Article Text
PDF
Abstract
Background MicroRNA (miRNA)-200c and miRNA-141 are tumour suppressors, which regulate epithelial-mesenchymal transition (EMT), leading to tumour invasion and metastasis in various malignancies. miRNA-200c and miRNA-141 maintain the epithelial phenotype by post-transcriptionally inhibiting the E-cadherin repressors, zinc finger E-box binding homeobox (ZEB)1 and ZEB2. The present study was performed to determine the prognostic significance of miRNA-200c and miRNA-141, and their association with EMT markers ZEB1, ZEB2 and E-cadherin in eyelid sebaceous gland carcinoma (SGC).
Methods Expression levels of miRNA-200c and miRNA-141 were determined in 42 eyelid SGC cases by quantitative real-time PCR (qPCR). Their association with ZEB1, ZEB2 and E-cadherin was determined by qPCR and immunohistochemistry. Kaplan-Meier plots and Spearman's rank correlation tests were applied to analyse the data. Patients were followed up for 7–44 months.
Results Low expression levels of miRNA-200c and miRNA-141 were seen in 36/42 (86%) and 28/42 (67%) cases, respectively. Low miRNA-200c correlated significantly with large tumour size (p=0.03) and poor differentiation (p=0.03). Low miRNA-141 correlated significantly with large tumour size (p=0.02) and lymph node metastasis (p=0.04). Survival analysis revealed that patients with low miRNA-200c (p<0.05) and miRNA-141 expression (p=0.07) had shorter disease-free survival. There was a significant association of both miRNA-200c and miRNA-141 with E-cadherin and ZEB2 expression.
Conclusions Low levels of miRNA-200c and miRNA-141 in patients with eyelid SGC facilitates tumour progression by promoting EMT and miRNA-200c has emerged as a novel potential predictor of survival.
- Eye Lids
- Pathology
- Experimental – laboratory
- Neoplasia
Statistics from Altmetric.com
Introduction
Sebaceous gland carcinoma (SGC) is a rare and potentially aggressive cutaneous malignancy. It is primarily periorbital (commonly in meibomian and Zeis glands) and is one of the most common neoplasms of the eyelid. It occurs more frequently in Asian populations and in women.1 SGC is known to masquerade as a variety of other ocular conditions, which leads to its delay in diagnosis. Risk factors include advanced age, previous irradiation to head and neck and a genetic predisposition for Muir-Torre syndrome.2 ,3 Poor prognostic factors include both upper and lower eyelids involvement, lymphovascular invasion, poorly differentiated tumours, pagetoid spread, multicentricity and large tumour size (≥20 mm).2
microRNAs (miRNAs) are a class of single-stranded, endogenous, small (17–25 nucleotides in length) non-coding RNAs, which regulate expression at the post-transcriptional level.4 miRNAs are first synthesised as long primary transcripts by RNA polymerase II, in the form of precursors of miRNAs (pri-miRNAs). These pri-miRNAs are cleaved into pre-miRNAs by Drosha and are transported out of the nucleus by exportin-5. In the cytoplasm, pre-miRNAs are further processed by Dicer in the form of mature miRNAs. Mature miRNAs negatively regulate gene expression through base pairing to the 3′ untranslated region (3′UTR) of mRNA, resulting in suppression of translation or degradation of mRNAs. miRNAs regulate various biological processes including cellular differentiation, proliferation and angiogenesis. The contribution of miRNAs towards cancer progression was recently investigated. miRNAs can either act as oncogenes or tumour suppressors through their interactions with target genes critical to the development and progression of various cancers.5 Therefore, miRNAs can be used as novel therapeutic, diagnostic and prognostic markers.
