Akagi et al. BMC Cancer 2014, 14:822 http://www.biomedcentral.com/1471-2407/14/822 RESEARCH ARTICLE Open Access ROCK1 as a novel prognostic marker in vulvar cancer Erica M Akagi1, André M Lavorato-Rocha1, Beatriz de Melo Maia1, Iara S Rodrigues1, Kátia C Carvalho2, Monica M Stiepcich3, Glauco Baiocchi4, Yukie Sato-Kuwabara5, Silvia R Rogatto6, Fernando A Soares1,5 and Rafael M Rocha1,7* Abstract Background: Vulvar carcinoma is an infrequent tumour, accounting for fewer than 3% of all malignant tumours that affect women, but its incidence is rising in the past few decades. In young women, the manifestation of the vulvar carcinoma is often linked to risk factors such as smoking and HPV infection, but most cases develop in women aged over 50 years through poorly understood genetic mechanisms. Rho-associated coiled-coil-containing protein kinase 1 (ROCK1) has been implicated in many cellular processes, but its function in vulvar cancer has never been examined. In this study, we aimed to determine the prognostic value of ROCK1 gene and protein analysis in vulvar squamous cell carcinoma (VSCC). Methods: ROCK1 expression levels were measured in 16 vulvar tumour samples and adjacent normal tissue by qRT-PCR. Further, 96 VSCC samples were examined by immunohistochemistry (IHC) to confirm the involvement of ROCK1 in the disease. The molecular and pathological results were correlated with the clinical data of the patients. Sixteen fresh VSCC samples were analyzed by array-based comparative genomic hybridization (aCGH). Results: In each pair of samples, ROCK1 levels were higher by qRT-PCR in normal tissue compared with the tumour samples (p = 0.016). By IHC, 100% of invasive front areas of the tumour and 95.8% of central tumour areas were positive for ROCK1. Greater expression of ROCK1 was associated with the absence of lymph node metastasis (p = 0.022) and a lower depth of invasion (p = 0.002). In addition, higher ROCK1 levels correlated with greater recurrence-free survival (p = 0.001). Loss of ROCK1 was independently linked to worse cancer-specific survival (p = 0.0054) by multivariate analysis. This finding was validated by IHC, which demonstrated enhanced protein expression in normal versus tumour tissue (p < 0.001). By aCGH, 42.9% of samples showed a gain in copy number of the ROCK1 gene. Conclusions: ROCK1 is lower expressed in tumour tissue when compared with adjacent normal vulvar epithelia. In an independent sample set of VSCCs, lower expression levels of ROCK1 correlated with worse survival rates and a poor prognosis. These findings provide important information for the clinical management of vulvar cancer. Keywords: Vulvar carcinoma, ROCK1, qRT-PCR, Immunohistochemistry, aCGH, Prognosis * Correspondence: rafael.malagoli@gmail.com 1Molecular Morphology Laboratory, Investigative Pathology, AC Camargo Cancer Center, São Paulo, SP, Brazil 7Molecular Morphology Laboratory, AC Camargo Cancer Center, Rua Antônio Prudente, 109. 1o Andar, Patologia Investigativa, Liberdade, São Paulo, SP CEP: 01509-900, Brazil Full list of author information is available at the end of the article © 2014 Akagi et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. mailto:rafael.malagoli@gmail.com http://creativecommons.org/licenses/by/4.0 http://creativecommons.org/publicdomain/zero/1.0/ Akagi et al. BMC Cancer 2014, 14:822 Page 2 of 10 http://www.biomedcentral.com/1471-2407/14/822 Background Vulvar carcinoma is an infrequent tumour, accounting for 3% to 5% of all cancers of the female genital system [1-3]. Its incidence rises with age, peaking in women aged between 65 and 75 years [4,5]. Vulvar squamous cell carcinoma (VSCC) constitutes 90% of all malignant vulvar tumours and has 2 clinico- pathological types. The first type arises primarily in youn- ger patients and is associated with human papillomavirus (HPV) infection; the other form is seen mostly in elderly patients and appears to develop independently of HPV in- fection. These types of VSCC have disparate epidemio- logical, clinical, pathological, and molecular characteristics [3,4,6]. Despite its rarity, the incidence of VSCC has been rising in the past several decades, necessitating the identi- fication of predictive factors of its prognosis. Changes in cellular dynamics induce morphological al- terations in cells, due to reorganization of the actin cyto- skeleton. The Rho family of small GTPases are central regulators of the dynamics and reorganization of the actin cytoskeleton, mediating the formation of stress fi- bers and focal adhesions [7-9]. Certain members of the Rho family, such as RhoA and RhoC, interact with down- stream targets, culminating in various cellular responses. Their principal activity is to promote actomyosin contract- ility by phosphorylating a specific serine/threonine kinase, Rho-kinase associated coiled-coil (ROCK). ROCK1 and 2 have been implicated in many cellular processes and pathologies, particularly in metastatic pro- cesses of cell lines and in the cardiovascular and nervous systems. Based on their oncogenic activity, ROCKs are being examined as therapeutic targets in various tumours, such as non-small-cell lung tumours [10]; glioblastoma [11]; osteosarcoma [12]; and prostate [13,14], breast [15], ovarian [16], hepatocellular [17], and bladder cancers [18]. Human ROCK1 maps to chromosome 18 (18q11.1) [19-22] and performs its functions by