3), an important vascular change observed
during inflammatory processes (4, 5). VEGF also plays a critical role in angiogenesis
and neovascularization, which actually can increase and extend the severity of the
inflammatory processes because of the increased transport of inflammatory cells, nu-
trients, and oxygen to the site of inflammation (5, 6). Recently, VEGF has been found to
be augmented in inflamed pulp tissue (4) and periapical lesions (7).
Prostaglandins (PG) are important mediators of inflammation, the synthesis of
which is initiated by release of arachidonic acid from cell membranes. Cyclooxygenase
(COX) catalyzes the conversion of arachidonic acid to PGH2 (8), and is the key and rate
limiting enzyme in the biosynthesis of prostanoid (8, 9). Two isoforms and one variant
of COX (COX-1, COX-2, and COX-3, respectively) have been cloned and characterized
(10). COX-1 is constitutively expressed in cells and is considered a housekeeping
enzyme (8). The inflammatory and inducible effects of COX are mediated through
COX-2, which is up-regulated in inflamed tissues and is at low or undetectable levels in
healthy tissues (9, 11). COX-2 has been recently shown to play an important role in the
pathogenesis of pulpal inflammation (8, 12).
COX-2 was originally identified as an inducible enzyme expressed at the site of
inflammation (13), whereas recent evidence demonstrates that the production of pro-
stanoids by COX-2 promotes the expression of VEGF and subsequent angiogenesis
(14, 15). However, little is known about how the level of prostanoids change, and how
the expression and activity of the synthases involved in the production of angiogenic
factors are regulated (15). Induction of VEGF in infected human pulp fibroblasts have
been shown in vitro (12), and the possibility that COX-2 may be involved in pathological
angiogenesis may have strong implications for the study of pulpal disease as well as
future pharmacological strategies for the treatment of inflamed dental pulp. Conse-
quently, this study aimed to investigate the immunohistological expression of COX-2 and
VEGF in inflamed human pulp, in conjunction with expression of CD34, a transmem-
brane glycoprotein used to measure angiogenesis (16).
Materials and Methods
There were 24 pulpal tissue specimens (12 inflamed and 12 normal pulps) ob-
tained from extracted third molars of 21 patients (mean age 20 years; range, 18 –25
years). Medical histories of all patients were noncontributory. The inflamed pulps were
obtained from teeth with profound caries and showing spontaneous pain. A positive
pain history was recorded and was considered to be consistent with irreversible pulpitis
when the patients reported that they had experienced toothache or suffered sleep loss
attributed to dental pain during the past 48 hours. Patients who had taken analgesic
medicaments during the past 24 hours were excluded from the study. Asymptomatic
pulps, which served as controls, were obtained from noncarious teeth extracted for
orthodontic reasons. Informed consent was obtained from patients before extractions,
JOE — Volume 33, Number 1, January 2007

and both the consent procedure and the research protocol were per-
formed upon approval by the institutional human subject review com-
mittee.
Preparation of Tissue Specimens
Immediately after extraction, each tooth was halved lengthways
using a diamond instrument and a hammer. The dental pulp was, then,
extracted carefully using an explorer and/or a spoon excavator. After
fixation with 10% buffered formalin overnight, the specimens were de-
hydrated in an ascending series of graded alcohol and embedded in
paraffin (8). Consecutive 5-␮m thick sections were cut on a rotary
microtome, and collected on poly-L-lysine-coated slides. One of the
sections was stained with hematoxylin and eosin (H&E). Other sections
were immunohistologically stained in a sequence to observe the pres-
ence and localization of COX-2, VEGF, and CD34.
