Expression and clinicopathological significance of FSIP1 in breast cancer.

AIM
To investigate the clinicopathological significance of the expression of fibrous sheath interacting protein 1 (FSIP1) in breast cancer, serum samples, and wound fluid from patients with breast cancer.


METHODS
Wound fluid and serum samples from female patients with primary breast cancer, recurrent and metastatic breast cancer, and benign tumors were analyzed for FSIP1 expression using ELISA. 286 paraffin-embedded surgical specimens from breast cancer patients with at least 5 years of follow-up were included for FSIP1 expression assay using immunohistochemistry.


RESULTS
Expression of FSIP1 protein was significantly higher in breast cancer tissues compared to tumor-adjacent tissues (p = 0.001). Strong correlation was observed between FSIP1 expression and human epidermal growth factor receptor 2 (Her-2) or Ki67 expression in breast cancer (p = 0.027 and 0.002, respectively). Similarly, serum level of FSIP1 was higher in patients with recurrent and metastatic breast cancer compared to that of primary breast cancer (7, 713 ± 3, 065 vs. 4, 713 ± 3, 065 pg/ml, p = 0.003). Finally, patients with high FSIP1 expression showed a worse post-operative disease-specific survival (p = 0.024).


CONCLUSION
FSIP1 may play an important role in the tumorigenesis and invasion of breast cancer and is a potential biomarker for breast cancer diagnosis or prognosis.


BACKGROUND
Currently, the primary treatment for breast cancer is surgery followed by chemotherapy, radiotherapy or endocrine therapy, while targeted treatments are employed to eliminate residual tumor cells and thus reduce the risk of recurrence and metastasis [1][2][3]. Some patients, however, still show relapse or metastasis after postoperative therapy. The reason that post-operative therapies failed or not all patients responded to targeted therapy remains elusive. Therefore, it is urgent to identify novel biomarkers that can discriminate these refractory patients. More importantly, it is crucial to discover new therapeutic targets with high specificity via further understanding on the molecular mechanisms underlying tumorigenesis and metastasis of breast cancer [4,5].
FSIP1 is a recently discovered gene that encodes fibrous sheath interacting protein 1 (FSIP1). Expression of FSIP1 is known to be regulated by amyloid beta precursor protein [6]. FSIP1 is a potential target for cancer therapy since its mRNA level is undetectable in most normal tissues and its expression is elevated in breast tumors. However, the previous study only included a small sample size and did not correlate FSIP1 expression level with prognosis [7]. Therefore, further study with large sample size is required to clarify the role of FSIP1 in breast cancer.
The aim of this study was to investigate the protein expression of FSIP1 in breast cancer, and to build up the correlation between FSIP1 expression and the clinicopathological features and prognosis of breast cancer.

FSIP1 expression in breast cancer and its correlation with clinicopathological characteristics
In total, 45.45% of the cases showed high FSIP1 expression in breast tumor tissue with no expression in tumor-adjacent tissues (p = 0.001; Figure 1 and Table 1). FSIP1 protein was expressed at higher levels in human epidermal growth factor receptor 2 (Her-2) positive breast cancer tissues compared to Her-2 negative tissues (p = 0.029). Similarly, FSIP1 expression level was considerably higher in samples with more than 14% Ki67 expression compared to those with less than 14% (p = 0.002). No correlation between FSIP1 expression and age, tumor size, estrogen receptor (ER) status, or progesterone receptor (PR) status was observed (Table 1). Spearman correlation analysis revealed strong correlations between lymph node metastasis, Her-2 and Ki67 expression status and FSIP1 expression (p = 0.009, 0.027 and 0.002, respectively; Table 2).
We also checked the expression level of FSIP1 in four typical breast cancer cell lines, including MCF-7 (luminal A), BT-474 (luminal B), MD-231 (triple negative), and SK-BR3 (Her-2 over expression). Intracellular FSIP1 protein level was apparently higher in the SK-BR3 cell line than in the other three cell lines (Figure 3), which was consistent with the observation on immunohistochemistry (Figure 1 and Table 1) and Spearman correlation regression analysis ( Table 2).

