Profiling of Radicular Cyst and Odontogenic Keratocyst Cytokine Production Suggests Common Growth

(1)

Profiling of Radicular Cyst and Odontogenic Keratocyst Cytokine Production Suggests Common Growth

Mechanisms

Makoto Hayashi, DDS, PhD,*

^†

Takafumi Ohshima, DDS, PhD,*

Mitsuhiro Ohshima, DDS, PhD,

^‡§

Yoko Yamaguchi, BS,

^‡§

Hirofumi Miyata, DDS, PhD,*

Osamu Takeichi, DDS, PhD,*

^†

Bunnai Ogiso, DDS, PhD,*

^†

Koichi Ito, DDS, PhD,*

^†

Arne Östman, PhD, ^储 and Kichibee Otsuka, DDS, PhD

^‡§

Abstract

The aim of this study was to compare the cytokine expression profiles of cyst fluids (CFs) and tissue culture supernatants (SUPs) from 7 radicular cysts (RCs) and 7 odontogenic keratocysts (OKCs) by using Human Cyto- kine Antibody Array to identify the specific cytokines involved in formation and expansion of RCs and OKCs, respectively. There were significant differences in relative expression levels of IL-1␤, MCP1, MIP1␤, FGF-9, GDNF, HGF, IGFBP-3, Ang, IP-10, MIF, OPG, and TGF-␤2 between RC-CF and OKC-CF (P ⬍ .05). On the other hand, the cytokine expression patterns of RC-SUP (HGF, IL-8, NAP-2, IL-6, TIMP-1 and 2, GRO, IP-10, and Ang) were similar to those of OKC-SUP. Only the relative expression level of GRO differed between RC-SUP and OKC-SUP (P ⬍ .05). The similarities of cytokine production by tissue cultures derived from RC and OKC indicate that the expansion mechanisms of RC and OKC might involve similar biologic mechanisms other than infection.(J Endod 2008;34:14 –21)

Key Words

Cytokine profile, odontogenic keratocyst, radicular cyst

O

dontogenic cysts are considered to be initiated by the proliferation of residues of tooth-forming epithelia (1). They are divided into 2 common groups, developmen- tal cysts such as odontogenic keratocysts (OKCs) and inflammatory cysts such as radicular cysts (RCs) (2).

RCs form in the apical region and are the only infectious cyst and are most frequently seen in all kinds of odontogenic cysts (3–5). RCs are lined by nonkeratinized stratified squamous epithelium of variable thickness. The underlying supportive connective tissue might be focally or diffusely infiltrated with mixed inflammatory cell populations. It has been considered that endotoxins released by anaerobes in the infected necrotic tooth pulp induce inflammation around the apical area (5). There- fore, it is expected that the RCs might be healed by removing the antigenic factors with root canal treatment. However, RCs sometimes continuously expand after root canal therapy and eventually have to be removed by surgical procedure. This phenomenon suggests that processes other than those induced by infection contribute to the growth of these cysts.

On the other hand, it is thought that OKCs originate from dental lamina (6). Very recently, OKCs were classified as odontogenic tumors in the World Health Organization classification because of their aggressive biologic behavior (7). Microscopic observa- tion of OKC sections showed the cystic structure to be lined by a thin uniform layer of parakeratotic stratified squamous epithelium (6).

It is noteworthy that bone degradation is a common phenotype of both kinds of cysts. Bone-resorbing factors such as prostaglandins, interleukins, and proteinases, secreted by inflammatory cells and mesenchymal cells of the cysts, have been reported to be involved in cyst enlargement (5, 8, 9).

There have been few reports on the causes of odontgenic cysts that describe the relationship between epithelial tissue and surrounding connective tissue. Therefore, we hypothesized that profiling of cytokine in the cyst fluids (CFs) together with the cyst tissue culture supernatants (SUPs) might be a useful strategy in analyzing the interaction among the cyst-forming cells including infiltrating hematopoietic cells. We also predicted that the results of the cytokine profile would assist in exploring the mechanisms involved in the formation of RCs and OKCs, as well as in their diagnosis for clinical benefit.

In this study we have therefore compared the cytokine expression profiles between CFs and SUPs of RCs and OKCs by using Human Cytokine Antibody Array (Ray Biotech Inc, Norcross, GA), which allows semiquantitative analysis of 79 cytokines.

