J Cancer Res Clin Oncol (2009) 135:837–846 DOI 10.1007/s00432-008-0520-1

ORIGINAL PAPER

Expression of stathmin/op18 as a signiWcant prognostic factor for cervical carcinoma patients Wang Xi · Wang Rui · Lin Fang · Dong Ke · Gao Ping · Zhang Hui-Zhong

Received: 9 May 2008 / Accepted: 10 November 2008 / Published online: 26 November 2008 © Springer-Verlag 2008

Abstract Purpose Stathmin (Oncoprotein18), a ubiquitous and highly conserved 19-kDa cytosolic phosphoprotein, has been reported to play a critical role in mitosis and possibly other cellular processes, which is associated with tumor carcinogenesis and development. The purpose of this study was to examine the involvement of stathmin in human cervical carcinogenesis and to evaluate its prognostic signiWcance in human cervical carcinoma. Methods Using semiquantitative RT-PCR and Western blotting, we detected the expression of stathmin in human normal cervical epithelial cell line, immortalized cervical epithelial cell lines, and cervical carcinoma cell lines. Additionally, we also detected the expression of stathmin protein in 15 cases of cervical carcinoma tissues and adjacent non-carcinomous margin tissues. Furthermore, specimens from 148 patients with diVerent grade and stage cervical carcinoma were investigated by immunohistochemistry for stathmin expression. Correlations between the expression of stathmin and various clinicopathological factors were studied, while statistical analyses were performed to evaluate prognostic and diagnostic associations. Results The levels of stathmin mRNA and protein expression were signiWcantly higher in cervical carcinoma cells W. Xi and W. Rui contributed equally to this study, should be regarded as joint First authors.

and immortalized cervical epithelial cells than in normal cervical epithelial cells (P < 0.05). Moreover, Western blotting revealed high stathmin protein expression in 73.3% (11/15) cervical carcinoma tissues, while stathmin were overexpressed in tumor tissues as compared with adjacent non-carcinomous margin samples (P = 0.017). In addition, immunohistochemical staining revealed stathmin immunoreactivity in 81.1% (120/148) of cervical carcinoma tissues and high stathmin expression was signiWcantly correlated with clinical stage (P = 0.006), T classiWcation (P = 0.012), regional lymph node metastasis (P = 0.005) and hematogenous metastasis (P = 0.021). Kaplan–Meier analysis showed that high stathmin positivity was signiWcantly associated with a shorter survival time (P < 0.001). Clinical stage (P = 0.0022), T classiWcation (P = 0.0035), regional lymph node (P = 0.0008) or hematogenous metastasis (P = 0.0015) were also associated with survival time. Furthermore, by Cox multivariate analysis, only lymph node (P = 0.0052) or hematogenous metastasis (P = 0.0046) maintained their signiWcance as independent prognostic factors, although stathmin was not an independent prognostic factor (risk ratio: 1.45; P = 0.0624). Conclusions Stathmin expression correlates with cervical carcinogenesis and tumor progression. This molecule is a valuable prognostic marker in patients with cervical carcinoma. Keywords Stathmin · Op18 · Immunohistochemistry · Prognosis · Cervical carcinoma

W. Xi · L. Fang · D. Ke · Z. Hui-Zhong Department of Clinical Diagnosis, Tangdu Hospital, Fourth Military Medical University, Xi’an, China

Introduction W. Rui · G. Ping · Z. Hui-Zhong (&) Laboratory Department, Tangdu Hospital, Fourth Military Medical University, Xinsi Road, 710038 Xi’an, Shaanxi, China e-mail: [email protected]

Cervical carcinoma is the third most common gynecologic malignancy around the world and is typically seen in younger women, often with serious consequences (Jemal

123

838

et al. 2003). Although many clinicopathological parameters including clinical or pathological stage of disease, tumor volume or size, lymphatic and metastatic status, and tumor microvessel intensity have been reported to be associated with prognosis, it is still necessary to search so better prognostic markers to improve clinical management of cervical carcinoma patients (Kawagoe et al. 1999; Martin-Loeches et al. 2002; Gasinska et al. 2002). The advent of genomic and proteomic technologies has made it possible to identify novel biomarkers which can predict the presence and prognosis of cancers. Many studies have revealed diVerent signal pathways during the process of cervical carcinogenesis and development, but no biomarker has yet been identiWed to be obviously related to tumor progression or clinical outcome in cases of cervical carcinoma. Stathmin, also called Oncoprotein18 (Op18), is a founding member of the family of microtubule-destabilizing proteins. It is a ubiquitous, highly conserved 19 kDa cytosolic phosphoprotein that regulates microtubule dynamics during the assembly of the mitotic spindle (Rubin and Atweh 2004). The activity of stathmin itself is regulated by changes in its state of phosphorylation during the transition from interphase to metaphase. The initial clue that stathmin might have a direct role in the regulation of mitosis came from genetic studies that showed that manipulation of stathmin expression interferes with the progression of cells through mitosis, so it may play an important role in the malignant phenotype (Hanash et al. 1988). Many reports have shown that stathmin is overexpressed in a variety of human malignancies such as acute leukemia, breast carcinoma and lung cancer (Melhem et al. 1997; Bièche et al. 1998; Chen et al. 2003). In our previous study, stathmin was also found to be overexpressed in osteosarcoma and ovarian carcinoma cells, and the overexpression of this gene signiWcantly correlated with prognosis of patients (Zhang et al. 2004; Wei et al. 2008). Moreover, we found that small interfering RNA (siRNA) or antisense-mediated downregulation of stathmin expression could lead to proliferation inhibition and chemosensitivity enhancement in human osteosarcoma cells (Zhang et al. 2006; Wang et al. 2007). Therefore, stathmin might be involved in the tumor carcinogenesis, proliferation, metastasis and chemmoresistance, while the high levels of this molecule are signiWcantly correlated with tumor progression and patient outcome (Belletti et al. 2008; McGrogan et al. 2008) . At present, no data are available concerning the correlation between stathmin expression and cervical carcinoma progression. To clarify the prognostic signiWcance and the relationship between common clinicopathological factors and stathmin expression in human cervical carcinoma, the expression of stathmin mRNA and protein was examined in normal cervical epithelial cell line (NCEC), immortalized cervical epithelial cell lines (E6/E7-c or hTERT-c) and