Immunohistological Procedure
Immunoperoxidase staining for COX-2, VEGF, and CD34 was per-
formed by the biotin-streptoavidin method (17). Briefly, sections were
dewaxed and microwave treated at 500W for 5 minutes twice in 10 mM
sodium citrate (pH:6.0). Endogenous peroxidase was blocked by incu-
bation in 0.03% hydrogen peroxide in absolute methanol for 30 minutes
at room temperature and washed with PBS. The antibodies used were:
an epitope-specific rabbit antibody against COX-2 (RB-9072-R7, 1/200
dilution, Lab Vision Corp., Fremont, CA), a mouse monoclonal antibody
against VGEF protein (C-1: sc-7269, 1/100 dilution, Santa Cruz Biotech-
nology, Santa Cruz, CA); and a mouse monoclonal antibody against
CD34 (QBEnd/10; 1/100 dilution; NeoMarkers, Fremont CA). After
30 minutes of incubation with blocking solution, sections were incu-
bated with the primary antibodies at 4°C overnight, followed by sequen-
tial 30-minute incubations with biotinylated secondary antibodies and
streptoavidin. Immunohistochemistry was performed using a biotin-
streptoavidin universal kit (NeoMarkers, Fremont, CA). The sections
were counterstained with hematoxylin solution and mounted with glyc-
erol-gelatin. Negative controls for each tissue section were prepared by
omitting the primary antibody.
Figure 1. Immunoreactivity of VEGF, COX-2, and CD34 in healthy pulp speci-
mens (original magnification, ⫻100). A, A typical Hematoxylin-eosin-stained
section showing normal vasculature and stroma. B, Moderate expression of
VEGF in the endothelium (arrows) and stromal cells (arrowheads). C, No
expression of COX-2 is evident. D, Intense immunostaining of the endothelial
lining cells of the vessels with CD34.
JOE — Volume 33, Number 1, January 2007
Basic Research—Biology
Figure 2. There were 5-␮m thick consecutive sections of inflamed pulp, dem-
onstrating immunoreactivity of VEGF, COX-2, and CD34, low-magnification
views (original magnification, ⫻100). A, H&E stained section showing marked
infiltration of inflammatory cells: neutrophils (arrow), eosinophils (arrow-
head), and lymphocytes (asterisk). B, Intense immunostaining of the vessels
(arrows) and, to a lesser extent, inflammatory cells (arrowheads) with VEGF. C,
Intense positivity of the vessels (arrows) and inflammatory cells (arrowheads)
to COX-2. D, Positivity of the blood vessels to CD34.
Results
Healthy Pulp
H&E stained sections of the healthy pulp specimens showed a
normal morphology and regular distribution of the vessels and stromal
cells (Fig. 1A). Only a few of the vessels demonstrated moderate expres-
sion of VEGF, indicative of physiological angiogenesis (Fig. 2B). In
addition to the VEGF-positive endothelial lining cells, a few of the VEGF-
expressing stromal cells were also detected. There was no sign of COX-
2-positive cells in either the stroma or the endothelium (Fig. 1C). All
specimens demonstrated intense immunoreactivity to CD34 in the entire
endothelium (Fig. 1D).
Inflamed Pulp
As evidenced by H&E staining, the inflammatory infiltrate was com-
posed of neutrophils, eosinophils, and lymphocytes (Figs. 2A and 3A).
The endothelial lining cells, and to a lesser extent; the inflammatory
cells, were positively stained for VEGF (Figs. 2B and 3B). In consecutive
sections, similar findings were evident for COX-2 in the endothelium,
with a relatively greater number of positively stained inflammatory cells
(Figs. 2C and 3C). As with healthy pulp specimens, intense labeling of
endothelial cells with CD34 was observed consistently (Figs. 2D and 3D).
Discussion
Although indispensable in development and wound healing (18),
upregulation of angiogenesis in adults is usually linked with disease
(15) or inflammation (6). Thus, the blocking of angiogenesis often
relieves severity of the disease (19, 20). The role of cyclooxygenases,
prostanoids (prostaglandins and tromboxanes), and angiogenic factors
in diseases linked with excessive or pathologic angiogenesis are just
beginning to be understood. Accumulating evidence links COX-2, an
enzyme already known to be involved in inflammation and cancer, with
angiogenesis, suggesting that target COX-2 and related signaling cas-
cades could be used as anti-angiogenic agents (14, 15, 20).
