This article is mentioned in:

Introduction

Intracellular and extracellular signal transduction is capable of driving tumor formation. The upregulated signaling of tyrosine kinases plays crucial roles in the regulation of cell growth and differentiation, and exhibits direct correlations with cancer progression. c-Kit and platelet-derived growth factor receptor (PDGFR) belong to the class III receptor tyrosine kinase (RTK) family. Previous studies reported that c-Kit and PDGFR were associated with cell proliferation, invasiveness and metastasis of malignant tumors (1,2).

The expression and activation of c-Kit are well documented in a variety of human tumors including myeloid leukemia, GIST, lung cancer and breast carcinoma (3–5). The expression of c-Kit in epithelial ovarian tumors was previously described. However, the results obtained from human tissues showed specific new information to add to previous studies (6). At present, there are few reports on c-Kit expression in ovarian stromal tissues.

PDGFR α and β are involved in a variety of physiological and pathological processes, such as cell growth and survival, transformation, migration, vascular permeability, stroma modulation and wound healing. The expression of PDGFR in ovarian tumors was previously reported (2,6). PDGFRα exhibits different expression profiles in ovarian carcinoma and normal ovarian epithelium. Its expression is often linked to poor prognosis (7) and aggressive tumor characteristics (8). PDGFR expression in epithelial ovarian tumor cells and tumor stroma has not been well studied.

We previously reported the overexpression of c-Kit and PDGFRα in epithelial ovarian tumors (9). In the present study, we systematically investigated c-Kit and PDGFRα expression in malignant and benign epithelial ovarian tumors, normal tissues and tumor stroma. Our hypothesis is that the c-Kit and PDGFR-dependent tumor-stromal interactions may play significant roles in the growth of normal epithelial cells and the development of tumorigenic tissues as well.

Materials and methods

Tissue sources

This study was approved by the Institutional Review Board of the First Affiliated Hospital, Yangtze University, China. Written informed consent was obtained prior to enrollment. Tissues from 71 malignant and 20 benign epithelial ovarian tumors were collected between January 1997 and January 2006. The age of the patients ranged from 28 to 65 years, with a mean age of 43 years. The benign ovarian epithelial tumors included 10 serous and 10 mucinous cystadenomas, and the malignant ovarian epithelial tumors included 59 serous and 12 mucinous cystadenomas. These 71 malignant ovarian tumors were composed of 45 low-grade (well-differentiated) and 26 high-grade (poorly-differentiated) tumors. According to the Federation of Gynecology and Obstetrics (FIGO) of 1998, 30 cases were at stage I–II, and 41 were at stage III–IV. A total of 20 normal ovarian tissues were collected as controls from patients following surgery for uterine fibroids.

Immunohistochemistry

Tissue sections (4 μm) were cut from paraffin-embedded tissue blocks. The sections were deparaffinized in xylene, rehydrated in an ethanol gradient and rinsed in phosphate-buffered saline (PBS). Sections were then heated in EDTA buffer at 95°C for 8 min for antigen retrieval. Endogenous peroxidase activity was quenched by incubation in 3% hydrogen peroxide at 37°C for 10 min. The sections were incubated with primary mouse monoclonal antibody against human c-Kit or PDGFRα (Fuzhou Maxim Biotech Co., China) at room temperature for 60 min in a humid chamber. The sections were stained using a commercial kit (Fuzhou Maxim Biotech Co.) and visualized using freshly prepared solutions of diaminobenzidine (DAB) (Maxim Biotech Inc.). The specimens were counterstained with hematoxylin and mounted with mounting medium. Positive controls (including GIST for c-Kit and breast carcinoma for PDGFRα) were also performed to confirm antibody staining.

Evaluation of immunohistochemical staining

Staining was considered as positive when cytoplasmic or membranous staining was observed. Semi-quantitative analysis of the immunostaining for c-Kit or PDGFRα was performed according to the procedures described previously (9). The percentage of immunoreactive cells for each area was determined using a score of 0–3 (0, 0–5% ; 1, 6–25% ; 2, 26–50%; and 3, 51–100% stained cells). The staining intensity was also scored as 0–3 (0, negative; 1, weak; 2, moderate; and 3, strong). In each slide, the expression level scores of c-Kit or PDGF were achieved by amplifying the scores of staining intensity to the number of stained cells. The expression levels of c-Kit or PDGFRα were evaluated according to following scoring criteria: grade 0 or - (score 0–1); grade 1 or + (scores 2–3); grade 2 or ++ (scores 4–6); and grade 3 or +++ (scores 6–9). Specimens with grade 1 were regarded as having a weakly positive expression, whereas grades 2 and 3 were defined as having a strongly positive expression or overexpression.