123

J Cancer Res Clin Oncol (2009) 135:837–846

human primary cervical carcinoma cell lines (HeLa, SiHa, Caski and C33A). Moreover, we retrospectively examined 148 primary cervical carcinoma samples by immunohistochemical staining, and we investigated correlations among the expression of stathmin protein and several clinicopathological characteristics, including prognosis.

Materials and methods Cell culture The HeLa, SiHa, Caski and C33A human cervical carcinoma cell lines were preserved in our lab; A normal cervical epithelial cell line (NCEC) was established and preserved by our lab; Two immortalized cervical epithelial cell lines (E6/E7-c or hTERT-c) were established and preserved by our lab. All cell lines were maintained in RPMI 1640 medium (Invitrogen, USA) supplemented with 10% fetal calf serum (FCS). Patients and tissue sources During the period from 2004 to 2007, we collected a total of 148 biopsy samples from cervical carcinoma patients at the Department of Gynaecology and Obstetrics of Tangdu Hospital, Fourth Military Medicine University. All tissue samples were embedded in paraYn. Written informed consent was obtained from the patients to perform surgery and to use resected tissue samples for research. Clinical characters of the samples is described in detail in Table 1. The average age of 148 female patients was 47.4 years old (range 36–64 years). All cases were with no metastasis at its presence at original presentation with cervical carcinoma. Each patient was clinically staged according to the International Federation of Gynecology and Obstetrics (FIGO) staging system. None of the patients had received chemotherapy or radiation therapy. Extraction of total RNA and reverse transcription PCR Total RNA was extracted from cervical carcinoma cell lines, immortalized cervical epithelial cell line and normal cervical epithelial cell line using Simply P Total RNA isolation Kit (BioFlux, HZ, China) according to the manufacturer’s instructions. RNA pellets were dissolved in nuclease-free sterile water, and the concentrations were determined by measuring absorbance at 260 nm. RNA was reverse-transcribed into cDNA in a 20
L reaction system using Superscript First-Strand Synthesis Kit for RT-PCR (Promega Inc) under conditions described by the supplier. One microliter of cDNA was used to amplify the stathmin gene fragment with 1£ PCR buVer, 1.5 mM MgCl2, 200
M of each dNTP, 200
M of primers and 2u Taq

J Cancer Res Clin Oncol (2009) 135:837–846

839

T1/T2

38/54 (25.7/36.5)

T3/T4

36/20 (24.3/13.5)

1 mM phenylmethylsulfonylXuoride, 1
g/mL leupeptin, 1
g/mL pepstatin, 1
g/mL aprotinin, 1% Triton X-100 in 12.5 mM Tris–HCl buVer, pH 7.0) at 4°C for about 30 min. An equal volume of lysate was electrophoresed with SDS polyacrylamide gel electrophoresis (PAGE; 12%). The separated proteins in the gel were transferred to the nitrocellulose membrane. The membranes were blocked with TrisbuVered saline plus 0.1% Tween-20 (TTBS) containing 5% non-fat milk for 1 h, incubated with the appropriate primary antibodies [anti-stathmin (Santa Cruz, USA); anti--actin (Sigma, USA)] in TTBS containing 5% non-fat milk for 2 h, followed by incubation with horseradish peroxidaseconjugated anti-rabbit immunoglobulin antibodies (Sigma, USA) for 45 min. The immunocomplexes was visualized by enhanced chemoluminescence using a ECL™ kit (Amersham, NJ, USA), followed by exposure to X-ray Wlms.