Materials and Methods

Patients and Tissue Samples

The experimental protocol was approved by the Ethics Committee of Nihon Uni- versity School of Dentistry, and all patients gave informed consent in writing to this study.

Cysts were obtained from patients undergoing surgery at the Nihon University School of Dentistry Dental Hospital. All the patients were free from systemic disease.

From the *Department of Endodontics and^‡Department of Biochemistry, Nihon University School of Dentistry, Tokyo, Japan;^†Division of Advanced Dental Treatment and^§Division of Functional Morphology, Dental Research Center, Nihon University School of Dentistry, Tokyo, Japan; and ^储Cancer Center Karolinska, Department of Pathology-Oncology, Karo- linska Institute, Stockholm, Sweden.

Address requests for reprints to Dr Mitsuhiro Ohshima, Department of Biochemistry, Nihon University School of Den- tistry, 1-8-13, Kanda-Surugadai, Chiyoda-ku, Tokyo 101-8310, Japan. E-mail address: oshima-m@dent.nihon-u.ac.jp.

0099-2399/$0 - see front matter

doi:10.1016/j.joen.2007.08.020

(2)

TABLE 1. Case Information Reported by the Present Study

Case No. Age

(y) Sex Tooth No. Diagnosis Symptom

(Pain)

Size of Cysts (mm)

RC1 51 F 21 Radicular cyst No 10⫻ 8 ⫻ 7

RC3 56 N 13 Radicular cyst No 8⫻ 8 ⫻ 4

RC6 45 M 23 Radicular cyst No 17⫻ 17 ⫻ 7

OKC1 35 M Left mandibular angle Keratocyst No 20⫻ 10 ⫻ 7

OKC2 24 F Right mandibular

angle

Keratocyst No 12⫻ 7 ⫻ 6

OKC5 21 F Right mandibular

angle

Keratocyst No 30⫻ 25 ⫻ 20

OKC6 23 M Molar region in maxilla Keratocyst No 13⫻ 10 ⫻ 2

TABLE 2. Cytokines Present on the Array

Abbreviation Formal Name Sensitivity Abbreviation Formal Name Sensitivity

Ang Angiogenin 10 IL-13 Interleukin 13 100

BDNF Brain-derived neurotrophic factor

100 IL-15 Interleukin 15 100

BLC B-lymphocyte chemoattractant 10 IL-16 Interleukin 16 1

Ck␤8-1 Ck beta 8-1 1000 IP-10 IPN-␥ inducible protein 10 10

EGF Epidermal growth factor 1 Leptin Leptin 100

ENA-78 Epithelial neutrophil activating protein 78

1 LIF Leukemia inhibitory factor 1000

Eot Eotaxin 1 LIGHT Lymphotoxin-like inducible protein that competes

with glycoprotein D for binding herpesvirus entry mediator on T cells

1

Eot-2 Eotaxin 2 1

Eot-3 Eotaxin 3 320 MCP-1 Monocyte chemoattractant protein 1 3

FGF-4 Fibroblast growth factor 4 1000 MCP-2 Monocyte chemoattractant protein 2 100