The miRNA-200 family comprises miR-200a, miRNA-200b, miRNA-200c, miRNA-141 and miRNA-429. They are expressed as two different polycistronic pri-miRNA transcripts miR-200b-200a-429 and miR-200c-141. Both miR-200b-200a-429 and miR-200c-141 are located on chromosomes 1 and 12, respectively.6 Recent studies have shown that dysregulation of miRNA-200s plays an essential role in tumour progression by promoting epithelial-mesenchymal transition (EMT).6 EMT is a phenomenon by which epithelial cells lose their cell polarity and cell-cell adhesion, and gain migratory and invasive properties to acquire a mesenchymal phenotype. miRNA-200s repress EMT by directly targeting and downregulating zinc finger E-box binding homeobox (ZEB) family members ZEB1 and ZEB2, via miRNA-200 binding sites located within their 3′UTR regions, resulting in E-cadherin expression and inhibition of cancer progression.7
Downregulation of miRNA-200c and -141 has been earlier reported in various malignancies including breast, liver, ovarian, lung, gastric and renal cancer.8–12 The role of EMT regulators ZEB2 and E-cadherin proteins in eyelid SGC has been reported earlier, where overexpression of ZEB2 and loss of E-cadherin was associated with poor prognosis.13 E-cadherin gene silencing by promoter hypermethylation has also been reported in ocular SGC.14 The present study was designed to determine the status and prognostic significance of miRNA-200c and miRNA-141 in eyelid SGC, and to correlate with clinicopathological high-risk features. Correlation of expression levels of miRNA-200s with E-cadherin, ZEB1 and ZEB2 was also undertaken.
Materials and methods
Patients and tissue samples
Forty-two samples of histopathologically proven eyelid SGC who received surgical treatment between 2011 and 2015 were collected from operation theatre of Dr. Rajendra Prasad Centre for Ophthalmic Sciences, All India Institute of Medical Sciences (AIIMS), New Delhi, India. Control samples were six adjacent normal epidermis resected about 10 cm far from the main tumour. Patients who received preoperative treatment such as chemotherapy or radiotherapy were excluded from the study. Both tumour samples and controls were stored in RNA later solution for overnight and subsequently stored at −80°C until use. The study was conducted after approval from Institute's Ethical Committee, AIIMS (Ref. No. IESC/NP-221/2012) and was carried out in accordance with the Declaration of Helsinki principles. Written consent was obtained from all the patients enrolled in the study and their clinical and radiological details were noted. H&E-stained sections from formalin-fixed paraffin-embedded tumours were reviewed to determine the histological pattern and intraepithelial (pagetoid) spread. Patients were assigned a tumour, node, metastases stage (American Joint Committee on Cancer staging system, 2009, 7th Ed) and were followed-up for 7–44 months (mean 19.81±9.59 months).
RNA isolation
Total RNA was extracted from tumour samples and controls using mirVana miRNA isolation kit (Ambion, Austin, Texas, USA) according to the manufacturer's protocol. RNA concentration and purity of RNA was determined by optical density measurement using a nanodrop spectrophotometer (Thermo Scientific, Wilmington, Delaware, USA). One per cent agarose gel electrophoresis was performed to assess the integrity of RNA samples.
Reverse transcription
To detect the expression levels of miRNA-200c and miRNA-141 in all the tumour samples and normal epidermis, 10 ng of small RNA was reverse transcribed using TaqMan miRNA Reverse Transcription Kit (Applied Biosystems, Foster City, California, USA). RNU48 was used as reference gene. miRNA-specific stem-loop reverse transcription primers for miRNA-200c, miRNA-141 and RNU48 were purchased from Applied Biosystems for complementary DNA (cDNA) synthesis.
cDNA for mRNA analysis of E-cadherin, ZEB1 and ZEB2 was synthesised using SuperScript III First-Strand Synthesis System (Invitrogen, Carlsbad, California, USA), according to manufacturer's protocol. β-Actin was used as reference gene. For this purpose, 1 µg of RNA was used as the starting material. Reverse transcriptase reactions were performed using oligo (dT) primers and SuperScript III reverse transcriptase enzyme (Invitrogen) in 20 μL reaction volume.
Quantitative real-time PCR
qPCR on miRNA-200c and miRNA-141 was performed with TaqMan 2× Universal Master Mix (with no AmpErase uracil N-glycosylase) and TaqMan miRNA assays for mature miRNAs (Applied Biosystems). Assays included hsa-mir-200c (assay ID; 002286), hsa-mir-141 (assay ID; 000463) and reference gene RNU48 (assay ID; 001006). Thermal cycling conditions were as follows: 95°C for 10 min; followed by 40 cycles of 95°C for 15 s and 60°C for 1 min.
