SSH1 expression is associated with gastric cancer progression and predicts a poor prognosis

Gastric cancer (GC) is a common type of malignancy and the second leading cause of cancer-related death in China [1]. Unfortunately, the incidence of GC has been increasing in recent years; there were 1313 thousand incident cases and 819 thousand deaths in 2015 [2]. Therapeutic approaches to GC include surgical treatment, chemotherapy, radiotherapy, and gene therapy. However, the prognosis for GC patients remains poor [3],patients with early-stage GC have nonspecific symptoms, and lymph node invasion or distant metastasis has already occurred by the time they are diagnosed [4]. Therefore, it is important to identify biomarkers that are related to early diagnosis and prognosis.

Cancer cell migration and invasion lead to GC metastasis, which accounts for most cancer-related deaths. The metastasis of GC is a complex, multistep pathological process that involves several signalling pathways. Emerging evidence has shown that the balance between protein phosphorylation and dephosphorylation is crucial in the regulation of cell signalling, and failure to maintain an appropriate balance is known to play a critical role in cancer development [4].

In mammalian cells, SSH phosphatases are encoded by three genes (SSH1, SSH2, and SSH3) that encode members of a family of serine/threonine protein kinases that dephosphorylate the phospho-serine residue of cofilin [5, 6] and reactivate cofilin through phosphorylation at serine-3 [7]. SSH1 was found to be overexpressed in pancreatic cancer (PC) and to contribute to tumour cell migration [8]. Recent studies have demonstrated that SSH1 is a key regulator of vascular smooth muscle cell migration [9]. When cells are stimulated by growth factors or chemokines, SSH1 accumulates in the F-actin-rich lamellipodium and likely plays a critical role in polarized cell migration by maintaining the local activation of cofilin and the rapid turnover of actin filaments at the leading edge of the migrating cell [10]. In addition, there is evidence suggesting that SSH1 plays an important role in cancer development. However, there is currently no evidence suggesting that SSH1 expression correlates with GC prognosis. Thus, studies are needed to determine whether SSH1 is a prognostic biomarker for GC. We designed a tissue microarray (TMA) covering 100 patients with 8 years of follow up. We performed IHC analysis to investigate the correlation between SSH1 expression and clinico-pathological characteristics and to determine whether SSH1 can serve as a prognostic marker.

This study was approved by the hospital ethics committee, and written informed consent was received from the patients. We used a retrospective design to analyse GC tissue samples and all patients with GC had pathologically proven disease and underwent surgical operation between July 2006 and April 2007. All patients were followed up for 19–105 months (average, 49.74 months). The GC tissue and adjacent-to-carcinoma TMAs with a total of 100 cases of GC and 80 cases of adjacent-to-carcinoma paraffin-embedded tissue specimens were purchased from the National Engineering Center for Bio Chips, Shanghai, People’s Republic of China. Complete histological data for the patients with GC were available.

The patients’ clinical characteristics, such as gender, age, tumour location, tumour size, pathological type, pathological grading, tumour infiltration, lymph node metastasis, total lymph nodes dissected, total number of positive lymph nodes (pathology), and distant metastasis, were obtained from their medical records (Table 1). Metastasis status was expressed using the pathologic stage of the disease determined according to the seventh edition of the American Joint Committee on Cancer (AJCC)/International Union Against Cancer TNM classification system. TMA and IHC staining of SSH1 in GC tissue (100 cases) and adjacent-to-carcinoma tissue (80 cases) was performed in our study (Fig. 1a).

The detailed clinico-pathological characteristics of these 100 patients are provided in Table 1. According to the total lymph nodes dissected and positive lymph nodes results, which were confirmed by pathology, we calculated the positive lymph node ratio, which defined as ratio of positive lymph nodes to all lymph nodes removed, is a powerful prognostic factor in cancer [11], and identified 43 cases with a positive lymph node ratio ≥ 0.5 and 57 cases with a positive lymph node ratio < 0.5. Tumour infiltration was at T1 in 8 cases, T2 in 7 cases, and T3 in 64 cases and T4 in 20 cases; 1 case was unknown. Lymph node metastasis (LNM) was positive in 79 cases, and 21 cases had no LNM. Regarding pathological grade (I-IV), there were 15 cases of grade II, 74 cases of grade III (II-III, III), and 11 cases of grade IV (III-IV). According to the seventh edition of the AJCC staging system, 10 cases were phase I, 33 were phase II, 49 were phase III, 7 were phase IV, and 1 was unknown.