FGF-9 Fibroblast growth factor 9 100 MCSF Macrophage colony stimulating factor 1

Flt-3 FMS-like tyrosine kinase 3 ligand 1 MDC Macrophage-derived chemokine 1000

Fract Fractalkine 1600 MIF Macrophage migration inhibitory factor 100

GCP-2 Granulocyte chemotactic protein 2

100 MIG Monocyte induced by IFN-␥ 1

GCSF Granulocyte colony stimulating factor

2000 MIP-1␤ Macrophage inflammatory protein 1 beta 10

GDNF Glial derived neurotrophic factor

100 MIP-1␦ Macrophage inflammatory protein 1 delta 100

GM-CSF Granulocyte macrophage colony stimulating factor

100 MIP-3␣ Macrophage inflammatory protein 3 alpha 100

GRO Growth related oncogene 1000 NAP-2 Neurotrophil activating peptide 2 100

GRO-␣ Growth related oncogene alpha 1000 NT-3 Neurotrophin 3 20

HGF Hepatocyte growth factor 200 NT-4 Neurotrophin 4 2

I-309 T lymphocyte-secreted protein I-309

1000 OPG Osteoprotegerin 100

IFN-␥ Interferon gamma 100 OSM Oncostatin M 100

IGFBP-1 Insulin-like growth factor binding protein 1

1 PARC Pulmonary and activation-regulated protein 1000

10 PDGF-B Platelet-derived growth factor B 1000

1000 PIGF Placenta growth factor 100

1000 RANTES Regulated on activation, normal T cell expressed and secreted

2000

IGF-I Insulin-like growth factor I 10 SCF Stem cell factor 10

IL-1␣ Interleukin 1 alpha 1000 SDF-1 Stromal cell-derived factor 1 2000

IL-1␤ Interleukin 1 beta 100 TARC Thymus and activation-regulated chemokine 100

IL-2 Interleukin 2 25 TGF-␤1 Transforming growth factor beta 1 200

IL-5 Interleukin 5 1 TIMP-1 Tissue inhibitor of metalloproteinase 1 100

IL-6 Interleukin 6 1 TIMP-2 Tissue inhibitor of metalloproteinase 2 1

IL-7 Interleukin 7 100 TNF-␣ Tumor necrosis factor alpha -

IL-8 Interleukin 8 1 TNF-␤ Tumor necrosis factor beta 1000

IL-10 Interleukin 10 10 TPO Thrombopoetin 100

IL-12 Interleukin 12 1 VEGF Vascular epitherial growth factor 100

The profiles of cytokine expression were determined with the Human Cytokine Array V. The formal name and abbreviations of cytokines and detection limits (pg/mL) specified by the manufacturer are shown.

(3)

Seven RCs and 7 OKCs were used in this study, and more than half of all the specimens were subjected to pathologic examination.

Diagnosis of RC or OKC

A professor of Department of Pathology in Nihon University School of Dentistry who has a license of pathologist diagnosed all the specimens as 7 for RCs and other 7 for OKCs (Table 1). All the teeth with RCs were nonvital, and the microscopic observations of pathologic sections revealed that the cystic wall was composed of lining epithelium, inflamed granulation layer, and the most outside fibrous connective tissue, and cholesterol crystals were sometimes found in the cysts. On the other hand, microscopic observations that the cystic wall was lined by para- keratinized stratified squamous epithelium about 5– 8 cell layers with fibrous connective tissue and the cyst sometimes contained keratin sub- stances inside the cavity were diagnosed as OKC.

Sample Preparation

CF samples were collected from the lesions by aspirative punc- tures and centrifuged at 10,000g for 5 minutes at 4°C, and the

supernatant fractions were then stored at – 80°C until analyzed.

Tissue fragments (65–210 mg) of RCs or OKCs were cultured in fresh serum-free Dulbecco modified Eagle medium supplemented with antibiotics for 24 hours at 37°C in 5% CO₂/95% air and hu- midified atmosphere. Cyst SUPs were collected and stored at – 80°C until analyzed.

Determination of Cytokine Profiles in RCs and OKCs

The profiles of cytokine expression in CFs and SUPs of RCs or OKCs were determined by using Human Cytokine Array V, according to the manufacturer’s protocol. As shown inTable 2, this kit is capable of detecting 79 cytokines in one membrane. Cytokine array membranes that were placed into each container of an 8-well tray were blocked in 2 mL of blocking buffer for 30 minutes at room temperature and then incubated with 1 mL of SUP or CF plus 1 mL of blocking buffer at 4°C overnight. The mixtures of samples and buffer were aspirated from each container, and membranes were washed 3 times with 2 mL of wash buffer 1, followed by washing twice with 2 mL of wash buffer 2. Mem- branes were incubated in 1:250 dilution of biotin-conjugated primary Figure 1. (a) Exemplary presentation of an x-ray image of the cytokine array.

(b) The cytokine expression levels were proportional to the intensity of the spot.

Negative controls and positive controls were used to calculate RELs according to the given formula.

Figure 2. A typical result of a cytokine profile in RC-CF, OKC-CF, RC-SUP, and OKC-SUP.