Nakanishi et al. (8) studied the immunohistological characteris-
tics of COX-2-expressing cells in healthy and inflamed human dental
Co-Expression of COX-2 and VEGF in Inflamed Human Pulp 19
Basic Research—Biology
Figure 3. Development of VEGF and COX-2 immunoreactivity, consecutive high-
magnification views (original magnification, ⫻200). A, H&E stained section
showing marked infiltration of inflammatory cells. B, Strong cytoplasmic posi-
tivity to VEGF of the endothelial lining cells of blood vessels and inflammatory
cells. C, Intense immunostaining of the vessels and inflammatory infiltrate with
COX-2. D, Blood vessels, intensely labeled with CD34.
pulp (8). Our findings confirm and extend their work, demonstrating
the presence of COX-2 immunoreactivity in the endothelial lining cells.
This could be a key finding provided that COX-2-mediated VEGF expres-
sion in the endothelium of inflamed pulp vessels is substantiated by
further research. To our knowledge, expression of COX-2 in newly formed
microvessels was only demonstrated in the rat model previously (14).
CD34 is a cell adhesion molecule, and by immonuhistochemistry,
it has been found to be expressed strongly in endothelial cells (21). In
the present study, CD34 staining was utilized as a control staining to
verify normal vasculature and angiogenesis in both normal and patho-
logic conditions of the pulp. However, VEGF was only positively stained
in inflamed pulp samples with only a few exceptions seen in normal
pulp, presumably indicative of physiological angiogenesis. This finding
confirms those of Artese et al. (4) who demonstrated that both the
vessels and the stromal cells of inflamed human pulp samples were
strongly stained for VEGF.
Although qualitative co-expression of both COX-2 and VEGF was
clearly evident in inflamed pulps compared with the healthy pulp spec-
imens, absence of quantitative comparisons has to be considered a
limitation in the interpretation of our results. Also, the study was entirely
limited to one technique, immunostaining, without additional tech-
niques such as RT-PCR to show the co-expression of COX-2 and VEGF in
inflamed pulp.
Based on the limitations of this study and the literature discussed,
co-immunostaining of COX-2 and VEGF in inflamed pulp suggests a
possible induction of VEGF by COX-2, as reported in several other sites
of the human organism. However, it may be necessary to block COX-2
and detect no induction of VEGF to confirm this possibility (9). Studies
are now underway to substantiate this possible mechanism in inflamed
human pulp. Also of interest is whether induced VEGF up-regulates (22)
or inhibits COX-2 expression in the inflamed pulp, because anti-angio-
20 Güven et al.
genic factors are also expressed by dental pulp cells that may counteract
the possible COX-2-mediated VEGF expression and keep the dental pulp
within normal limits. (19).
References
1. Okabe E, Todoki K, Ito H. Direct pharmacological action of vasoactive substances on
pulpal blood flow: an analysis and critique. J Endod 1989;5:473–7.
2. Kim S. Neurovascular interactions in the dental pulp in health and inflammation.
J Endod 1990;16:48 –53.
3. Senger DR, Galli SJ, Dvorak AM, Peruzzi CA, Harvey VS, Dvorak HF. Tumor cells
secrete a vascular permeability factor that promotes accumulation of ascites fluid.
Science 1983;219:983–5.
4. Artese L, Rubini C, Ferrero G, Fioroni M, Santinelli A, Piatelli A. Vascular endothelial
growth factor (VEGF) expression in healthy and inflamed dental pulps. J Endod
2002;28:20 –3.
5. Chu SC, Tsai HC, Yang SF, et al. Induction of vascular endothelial growth factor gene
expression of proinflammatory cytokines in human pulp and gingival fibroblasts. J
Endod 2004;30:704 –7.
6. Johnson RB, Serio FG, Dai X. Vascular endothelial growth factor and progression of
periodontal diseases. J Periodontol 1999;70:848 –52.