E-cadherin, ZEB1 and ZEB2 mRNA levels were quantified by qPCR using SYBR Green Master Mix (Invitrogen). Specific primer pair for E-cadherin, ZEB1, ZEB2 and reference gene β-actin was designed by PrimerBank (Anne T Ferguson) (see online supplementary table S1). All PCR reactions were carried out in a final volume of 20 µL in the presence of ≤1 µg total RNA, 10 µL of SYBR Green Master Mix and 10 pmol of each primer. The PCR conditions of each target are mentioned in online supplementary table S1. Each PCR reaction was followed by continuous melt curve analysis.
supplementary table
Primers and qPCR conditions for E-cadherin, ZEB1 and ZEB2 analysis
All samples were run in triplicates and to assess contamination, a no template control was included in each PCR run. Reactions were performed on StepOne Real-Time PCR Systems (Applied Biosystems). Data were normalised to the reference gene in all the cases and the results were compared with the normalised expression in control (normal skin) to calculate a fold change value.
The 2(−ΔΔCT) method was used to quantify relative amount of miRNA-200c and miRNA-141, where ΔΔCT=ΔCT1–ΔCT2.
ΔCT1=CT of miRNA target in tumour sample−CT of reference gene in tumour sample.
ΔCT2=CT of miRNA target in control−CT of reference gene in control.
Immunohistochemistry
Unstained sections of 3–4 μm thickness were cut on poly-l-lysine-coated slides from formalin-fixed paraffin-embedded blocks and stained immunohistochemically using avidin-biotin indirect method. In brief, after deparaffinisation and rehydration, antigen retrieval was performed in citrate buffer solution (pH 6.0) at 100°C for 20–30 min. Thereafter, endogenous peroxidase activity was blocked by treating the slides with 0.3% hydrogen peroxide in absolute methanol for 30 min, followed by treatment of sections with 1% bovine serum albumin for 30 min to block non-specific reactions. The sections were then incubated with antibody against ZEB2 (Clone AP01369PU-N; 1:400 dilution; Acris Antibodies GmbH, Herford, Germany), ZEB1 (Clone NP_110378.3; 1:250 dilution; Bethyl Laboratories, Montogomery, Texas, USA) and E-cadherin (Clone HECD-1; 1:50 dilution; Abcam, Cambridge, UK). Subsequent incubations were performed using biotinylated secondary antibody and peroxidase-labelled streptavidin (LSAB+System-HRP kit; Dako Cytomation, Glostrup, Denmark). Following incubation, sections were visualised using 3,3′-diaminobenzidine substrate for 3–4 min, counterstained with haematoxylin and visualised by light microscopy. All tests were carried out using appropriate positive and negative controls. High-grade glioma was used as positive control for ZEB213 ,15 and breast carcinoma served as positive control for both ZEB116 and E-cadherin.17 Normal epidermis and normal sebaceous glands within the test samples also served as internal positive controls for E-cadherin immunostaining.
The immunoreactivity scores for ZEB2 and E-cadherin expression were determined as described previously.13 ZEB1 immunostaining was considered positive when >10% tumour cells showed nuclear/cytoplasmic positivity18 and/or >25% stromal cells showed positivity in 10 high-power fields.16
Statistical analysis
Statistical analyses were performed using SPSS V.18.0 for Windows (SPSS, Chicago, Illinois, USA). The expression of all EMT markers were split into high and low expression groups based on relative ratio (R) values, which represents n-fold change in gene expression normalised to an endogenous reference gene and relative to the control. Value of R<2.0 represents low expression of miRNA-200c, miRNA-141 and E-cadherin, whereas R>2.0 represents high expression of ZEB1 and ZEB2 in eyelid SGC cases relative to six controls.19–22 The association between miRNA-200c, miRNA-141 expression levels and clinicopathological variables was analysed using χ2 test. Disease-free survival (length of time after primary treatment that the patient survived without any signs or symptoms like local recurrence or lymph node metastasis) was determined using Kaplan-Meier product-limit method and analysed by log-rank test. Spearman's rank correlation coefficient was used to assess the correlation between gene expression levels of miRNA-200s, E-cadherin, ZEB1 and ZEB2. Results with p<0.05 were considered to be statistically significant and all the tests were two-sided.