TABLE 3. Specific Primers for Polymerase Chain Reaction (PCR) Name of

Target Gene

5= Primer Sequence Product Size (bp)

PCR Cycles

Reference, PrimerBank ID*

or Designed by Primer3†

3= Primer Sequence

TIMP-1 CTTCTGCAATTCCGACCTCGT 127 35 4507509a1*

CCCTAAGGCTTGGAACCCTTT

TIMP-2 AAGCGGTCAGTGAGAAGGAAG 136 35 4507511a3*

GGGGCCGTGTAGATAAACTCTAT

Ang TGGGCGTTTTGTTGTTGGTCTTC 366 35 Etoh et al²⁸

CGTTTCTGAACCCCGCTGTGG

HGF AGTACTGTGCAATTAAAACATGCG 384 35 Ljubimova et al²⁹

TTGTTTGGGATAAGTTGCCCA

NAP-2 CAGCAACTCACCCTCACTCA 188 35 Primer3^†

GTTTGTCCTTTGGTGGAGGA

IP 10 AGGAACCTCCAGTCTCAGCA 192 35 Primer3^†

CAAAATTGGCTTGCAGGAAT

GRO␣ AGGGAATTCACCCCAAGAAC 171 35 Primer3^†

TGGATTTGTCACTGTTCAGCA

GRO␤ GCAGGGAATTCACCTCAAGA 172 35 Primer3^†

GGATTTGCCATTTTTCAGCA

GAPDH ATCACCATCTTCCAGGAG 318 30 Ozawa et al³⁰

CTCATGACCACAGTCCCAT

*http://pga.mgh.harvard.edu/primerbank/

†http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi

(4)

Figure 3. Frequency (y-axis) of positive cytokine detection in the CF. Columns represent the frequency of positive cytokine detection from 7 samples. Positive detection was defined if the REL exceeded 40%. (a) RC, (b) OKC.

(5)

antibodies at 4°C for 24 hours and were then washed as described above before incubation in 1:1000 dilution of horseradish peroxidase (HRP)– conjugated streptavidin. After incubation in HRP-conjugated streptavidin at room temperature for 60 minutes, membranes were washed similarly and exposed to a peroxidase substrate (Immunostar Reagents; Wako, Osaka, Japan) for 1 minute before imaging. Mem- branes were exposed to x-ray film (Hyperfilm ECL; Amersham Bio- sciences UK, Buckinghamshire, UK) for 5–30 minutes before development.

Data Analysis

The images of x-ray films were captured by using an optical scan- ner (GT9000; Seiko Epson Corporation, Nagano, Japan) and saved as 8-bit gray-scale pictures (Fig. 1). The signal intensities were quantified with Scion Image (Scion Corporation, Frederick, MD) to compare intensities of the same cytokine from different samples. Dots were measured in the constant range from their center, and the measured values of each dot were obtained on the scale from 0 –255. To normalize the results, the measured values of each cytokine on the same membrane were revised by subtracting the average of negative control values from the intensity of each dot, including positive controls. The relative expression levels (RELs) of the cytokines were calculated according to the formula given inFig. 1and were shown as average expression levels of all patients in percentages. Finally, to describe the frequency of positive detection, positive cytokine expression was defined if the REL was higher than 40% per patient. We analyzed RELs and then compared RELs in RC-CF groups with KC-CF groups by Mann-Whitney U test. We also analyzed RELs and compared RELs in RC-SUP groups with KC-SUP groups by Mann-Whitney U test.

Reverse Transcriptase Polymerase Chain Reaction

Reverse transcriptase polymerase chain reaction (RT-PCR) was carried out to confirm the gene expression of several detected cytokines on the membrane. Total RNA was extracted from each sample of RC

tissue by using Trizol reagent (Gibco BRL; Life Technologies, Rockville, MD), according to the manufacturer’s protocol. The harvested RNA was then reverse-transcribed and amplified with a Gene Amp RNA PCR Core Kit (Applied Biosystems, Edison, NJ). The cDNA was amplified with specific primer pairs for various growth factors and chemokines, as shown inTable 3, in a reaction mixture containing Taq polymerase. PCR was performed in a DNA thermal cycler (Takara Bio Inc, Otsu, Japan) for 35 cycles of 30 seconds at 94°C, 30 seconds at 55°C, 58°C, or 59°C, and 30 seconds at 72°C. The amplified products were visualized on 1.5% agarose gels stained with ethidium bromide and photographed under ultraviolet light.

Results

Human Cytokine Antibody Array

Typical images in RC-CF, OKC-CF, RC-SUP, and OKC-SUP are shown inFig. 2. There was a clear difference in the cytokine profile between RC-CF and OKC-CF.

