The Expression of Egfl7 in Human Normal Tissues and Epithelial Tumors
Abstract
Background and aims
To investigate the expression of Egfl7 in normal adult human tissues and human epithelial tumors.
Methods
RT-PCR and Western blot were employed to detect Egfl7 expression in normal adult human tissues and 10 human epithelial tumors including hepatocellular carcinoma (HCC), lung cancer, breast cancer, prostate cancer, colorectal cancer, gastric cancer, esophageal cancer, malignant glioma, ovarian cancer and renal cancer. Immunohistochemistry and cytoimmunofluorescence were subsequently used to determine the localization of Egfl7 in human epithelial tumor tissues and cell lines. ELISA was also carried out to examine the serum Egfl7 levels in cancer patients. In addition, correlations between Egfl7 expression and clinicopathological features as well as prognosis of HCC and breast cancer were also analyzed on the basis of immunohistochemistry results.
Results
Egfl7 was differentially expressed in 19 adult human normal tissues and was overexpressed in all 10 human epithelial tumor tissues. The serum Egfl7 level was also significantly elevated in cancer patients. The increased Egfl7 expression in HCC correlated with vein invasion, absence of capsule formation, multiple tumor nodes and poor prognosis. Similarly, upregulation of Egfl7 in breast cancer correlated strongly with TNM stage, lymphatic metastasis, estrogen receptor positivity, Her2 positivity and poor prognosis.
Conclusions
Egfl7 is significantly upregulated in human epithelial tumor tissues, suggesting Egfl7 to be a potential biomarker for human epithelial tumors, especially HCC and breast cancer.
Introduction
The global incidence and mortality of cancer have risen sharply in recent years. Almost 5% of the population is diagnosed with cancer each year, and this disease is responsible for about 25% of all deaths (1). Despite progress in clinical diagnosis, surgical techniques, preoperative management and comprehensive treatment of cancer, the 5-year survival rate of patients with cancer remains unsatisfactory mainly because of a scarcity of useful tumor biomarkers (2, 3). Thus, research into novel biomarkers for cancer detection and prognosis assessment has been undertaken to reduce cancer deaths (4-8).
Epidermal growth factor-like domain 7 (Egfl7) is a novel gene that was first identified in endothelial cells (ECs) of the developing mouse embryo (9). It was initially regarded as an endothelial cell-specific gene and an important regulator in tubulogenesis during embryonic development (9-11). However, recent studies have demonstrated that Egfl7 can also be found in other cells such as primordial germ cells (12). The Egfl7 gene encodes a 30-kDa secreted protein that contains a signal sequence, an EMI domain at the amino terminus, followed by 2 EGF-like domains and a leucine- and valine-rich carboxyterminal region (9, 10). Egfl7 protein has been found to act as a chemoattractant for cell migration and could regulate collective migration of ECs, indicating that Egfl7 may be important in cell motility (13, 14). Egfl7 is conserved across species (9). In mouse, Egfl7 expression is high during embryonic development, downregulated in almost all mature tissues except for lung, heart and kidney, and strongly upregulated in many proliferative tissues, including tumors (9, 10). This special expression pattern, which is similar to the embryo proteins alpha-fetoprotein (AFP) and carcinoembryonic antigen (CEA), indicates the potential of Egfl7 as a novel tumor biomarker (15-17).
In our previous study, we showed that Egfl7 was significantly upregulated in human hepatocellular carcinoma (HCC) tissues and the increase in Egfl7 expression was closely related to the poor prognosis of HCC (18). Interestingly, Egfl7 was found predominantly in HCC cells but not in ECs. Recently, increased Egfl7 expression was also reported to significantly increase in human colon cancer and laryngeal squamous cell carcinoma and closely correlate with lymph node metastasis, advanced tumor stage and poor prognosis (19-21). These results further indicated Egfl7 as a potential tumor biomarker and shed light on the role of Egfl7 in other human tumors. However, the expression patterns of Egfl7 in human tumor tissues as well as normal tissues remain unclear.