Results
Clinicopathological characteristics and outcome
Twenty males and 22 females with a mean age of 58.7±13.9 years (range 30–83 years) were enrolled in the present study. The tumour was most commonly located in the upper eyelid in 30 (71%) patients, and less frequently in the lower eyelid in 11 (26%) patients. Both lids were involved in one case. Twenty-seven (64%) cases presented at early tumour stages IB–II (largest tumour dimension <2 cm) and these underwent excision biopsy. Regional/systemic metastasis was absent in these cases. Fifteen (36%) cases belonged to advanced tumour stages (stage IIIA–IV).
Of the 42 eyelid SGC cases, tumours were graded as large (≥2 cm) in 24 (57%) cases. On radiological examination, the tumour was extended into the orbit in 10 (24%) cases. Frozen section-guided tumour excision was performed to assess the margins for any residual tumour and it was the most common treatment modality in 36 (86%) cases. Map biopsies were performed for all the suspected cases of eyelid SGC with blepheroconjunctivitis-like presentation. Six patients with orbital invasion underwent exenteration. On histopathological examination, 18 (43%) cases showed poor differentiation of tumours and 14 (33%) cases demonstrated pagetoid spread (table 1). None of the patients had any documented risk factors including clinical history of Muir-Torre syndrome or history of irradiation.
Clinicopathological features of eyelid sebaceous gland carcinoma
Patients were followed-up at 3 monthly intervals for 7–44 months (mean 39.21±13.35 months). Ten patients (24%) were found to have regional nodal metastases at the time of presentation. Local recurrence developed in 13 (31%) patients and the resected margins were uninvolved in all the recurrent cases. One patient died of systemic metastasis (table 1).
Expression levels of miRNA-200c and miRNA-141 in eyelid SGC
Expression analysis of tumour suppressors miRNA-200c and miRNA-141 was performed by qPCR. The expression levels of both miRNAs were classified as low or high in relation to R cut-off value of 2.0. Low expression levels of miRNA-200c and miRNA-141 were seen in 36/42 (86%) and 28/42 (67%) cases, respectively. Low expressions of both miRNA-200c and miRNA-141 expression were observed in 25/42 (60%) cases.
Association of miRNA-200c and miRNA-141 expression with clinicopathological parameters of eyelid SGC
Low miRNA-200c expression significantly associated with clinicopathological high-risk features including large tumour size (p=0.03) and poor differentiation of tumours (p=0.03). A borderline significant association of low miRNA-200c expression was also observed with advanced tumour stages (p=0.05) and pagetoid spread (p=0.08) (table 2). Low expression of miRNA-141 significantly correlated with large tumour size (p=0.02) and lymph node metastasis (p=0.04) (table 2).
Association between miRNA-200c, miRNA-141 and clinicopathological characteristics of eyelid SGC
Prognostic value of low miRNA-200c and miRNA-141 expression in eyelid SGC
Disease-free survival curves were plotted according to miRNA-200c and miRNA-141 expression levels by Kaplan-Meier method (figure 1A, B). Patients with low miRNA-200c expression had significantly reduced disease-free survival (p<0.05). However, low miRNA-141 expressing group of patients showed a trend towards reduced disease-free survival (p=0.07).
Kaplan-Meier estimates of disease-free survival in patients with eyelid sebaceous gland carcinoma. (A) Disease-free survival based on microRNA (miRNA)-200c expression. (B) Disease-free survival based on miRNA-141 expression.