phosphorylating substrates, such as myosin light chain (MLC), the MLC phosphatase subunit (MYPT-1), and LIM kinase; many other substrates continue to be reported. These sub- strates catalyze many processes during morphological changes and metastasis, including structural rearrange- ment, adhesion, alterations in cellular polarity, migra- tion, invasion, transformation, proliferation, cytokinesis, and apoptosis [9,22,23]. The precise function of ROCK1 in carcinogenesis and in the architectural rearrangement of tumour cells during metastasis remains debated [24]. ROCK1 ap- pears to be implicated in a complex balance between oncogene function and proapoptotic responses, de- pending on the cell type. Based on its involvement in cell migration in other tumours and the lack of data on its function in vulvar carcinomas, we selected ROCK1 for further study. We aimed to examine the function of ROCK1 in the progression of vulvar carcinoma. In this study, we mea- sured ROCK1 mRNA and protein levels and analyzed the data on ROCK1 copy number alterations from a parallel project of our group. The transcript and protein results were correlated with clinicopathological characteristics to determine the prognostic value of ROCK1 in vulvar cancer. Methods Patient and sample selection A total of 96 invasive vulvar carcinoma samples were randomly and retrospectively selected from the archives of the AC Camargo Cancer Center Anatomic Pathology Department from January 1990 to December 2010 and analyzed by immunohistochemistry. All samples were formalin-fixed and paraffin-embedded (FFPE), and their HPV status has been reported [2,5,25]. Sixteen fresh fro- zen tumour samples and 11 adjacent nontumour sam- ples were also obtained from the AC Camargo Cancer Center Biobank for mRNA expression and DNA copy number analysis. The inclusion criteria were patients who had under- gone surgery or biopsy in this hospital and were diag- nosed with invasive vulvar squamous cell carcinoma. All cases were H&E-stained and reviewed by experienced pathologists to confirm the previous diagnosis and adapt the reports to updated nomenclature. The clinical data on all patients were obtained from their medical records. In situ carcinomas, cases in which neoadjuvant radio- therapy and/or chemotherapy were performed, and cases that lacked sufficient material or clinical information for the analyses were excluded from the study. This work was approved by the ethics committee at AC Camargo (Research Ethics Committee number 1672/ 12) and was performed per the Helsinki Declaration. RNA extraction from fresh frozen samples The RNeasy Mini Kit RNA Extraction Kit (QIAGEN, Austin, TX, USA) and a Precellys® 24 homogenizer (Stretton Scientific, Stretton, UK) were used to extract RNA from the fresh frozen samples per the manufac- turer’s instructions. Prior to the extraction, the H&E slides from all samples were reviewed by the Biobank’s chief pathologist (Dr. AHJFMC). Aliquots of RNA were stored at -80°C until cDNA synthesis. Quantitative real-time RT-PCR (RT-qPCR) Gene expression was analyzed by RT-qPCR on an Applied Biosystems 7900HT Fast Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) using the TaqMan Universal PCR Master Mix detection system (Applied Biosystems), according to the supplier’s specifica- tions. Primers and probes for ROCK1 (Hs01127688_m1) Akagi et al. BMC Cancer 2014, 14:822 Page 3 of 10 http://www.biomedcentral.com/1471-2407/14/822 were purchased from Applied Biosystems. HPRT was used as an endogenous control. Data analyses were per- formed, comparing adjacent normal and tumorous vul- var samples. The Pfaffl [26] method was used to obtain relative quantification (RQ) values and determine gene expression levels [26]. Immunohistochemistry Four-micrometer-thick FFPE samples were placed on StarFrost® electrically charged slides (Braunschweig, Germany). All reactions were performed on whole-tissue slides using the Advance Kit Protocol (DAKO). Antigen recovery was performed using Tris-EDTA (pH 9.0) in a water bath (96°C). The primary antibody was anti- ROCK1 (ABCAM, Cat.#1761-1, Clone EP786Y), diluted 1:100. At the end of the reaction, the slides were washed with tap water, dehydrated sequentially in alcohol and xylene, and mounted manually. Evaluation of immunohistochemistry Slides were digitalized on an APERIO® scanner and scored visually. IHC expression patterns were evaluated quantitatively, wherein expression levels were scored by the percentage of positive cells and the intensity of im- munostaining [HScore = Σ (ix Pi) and Pi: percentage of positive pixels, ranging from 0% to 100% and color in- tensity of the pixel i =0, 1, 2, or 3], ranging between 20 and 250 per Rodrigues et al. [25]. Final HScores were defined as HScore =1 when the positivity was weak, with staining intensity ranging from 20 to 149 and HScore =2 for strong staining and a staining intensity of ≥150. ROCK1 immunostaining was present in all samples. Two areas for each case—the central tumour and inva- sive front—were examined for ROCK1 expression. As described (Rodrigues et al. [25]), the central tumour was considered as the largest area of extension of the tumour; at least 3 areas were selected and analyzed. The invasive front was defined as a group of up to 5 cells that detached from the main tumour mass, which usu- ally infiltrated the adjacent stroma; 10 fields were se- lected [25]. Array based-comparative genomic hybridization array (aCGH) Based on our ROCK1/mRNA and protein data, we ex- amined ROCK1 copy number alterations in vulvar car- cinoma samples by array-CGH using data from a parallel study. A total of 200 ng each of tumour DNA and normal commercially available DNA (Human Genomic DNA: Female; Promega, Madison, USA) were analyzed compared on an 8 × 60 K Agilent platform for aCGH (Agilent Technologies®, Santa Clara, USA). The labeling, hybridization, and washes were per- formed per the Agilent Oligonucleotide Array-Based CGH for Genomic DNA Analysis – Enzymatic Labeling kit protocol (Agilent Technologies®, Santa Clara, USA). The slides were scanned on a DNA microarray scanner with Surescan High-Resolution Technology (Agilent Tech- nologies®, Santa Clara, USA), based on HG19, and the results were extracted using Feature Extraction, v10.7.3.1 (Agilent Technologies®, Santa Clara, USA). Copy number analysis was performed using Nexus Copy Number Soft- ware, v6.0 (Biodiscovery, El Segundo, USA). A copy number alteration was defined as exceeding the significance threshold of 1 × 10−6 in a minimum of 5 con- secutive probes and in more than 30% of the samples. Thresholds were defined as the average log2 CGH fluorescence ratio for copy gains ≥0.3, high copy number gains defined as ≥0.6, losses defined as ≤ -0.3, and homo- zygous losses defined as ≤ -1.0. Nonrandom genomic copy number alterations were identified using the Fast Adaptive States Segmentation Technique 2 (FASST2) al- gorithm and the Significance Testing for Aberrant Copy number (STAC) statistical method [27,28]. Alterations that were detected in at least 42.9% of samples were ex- amined in greater detail. Statistical analysis Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS, IBM), version 20.0. Protein expression in the tumour center and invasive front was compared by Wilcoxon signed-rank test. Mann- Whitney test and student t-test were used to analyze the association between protein expression and clinicopatho- logical parameters, and the Kaplan-Meier method was used to examine specific cancer survival and recurrence- free survival rates. The difference between survival curves was assessed by log-rank test. Multivariate analyses were performed using the Cox proportional hazards regression model. Statistical significance was set to p ≤0.05. Results Demographic and clinicopathological features The mean age of the 96 patients was 75 years, ranging from 30 to 103 years. The mean age at menopause was 50 years, ranging between 38 and 60 years. Most pa- tients were Caucasian (83.3%) and did not consume al- cohol (88.5%); 15.6% of patients were current or past smokers. Forty eight percent of the patients were HPV-positive, most of whom had the subtypes HPV16 (48%), HPV33 (24%), and HPV18 (15%). Associated lesions were present, such as vulvar intraepithelial neoplasias (VINs; 13.5% of patients) and lichen sclerosus (6.25%). Based on the histo- logical diagnosis, moderately differentiated squamous cell carcinomas (SCC2) was the most prevalent form (46%), followed by SCC1 (34%), basaloid (9%), SCC3 (8%), sarco- matoid (2%), and verrucous carcinoma (1%). Most tumours Akagi et al. BMC Cancer 2014, 14:822 Page 4 of 10 http://www.biomedcentral.com/1471-2407/14/822 were classified as FIGO stage IB (53.1%), followed by stage IIIB (20.3%), IIIA (12.1%), II (6.2%), IIIC (6.2%), and IVA (2.1%). Of the 96 patients with VSCC, 34.4% died due to the cancer, and 46.9% expired due to other causes. ROCK1 mRNA expression By RT-qPCR, ROCK1 was overexpressed in normal adja- cent samples compared with the tumour tissue (p = 0.0167, Figure 1A). Also, ROCK1 HScores were higher in normal epithelium versus the tumour areas in a subset of samples (n = 21) (p < 0.001) (Figure 1B, C, and D). ROCK1 immunostaining ROCK1 immunostaining was heterogeneous and cyto- plasmic in all tumour extensions (Figure 2) and positive in the invasive front of all cases (100%) and in 92 central tumours (95.8%). There was a significant positive correl- ation between central tumour and invasive front expres- sion of ROCK1 (p <0.001; Figure 3A). In the statistical analysis, greater expression of ROCK1 in central tumours and the invasive front correlated significantly with the absence of lymph node metastasis (p = 0.036 and p = 0.022, respectively), the presence of inflammatory infiltrate (p =0.010 and p = 0.009, respectively), and a lower depth of invasion (p = 0.048 and p = 0.002), as shown in Figure 3B. There was no association Figure 1 ROCK 1 mRNA and protein are overexpressed in normal tiss samples compared with tumour samples (p =0.016, A). Increased ROCK1 im a subset of samples (n = 21) (B); p <0.001. Representative images of norma captured at 400× magnification. between ROCK1 positivity and HPV infection, histological type, FIGO stage, recurrence, or vascular invasion. ROCK1 expression and patient survival Patients with lower expression of ROCK1 in the central tumour and invasive front had lower recurrence-free survival rates (p = 0.004 and p = 0.001, respectively; Figure 4A and B), and those with weak ROCK1 expression in the invasive front experienced lower cancer-specific survival (p <0.001; Figure 4C and D). By multivariate analysis, high ROCK1 expression in the invasive front was independently associated with greater cancer-specific survival (HR 0.3, 95% CI 0.11–0.84, p = 0.0054; Table 1). The clinicopathological characteristics were analyzed by Cox regression. Lymph node metastasis (data not shown) and vascular invasion were independently asso- ciated with decreased of the survival (p = 0.0074 and 0.0365, respectively). aCGH analysis By aCGH analysis, 29 regions underwent significant copy number alterations: 9 were associated with copy number loss (8p23.3, 5q11.1-q11.2, 3p11.1-q11.1, 9p23, 21p11.2- p11.1, Xq28, 7q36.3, 19p13.3, and 21p11.2), and 20 had gains (1q22, 20q11.21-q11.23, 1p36.23-p36.22, 11q13.3, 19q13.12, 19q13.32, 7q11.21, 7q11.22, 7q11.23, 7q11.23, 7q22.1, 11q12.2-q12.3, 11q12.3, 11q13.2, 16q22.1, 18q11.1- ues. ROCK1 is overexpressed by qRT-PCR analysis in normal adjacent munostaining HScore in normal epithelium compared with tumour in l epithelium (C) and tumour (D) immunostaining from the same case, Figure 2 Immunohistochemical staining of ROCK1 in vulva carcinoma. Representative image of weak-positive staining in central tumour (A) and invasive front (B, arrows); moderate staining in the central tumour (C) and invasive front (D, arrows); strong-positive staining in central magnification. Images A and B were captured at 200× magnification. Images C, D, E, and F were captured at 400× magnification. Akagi et al. BMC Cancer 2014, 14:822 Page 5 of 10 http://www.biomedcentral.com/1471-2407/14/822 q11.2, 18q11.2, 7p22.2-p22.1, 12q24.31, and 15q11.1- q11.2). Of the latter, region 18q11.1-q11.2, which harbors ROCK1, had more copies than the reference DNA in 42.9% of samples (Figure 5). To identify their function and the processes that they mediate, the genes that were selected in the copy number analysis were included in an in silico functional analysis, performed by Ingenuity Pathways Analysis (IPA). ROCK1 appeared in 2 of the top 5 canonical pathways with the highest ratios—RhoGDI and Rho GTP signaling—both of which are associated with cell migration. Discussion ROCK has significant functions in cancer progression and metastasis, rendering it a potential therapeutic target [9]. In this study, we examined the function of the Rho- associated protein kinase ROCK1 in human vulvar carcin- oma. Our data showed that aspects that are related to a good prognosis, such as the absence of lymph node me- tastasis, lower depth of invasion, and better survival, cor- related with ROCK1 immunoexpression, suggesting that ROCK1 is a marker of good prognosis in vulvar cancer. ROCK1 immunopositivity was observed in the tumour invasion fronts of all cases and in nearly all central tumour areas. Also, there was no difference in ROCK1 expression levels between central tumour versus invasive front areas, in contrast to what we have reported con- cerning the variability of epithelial to mesenchymal transition markers [26], EGFR [29], and c-Kit [5] in tu- mours. Nevertheless, vulvar carcinomas can be highly heterogeneous [29,30], and cytoplasmic immunostain- ing for ROCK1 protein was heterogeneous in all tumour extensions in our cases, reflecting a disadvan- tage of ROCK-targeting therapies in this tumour type. In this study, we performed a global evaluation of ROCK1 expression and its relationship with clinical data and the prognosis. ROCK has a significant influence on cancer progression [8,9,11,31], metastasis [17,21,32,33], and apoptosis [12,23,34]. Recent evidence suggests that ROCK phosphorylates PTEN [34,35], a negative regula- tor of the PI3-K/Akt pathway, with roles in cell survival and apoptosis [34,36]. Inhibition of ROCK/Rho-kinase in Ras-transformed cells is insufficient to effect a motile phenotype in them, Figure 3 Association between clinicopathological features and ROCK1 in vulvar carcinoma. Abbreviations: ≤2 = 2 or fewer lymph nodes involved; >2 =more than 2 lymph nodes involved; SMD = superficial and mid-dermis; DDA = deep dermis and subcutaneous tissues. *Statistically significant, p < 0.05. Akagi et al. BMC Cancer 2014, 14:822 Page 6 of 10 http://www.biomedcentral.com/1471-2407/14/822 Figure 4 Kaplan-Meier survival curves for ROCK1 staining. Increased expression of ROCK1 in the central tumour (p =0.004; A) and in the invasive front (p =0.001; B) correlates with better recurrence-free survival and lower cancer-specific survival in the central tumour (p =0.081; C) and in the invasive front (p <0.001; D). Abbreviations: CT = central tumour; IF = invasive front. Table 1 Multivariate analysis of ROCK1 expression and clinicopathological characteristics in patients with vulvar SCC Variables Category n Hazard ratio for survival 95.0% CI p Histologic types SCC1, SCC2, Verrucous Ca (1) 76 1 1 0.6177 SCC3, Basaloid Ca, Sarcomatoid Ca(2) 20 0.78 0.30 - 2.05 FIGO stage IA, IB, II (1) 54 1 1 0.2083 IIA, IIIB, IIIC, IVA, IVB (2) 40 1.62 0.76 - 3.47 HPV Absent (0) 50 1 1 0.2238 Present (1) 46 0.63 0.30 - 1.33 Inflammatory infiltrate Absent (0) 38 1 1 0.4964 Present (1) 58 0.77 0.36 - 1.64 Vascular Invasion Absent (0) 74 1 1 0.0365* Present (1) 17 2.22 1.03 - 4.77 Perineural invasion Absent (0) 76 1 1 0.4178 Present (1) 12 1.43 0.60 - 3.38 ROCK1 CT Weak expression 22 1 1 0.6653 Moderate expression 55 0.70 0.28 - 1.75 Strong expression 19 0.95 0.34 - 2.67 ROCK1 IF Weak expression 8 1 1 0.0054* Moderate expression 53 0.22 0.08 - 0.60 Strong expression 35 0.30 0.11 -0.84 Abbreviations: SCC squamous cell carcinomas, CI confidence interval, FIGO International Federation of Gynecology and