Erosio type

26 (17.6)

Immunostaining of stathmin protein

Exogenesis fungating type

63 (42.6)

Endogeny inWltrating type

32 (21.6)

Ulcer type

27 (18.2)

Table 1 Patient characteristics (2004–2007) Characteristics

Number of patients (%)

Age <55

72 (48.6)

¸55

76 (51.4)

Tumor size <4 cm

66 (44.6)

¸4 cm

82 (55.4)

Clinical stage 0

8 (5.4)

I/II

24/32 (16.2/21.6)

III/IV

53/31 (35.8/21.0)

T classiWcation

Pathological shape

Histological type Squamous cell carcinoma

109 (73.6)

Adenocarcinoma

23 (15.5)

Adeno-squamous carcinoma

16 (10.9)

Regional lymph nodes N0

35 (23.6)

N1/N2/N3

62/34/17 (41.9/23.0/11.5)

Hematogenous metastasis No

102 (68.9)

Yes

46 (31.1)

Vital status(at follow up) Alive

112 (75.7)

Dead of cervical carcinoma

23 (15.5)

Dead of other disease

13 (8.8)

DNA polymerase. We used house-keeping gene (-actin) as an internal marker The PCR primer sequences were designed according to the stathmin and -actin gene sequences reported in GeneBank and chemically synthesized and the primers were as follows: stathmin forward: 5⬘-ATGGCTTCTTCTGATATCCAG-3⬘, reverse: 5⬘-TTAGT CAGCTTCAGTCTCGTC-3⬘(yielding 450 bp);
-actin forward: 5⬘-CTACAATGAGCTGCGTG-3⬘, reverse: 5⬘-GGT CTCAAACATGATC-3⬘(yielding 358 bp). PCR products were visualized by ethidium bromide staining of a 1.5% agarose gel. Western blot analysis Cell and tissue extracts were prepared with lysis buVer (1 mM dithiothreitol, 0.125 mM EDTA, 5% glycerol,

Immunohistochemistry was done to study altered protein expression in 148 human cervical carcinoma tissue specimens. Sections (4
m), cut from the original paraYn blocks, were deparaYnized in xylene and rehydrated in graded alcohols and distilled water. After inhibition of endogenous peroxidase activity for 30 min with methanol containing 0.3% H2O2, the sections were blocked with 10% normal goat serum (Invitrogen, USA) for 20 min and incubated overnight with rabbit anti-human-stathmin polyclonal antibody (diluted 1:300, Santa Cruz Biotechnology, CA, USA) at 4°C. The sections were then incubated with biotinylated anti-rabbit IgG for 30 min at room temperature, followed by incubation with peroxidase-conjugated avidin/ biotin complexes and stained with 3, 3-diaminobenzidine tetrahydrochloride (DAB). Finally, the sections were counter-stained with hematoxylin. Normal rabbit serum was used as a negative control for the staining reactions. Evaluation of immunotaining Stathmin immunoreactivity was evaluated in Wve areas of each slide for correlation and conWrmation of the tissue diagnosis, and the intensity of the immunoreactivity (intensity score) was stratiWed into four categories: no IR (¡); weak IR (+); moderate IR (++); and strong IR (+++). The percentage of positive cells (more than 50% of cells staining positive, 3+; 5–50% of cells staining positive, 2+; less than 5% of cells staining positive, 1+; and no cells staining positive, 0), as previously described. Slides were scored in the absence of any clinical data, and the Wnal immunostaining score reported was the average of three observers. If diVerent grades were assigned, Wnal agreement was obtained after careful review of the images on the same digital monitor screen.

123

840

J Cancer Res Clin Oncol (2009) 135:837–846

Statistical analysis Statistical analyses were performed using SPSS 10.0 statistical software (SPSS, Inc., Chicago, IL, USA). Mann–Whitney U test was used to analyze the relationship between stathmin expression and clinicopathologic characteristics. Survival curves were calculated using the Kaplan–Meier method and diVerences were analyzed using the log rank test. The signiWcance of various variables for survival was analyzed by the Cox proportional hazards model in the multivariate analysis. P < 0.05 was considered to indicate a signiWcant diVerence.

Results Detection of stathmin expression in cell lines by RT-PCR and Western blotting To determine the levels of stathmin mRNA and protein expression in cervical epithelial cell lines, semiquantitative RT-PCR and Western blotting assays were performed in the following cell lines: normal cervical epithelial cells (NCEC), cervical carcinoma cell lines (HeLa, SiHa, Caski and C33A), and two immortalized cervical epithelial cell lines (E6/E7-c and hTERT-c). As shown in Figs. 1 and 2, the levels of stathmin mRNA and protein expression in all four cervical carcinoma cell lines and the immortalized cervical epithelial cell lines were obviously higher than that in the normal cervical epithelial cell line (P < 0.01).

Fig. 2 Western blotting detection of the stathmin protein expression in normal cervical epithelial cell line (1 NCEC), two immortalized cervical epithelial cell lines (2 E6/E7-c, 3 hTERT-c) and cervical carcinoma cell lines (4 HeLa, 5 SiHa, 6 Caski, 7 C33A)

Fig. 3 Western blotting analysis of stathmin protein expression in cervical carcinoma. T: cervical carcinoma tissue samples (T12 and T14: none expression; T2 and T7: lower expression; Other T: higher expression); N: adjacent noncarcinomous tissue samples

Detection of stathmin protein in cervical carcinoma tissues by Western bloting To evaluate the relative content of stathmin in human cervical carcinoma, we also detect the protein levels of stathmin

by Western blotting. Of the 15 cervical carcinoma patients, 11 cases were detected to have the higher expression of stathmin protein and 2 cases were detected to have the lower expression of stathmin protein, the molecular weight of which was 19 kDa. Only two cases were not found to have stathmin protein expressed (Fig. 3). Stathmin were overexpressed in tumour samples as compared with adjacent non-carcinomous margin samples (t = 3.114, P = 0.017). Immunistaining of stathmin protein in cervical carcinoma tissues