Tumour development is a complex process, and multiple factors and steps participate in tumour malignant behaviour. The actin cytoskeleton plays important roles in cell migration, affecting such processes as cell-substrate adhesion, protrusion, phagocytosis and cytokinesis [12], and actin proteins form a large family and are central players in cell shape and movement [13]. SSH1-mediated cofilin activation is an essential regulator of actin dynamics [14]; SSH1 and cofilin are highly expressed or highly activated in various malignant tumours, increasing the rate of invasion and metastasis of tumour cells through different mechanisms and pathways [15, 16]. However, there is currently little evidence suggesting that SSH1 expression correlates with GC prognosis. The results of this study showed that SSH1 expression in GC tissues predicted GC lymph node metastasis based on TNM stage (P = 0.032; Table 3) and poor survival (P = 0.016; Table 4 & Fig. 2). SSH1 expression is indeed a useful prognostic factor for GC and is significantly associated with a poor prognosis.

SSH1 dephosphorylates and activates cofilin, which is a member of the actin-depolymerizing factor/cofilin family of proteins [12, 17] that regulates formation of the actin cytoskeleton. Tumour invasion and metastasis, the main causes of cancer-related death, are directly associated with cofilin activity [18, 19]. Yusufu et al. [20] reported that the cofilin phosphorylation status plays a critical role in breast cancer, and dephosphorylated cofilin is related to lower overall survival. It has been reported that cofilin is related to GC progression [21]. Wang et al. [8] suggested that SSH1 may be a potential target to prevent or treat PC invasion and metastasis by regulating cofilin phosphorylation. Combined with our results, we speculated that SSH1 is a key protein that associates with F-actin, which plays a key role in the malignant biological behaviour of GC cells.

According to our IHC analysis and microarray data, SSH1 expression was not different between GC and adjacent-to-carcinoma tissues (P = 0.769; Table 2). However, in this study, SSH1 was expressed in GC tissue, and its expression level was positively correlated with tumour LNM and poor prognosis; multivariate regression analysis clearly indicated that SSH1 expression in GC tissues was an exceptional predictor of poor prognosis (P = 0.030; Table 4). SSH1 affects tumour migration by altering cofilin activity, and researchers have proposed studying this interaction to gain better insight into the pathogenesis of digestive system tumours. Wang et al. [8] designed a study to explore the role of SSH1 in the development of PC; their results showed that SSH1 overexpression is related to cofilin activity and contributes to LNM and tumour cell migration, findings that were completely consistent with our results.

Not a single gene was activated within the migrating and invading tumour cells. Tumour invasion and metastasis, the main cause of cancer-related death, is directly associated with cofilin activity [18, 22]. Tan [16] et al. reported that the poor prognosis of patients with colorectal cancer is related to regulation of F-actin polymerization through the SSH/cofilin pathway. The function of the SSH/cofilin signalling pathway in actin polymerization and dynamics was studied using small interfering RNA (siRNA) silencing of SSH1,down-regulation SSH1 led to decreases in vascular cell migration and metastatic progression [9, 23]. Cell migration is associated with pathological remodelling at the site of vascular injury and with the invasive capacity of cancer cells. Our results showed that SSH1 expression level in GC was related to poor patient survival (P = 0.016). Therefore, SSH1 is the most important factor in cancer cell migration, and the abovementioned findings may indicate why the regulation of SSH1 activity is a promising therapeutic strategy to prolong the survival of cancer patients. As a regulator of cofilin, SSH1 may be an important regulatory factor in GC cell invasion and metastasis and may be closely related to the occurrence and development of GC. Therefore, the regulation of SSH1 activity is a promising therapeutic strategy for GC.

At present, there are three important limitations to our study. First, we used a retrospective design to analyse GC tissue samples from patients treated at a single centre and selection bias was unavoidable. Second, we analysed only the correlation with overall survival and could not access data regarding recurrence-free and disease-free survival. Third, our study did not include any molecular experiments.

In summary, this study clarified the clinical significance of SSH1 expression level in GC patients and showed that SSH1 activation plays an important role in GC progression and prognosis. Importantly, SSH1 dephosphorylates and activates cofilin, and SSH1 may play an important role in cell migration and cancer development through SSH/cofilin pathways. Our study indicated that SSH1 expression is a promising biomarker for the prevention and/or treatment of GC.