7. Leonardi R, Caltabiano M, Pagano M, Pezzuto V, Loreto C, Palestro G. Detection of
vascular endothelial growth factor/vascular permeability factor in periapical lesion.
J Endod 2003;29:180 –3.
8. Nakanishi T, Shimizu H, Hosokawa Y, Matsuo T. An immunohistological study on
Cyclooxygenase-2 in human dental pulp. J Endod 2001;27:385– 8.
9. Holt CI, Hutchins MO, Pileggi R. A real time quantitative PCR analysis and correlation
of COX-1 and COX-2 enzymes in inflamed dental pulps following administration of
three different NSAIDs. J Endod 2005;31:799 – 804.
10. Chandrasekharan NV, Dai H, Roos KL, Evanson NK, Tomsik J, Elton TS. COX-3, a
cylooxygenase-1 variant inhibited by acetaminophen and other analgesic/antipyretic
drugs: cloning, structure and expression. Proc Natl Acad Sci USA 2002;99:
13926 –31.
11. Huang FM, Tsai CH, Ding SJ, Chang YC. Induction of cyclooxygenase-2 expression in
human pulp cells stimulated by dentin bonding agents. Oral Surg Oral Med Oral
Pathol Oral Radiol Endod 2005;100:501– 6.
12. Chang YC, Huang FM, Yang SF, et al. Induction of cyclooxygenase-2 mRNA and protein
expression in human pulp cells stimulated with black-pigmented bacteroides. J
Endod 2003;29:240 –3.
13. O’Banion MK, Winn VD, Young DA. cDNA cloning and functional activity of a glu-
cocorticoid-regulated inflammatory cyclooxygenase. Proc Natl Acad Sci USA
1992;89:4888 –92.
14. Majima M, Hayashi I, Muramatsu M, Katada J, Yamashina S, Katori M. Cyclo-oxygen-
ase-2 enhances basic fibroblast growth factor-induced angiogenesis through induc-
tion of vascular endothelial growth factor in rat sponge implants. Br J Pharmacol
2000;130:641–9.
15. Iniguez MA, Rodriguez A, Volpert OV, Fresno M, Redondo JM. Cyclooxygenase-2: a
therapeutic target in angiogenesis. Trends Mol Med 2003;9:73– 8.
16. Ramani P, Bradley NJ, Fletcher CD. QBEND/10, a new monoclonal antibody to en-
dothelium: assessment of its diagnostic utility in paraffin sections. Histopathology
1990;17:237– 42.
17. Rissanen TT, Vajanto I, Hiltunen MO, et al. Expression of vascular endothelial growth
factor and vascular endothelial growth factor receptor-2 (KDR/Flk-1) in ischemic
skeletal muscle and its regeneration. Am J Pathol 2002;160:1393– 403.
18. Roberts-Clark DJ, Smith AJ. Angiogenic factors in human dentine matrix. Arch Oral
Biol 2000;45:1013– 6.
19. Leahy KM, Koki AT, Masferrer JL. Role of cyclooxygenases in angiogenesis. Curr Med
Chem 2000;7:1163–70.
20. Miura S, Tatsuguchi A, Wada K, et al. Cyclooxygenase-2-regulated vascular endothe-
lial growth factor release in gastric fibroblasts. Am J Physiol Gastrointest Liver Physiol
2004;287:444 –51.
21. Parra ER, Silverio da Costa LR, Ab’Saber A, et al. Nonhomogeneous density of CD34
and VCAM-1 alveolar capillaries in major types of idiopathic interstitial pneumonia.
Lung 2005;183:363–73.
22. Akarasereenont PC, Techatraisak K, Thaworn A, Chotewuttakorn S. The expression of
COX-2 in VEGF-treated endothelial cells is mediated through protein tyrosine kinase.
Mediators Inflamm 2002;11:17–22.
JOE — Volume 33, Number 1, January 2007