Fig. 1 RT-PCR detection of the stathmin mRNA expression in normal cervical epithelial cell line (1 NCEC), two immortalized cervical epithelial cell lines (2 E6/E7-c, 3 hTERT-c) and cervical carcinoma cell lines (4 HeLa, 5 SiHa, 6 Caski, 7 C33A)

123

In the present study, we employed immunohistochemistry to analyze expression and subcelluar localization of stathmin protein in the 148 paraYn-embedded cervical carcinoma tissue samples. Stathmin staining was observed in the cytoplasm of cervical carcinoma cells (Fig. 4b–d). Twenty-eight tumors showed negative expression (¡), 32 tumors showed weak or focal expression (1+), 50 tumors showed moderate expression with focal strong expression

J Cancer Res Clin Oncol (2009) 135:837–846

841

Fig. 4 Expression analysis of stathmin protein by immunohistochemistry. Stathmin expression was mainly localized within cytoplasma, and its expression was observed in cancer cells and areas of metaplasia. However, it was not expressed in normal epithelial cells. a Negative expression(¡) (£200); b weak expression (1+) (£200), c:moderate expression with focal strong expression (2+) (£200); d strong expression (3+) (£200)

(2+), and 38 tumors showed strong expression (3+). No speciWc stathmin staining was observed in normal cervical epithelial cells (Fig. 4a) and in the surrounding stroma cells. Correlation between stathmin protein expression and clinicopathological characters We also evaluated the correlation between the level of stathmin protein intensity and various clinicopathological characters of patients (Table 2). The level of stathmin protein expression was signiWcantly associated with clinical stage (P = 0.006), T classiWcation (P = 0.012), regional lymph node metastasis (P = 0.005) and hematogenous metastasis (P = 0.021). However, there were no obvious correlations between the level of stathmin protein expression and age, tumor size, pathological shape and histological type. Moreover, we combined the patients with stathmin (1+), stathmin (2+) and stathmin (3+) tumors into one group (stathmin positive) and reevaluated the clinical signiWcance of stathmin protein expression (Table 3). As a result, we found that the positivity of stathmin expression was obviously correlated with clinical stage (P = 0.023), T classiWcation (P = 0.009), regional lymph node metastasis (P = 0.004) and hematogenous metastasis (P = 0.013). All these results showed that higher clinical stage and T classiWcation or metastasis correlated with higher level of stathmin protein expression.

Association of stathmin expression with patient survival The level of stathmin protein expression in cervical carcinoma was obviously associated with patients’ survival months (P < 0.001) and the correlation coeYcient was ¡0.32, suggesting that higher levels of stathmin protein expression was associated with shorter survival months. Kaplan–Meier analysis indicated that the prognosis of patients with tumors expressing stathmin protein was signiWcantly worse than that of those with stathmin proteinnegative tumors (Fig. 5). The log-rank test showed that the survival time was obviously diVerent between two groups (log-rank P = 0.022, Wilcoxon P = 0.013). The cumulative 5-year survival rate was 81.3% (95% CI, 0.713–0.901) in the stathmin expression-negative group, however, it was only 55.4% (95% CI, 0.522–0.701) in the stathmin expression-positive group. Moreover, by Kaplan–Meier analysis and log-rank test, we also observed that clinical stage, T classiWcation, regional lymph node metastasis and hematogenous metastasis were also associated with survival time (for clinical stage P = 0.0022; T classiWcation, P = 0.0035; regional lymph node metastasis P = 0.0008; hematogenous metastasis, P = 0.0015). Univariate and multivariate analysis for prognosis of patients with cervical carcinoma In univariate survival analysis, all clinicopathological factors were used to evaluate their role in prognosis.

123

842 Table 2 Relationship between stathmin intensity and clinicopathologic characteristics in all patients

J Cancer Res Clin Oncol (2009) 135:837–846

Characteristics

Stathmin (¡)

Stathmin (1+)

Stathmin (2+)

Stathmin (3+)

(n = 28)

(n = 32)

(n = 50)

(n = 38)

<55 year

15

22

24

11

¸ 55 year

13

10

26

27

<4 cm

16

8

28

14

¸4 cm

12

24

22

24

0

2

3

1

2

I

5

9

6

4

II

8

10

8

6

III

8

6

23

16

IV

5

4

12

10

T1

7

6

14

11

T2

10

8

18

18

T3

6

12

12

6

T4

5

6

6

3

6

8

7

5

Exogenesis fungating type

12

16

23

12

Endogeny inWltrating type

6

5

12

9

Ulcer type

4

3

8

12

16

22

39

32

P

Age 0.21

Tumor size 0.15

Clinical stage 0.006

T classiWcation 0.012

Pathological shape Erosio type

0.32

Histological type Squamous cell carcinoma Adenocarcinoma

4

8

7

4

Adeno-squamous carcinoma

8

2

4

2

N0

12

5

13

5

N1

6

10

24

22

N2

7

12

7

8

N3

3

5

6

3

No

18

20

34

30

Yes

10

12

16

8

0.08

Regional lymph nodes 0.005

Hematogenous metastasis

¡ Negative, + Positive

From data shown in Table 4, we observed that clinical stage, T classiWcation, regional lymph node metastasis, hematogenous metastasis and stathmin score, were independent prognostic factors inXuencing both disease-free survival (DFS) and overall survival (OS) rates. In multivariate survival analysis by the Cox model, only lymph node or hematogenous metastasis maintained their signiWcance as independent prognostic factors for both DFS and OS rates. Moreover, by Cox multivariate analysis, we also concluded that stathmin was not an independent prognostic factor (risk ratio: 1.45; P = 0.0624) (Table 5).