Thus, in the present study, we have examined the expression of Egfl7 in 19 normal adult human tissues, 10 human epithelial cancer tissues and the corresponding tumor cell lines. Serum of healthy donors and patients with these 10 epithelial tumors was also assayed for Egfl7 levels. Furthermore, we have analyzed the relationship between Egfl7 expression and clinicopathological features as well as prognosis of HCC and breast cancer, which provides novel evidence for Egfl7 to be a potential human tumor biomarker.
Materials and Methods
Patients and Samples
We obtained fresh tissue specimens from 36 HCC, 30 lung cancer, 30 breast cancer, 4 prostate cancer, 30 colorectal cancer, 30 gastric cancer, 5 esophageal cancer, 3 malignant glioma, 3 ovarian cancer and 6 renal cancer cases by surgical resection at the Xiangya Hospital from October 2007 to December 2008. All tumor samples were matched with adjacent nontumorous tissues (ANT). The specimens were frozen in liquid nitrogen immediately after resection, and then stored at −80°C for RT-PCR and Western blotting analysis. In addition, paraffin-embedded specimens from 132 cases of HCC and 102 cases of breast cancer were obtained from June 2000 to June 2007 and from June 1994 to June 2007, respectively. Paraffin-embedded specimens from 30 cases of lung cancer, 8 cases of prostate cancer, 30 cases of colorectal cancer, 30 cases of gastric cancer, 10 cases of esophageal cancer, 8 cases of malignant glioma, 6 cases of ovarian cancer and 16 cases of renal cancer were also obtained from June 2006 to June 2008. All these paraffin-embedded specimens were obtained from patients who underwent surgical resection at the Xiangya Hospital and diagnoses were verified by histopathological examination. Nineteen normal adult human tissues (brain, retina, heart, lung, thyroid, liver, pancreas, spleen, stomach, duodenum, intestine, colon, ureter, bladder, prostate, testicle, muscle, portal vein and aorta) from 4 postmortem cases within 3 hours of death were also obtained from the Xiangya Hospital. This study and the use of all clinical tissue specimens mentioned were approved by the Ethical Committee of Xiangya Hospital.
RNA isolation and RT-PCR
RNA from tissues was isolated using TRIzol reagent (Invitrogen, CA, USA). RT-PCR was used to determine the expression of mRNA for Egfl7. Total RNA (2 μg) was reverse transcribed to cDNA by M-MLV reverse transcriptase with oligo-dT primers (Promega, Madison, USA). PCR was performed using the resulting cDNA, with the following conditions: 35 cycles with 95°C denaturation for 5 minutes, 95°C denaturation for 30 seconds, 55°C annealing for 30 seconds, 72°C elongation for 30 seconds and 72°C extension for 5 minutes. The following Egfl7 primers were used: forward, 5′-CAACCCGACAGGAGTGGACAGT-3′; reverse, 5′-TCACGAGTCTTTCTTGCAGGAGC-3′. GAPDH expression was determined as a control using the following primers: forward, 5′-CTGCAGCATCTTCTCCTTCC-3′; reverse, 5′-CAAAGTTGTCATGGATGACC-3′. The PCR product (5 μL) was then electrophoresed on 2.0% agarose gel and the intensity of bands was quantified by the Eagle Eye II laser densitometry program (Strategene, La Jolla, USA). Egfl7 gene expression was compared by the relative yield of PCR product from the target sequence to that from GAPDH gene. PCR experiments were done in triplicate.
Western blotting
We performed immunoblotting to determine the expression levels of Egfl7 proteins. The total protein was extracted from fresh tissues and cells. SDS-PAGE (5% stacking gel and 12% separating gel) was used to separate 100 μg of total protein. Proteins were transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA). Rabbit antihuman Egfl7 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, 1:150) was used as the primary antibody. Horseradish peroxidase-conjugated goat antirabbit secondary antibody (KPL, Gaithersburg, MD; 1:6000) was used as secondary antibody. Beta-actin protein was also determined by using the specific antibody (Sigma, St Louis, MO) as a loading control. Antigen-antibody complexes were visualized by the enhanced chemiluminescence (ECL) system (Pierce, Rockford, IL). All experiments were carried out in triplicate.