Statistical analysis

The differences in c-Kit and PDGFRα expression among benign and malignant epithelial ovarian tumors, normal tissues and tumor stroma were analyzed using the Chi-square test. P≤0.05 was considered to be statistically significant.

Results

Expression of c-Kit and PDGFRα in normal ovarian, benign and malignant ovarian tumor tissues

Based on the immunohistochemical staining, 10.0% of the normal ovarian tissue samples and 20.0% of the benign ovarian epithelial tumor samples were positive for c-Kit expression (Table I). No significant difference in expression levels of c-Kit in normal ovarian tissues and benign ovarian epithelial tumors was observed (P>0.05). In contrast, c-Kit was positively expressed in 50.7% of the malignant ovarian tumor tissues (Table I). The expression level of c-Kit in malignant ovarian tumors was found to be significantly higher (P<0.01) when compared to the normal ovarian tissues and benign ovarian tumors.

Table I Expression level of c-Kit and PDGFRα in different types of tissues.

Table I

Expression level of c-Kit and PDGFRα in different types of tissues.

		c-Kit expression		PDGFRα expression
Tissue types	Case (n)	Positive (n)	Positive rate (%)	Positive (n)	Positive rate (%)
Normal ovarian tissues	20	2	10.0	3	15.0
Benign ovarian tumors	20	4	20.0a	5	25.0a
Malignant ovarian tumors	71	36	50.7b	45	63.4b

a P>0.05 (benign ovarian tumor tissues vs. normal ovarian tissues);

b P<0.01 (malignant ovarian tumors vs. benign ovarian tumor tissues or normal ovarian tissues).

Similarly, no significant difference (P>0.05) was found in PDGFRα expression in normal ovarian tissues and benign ovarian tumors. In the normal ovarian tissues and benign ovarian tumors, 15.0 and 25.0% of the tissues and tumors, respectively, were positive for PDGFRα expression (Table I). PDGFRα was positively expressed in 63.4% of the malignant ovarian tumor tissues (Table I). The expression level of PDGFRα in malignant ovarian tumor was also found to be significantly higher (P<0.01) when compared to the normal ovarian tissues and benign ovarian tumors.

Correlation of c-Kit and PDGFRα expression with clinicopathological parameters

To characterize the grade-specific expression of these kinases, tumors were subdivided into two categories: low-grade (Grade 1 and 2) and high-grade (Grade 3). c-Kit or PDGFRα was expressed more highly in the high-grade tumors, and the difference between groups was statistically significant (P<0.05). Although c-Kit and PDGFRα were expressed more highly in the FIGO clinical stages III–IV than in stages I–II, the difference between groups was not statistically significant (P>0.05). Compared with mucinous carcinomas, the expression of c-Kit in the serous carcinomas was higher (P<0.05), but the expression of PDGFRα showed no difference between serous and mucinous carcinomas (P>0.05).

Expression of c-Kit and PDGFRα in malignant epithelial ovarian tumor and stroma

The expression level of c-Kit in tumor stroma was significantly lower than that of malignant epithelial ovarian tumor tissues (P<0.01). In tumor stroma, only 4.2% of the tissues were positive for c-Kit expression, whereas 50.7% of the malignant ovarian tumors were positive (Table II). As shown in Table II, although PDGFRα was highly expressed in malignant epithelial ovarian tumor tissues (63.4%) and tumor stroma (87.3%), the expression level of PDGFRα in tumor stroma was significantly higher than that in malignant epithelial ovarian tumors (P<0.01). The representative immnuohistochemical staining images are shown in Fig. 1.

Figure 1 Immunohistochemical staining of c-Kit or PDGFRα in malignant ovarian tumor and tumor stroma. (A) Positive expression of c-Kit in the tumor tissue and negative expression of c-Kit in tumor stroma (magnification, ×400). (B) Positive expression of PDGFRα in the tumor tissue and the tumor stroma (magnification, ×200).

Table II Expression of c-Kit and PDGFRα in malignant epithelial ovarian tumors and tumor stroma.

Table II

Expression of c-Kit and PDGFRα in malignant epithelial ovarian tumors and tumor stroma.

		c-Kit expression		PDGFRα expression
Tissue types	Case (n)	Positive (n)	Positive rate (%)	Positive (n)	Positive rate (%)
Epithelial ovarian tumors	71	36	50.7	45	63.4
Tumor stroma	71	3	4.2a	62	87.3a

a P<0.01 (epithelial ovarian tumors vs. tumor stroma).