123

0.021

Discussion Cervical carcinoma remains the second most common cancer in women worldwide and the leading cancer in developing countries, where 83% of cases occur (Parkin et al. 2005). Many diagnostic markers and markers for disease follow-up have been identiWed and investigated for cervical carcinoma, but a useful screening marker for cervical carcinoma has not been clearly established up to now. Tumor markers in cervical carcinoma are related to histology, with SCC being the most sensitive marker in squamous tumors and CEA in adenocarcinoma, while the sensitivity for

J Cancer Res Clin Oncol (2009) 135:837–846 Table 3 Correlations of stathmin expression and clinicopathological characteristics in all patients

Characteristics

843

Total (n = 148)

Stathmin

P value

Negative (n = 28)

Positive (n = 120)

n%

n%

Age <55 year

72

15 (20.8)

57 (79.2)

¸55 year

76

13 (17.1)

63 (82.9)

<4 cm

66

16 (24.2)

50 (75.8)

¸4 cm

82

12 (14.6)

70 (85.4)

0

8

2 (25.0)

6 (75.0)

I

24

5 (20.8)

19 (79.2)

II

32

8 (25.0)

24 (75.0)

0.568a

Tumor size 0.139a

Clinical stage

III

53

8 (15.1)

45 (84.9)

IV

31

5 (16.1)

26 (83.9)

0.023

T classiWcation T1

38

7 (18.4)

31 (81.6)

T2

54

10 (18.5)

44 (81.5)

T3

36

6 (16.7)

30 (83.3)

T4

20

5 (25.0)

15 (75.0)

0.009

Pathological shape Pathological shape

26

6 (23.1)

14 (76.9)

Exogenesis fungating type

63

12 (19.0)

51 (81.0)

Endogeny inWltrating type

32

6 (18.8)

26 (81.2)

Ulcer type

27

4 (14.8)

23 (85.2)

0.274

Histological type Squamous cell carcinoma

109

16 (14.7)

16 (14.7)

Adenocarcinoma

23

4 (17.3)

19 (82.7)

Adeno-squamous carcinoma

16

8 (50.0)

8 (50.0)

N0

35

12 (34.2)

23 (65.8)

N1

62

6 (9.7)

56 (91.3)

N2

34

7 (20.1)

27 (79.9)

N3

17

3 (17.6)

14 (82.4)

No

102

18 (17.6)

84 (82.4)

Yes

46

10 (21.7)

36 (78.3)

0.451

Regional lymph nodes 0.004

Hematogenous metastasis a

Spearman rank correlation test P value

early-stage disease before clinical detection remains questionable (Scambia et al. 1994; Duk et al. 1996). Thus, the identiWcation of new molecular prognostic and predictive markers in cervical carcinoma may help to exactly evaluate the prognosis of patients and further stratify patients into diVerent risk groups for whom speciWc adjuvant therapies may be appropriate, thereby increasing survival time. Stathmin (Op18) was Wrstly identiWed as a 17-kDa cytosolic phosphoprotein that is expressed at moderately high levels in human solid tumors and at very high levels in leukemias and lymphomas (Ghosh et al. 1993). It is a founding

0.013

member of a family of microtubule-destabilizing proteins that play critical roles in the regulation of mitosis, which is a major cellular substrate for p34cdc2 kinase, mitogen-activated protein kinase, and other kinases that are important for cellular proliferation and diVerentiation (Luo et al. 1994; Marklund et al. 1994). Additionally, stathmin is a conserved cytosolic protein that has been studied in various cellular systems such as oncoprotein 18, prosolin, p19, 19 K, p18, and op18 (Belmont and Mitchison 1996; Laird and Shalloway 1997). Considerable researches have shown that stathmin plays a critical role in mitosis and possibly

123

844

Fig. 5 Survival of patients stratiWed by stathmin protein expression. The prognosis of patients with stathmin-positive tumors was signiWcantly worse than that of patients with stathmin-negative tumors (log rank test, P = 0.022; Wilcoxon test, P = 0.013). Moreover, the level of stathmin expression was obviously associated with the prognosis among the patients with stathmin-positive tumors (1+, 2+ and 3+)