The analyses of RC-CF showed that GRO, IL-8, NAP-2, TIMP-1, and TIMP-2 were detected in all cases (Fig. 3a). In addition, MIP-1␤, TNF-␤, Ang, IP-10, MIF, TGF-␤2, IL-1␤, HGF, IGFBP-2, LIF, and OPG were detected in 5 or more cases among the 7 cases in RC-CFs (Fig. 3a).

Concerning the OKC-CFs, IL-8 and NAP-2 were identified in almost all cases (Fig. 3b).

There were significant differences in RELs of IL-1␤ (P ⫽ .006), MCP1 (P⫽ .041), MIP1␤ (P ⫽ .015), FGF-9 (P ⫽ .030), GDNF (P ⫽ .040), HGF (P⫽ .021), IGFBP-3 (P ⫽ .010), Ang (P ⫽ .015), IP-10 (P

⫽ .003), MIF (P ⫽ .021), OPG (P ⫽ .021), and TGF-␤2 (P ⫽ .030) between RC-CF and OKC-CF (Table 4).

On the other hand, IL-8 and Ang were detected in all cases in RC-SUP (Fig. 4a). In addition, the analysis of RC-SUP showed that IL-6, HGF, IP-10, NAP-2, TIMP-1, GRO, and TIMP-2 were detected in almost cases (Fig. 4a).

GRO, IL-6, IL-8, Ang, and NAP-2 were detected in all cases in OKC-SUP (Fig.

4b), and HGF and TIMP-1 were identified in almost all cases in OKC-SUP TABLE 4. Comparison of Cytokine Expression between RC and OKC

RC-CF (Mean%)

OKC-CF

(Mean%) P Value RC-SUP

(Mean%)

OKC-SUP

(Mean%) P Value

GRO 73.90 25.28 .064 GRO 55.89 84.45 .012*

IL-1␤ 66.80 13.00 .006** IL-1␤ 18.22 45.80 .085

IL-8 100.00 68.53 .015* IL-6 85.06 100.00 .085

MCP-1 46.46 12.90 .041* IL-8 98.98 100.00 .338

MIP-1␤ 75.94 13.43 .015* MCP-1 29.52 53.77 .224

TNF-␤ 68.02 29.18 .055 MIP-1

␤ 26.11 50.72 .180

BDNF 44.64 9.22 .021* TNF-␤ 40.93 27.22 .401

FGF-9 40.70 11.49 .030* Ang 84.55 90.03 .112

GDNF 59.58 23.18 .040* GDNF 41.06 23.30 .401

HGF 64.97 20.80 .021* HGF 71.00 78.37 .655

IGFBP-2 62.31 22.03 .074 IP-10 60.00 46.05 .338

IGFBP-3 44.35 8.19 .010* NAP-2 78.36 80.97 .655

Ang 87.45 27.67 .015* TIMP-1 75.08 63.52 .277

IP-10 79.86 29.98 .003** TIMP-2 67.56 56.12 .482

LIF 63.94 11.60 .074

MIF 78.85 27.19 .021*

NAP-2 90.82 64.85 .074

OPG 57.16 18.54 .021*

PARC 55.59 57.92 .898

TGF-␤ 2 82.05 36.65 .030*

TIMP-1 90.86 47.31 .160

TIMP-2 100.00 57.97 .201

The REL of cytokines in the CF and SUP were compared with the RC and OKC by Mann-Whitney U test.

*P⬍ .05.

**P⬍ .01.

(6)

Figure 4. Frequency (y-axis) of positive cytokine detection in the cyst SUPs. Columns represent the frequency of positive cytokine detection from 7 samples. Positive detection was defined if the REL exceeded 40%. (a) RC, (b) OKC.

(7)

(Fig. 4b). Only in the RELs of GRO did the RC-SUPs and OKC-SUPs show significant differences (P⫽ .012) (Table 4).

In summary, the cytokine expression pattern in RC-SUP was very similar to that of OKC-SUP.

RT-PCR

The mRNA expression of HGF, NAP-2, TIMP-1, TIMP-2, GRO, IP- 10, and Ang in 4 samples of RC tissues is shown inFig. 5.

We detected Ang, TIMP-1, TIMP-2, and GRO␣ mRNA expression in all 4 cases. HGF and GRO were also detected in expression in 3 of 4 cases. NAP-2 and IP-10 were found only in 2 cases.