Association between miRNA-200c, miRNA-141 and other EMT markers
Expression analysis of E-cadherin, ZEB2 and ZEB1, which are known targets of miRNA-200s during EMT, was performed by qPCR using comparative 2(−ΔΔCT) method and immunohistochemistry. Low E-cadherin and high ZEB2 mRNA expression was observed in 27/42 (64%) and 30/42 (71%) cases, respectively. High expression levels of ZEB1 mRNA was observed only in 6/42 (14%) of eyelid SGC cases. Immunohistochemical analysis of ZEB2 revealed cytoplasmic overexpression in 29/42 (69%) (figure 2A) as compared with the normal sebaceous glands, which showed absence of ZEB2 immunoexpression (figure 2B). This positivity was observed in the central as well as the peripheral part of the tumour lobules. On the contrary, ZEB1 immunoexpression was observed in the stromal cells surrounding the tumour in 9/42 (21%) cases (figure 2C). Furthermore, membranous loss of E-cadherin was observed in 28/42 (67%) cases (figure 2D) as compared with the normal epidermis and normal sebaceous glands, which showed strong membranous positivity for E-cadherin immunoexpression (figure 2E, F).
Immunoexpression of epithelial-mesenchymal transition markers using avidin-biotin method. (A) Strong cytoplasmic positivity of zinc finger E-box binding homeobox (ZEB)2 in an eyelid sebaceous gland carcinoma (SGC) case. (B) Absence of ZEB2 immunoexpression in normal sebaceous glands (C) ZEB1 immunopositivity in stromal cells of eyelid SGC. (D) Loss of E-cadherin expression in a case of eyelid SGC. (E) Membranous positivity of E-cadherin in normal epidermis. (F) Membranous positivity of E-cadherin in normal sebaceous glands. Original magnification: (A and B) ×200; (C–F) ×400.
Spearman's correlation was calculated to determine the association between miRNA-200c, miRNA-141 and their mRNA targets ZEB1, ZEB2 and E-cadherin. Both miRNA-200 family members, miRNA-200c and miRNA-141, showed significant positive correlations (r=0.353, p=0.02) with each other (table 3). Notably, E-cadherin correlated with miRNA-200c (r=0.42; p=0.006) and miRNA-141 (r=0.46; p=0.002) depicting their regulatory effect on E-cadherin expression. Also, a significant inverse correlation was observed between miRNA-141 and ZEB2 expression (r=−0.327, p=0.03) suggesting a role of ZEB2 in direct suppression of miRNA-200s (table 3). However, ZEB1 expression did not correlate with miRNA-200c or miRNA-141 expression.
Spearman's rank correlation coefficient and associated p values between the expression of miRNA-200c, miRNA-141, ZEB family and E-cadherin
Furthermore, on correlating miRNA-200c and miRNA-141 expression with protein levels of other EMT markers, both miRNA-200c (p=0.03) and miRNA-141 (p<0.001) showed significant inverse association with ZEB2 immunoexpression, further validating an important role of ZEB2 expression in regulating miRNA-200c and miRNA-141 in eyelid SGC (table 4). Also, a direct significant association between miRNA-200c expression and E-cadherin protein expression (p=0.03) was observed (table 4). These results indicate that the expression of miRNA-200c and miRNA-141 play a central role in the regulation of EMT and in the malignant progression of eyelid SGC.
Association of miRNA-200c and miRNA-141 with other EMT markers
Discussion
The present study demonstrates aberrant expression of miRNA-200 family in eyelid SGC. The miRNA-200 family exhibits a tumour suppressor function and is a powerful regulator of the EMT process, which plays an essential role in tumour progression and metastasis. Downregulation of this family has been associated with cancer invasion and metastasis in several malignancies.8–12 The present study reveals low expression levels of miRNA-200c and miRNA-141 in 86% and 67% cases of eyelid SGC, respectively. This observation is in line with recent studies suggesting a lower expression of miRNA-200c in renal cell carcinoma and colorectal cancer.12 Similar studies suggesting low expression of miRNA-141 are available for bladder cancer23 and renal cell carcinoma.24
Low miRNA-200c expression in eyelid SGC correlated with clinicopathological high-risk features including large tumour size (≥2 cm), poor differentiation of tumour, advanced tumour stages and pagetoid spread. Similar significant association of low miRNA-200c expression levels with tumour stage and large tumour size is also reported in gastric cancer and bladder cancer.11 ,25 Low expression of miRNA-141 in the present study also had a significant association with large tumour size and lymph node metastasis. Analogous observation has been recently reported for hepatocellular carcinoma, where low miRNA-141 expression showed a significant association with lymph node metastasis.23