other cellular processes. High levels of stathmin expression have been reported in various cancer cell lines and cancer tissues, including oral squamous-cell carcinoma, ovarian carcinoma, breast carcinoma and prostate cancers, and are found to be associated with a more aggressive behavior (Koike et al. 2005; Price et al. 2000; Brattsand 2000; Friedrich et al. 1995). Furthermore, stathmin overexpression is suYcient to induce tumorigenesis, tumor invasion and metastasis, and chemoresistance. Takahashi et al. showed that stathmin, which is involved in the microtubule system, was highly expressed in high-stage Wilms’ tumours (Takahashi et al. 2002). Kouzu et al. also reported that stathmin was overexpressed in oral squamous-cell carcinoma: correlation with tumour progression and poor prognosis (Kouzu1 et al. 2006). However, stathmin expression has not been studied previously in human cervical carcinoma and the actual role of stathmin remains unclear. Here, we Wrstly reported that stathmin was overexpressed at both mRNA and protein levels in cervical carcinoma cells. Moreover, we also found that stathmin protein expression was positive in 81.1% of cervical carcinoma tissues, while the level of stathmin protein expression was signiWcantly correlated with progression of primary cervical carcinoma and the prognosis of cervical carcinoma patients. In the present study, we Wrstly showed that the level of stathmin expression was signiWcantly higher in cervical carcinoma cell lines compared with normal cervical epithelial cell line at both transcriptional and translational levels. Intriguely, both of the two immortalized cell lines (E6/E7-c and hTERT-c) also showed a higher level of stathmin mRNA and protein expression. Followingly, we evaluated the status of stathmin protein expression in cervical carcinoma

123

J Cancer Res Clin Oncol (2009) 135:837–846

tissues by Western blotting and immunostaining. Western blotting showed that stathmin was more frequently and abundantly expressed in cervical carcinoma than in noncarcinomous margin tissues. Immunohistochemical analysis conWrmed that many tumor cells exhibited moderate to strong stathmin staining mainly in the cytoplasm, whereas positive stathmin staining was not observed in normal cervical tissues. Considering the oncogenic properties of the stathmin proteins and their ability to promote proliferation in diVerent cell types in vitro, the high expression of stathmin gene observed in cervical carcinoma supports its involvement in the progression of human cervical carcioma in vivo. Next, we analyzed the correlation between stathmin expression and clinical characteristics of cervical carcinoma patients. By statistical analysis, we found that immunohistochemical stathmin positivity was signiWcantly associated with clinical stage (P = 0.006), T classiWcation (P = 0.012), regional lymph node metastasis (P = 0.005) and hematogenous metastasis (P = 0.021), while we also found that there were no obvious correlations between stathmin expression and other clinical factors including age, tumor size, pathological shape and histological type (P > 0.05). All these results strongly suggested that stathmin may not only have a role in promoting growth of the primary tumor, but also in the development of lymph node or hematogenous metastasis. The development process of cervical carcinoma usually includes cervical epithelial dysplasia (CIN), carcinoma in situ and inWltrating carcinoma. Approximately 50% lowgrade cervical epithelial dysplasias naturally subside, and only less than 2% of them develop into carcinoma in situ in the end (Duggan 2002). It is also reported that at least 20% cervical epithelial dysplasia III (CIN III) might develop into inWltrating carcinoma during the next ten years (Arends et al. 1998). Development of human cervical carcinoma is a result of the accumulation of multiple genetic alterations leading to the evolution of clonal outgrowth and malignant transformation for a long time (Huang et al. 2005). In our study, we observed that two immortalized cervical epithelial cells which were induced by HPV16 oncogenic E6/E7 gene and the telomerase gene (hTERT) showed higher levels of stathmin expression compared with normal cervical epithelial cell line. Cell immortalization is the early stage of malignant transformation, so we concluded that the upregulated levels of stathmin might contribute to early malignant transformation of human cervical epithelial cells. Furthermore, the present study indicated that the survival time in patients with stathmin-positive tumors was far lower than that in patients with stathminnegative tumors. By multivariate analysis, we demonstrated the role of stathmin along with other clinicopathological factors as independent prognostic factors inXuencing both disease-free survival (DFS) and overall survival (OS) rates,

J Cancer Res Clin Oncol (2009) 135:837–846 Table 4 Univariate analysis of clinicopathological characteristics for DFS and OS of 148 patients with cervical carcinoma

845

Characteristics

No. of patients

DFS rate (%)

P value

NS

OS rate (%)