Level of FSIP1 in wound fluid of breast cancer patients after surgery
No difference in the levels of FSIP1 between the wound fluid and serum sample of patients with primary breast cancer was observed (4, 613 ± 3, 612 pg/ml vs. 4, 713 ± 3, 065 pg/ml, p = 0.0613) ( Figure 4A). However, FSIP1 expression in the wound fluid from patients with negative lymph nodes was significantly higher than in wound fluid from patients with positive lymph nodes (4, 937 ± 2, 914 pg/ml vs. 3, 273 ± 2, 647 pg/ml, p = 0.0384; Figure 4B).

Prognosis analysis
Patients with high FSIP1 expression in tumors tended to have worse post-operative disease-specific survival (p = 0.022; Figure 5A). When the data were analyzed according to the expression status of ER, PR, Her-2, and Ki67 in each tumor ( Figure 5B-5E), significant survival differences were observed between FSIP1positive status and FSIP1-negative status in patients with ER-positive and Her-2 negative tumors (p = 0.016 and 0.009, respectively; Figures 5F and 5G). The hazard ratio for death was 1.578 (95% CI, 1.062-2.345; p = 0.024) The p value was calculated using chi-square test or fisher's extract test.

DISCUSSION
FSIP1 is a component of fibrous sheath in sperm flagellum that assembles AKAP4, which was the original X-linked CT antigen detected in breast cancer [9]. It is well-known that protein kinase A (PKA) plays an     important role in tumor proliferation, angiogenesis, and chemoresistance [10][11][12][13]. AKAP4 has been reported to be one of the scaffolding proteins associated with cAMPdependent PKA [10]. Multiple studies have shown that AKAP4 is strongly expressed in several types of cancer [9,14]. As a component of AKAP4, FSIP1 may play a role in tumorigenesis and could therefore be a target for cancer immunotherapy. It has also been shown that FSIP1 functions in the regulation of chromosome segregation in tumor cells [15]. In addition, there is evidence to support that FSIP1 is a target of steroid receptor coactivator-3 [16], which is an oncogene associated with breast cancer [17] and a coactivator for nuclear receptors, such as ER-α [18]. However, the expression levels and clinical implications of FSIP1 expression in breast cancer and especially in the serum and wound fluid were still unclear.
In the present study, a cohort of 286 breast cancer samples was assayed for FSIP1 expression. The results indicated that FSIP1expression was significantly higher in breast cancer tissues compared to benign tissues, and FSIP1 expression in breast cancer was found to be correlated with a worse post-operative disease-specific survival. Moreover, FSIP1 expression was significantly correlated to Her-2 and Ki-67 expression but not to ER or PR level. However, Chapman et al reported higher FSIP1 expression in ER-positive breast tumors compared to ER-negative breast tumors [7], such conflict may due to different methods for FSIP1 quantification, number of samples verified with FSIP1 protein expression or different ethnic origins. We tested the level of FSIP1 protein in breast cancer cell lines in order to confirm the outcomes of the clinical data. The Her-2 positive cell line SK-BR3 expressed higher level of FSIP1 compared to the other cell lines including MCF-7 (an ER positive cell line). We are now studying the function of FSIP1 in Her-2 positive breast cancers.
We also quantified FSIP1 expression in serum and wound fluid to determine whether FSIP1 could be secreted into the wound or blood, which may have an impact on the dissemination of residual tumor cells after surgery. We observed that FSIP1 was highly expressed in the serum of recurrent and metastatic breast cancer compared to primary breast cancer. Furthermore, FSIP1 expression level in the tumor significantly predicted distant metastasis in prognosis analysis. Further functional study is needed to address how FSIP1 might regulate tumor metastasis.

Patients and samples
Blood serum samples from 122 female patients with primary breast cancer (mean age: 52.5 ± 8.3 years), 112 patients with recurrent and metastatic breast cancer (mean age: 56.3 ± 11.6 years ) and 38 patients with benign tumor (mean age: 37.2 ± 10.8 years) were included in this study. Serum samples were collected within 1 week before surgery and 3 days after surgery, wound fluid was collected 2 and 3 days after surgery. For the 122 patients with primary breast cancer, no mastectomy, breastconserving surgery or systemic treatment for breast cancer was performed before undergoing primary breast cancer surgery at the 2nd affiliated hospital of Dalian Medical University between 2011 and 2013. In addition, 286 paraffin-embedded breast cancer tissues from patients with at least a 5-year follow-up were assayed for FSIP1 protein expression using immunohistochemistry and included in prognosis analysis.
The diagnosis of all patients met the criteria of modified National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology-Breast Cancer Guideline 2012. The present study was approved by the ethics committee of the 2nd affiliated hospital of Dalian Medical University, according to the Declaration of Helsinki. All individuals provided written consent for participation in the study.