Discussion

The merits of the cytokine antibody array that is capable of profiling the 79 kinds of protein to analyze RC-derived or OKC-derived CF and SUP might increase our knowledge concerning the development of each type of cyst. Because it is impossible to use cDNA microarray to analyze cytokine profiling of secreted fluid, it might be a considerable advantage to use this cytokine array kit.

In this study we have compared the cytokine profile in CFs and in SUPs of RCs and OKCs. It is predicted that the CF will reflect the cytokine production of the cyst tissue, as well as of inflammatory cells, whereas the SUP will contain only cytokines produced by the cyst tissue, without any contribu- tions from inflammatory cells. The profiling of the CF will possibly assist in identifying potential diagnostic markers. Importantly, the profiling of cyst SUP should indicate to what extent the tissue properties of these 2 types of cyst are similar or different.

The present investigation revealed that it is easy to distinguish RC-CF and OKC-CF because RC-CF contains so many kinds of cytokines that might be involved in host defense. Hahn and Liewahr (10) and Wisithphrom and Windsor (11) have reported that the role of proin- flammatory cytokines such as IL-1, IL-6, and TNF-␣ was important in the

pathogenesis of pulpitis. In addition, the profile of RC-CF was very similar to that of gingival crevicular fluid such as GRO, Ang, and IP-10 (12).

These might imply that RC-CF contains the results of inflammatory response against antigens from root canals. Previous studies revealed that RC-CF contains IL-1, IL-6, and GM-CSF (5, 13). We have detected addi- tional cytokines, chemokines, and growth factors such as GDNF, Ang, TNF-␤, IP-10, HGF, and NAP-2 for the first time in this fluid. Whether one of these cytokines might be a marker protein for RC requires further study. As expected, OKC-CF contained fewer kinds of cytokines compared with those of RC-CF. Only IL-8 and NAP-2, which were also detected in GCF from healthy gingiva, are predominant in OKC-CF. This might suggest that OKCs are usually not in an inflammatory condition.

Surprisingly, the cytokine profile of RC-SUP was very similar to that of OKC-SUP. Both types of cysts commonly secreted GRO, IL-6, IL-8, Ang, HGF, NAP-2, IP-10, TIMP-1, and TIMP-2. They consisted of chemokines, cytokines, growth factors, and matrix metalloproteinase (MMP) inhib- itors. The CXC chemokines, GRO and IL-8, are typical neutrophil che- moattractants; however, they are now also known as autocrine growth factors for cancer cells and stimulate tumor angiogenesis (14). NAP-2 stimulates endothelial cell chemotaxis in vitro and angiogenesis in vivo as GRO and IL-8. IL-6 has a function for bone-resorbing factor (15, 16), although originally reported as a B-cell growth and differentiation factor (17). Ang is an angiogenic factor first detected in cancer cells (18). HGF is mainly derived from mesenchymal cells and binds proto-oncogene product, c-Met, and stimulates epithelial cell proliferation and migration (19). We have suggested that HGF might play a central role as a chemoattractant for gingival epithelial cells in periodontitis (20, 21).

HGF might play a role in cyst expansion through inducing lining epithelial cell proliferation and its invasion into capsular connective tissue.

Furthermore, HGF has been reported to play a role in neovasculariza- tion through tubule formation (22).

Figure 5. Expression of mRNA of some cytokines in 4 RC tissues examined by RT-PCR.

(8)

Contrary to the above factors, CXC chemokine IP-10 has been reported as an inhibitor for angiogenesis in tumor (14). As Moore et al (23) proposed that angiogenesis in tumors is regulated by a complex balancing act between opposing angiogenic and angiostatic factors, this mechanism might be also applied to cyst expansion.

TIMP-1 and TIMP-2 inhibit proteolytic activity of MMPs, which play central roles in matrix degradation during pathologic and physiologic processes (24). However, TIMPs also possess other functions such as inhibition of apoptosis, induction of malignant transformation, stimulation of cell growth, and regulation of angiogenesis (25).

Previously we examined gene expression of cytokines, growth factors, and their receptors in lining epithelia of RCs by using RT-PCR (26).