P value

76.5

NS

Age 1. <55

72

81.2

2. ¸55

76

78.3

70.8

Tumor size 1. <4 cm

66

74.5

2. ¸4 cm

82

69.8

NS

75.2

1. 0–II

64

87.8

2. III–IV

84

43.3

1. T1–T2

92

77.6

2. T3–T4

56

46.5

26

93.4

2. Exogenesis fungating type

63

90.6

76.9

3. Endogeny inWltrating type

32

88.7

84.4

4. Ulcer type

27

80.5

79.8

109

93.3

2. Adenocarcinoma

23

83.1

76.5

3. Adeno-squamous carcinoma

16

79.2

72.3

35

84.4

113

45.9

NS

72.4

Clinical stage 0.0016

83.2

0.0028

39.8

T stage 0.0034

70.5

0.0012

41.2

Pathological shape 1. Erosio type

NSa

89.6

NSa

Histological type 1. Squamous cell carcinoma

NSa

86.3

NSa

Regional lymph nodes 1. N0 2. N1–N3

0.0008

79.7

0.0015

41.5

Hematogenous metastasis 1. No

102

88.9

2. Yes

46

42.2

1. Negative

28

81.3

2. Positive

120

55.4

0.0022

83.2

0.0031

39.3

Stathmin expression a

For 1 versus 2–4, 1–2 versus 3–4, or 1–3 versus 4

Table 5 Results of Cox multivariate analysis Characteristics

Risk ratio

95% CI

P value

Clinical stage

3.02

1.13–5.57

0.0017

T stage

2.21

1.34–6.32

0.0028

Regional lymph nodes

2.89

1.77–6.85

0.0002

Hematogenous metastasis

1.37

0.98–3.66

0.0006

Stathmin-positive

1.45

1.02–2.64

0.0024

CI conWdence interval

but the results of Cox multivariate analysis showed that this molecule was not an independent prognostic factor (P = 0.0624). As the patient number was relatively smaller, we could not have a clear look on the prognostic relevance of stathmin expression for cervical carcinoma patients as an independent factor. However, we concluded that stathmin

0.012

79.4

0.026

41.2

in combination with the other clinicopathological factors could be a valuable tool in stratiWcation and prognosis in cervical carcinoma patients. In summary, we Wrstly reported the expression of stathmin gene in cervical carcinoma cells as well as specimens, suggesting the clinical signiWcance of stathmin in cervical carcinoma. Stronger stathmin expression is obviously correlated with more aggressive behavior of cervical carcinoma, and the higher levels of stathmin expression is signiWcantly correlated with a shorter survival time and poorer prognosis. Thus, stathmin may be a valuable prognostic factor in patients with cervical carcinoma and stathmin plays its role as a useful marker of cervical carcioma initiation, development, progression and metastasis. Further investigations should focus on whether stathmin protein may be used as a therapeutic target for the treatment of human cervical carcinoma.

123

846 Acknowledgment Thanks to every one of the Department of Clinical Laboratory for their sincere help and excellent technical assitance.

References Arends MJ, Buckley CH, Wells M (1998) Aetiology, pathogenesis, and pathology of cervical neoplasia. J Clin Pathol 51:96–103 Belletti B, Nicoloso MS, Schiappacassi M, Berton S, Lovat F, Wolf K, Canzonieri V, D’Andrea S, Zucchetto A, Friedl P, Colombatti A, Baldassarre G (2008) Stathmin activity inXuences sarcoma cell shape, motility, and metastatic potential. Mol Biol Cell 19:2003– 2013. doi:10.1091/mbc.E07-09-0894 Belmont LD, Mitchison TJ (1996) IdentiWcation of a protein that interacts with tubulin dimers and increases the catastrophe rate of microtubules. Cell 84:623–631. doi:10.1016/S0092-8674(00) 81037-5 Bièche I, Lachkar S, Becette V, Cifuentes-Diaz C, Sobel A, Lidereau R, Curmi PA (1998) Overexpression of the stathmin gene in a subset of human breast cancer. Br J Cancer 78:701–719 Brattsand G (2000) Correlation of oncoprotein 18/stathmin expression in human breast cancer with established prognostic factors. Br J Cancer 83:311–318. doi:10.1054/bjoc.2000.1264 Chen G, Wang H, Miller CT, Thomas DG, Gharib TG, Misek DE, Giordano TJ, Orringer MB, Hanash SM, Beer DG (2003) Overexpression of oncoprotein 18 correlates with poor diVerentiation in lung adenocarcinomas. Mol Cell Proteomics 2:107–116. doi:10.1074/mcp.M200055-MCP200 Duggan MA (2002) A review of the natural history of cervical intraepithelial neoplasia. Gan To Kagaku Ryoho 29:176–193 Duk JM, Groenier KH, de Bruijn HW, Hollema H, ten Hoor KA, van der Zee AG, Aalders JG (1996) Pretreatment serum squamous cell carcinoma antigen: a newly identiWed prognostic factor in early-stage cervical carcinoma. J Clin Oncol 14:111–118 Friedrich B, Grönberg H, Landström M, Gullberg M, Bergh A (1995) DiVerentiation stage speciWc expression of oncoprotein 18 in human and rat prostatic adenocarcinoma. Prostate 27:102–109. doi:10.1002/pros.2990270207 Gasinska A, Urbanski K, Adamczyk A, Pudelek J, Lind BK, Brahme A (2002) Prognostic signiWcance of intratumour microvessel density and haemoglobin level in carcinoma of the uterine cervix. Acta Oncol 41:437–443. doi:10.1080/028418602320405023 Ghosh PK, Anderson J, Cohen N, Takeshita K, Atweh GF, Lebowitz P (1993) Expression of the leukemia-associated gene, p18, in normal and malignant tissues: inactivation of expression in a patient with cleaved B cell lymphoma/leukemia. Oncogene 8:2869–2872 Hanash SM, Strahler JR, Kuick R, Chu EH, Nichols D (1988) IdentiWcation of a polypeptide associated with the malignant phenotype in acute leukemia. J Biol Chem 263:12813–12815 Huang FY, Kwok YK, Lau ET, Tang MH, Ng TY, Ngan HY (2005) Genetic abnormalities and HPV status in cervical and vulvar squamous cell carcinomas. Cancer Genet Cytogenet 157:42–48. doi:10.1016/j.cancergencyto.2004.06.002 Jemal A, Murray T, Samuels A, Ghafoor A, Ward E, Thun MJ (2003) Cancer statistics, 2003. CA cancer. J Clin 53:5–26 Kawagoe T, Kashimura M, Matsuura Y, Sugihara K, Toki N, Aoki T (1999) Clinical signiWcance of tumor size in stage IB and II carcinoma of the uterine cervix. Int J Gynecol Cancer 9:421–426. doi:10.1046/j.1525-1438.1999.99059.x Koike H, Uzawa K, Nakashima D, Shimada K, Kato Y, Higo M, Kouzu Y, Endo Y, Kasamatsu A, Tanzawa H (2005) IdentiWcation