Obstetrics, HPV human papillomavirus, SMD superficial and mid dermis, DDA deep dermis and subcutaneous tissues, CT central tumour and IF invasive front. *Statistically significant, p < 0.05. Akagi et al. BMC Cancer 2014, 14:822 Page 7 of 10 http://www.biomedcentral.com/1471-2407/14/822 Figure 5 Representative image of aCGH analysis of chromosome 18 with emphasis on ROCK1. (A) The copy number gain (chr18:18,539,853-19,429,001) is in blue; exons of ROCK1 are illustrated at the bottom of the diagram. (B) Example highlighting the gains (≥0.3) in blue and high gains (≥0.6) in yellow. Gene regions covered by each probe can be seen as small dots. Akagi et al. BMC Cancer 2014, 14:822 Page 8 of 10 http://www.biomedcentral.com/1471-2407/14/822 suggesting that this cell type requires changes in other regulators of the cytoskeleton to increase its motility [32]. Notably, elongated cells, such as SW-962, a vulvar squamous cell carcinoma metastatic cell line do not re- quire Rho or ROCK function, unlike cells that move through rounded, or amoeboid, movement [21,37,38]. Thus, we hypothesize that elongated vulvar carcinoma cells move and migrate using mechanisms other than Rho/ROCK activation. The effectiveness of therapeutic agents against ROCK, such as fusadil and Y27632, might be limited when cells move through elongated morph- ology [37]. ROCK1 immunostaining was also associated with im- portant clinical features in vulvar cancer and with the most significant clinical property and prognostic factor in this tumour: lymph node metastasis. When overex- pressed, ROCK1 correlated inversely with lymph node metastasis in the central tumour and invasive front. To define groups for the statistical analysis regarding the clinical implications of the presence or absence of lymph node involvement, we considered positivity as metastasis when 2 or more lymph nodes were involved and negativ- ity when 1 or 0 lymph nodes were involved. This strat- egy was based on a previous study that demonstrated that 5-year survival for patients with negative or 1 posi- tive lymph node did not differ from each other [39]. Similarly, greater expression in the central tumour and invasive front was associated with lower invasive depth and higher recurrence-free survival. Clinicopathological characteristics analyzed by Cox re- gression demonstrated that lymph node metastasis and vascular invasion were independently associated with de- creased of the survival, indicating that these features are related to poor survival in patients with VSCC. Also, higher expression of ROCK1 was linked to greater survival, the absence of lymph node metastasis, and a lower depth of invasion. Moreover, elevated ROCK1 levels in the invasive front was an independent pro- tective factor (HR =0.22 for moderate expression, and HR =0.33 for strong expression) with regard to cancer- specific survival (p =0.0054). These results implicate ROCK1 as a good prognostic marker in vulvar cancer. In addition, patients with weak expression of ROCK1 in the invasive front, but not the central tumour, had lower cancer-specific survival rates, implying that this marker is protective during cancer progression. Conversely, ROCK1 has been largely re- ported as a marker of worse prognosis in many cancer types [12]. There is emerging evidence that ROCK governs the morphological events that take place during apoptosis (cell contraction, membrane blebbing, nuclear fragmen- tation, and disintegration of apoptotic cells) through cytoskeletal rearrangement and actomyosin contractility [40,41]. Other groups contend that ROCK1 is required for apoptotic fragmentation and phagocytosis of dying COS-7 cells [41]. Because ROCK is a proapoptotic regu- lator in various cell types, depending on the cell type and apoptotic stimulus [40], we believe that its overex- pression in vulvar cancer is associated with apoptotic stimuli and, thus, it can be associated with better prog- nosis; as demonstrated in other studies [41,42]. Although the relationship between apoptosis and the prognosis re- mains unknown, it could, at least in part, explain the as- sociation of ROCK1 with a good prognosis in vulvar cancer. ROCK1 copy number gains were detected in 42.9% of our samples. However, these data are controversial, be- cause ROCK1 was more highly expressed in normal tis- sue by IHC and RT-qPCR compared with tumour Akagi et al. BMC Cancer 2014, 14:822 Page 9 of 10 http://www.biomedcentral.com/1471-2407/14/822 samples. Although associations between copy number and gene expression comprise the concomitant amplifi- cation of the gene with enhancement of its expression, there remain other genes, approximately 50% [43], the amplification of which does not correspond to gene overexpression. We hypothesize that tissues that were used for normalization of the validation techniques were adjacent to the tumour. Despite careful morphological analysis by an experienced pathologist of nontumour tissue, the proximity of malignant cells could have influenced the surrounding microenvironment, upregulating various genes, including ROCK1, in normal epithelia. In addition, other mechanisms, such as microRNA regulation and epigenetic alterations (including methyla- tion and histone deacetylation), might cause the lack of correlation between the genomic and proteomic data found in our study. Previous reports have demonstrated the function of microRNAs (eg, microRNA-135a and microRNA124-3p) in ROCK1 regulation in prostate, gas- tric, and bladder carcinoma [44-46]. Also, despite the increase in copy number in tumour samples by aCGH analysis, we can not make any conclu- sions regarding gene integrity and the extent of its ab- normal. Our results do not allow us to conclude much concerning the amplification of ROCK1 in the tumour samples or on the possible generation of aberrant mRNA or truncated protein. The genomic mechanism that leads to gene amplification in tumour cells remains undefined, as do the molecular pathways that effect amplified gene expression. Conclusions This is the first report to demonstrate that ROCK1 cor- relates with a good prognosis in cancer. Although vulvar carcinomas are rare, this type of cancer can serve as a valuable model in the study of molecular alterations that can be transposed to other types of epithelial neoplasms. Further, novel biomarkers, such as ROCK1, are signifi- cant, because its evaluation by IHC in routine practice can help better establish prognosis and select more con- servative surgical approaches for this mutilating disease. Abbreviations ROCK1: Rho-associated coiled-coil containing protein kinase 1; VSCC (or SCC): vulvar squamous cell carcinoma; HPV: human papillomavirus; IHC: immunohistochemistry; qRT-PCR: quantitative real-time polymerase chain reaction; aCGH: array-based comparative genomic hybridization; MLC: myosin light chain; FIGO: International Federation of Gynecology and Obstetrics; FFPE: formalin-fixed paraffin-embedded; PTEN: phosphatase and tensin homolog deleted from chromosome ten. Competing interests All authors declare that they have no competing interests. Authors’ contributions Conceived and designed the research studies: RMR, EMA. Performed the experiments: EMA, AMLR, YSK, KCC. Analyzed and interpreted the data: EMA, AMLR, BMM, RMR. Contributed with reagents/materials/analysis tools: ISR, KCC, MMS, GB, YSK, SRR, FAS. Drafted the manuscript: EMA, BMM, AMLR, SRR, RMR. Pathological revision: MMS. All authors have read and approved the final manuscript. Acknowledgments We thank the AC Camargo Cancer Center Anatomic Pathology Department and Biobank for providing human specimens. This work was supported by the São Paulo Research Foundation (FAPESP) and the National Council of Research (CNPq). Author details 1Molecular Morphology Laboratory, Investigative Pathology, AC Camargo Cancer Center, São Paulo, SP, Brazil. 2Department of Obstetrics and Gynecology, School of Medicine of São Paulo University, São Paulo, SP, Brazil. 3Pathology Department, Fleury Institute, São Paulo, SP, Brazil. 4Department of Gynecology Oncology, AC Camargo Cancer Center, São Paulo, SP, Brazil. 5Department of Anatomic Pathology, AC Camargo Cancer Center, São Paulo, SP, Brazil. 6NeoGene Laboratory, AC Camargo Cancer Center, São Paulo and Department of Urology, Faculty of Medicine, UNESP, Botucatu, SP, Brazil. 7Molecular Morphology Laboratory, AC Camargo Cancer Center, Rua Antônio Prudente, 109. 1o Andar, Patologia Investigativa, Liberdade, São Paulo, SP CEP: 01509-900, Brazil. Received: 3 June 2014 Accepted: 10 October 2014 Published: 7 November 2014 References 1. De Melo MB, Munhoz Cestari F, Lavorato-Rocha AM, Sant’Ana Rodrigues I, Baiocchi G, Cardoso Guimarães G, Vassallo J, Stiepcich M, Soares FA, Malagoli Rocha R: Characterization of sociodemographic and clinicopathological features in Brazilian patients with vulvar squamous cell carcinoma. Gynecol Obstet Invest 2013, 75(1):53–60. 2. Lavorato-Rocha AM, De Melo Maia B, Rodrigues IS, Stiepcich MM, Baiocchi G, Da Silva Cestari FM, Carvalho KC, Soares FA, Rocha RM: Prognostication of vulvar cancer based on p14ARF status: molecular assessment of transcript and protein. Ann Surg Oncol 2013, 20(1):31–39. 3. Trietsch MD, Peters AA, Gaarenstroom KN, Van Koningsbrugge SH, Ter Haar NT, Osse EM, Halbesma N, Fleuren GJ: Spindle cell morphology is related to poor prognosis in vulvar squamous cell carcinoma. Br J Cancer 2013, 109(8):2259–2265. 4. Del Pino M, Rodriguez-Carunchio L, Ordi J: Pathways of vulvar intraepithelial neoplasia and squamous cell carcinoma. Histopathology 2013, 62(1):161–175. 5. De Melo MB, Lavorato-Rocha AM, Rodrigues IS, Baiocchi G, Cestari FM, Stiepcich MM, Chinen LT, Carvalho KC, Soares FA, Rocha RM: Prognostic significance of c-KIT in vulvar cancer: bringing this molecular marker from bench to bedside. J Transl Med 2012, 10:150. 6. Allen DG, Hutchins AM, Hammet F, White DJ, Scurry JP, Tabrizi SN, Garland SM, Armes JE: Genetic aberrations detected by comparative genomic hybridisation in vulvar cancers. Br J Cancer 2002, 86(6):924–928. 7. Morgan-Fisher M, Wewer UM, Yoneda A: Regulation of ROCK activity in cancer. J Histochem Cytochem 2013, 61(3):185–198. 8. Banyard J, Anand-Apte B, Symons M, Zetter BR: Motility and invasion are differentially modulated by Rho family GTPases. Oncogene 2000, 19(4):580–591. 9. Majid S, Dar AA, Saini S, Shahryari V, Arora S, Zaman MS, Chang I, Yamamura S, Chiyomaru T, Fukuhara S, Tanaka Y, Deng G, Tabatabai ZL, Dahiya R: MicroRNA-1280 inhibits invasion and metastasis by targeting ROCK1 in bladder cancer. PLoS One 2012, 7(10):e46743. 10. Vigil D, Kim TY, Plachco A, Garton AJ, Castaldo L, Pachter JA, Dong H, Chen X, Tokar B, Campbell SL, Der CJ: ROCK1 and ROCK2 are required for non-small cell lung cancer anchorage-independent growth and invasion. Cancer Res 2012, 72(20):5338–5347. 