Assay for FSIP1 level in wound fluid and serum
FSIP1 level in wound fluid and serum was measured using an enzyme-linked immunosorbent assay (ELISA) in accordance with the manufacturer's recommendation (R&D Systems, USA).

Immunohistochemistry
Procedure was the same as previously described with minor modifications [8]. Briefly, 4-μm breast tumor tissues were cut using a cryostat. Sections were mounted on microscope slides, fixed in a mixture of 50% acetone and 50% methanol after air dry. Samples were then dewaxed in xylene, gradually hydrated with gradient alcohol, and washed with phosphate buffered saline (PBS). After that, sections were incubated for 60 min with rabbit polyclonal FSIP1 antibody (1:500 dilution, Santa Cruz Biotechnology, USA). Following PBS wash, sections were further incubated for 30 min with the secondary biotinylated antibody (Multilink swine anti-goat/mouse/ rabbit immunoglobulin; Dako Inc., Denmark). Next, an avidin biotin complex (1:1000 dilution, Vector Laboratories Ltd., United Kingdom) was applied to the sections and incubated for 30-60 min at room temperature. The immunoreactive products were visualized by catalysis of 3, 3-diaminobenzidine with horseradish peroxidase (HRP) in the presence of H 2 O 2 . Last, sections were counterstained with Gill's hematoxylin and dehydrated in ascending grades of methanol, before clearing in xylene and mounting under a coverslip. As a negative control, staining was performed in parallel without primary antibody.
Reactivity of anti-FSIP1 antibody was showed as brown granules located at tumor cell membrane/ cytoplasm and graded as follows: 0, no staining; 1, partial staining of the membrane/cytoplasm; 2, mild to moderate circumferential staining of the membrane/cytoplasm; and 3, strong circumferential staining of the membrane/ cytoplasm. A score of 2 or 3 was considered positive for FSIP1 expression.

Western blot analysis
Total proteins were extracted using a protein extraction kit (ProMab, USA) followed by centrifugation. Protein concentration was quantified using BCA assay (Santa Cruz Biotechnology, USA), individual cell lysate (30 μg/lane) was then separated on sodium dodecyl sulfate polyacrylamide gel and transferred onto polyvinylidene fluoride membranes. Membrane was blocked with 5% fat-free dry-milk in TBST and incubated with rabbit anti-FSIP1 antibody (1:100 dilution; Abcam, USA) followed by rabbit antiβ-actin antibody (1:5000 dilution; Abcam) at 4°C overnight. Bound antibodies were detected with HRP-conjugated antirabbit, anti-mouse, or anti-goat immunoglobulin G (IgG) at room temperature for 1 h and visualized with enhanced chemiluminescence (Santa Cruz Biotechnology, USA). Purified mouse, rabbit, or goat IgG was used as a negative control. Relative levels of targeting protein to the control β-actin were determined using ImmuNe software.

Statistical analysis
Statistics was performed with SPSS Statistics software, version 16.0. Experimental data are presented as mean ± standard error. Continuous variables from the study were analyzed using ANOVA and/or Student t-test (with a parametric distribution) or Mann-Whitney U test (with a nonparametric distribution). Spearman correlation coefficient was applied to test for correlations between two variables. Survival analysis was performed using Kaplan-Meier method and comparisons were made by using logrank test. Hazard ratios and corresponding 95% confidence intervals (CIs) were calculated using Cox proportional hazards model. Statistic significance was set at p < 0.05.

CONCLUSION & FUTURE PERSPECTIVES
Our data support that FSIP1 is a potential biomarker for early diagnosis and prognosis of breast cancer, together with previous reports, FSIP1 maybe a novel target for breast cancer therapy. So far, little is known about the role of FSIP1 in cancer development, we are working on defining the mechanism by which FSIP1 contributes to tumorigenesis and metastasis.