123

J Cancer Res Clin Oncol (2009) 135:837–846 of diVerentially expressed proteins in oral squamous-cell carcinoma using a global proteomic approach. Int J Oncol 27:59–67 Kouzu Y, Uzawa K, Koike H, Saito K, Nakashima D, Higo M, Endo Y, Kasamatsu A, Shiiba M, Bukawa H, Yokoe H, Tanzawa H, Kouzu Y, Uzawa K, Koike H (2006) Overexpression of stathmin in oral squamous-cell carcinoma: correlation with tumour progression and poor prognosis. Br J Cancer 94:717–723 Laird AD, Shalloway D (1997) Oncoprotein signalling and mitosis. Cell Signal 9:249–255. doi:10.1016/S0898-6568(96)00176-3 Luo XN, Mookerjee B, Ferrari A, Mistry S, Atweh GF (1994) Regulation of phosphoprotein p18 in leukemic cells. Cell cycle regulated phosphorylation by p34cdc2 kinase. J Biol Chem 269:10312– 10318 Marklund U, Osterman O, Melander H, Bergh A, Gullberg M (1994) The phenotype of a “Cdc2 kinase target site-deWcient” mutant of oncoprotein18 reveals a role of this protein in cell cycle control. J Biol Chem 269:30626–30635 Martin-Loeches M, Ortí RM, Cazorla E, Asins E, Llixiona J (2002) Multivariate analysis of the morphometric characteristics of tumours as prognostic factors in the survival of patients with uterine cervix cancer treated with radical surgery. Eur J Obstet Gynecol Reprod Biol 105:170–176 McGrogan BT, Gilmartin B, Carney DN, McCann A (2008) Taxanes, microtubules and chemoresistant breast cancer. Biochim Biophys Acta 1785:96–132 Melhem R, Hailat N, Kuick R, Hanash SM (1997) Quantitative analysis of Op18 phosphorylation in childhood acute leukemia. Leukemia 11:1690–1695. doi:10.1038/sj.leu.2400792 Parkin DM, Bray F, Ferlay J, Pisani P (2005) Global cancer statistics, 2002. CA Cancer J Clin 55:74–108 Price DK, Ball JR, Bahrani-Mostafavi Z, Vachris JC, Kaufman JS, Naumann RW, Higgins RV, Hall JB (2000) The phosphoprotein Op18/stathmin is diVerentially expressed in ovarian cancer. Cancer Invest 18:722–730. doi:10.3109/07357900009012204 Rubin CI, Atweh GF (2004) The role of stathmin in the regulation of the cell cycle. J Cell Biochem 93:242–250. doi:10.1002/jcb. 20187 Scambia G, Benedetti Panici P, Foti E, Amoroso M, Salerno G, Ferrandina G, Battaglia F, Greggi S, De Gaetano A, Puglia G (1994) Squamous cell carcinoma antigen: prognostic signiWcance and role in the monitoring of neoadjuvant chemotherapy response in cervical cancer. J Clin Oncol 12:2309–2316 Takahashi M, Yang XJ, Lavery TT, Furge KA, Williams BO, Tretiakova M, Montag A, Vogelzang NJ, Re GG, Garvin AJ, Söderhäll S, Kagawa S, Hazel-Martin D, Nordenskjold A, Teh BT (2002) Gene expression proWling of favorable histology Wilms tumors and its correlation with clinical features. Cancer Res 62:6598–6605 Wang R, Dong K, Lin F, Wang X, Gao P, Wei SH, Cheng SY, Zhang HZ (2007) Inhibiting proliferation and enhancing chemosensitivity to taxanes in osteosarcoma cells by RNA interference-mediated downregulation of stathmin expression. Mol Med 13:567–575. doi:10.2119/2007-00046.Wang Wei SH, Lin F, Wang X, Gao P, Zhang HZ (2008) Prognostic signiWcance of stathmin expression in correlation with metastasis and clinicopathological characteristics in human ovarian carcinoma. Acta Histochem 110:59–65. doi:10.1016/j.acthis.2007.06.002 Zhang HZ, Gao P, Yan L, Lin F (2004) SigniWcance of stathmin gene overexpression in osteosarcoma cells. Ai Zheng 23:493–496 Zhang HZ, Wang Y, Gao P, Lin F, Liu L, Yu B, Ren JH, Zhao H, Wang R (2006) Silencing stathmin gene expression by survivin promoter-driven siRNA vector to reverse malignant phenotype of tumor cells. Cancer Biol Ther 5:1457–1461