In those studies we revealed the mRNA expression of IL-1␣, IL-1␤, IL-6, IL-8, IL-11, TGF-␤1, PDGF-A, aFGF, c-erbB, KGF receptor, c-met, and IL-6 receptor in all 5 lining epithelia of RCs. Therefore, IL-6 and IL-8 detected in RC-SUP might be derived from lining epithelial cells, and this might be similar in OKC-SUP. Kusumi et al (27) reported high expression and secretion of IL-6 in fibroblasts derived from RCs. Because epithelial lining of RCs expressed c-met and IL-6 receptor, these cells might respond to HGF from fibroblasts in capsule in paracrine manner and IL-6 from epithelial cells in both autocrine and paracrine manner.

We have thus found factors for angiogenesis (Ang, IL-8, NAP-2, GRO, HGF, and TIMPs), growth factors for epithelial cells (HGF), and bone-resorbing factor (IL-6) in both RC-SUP and OKC-SUP in similar frequency. The unexpected similarity in production of cytokines of RC and OKC cyst tissue is compatible with the hypothesis that both types of cysts share common mechanisms for expansion and bone resorption.

One implication of our study is that the bone-absorbing activity in RCs might be partially independent of inflammatory stimuli and could possibly explain the persistent growth of some of these cysts after root canal treatment. Furthermore, these findings suggest that elucidation of the mechanisms responsible for the up-regulation of these cytokines might provide leads for novel treatment regimens for both RCs and OKCs.

Conclusion

The similarities of cytokine profile of SUPs derived from RCs and OKCs indicate that the expansion mechanisms of RCs and OKCs might involve similar biologic mechanisms other than infection.

Acknowledgments

The authors are grateful to Professor C. S. Langham for care- fully reading and editing the manuscript. This study was supported in part by Grants-in-Aid for Scientific Research (A-#16209063, C-#19592395) from the Japan Society for Promotion of Science (M.O.), Nihon University Individual Research Grant for 2006 (M.O.), Nihon University Research Grant for Assistants and Young Researchers for 2006 (Y.Y.), Grant from Dental Research Center, Nihon University School of Dentistry for 2005 (O.T. and B.O.) and 2006 (M.O.), and the Promotion and Mutual Aid Corporation for Private Schools of Japan for 2005 and 2006 (T.O.).

References

1. Harris M, Toller P. The pathogenesis of dental cysts. Br Med Bull 1975;31:159 – 63.

2. Sciubba JJ, Fantasia JE, Kahn LB. Tumors and cysts of the jaws: atlas of tumor pathology. Washington, DC: Armed Forces Institute of Pathology, 2001;third series fascicle 29:11–3.

3. Bando Y, Henderson B, Meghji S, Poole S, Harris M. Immunocytochemical localiza- tion of inflammatory cytokines and vascular adhesion receptors in radicular cysts.

J Oral Pathol Med 1993;22:221–7.

4. Hosoya S, Matsushima K. Stimulation of interleukin-1 beta production of human dental pulp cells by Porphyromonas endodontalis lipopolysaccharide. J Endod 1997;23:39 – 42.

5. Meghji S, Qureshi W, Henderson B, Harris M. The role of endotoxin and cytokines in the pathogenesis of odontogenic cysts. Arch Oral Biol 1996;41:523–31.

6. Shear M. The aggressive nature of the odontogenic keratocyst: is it a benign cystic neoplasm? part 1: clinical and early experimental evidence of aggressive behaviour.

Oral Oncol 2002;38:219 –26.

7. Philipsen HP. Keratocystic odontogenic tumor. In: Barnes L, Eveson JW, Reichert P, Sidransky D, eds. World Health Organization classification of tumours: pathology and genetics of head and neck tumours. Lyon, France: IARC Press, 2005:306 –7.

8. Kawashima N, Stashenko P. Expression of bone-resorptive and regulatory cytokines in murine periapical inflammation. Arch Oral Biol 1999;44:55– 66.

9. Ohshima T, Ohshima M, Yamaguchi Y, et al. PGE2stimulates IL-6, HGF and PGE2

production by cultured human radicular cyst-derived fibroblast-like cells. Interna- tional Congress Series 2005;1284:223– 4.

10. Hahn CL, Liewehr FR. Relationships between caries bacteria, host responses, and clinical signs and symptoms of pulpitis. J Endod 2007;33:213–9.

11. Wisithphrom K, Windsor LJ. The effects of tumor necrosis factor-␣, interleukin-1␤, interleukin-6, and transforming growth factor-␤1 on pulp fibroblast mediated col- lagen degradation. J Endod 2006;32:853– 61.