Furthermore, low miRNA-200c in this study showed significant association with shorter disease-free survival, which is in accordance with previously published reports on gastric cancer, epithelial ovarian cancer, renal cell carcinoma, breast and colorectal cancer.9 ,11 ,24 ,26 However, low miRNA-141 expression in our patients with SGC showed a borderline significance with poor disease-free survival. A correlation between low miRNA-141 expression levels and worse prognosis has been suggested for renal cell carcinoma, breast cancer and hepatocellular carcinoma.24 ,26 ,27
Our study reveals a significant association between miRNA-200c and miRNA-141 expression levels in eyelid SGC. Wang et al12 and Yoshino et al28 also showed a significant correlation between both the miRNAs in renal cell carcinoma. The mechanism underlying the downregulation of miRNA-200c/miRNA-141 has been well studied in various malignancies. The double-negative feedback loop controlling ZEB1/ZEB2 expression is the best known mechanism of miR-200c/miR-141 regulation.11 In the present study, ZEB1 overexpression was found in 21% cases of eyelid SGC by immunohistochemistry. However, this positivity was only seen the stromal cells surrounding the tumour lobules. Similar findings with ZEB1 positivity in the stromal cells have been reported in breast carcinoma,16 pancreatic adenocarcinomas29 and low-grade endometrial adenocarcinomas.30 Since EMT is defined as a phenomenon where tumour cells lose their epithelial features, thus a proportion of the cells in the stromal tissue might represent process of transformation in the tumour microenvironment. Furthermore, ZEB1 expression did not show any significant role in regulation of miRNA-200c and miRNA-141 expression. Tuomarila et al31 also demonstrated similar findings in breast cancer, where ZEB1 expression did not correlate with miRNA-200c expression. However, ZEB2 mRNA expression showed a significant inverse correlation with miRNA-141 expression. Also, there was an inverse significant association of ZEB2 protein with miRNA-200c and miRNA-141 expression, suggesting that ZEB2 triggers a miRNA-mediated feedforward loop that stabilises EMT and promotes tumour progression and invasion. This aberrant expression of miRNA-200s and ZEB2 promotes EMT by post-transcriptionally repressing E-cadherin. Our study also showed a significant association between both the miRNAs and E-cadherin mRNA, which is in accordance with previously published report on renal cell carcinoma28 and ovarian cancer.7 The direct significant association of miRNA-200c expression with E-cadherin immunoexpression further validates their important regulatory function in promoting EMT in eyelid SGC. Other than ZEB1/ZEB2 controlling the expression of miRNA-200c/miRNA-141 by a double-negative feedback loop, transforming growth factor β/miRNA-200 signalling network also regulates the establishment and maintenance of EMT.32 Furthermore, DNA methylation of CpG islands in the promoter region miRNA-200c/miRNA-141 is associated with EMT.33
In conclusion, our results suggest an important role of EMT-related miRNA-200 family members (miRNA-200c and miRNA-141) as tumour suppressors; low expression of these is associated with tumour progression and poor prognosis of patients with eyelid SGC. This is the first study to explore the status of miRNAs in SGC of the eyelid. The association between miRNA-200s and other EMT regulators E-cadherin and ZEB2 suggests their important contribution in tumour progression of eyelid SGC. Low miRNA-200c expression emerged as a reliable marker to predict survival and prognosis in patients with eyelid SGC.
Acknowledgments
The authors are very grateful to Mr Vijay Kumar Singh (Department of Ocular Pathology, AIIMS) for his excellent technical assistance.
References
Footnotes
Contributors MB: data collection, analysis, interpretation and writing. SS, KC and AS: conception and design. SK, NP, MSB and SB: acquisition of clinical data.
Funding Supported by All India Institute of Medical Sciences (grant F.5-59/IRG/2010/RS) and Indian Council of Medical Research (grant 3/2/2/240/2014-NCD-III) for financial support.
Competing interests None declared.
Patient consent Obtained.
Ethics approval This study was conducted after approval from the Institute Ethics Committee, AIIMS, New Delhi (Ref. No. IESC/NP-221/2012) and carried out in accordance with Declaration of Helsinki principles. Informed consent was obtained from all patients participating in this study.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- At a glance