11. Zohrabian VM, Forzani B, Chau Z, Murali R, Jhanwar-Uniyal M: Rho/ROCK and MAPK signaling pathways are involved in glioblastoma cell migration and proliferation. Anticancer Res 2009, 29(1):119–123. 12. Liu X, Choy E, Hornicek FJ, Yang S, Yang C, Harmon D, Mankin H, Duan Z: ROCK1 as a potential therapeutic target in osteosarcoma. J Orthop Res 2011, 29(8):1259–1266. Akagi et al. BMC Cancer 2014, 14:822 Page 10 of 10 http://www.biomedcentral.com/1471-2407/14/822 13. Somlyo AV, Bradshaw D, Ramos S, Murphy C, Myers CE, Somlyo AP: Rho- Kinase inhibitor retards migration and in vivo dissemination of human prostate cancer cells. Biochem Biophys Res Commun 2000, 269(3):652–659. 14. Zhang C, Zhang S, Zhang Z, He J, Xu Y, Liu S: ROCK has a crucial role in regulating prostate tumor growth through interaction with c-Myc. Oncogene 2013, 1–10. doi:10.1038/onc.2013.505. [Epub ahead of print]. 15. Lane J, Martin TA, Watkins G, Mansel RE, Jiang WG: The expression and prognostic value of ROCK I and ROCK II and their role in human breast cancer. Int J Oncol 2008, 33(3):585–593. 16. Ohta T, Takahashi T, Shibuya T, Amita M, Henmi N, Takahashi K, Kurachi H: Inhibition of the Rho/ROCK pathway enhances the efficacy of cisplatin through the blockage of hypoxia-inducible factor-1α in human ovarian cancer cells. Cancer Biol Ther 2012, 13(1):25–33. 17. Wong CC, Wong CM, Ko FC, Chan LK, Ching YP, Yam JW, Ng IO: Deleted in liver cancer 1 (DLC1) negatively regulates Rho/ROCK/MLC pathway in hepatocellular carcinoma. PLoS One 2008, 3(7):e2779. 18. Kamai T, Tsujii T, Arai K, Takagi K, Asami H, Ito Y, Oshima H: Significant association of Rho/ROCK pathway with invasion and metastasis of bladder cancer. Clin Cancer Res 2003, 9(7):2632–2641. 19. Guan R, Xu X, Chen M, Hu H, Ge H, Wen S, Zhou S, Pi R: Advances in the studies of roles of Rho/Rho-kinase in diseases and the development of its inhibitors. Eur J Med Chem 2013, 70:613–622. 20. Lock FE, Ryan KR, Poulter NS, Parsons M, Hotchin NA: Differential regulation of adhesion complex turnover by ROCK1 and ROCK2. PLoS One 2012, 7(2):e31423. 21. Rath N, Olson MF: Rho-associated kinases in tumorigenesis: re-considering ROCK inhibition for cancer therapy. EMBO Rep 2012, 13(10):900–908. 22. Amano M, Nakayama M, Kaibuchi K: Rho-kinase/ROCK: A key regulator of the cytoskeleton and cell polarity. Cytoskeleton (Hoboken) 2010, 67(9):545–554. 23. Gurkar AU, Chu K, Raj L, Bouley R, Lee SH, Kim YB, Dunn SE, Mandinova A, Lee SW: Identification of ROCK1 kinase as a critical regulator of Beclin1- mediated autophagy during metabolic stress. Nat Commun 2013, 4:2189. 24. Nakashima M, Adachi S, Yasuda I, Yamauchi T, Kawaguchi J, Hanamatsu T, Yoshioka T, Okano Y, Hirose Y, Kozawa O, Moriwaki H: Inhibition of Rho-associated coiled-coil containing protein kinase enhances the activation of epidermal growth factor receptor in pancreatic cancer cells. Mol Cancer 2011, 10:79. 25. Rodrigues IS, Lavorato-Rocha AM, De M, Maia B, Stiepcich MM, De Carvalho FM, Baiocchi G, Soares FA, Rocha RM: Epithelial-mesenchymal transition-like events in vulvar cancer and its relation with HPV. Br J Cancer 2013, 109(1):184–194. 26. Pfaffl MW: A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001, 29(9):e45. 27. Diskin SJ, Eck T, Greshock J, Mosse YP, Naylor T, Stoeckert CJ, Weber BL, Maris JM, Grant GR: STAC: A method for testing the significance of DNA copy number aberrations across multiple array-CGH experiments. Genome Res 2006, 16:1149–1158. 28. Silveira SM, Villacis RAR, Marchi FA, Barros Filho MC, Drigo AS, Scapulatempo Neto C, Lopes A, Cunha IW, Rogatto SR: Genomic signatures predict poor outcome in undifferentiated pleomorphic sarcomas and leiomyosarcomas. PLoS One 2013, 8(6):e67643. 29. De Melo MB, Fontes AM, Lavorato-Rocha AM, Rodrigues IS, De Brot L, Baiocchi G, Stiepcich MM, Soares FA, Rocha RM: EGFR expression in vulvar cancer: clinical implications and tumor heterogeneity. Human Pathol 2014, 45(5):917–925. 30. Brustmann H: Epidermal growth factor receptor is involved in the development of an invasive phenotype in vulvar squamous lesions, but is not related to MIB-1 immunoreactivity. Int J Gynecol Pathol 2007, 26(4):481–489. 31. Mali RS, Ramdas B, Ma P, Shi J, Munugalavadla V, Sims E, Wei L, Vemula S, Nabinger SC, Goodwin CB, Chan RJ, Traina F, Visconte V, Tiu RV, Lewis TA, Stern AM, Wen Q, Crispino JD, Boswell HS, Kapur R: Rho kinase regulates the survival and transformation of cells bearing oncogenic forms of KIT, FLT3, and BCR-ABL. Cancer Cell 2011, 20(3):357–369. 32. Sahai E, Olson MF, Marshall CJ: Cross-talk between Ras and Rho signalling pathways in transformation favours proliferation and increased motility. EMBO J 2001, 20(4):755–766. 33. Hosono Y, Yamaguchi T, Mizutani E, Yanagisawa K, Arima C, Tomida S, Shimada Y, Hiraoka M, Kato S, Yokoi K, Suzuki M, Takahashi T: MYBPH, a transcriptional target of TTF-1, inhibits ROCK1, and reduces cell motility and metastasis. EMBO J 2012, 31(2):481–493. 34. Li G, Liu L, Shan C, Cheng Q, Budhraja A, Zhou T, Cui H, Gao N: RhoA/ ROCK/PTEN signaling is involved in AT-101-mediated apoptosis in human leukemia cells in vitro and in vivo. Cell Death Dis 2014, 5:e998. 35. Li GB, Chenq Q, Liu L, Zhou T, Shan CY, Hu XY, Zhou J, Liu EH, Li P, Gao N: Mitochondrial translocation of cofilin is required for allyl isothiocyanate- mediated cell death via ROCK1/PTEN/PI3K signaling pathway. Cell Commun Signal 2013, 11:50. 