12. Sakai A, Ohshima M, Sugano N, Otsuka K, Ito K. Profiling the cytokines in gingival crevicular fluid using a cytokine antibody array. J Periodontol 2006;77:856 – 64.

13. Gervsio AM, Silva DA, Taketomi EA, Souza CJ, Sung SS, Loyola AM. Levels of GM-CSF, IL-3, and IL-6 in fluid and tissue from human radicular cysts. J Dent Res 2002;81:64 – 8.

14. Frederick MJ, Clayman GL. Chemokines in cancer. Expert Reviews in Molecular Medicine. Available at www.expertreviews.org/01003301h.htm. Accessed July 18, 2001.

15. Ishimi Y, Miyaura C, Jin CH, et al. IL-6 is produced by osteoblasts and induces bone resorption. J Immunol 1990;145:3297–303.

16. Kwan Tat S, Padrines M, Théoleyre S, Heymann D, Fortun Y. IL-6, RANKL, TNF-alpha/

IL-1: interrelations in bone resorption pathophysiology. Cytokine Growth Factor Rev 2004;15:49 – 60.

17. Kishimoto T, Hirano T. Molecular regulation of B lymphocyte response. Annu Rev Immunol 1988;6:485–512.

18. Fett JW, Strydom DJ, Lobb RR, et al. Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells. Biochemistry 1985;24:5480 – 6.

19. Zarnegar R, Michalopoulos GK. The many faces of hepatocyte growth factor: from hepatopoiesis to hematopoiesis. J Cell Biol 1995;129:1177– 80.

20. Ohshima M, Noguchi Y, Ito M, Maeno M, Otsuka K. Hepatocyte growth factor secreted by periodontal ligament and gingival fibroblasts is a major chemoattractant for gingival epithelial cells. J Periodontal Res 2001;36:377– 83.

21. Ohshima M, Yokosuka R, Yamazaki Y, Tokunaga T, Maeno M, Otsuka K. Effects of serum on hepatocyte growth factor secretion and activation by periodontal ligament and gingival fibroblasts. J Periodontol 2002;73:473– 8.

22. Jiang WG, Hiscox SE, Parr C, et al. Antagonistic effect of NK4, a novel hepatocyte growth factor variant, on in vitro angiogenesis of human vascular endothelial cells.

Clin Cancer Res 1999;5:3695–703.

23. Moore BB, Keane MP, Addison CL, Arenberg DA, Strieter RM. CXC chemokine mod- ulation of angiogenesis: the importance of balance between angiogenic and angiostatic members of the family. J Investig Med 1998;46:113–20.

24. Birkedal-Hansen H, Moore WG, Bodden MK, et al. Matrix metalloproteinases: a review. Crit Rev Oral Biol Med 1993;4:197–250.

25. Chirco R, Liu XW, Jung KK, Kim HR. Novel functions of TIMPs in cell signaling. Cancer Metastasis Rev 2006;25:99 –113.

26. Ohshima M, Nishiyama T, Tokunaga K, Sato S, Maeno M, Otsuka K. Profiles of cytokine expression in radicular cyst-lining epithelium examined by RT-PCR. J Oral Sci 2000;42:239 – 46.

27. Kusumi A, Sakaki H, Fukui R, Satoh H, Kusumi T, Kimura H. High IL-6 synthesis in cultured fibroblasts isolated from radicular cysts. Arch Oral Biol 2004;49:643–52.

28. Etoh T, Shibuta K, Barnard GF, Kitano S, Mori M. Angiogenin expression in human colorectal cancer: the role of focal macrophage infiltration. Clin Cancer Res 2000;6:3545–51.

29. Ljubimova JY, Petrovic LM, Wilson SE, Geller SA, Demetriou AA. Expression of HGF, its receptor c-met, c-myc, and albumin in cirrhotic and neoplastic human liver tissue.

J Histochem Cytochem 1997;45:79 – 87.

30. Ozawa Y, Shimizu N, Abiko Y. Low-energy diode laser irradiation reduced plasmin- ogen activator activity in human periodontal ligament cells. Lasers Surg Med 1997;21:456 – 63.

Profiling of Radicular Cyst and Odontogenic Keratocyst Cytokine Production Suggests Common Growth