Discussion

Ovarian cancer is often variable in response to treatment. Effective molecular markers that are capable of stratifying patient tumors are essential to determine the best treatment options for patients. The type III tyrosine kinases c-Kit and PDGFR have been extensively investigated in human epithelial ovarian cancer as potential molecular markers. However, the previously reported c-Kit and PDGFR expression levels in ovarian tumors are inconsistent and occasionally even contradictory to each other. For example, Parrott et al (10) reported c-Kit expression in normal ovarian surface epithelium, but this finding has not been confirmed by other studies (6,11). In ovarian carcinomas, a high expression of c-Kit has been demonstrated (6,10,11). However, other studies have shown lower levels of c-Kit expression (12,13). Similarly, the reported level of PDGFRα expression in ovarian carcinomas ranges from 5 to 100% (7,14,15). Thus, further evaluation of c-Kit and PDGFR expression in ovarian cancer is required. In the present study, we showed a significantly higher expression of c-Kit and PDGFRα in malignant ovarian tumors when compared to normal ovarian tissues and benign ovarian tumors. No significant difference was observed in c-Kit and PDGFRα expression in normal ovarian tissues and benign ovarian tumors. These results suggest that a high expression of c-Kit and PDGFRα is associated with the malignant form of ovarian tumor. Therefore, it is possible to distinguish malignant ovarian tumors from benign tumors or normal tissues by examining c-Kit and PDGFRα expression.

In the current study, the positive staining of c-Kit and PDGFRα in the low-differentiated grade group was higher than that in the high- and middle-differentiated grade groups (Table III). The detection of c-Kit and PDGFRα proteins may be useful for examining the differential degree of malignancy and the prognosis of patients with epithelial ovarian carcinomas. The expression of c-Kit in the serous carcinomas was higher as compared with the mucinous carcinomas, but no difference was observed in the expression of PDGFRα between serous and mucinous carcinomas. The overexpression of c-Kit is actually more closely associated with tumorigenesis and development of the serous ovarian carcinomas.

Table III Expression of c-Kit and PDGFRα in samples of various clinical stages (FIGO) and histological subtypes of ovarian serous carcinomas.

Table III

Expression of c-Kit and PDGFRα in samples of various clinical stages (FIGO) and histological subtypes of ovarian serous carcinomas.

		c-Kit expression			PDGFRα expression
	Case	Positive (n)	Positive rate (%)	P-value	Positive (n)	Positive rate (%)	P-value
Grade				<0.05			<0.05
1,2	45	18	40.0		24	53.3
3	26	18	69.2		21	28.8
Stage				<0.05			<0.05
I–II	30	12	40.0		18	60.0
III–IV	41	24	58.5		27	65.9
Histologic subtype				<0.05			<0.05
Serous	59	34	57.6		39	66.1
Mucinous	12	2	16.7		6	50.0

The interactions between ovarian epithelial cells and stromal cells also exhibit crucial biological functions, such as involvement in the formation of a variety of extracellular matrix components (16). The ovarian epithelial tumorigenic tissues, as well as most ovarian tumors, have a close correlation with tumor stromal tissues. Therefore, the interactions between stromal and ovarian epithelial cells may play a key role in the growth of normal epithelial cells and the development of tumorigenic tissues. However, the role of tumor stromal tissues in the development of ovarian cancer is not well understood.

Findings of the present study demonstrated a high PDGFRα expression but a low c-Kit expression in the tumor stroma of the malignant ovarian tumors. These results suggest that the major functions of c-Kit and PDGFRα during the formation of epithelial ovarian cancer may be different. c-Kit is mainly expressed in ovarian cancer cells. Over-activated c-Kit results in the inhibition of apoptosis and cell proliferation in the tumor. By contrast, a high expression of PDGFRα in stromal tissues of ovarian cancer may be beneficial for the growth, invasion and metastasis of ovarian tumors by promoting cell proliferation and angiogenesis (17). PDGFR was found to be highly expressed in fibroblasts, pericytes and endothelial cells (18,19). Blocking PDGFR is capable of changing the tumor microenvironment by increasing stroma permeability and inhibiting angiogenesis, thus affecting tumor cells directly. Blocked PDGFR signaling may also enhance chemotherapy delivery (17). Whether c-Kit and PDGFR are capable of synergistically affecting the development of ovarian cancer requires further study.

Acknowledgements

This study was supported by the Research Foundation of the Health Department of Hubei Province (JX3C54).

References