36. Keniry M, Parsons R: The role of PTEN signaling perturbations in cancer and in targeted therapy. Oncogene 2008, 27:5477–5485. 37. Sahai E, Marshall CJ: Differing modes of tumour cell invasion have distinct requirements for Rho/ROCK signalling and extracellular proteolysis. Nat Cell Biol 2003, 5(8):711–719. 38. Nakayama M, Amano M, Katsumi A, Kaneko T, Kawabata S, Takefuji M, Kaibuchi K: Rho-kinase and myosin II activities are required for cell type and environment specific migration. Genes Cells 2005, 10(2):107–117. 39. Hacker NF, Berek JS, Lagasse LD, Leuchter RS, Moore JG: Management of Regional Lymph Nodes and Their Prognostic Influence in Vulvar Cancer. Obstet Gynecol 1983, 61(4):408–412. 40. Shi J, Wei L: Rho kinase in the regulation of cell death and survival. Arch Immunol Ther Exp 2007, 55(2):61–75. 41. Orlando KA, Stone NL, Pittman RN: Rho kinase regulates fragmentation and phagocytosis of apoptotic cells. Exp Cell Res 2006, 312(1):5–15. 42. Tanaka F, Kawano Y, Li M, Takata T, Miyahara R, Yanagihara K, Ohtake Y, Fukuse T, Wada H: Prognostic significance of apoptotic index in completely resected non-small-cell lung cancer. J Clin Oncol 1999, 17(9):2728–2736. 43. Hyman E, Kauraniemi P, Hautaniemi S, Wolf M, Mousses S: Impact of DNA amplification on gene expression patterns in breast cancer. Cancer Res 2002, 62(21):6240–6245. 44. Kroiss A, Vincent S, Decaussin-Petrucci M, Meugnier E, Viallet J, Ruffion A, Chalmel F, Samarut J, Allioli N: Androgen-regulated microRNA-135a decreases prostate cancer cell migration and invasion through downregulating ROCK1 and ROCK2. Oncogene 2014. doi:10.1038/ onc.2014.222. [Epub ahead of print]. 45. Shin JY, Kim YI, Cho SJ, Lee MK, Kook MC, Lee JH, Lee SS, Ashktorab H, Smooy DT, Ryu KW, Kim YW, Chhoi IJ: MicroRNA 135a suppresses lymph node metastasis through down-regulation of ROCK1 in early gastric cancer. PLoS One 2014, 9(1):e85205. 46. Xu X, Li S, Lin Y, Chen H, Hu Z, Mao Y, Xu X, Wu J, Zhu Y, Zheng X, Luo J, Xie L: MicroRNA 124-3p inhibits cell migration and invasion in bladdes cancer cells by targeting ROCK1. J Trans Med 2013, 11:276. doi:10.1186/1471-2407-14-822 Cite this article as: Akagi et al.: ROCK1 as a novel prognostic marker in vulvar cancer. BMC Cancer 2014 14:822. Submit your next manuscript to BioMed Central and take full advantage of: • Convenient online submission • Thorough peer review • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit Abstract Background Methods Results Conclusions Background Methods Patient and sample selection RNA extraction from fresh frozen samples Quantitative real-time RT-PCR (RT-qPCR) Immunohistochemistry Evaluation of immunohistochemistry Array based-comparative genomic hybridization array (aCGH) Statistical analysis Results Demographic and clinicopathological features ROCK1 mRNA expression ROCK1 immunostaining ROCK1 expression and patient survival aCGH analysis Discussion Conclusions Abbreviations Competing interests Authors’ contributions Acknowledgments Author details References << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /PageByPage /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Error /CompatibilityLevel 1.4 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails true /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams true /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages true /ColorImageMinResolution 300 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages true /GrayImageMinResolution 300 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages true /MonoImageMinResolution 1200 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier () /PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped /False /CreateJDFFile false /Description << /ARA /BGR /CHS /CHT /CZE /DAN /DEU /ESP /ETI /FRA /GRE /HEB /HRV /HUN /ITA /JPN /KOR /LTH /LVI /NLD (Gebruik deze instellingen om Adobe PDF-documenten te maken voor kwaliteitsafdrukken op desktopprinters en proofers. De gemaakte PDF-documenten kunnen worden geopend met Acrobat en Adobe Reader 5.0 en hoger.) /NOR /POL /PTB /RUM /RUS /SKY /SLV /SUO /SVE /TUR /UKR /ENU (Use these settings to create Adobe PDF documents for quality printing on desktop printers and proofers. Created PDF documents can be opened with Acrobat and Adobe Reader 5.0 and later.) >> /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ << /AsReaderSpreads false /CropImagesToFrames true /ErrorControl /WarnAndContinue /FlattenerIgnoreSpreadOverrides false /IncludeGuidesGrids false /IncludeNonPrinting false /IncludeSlug false /Namespace [ (Adobe) (InDesign) (4.0) ] /OmitPlacedBitmaps false /OmitPlacedEPS false /OmitPlacedPDF false /SimulateOverprint /Legacy >> << /AddBleedMarks false /AddColorBars false /AddCropMarks false /AddPageInfo false /AddRegMarks false /ConvertColors /NoConversion /DestinationProfileName () /DestinationProfileSelector /NA /Downsample16BitImages true /FlattenerPreset << /PresetSelector /MediumResolution >> /FormElements false /GenerateStructure true /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles true /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /NA /PreserveEditing true /UntaggedCMYKHandling /LeaveUntagged /UntaggedRGBHandling /LeaveUntagged /UseDocumentBleed false >> ] >> setdistillerparams << /HWResolution [2400 2400] /PageSize [595.440 793.440] >> setpagedevice