Immunohistochemistry
Sections from paraffin-embedded tissues were cut into 5-μm-thick sections. Antigen retrieval was exposed and repaired by microwave treatment for 10 minutes with EDTA buffer (1 mM, pH 8.0). Then the sections were blocked in goat serum at 37°C for 30 minutes before being incubated with the primary antibody, mouse anti-Egfl7 (Santa Cruz Biotechnology, 1:150) overnight at 4°C. Subsequently, the sections were incubated with biotinylated goat anti-mouse IgG at 37°C for 30 minutes, followed by streptavidin-biotin complex (Zhongshan Goldenbridge Biotechnology, Beijing, China) for an additional 30 minutes at 37°C. The peroxidase assay was performed by the addition of 0.5 mg/mL 3, 3-diaminobenzidine tetrahydrochloride (DAB, Zhongshan Golden Bridge Biotechnology) for 3 minutes. The sections were then counterstained with Mayer's hematoxylin. Negative control slides were probed with normal mouse serum under the same experimental conditions. The immunohistochemical staining was examined by 2 independent pathologists and scored to using a 4-point scale according to the percentage of positive cancer cells (18): 0 (≤10% positive); 1+ (11%-25% positive); 2+ (26%-50% positive); 3+ (≥51% positive). The protein expression of Egfl7 was thus defined as low expression if scored 0 or 1+ while a score of 2+ or 3+ was defined as high expression.
Cell culture
The human cancer cell lines used in this study included HCC cell HCCLM3, lung carcinoma cell A549, breast cancer cell MCF-7, prostatic carcinoma cell PC-3, colon carcinoma cell SW480, gastric adenocarcinoma cell MGC803, esophageal cancer cell Eca-109, malignant glioma cell U251, ovarian cancer cell HO8910 and renal carcinoma cell KC. HCCLM3, A549, MGC803, U251 and PC-3 were preserved in our laboratory; MCF-7, SW480, HO8910 and KC cells were generously provided by the Cell Center of Xiangya Medical School, Central South University; Eca-109 was purchased from the Cell Bank of the Chinese Academy of Sciences. All cells were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM; GIBCO BRL, Gaithersburg, MD) supplemented with 10% fetal bovine serum (FBS; HyClone, Logan, UT) and incubated in 5% CO2 at 37°C.
Cytoimmunofluorescence
Tumor cells were washed 3 times with phosphate-buffered saline (PBS) (pH 7.4) and subsequently fixed in 4% paraformaldehyde at room temperature for 30 minutes, followed by washing twice with PBS. Afterwards, cells were permeabilized with 1% Triton X-100 in PBS at room temperature for 30 minutes, after which we blocked the cells with normal goat serum for 30 minutes and incubated them with rabbit anti-Egfl7 polyclonal antibody (Santa Cruz Biotechnology, CA, 1:50) for 1 hour at 37°C. After incubation, cells were washed 3 times with PBS and then incubated with fluorescein isothiocyanate-conjugated goat anti-rabbit secondary antibody (Jingmei Biotechnology, China, 1:2000) for 1 hour at room temperature. SlowFade mounting medium was added to the slides and fluorescence was analyzed by fluorescent microscopy (Nikon, Tokyo, Japan).
Serum samples
Serum samples were collected with informed consent from 286 healthy donors (obtained from the Changsha Blood Center) from October 2005 to March 2009. There were also 390 patients with epithelial tumors including 78 cases of HCC, 47 cases of breast cancers, 46 cases of lung cancers, 65 cases of gastric cancers, 97 cases of colorectal cancers, 29 cases of renal cancers, 11 cases of malignant glioma, 18 cases of esophageal cancers, 17 cases of prostate cancers and 15 cases of ovarian cancers (obtained from Xiangya Hospital) from October 2005 to March 2009. Blood samples were drawn from subjects prior to initiation of treatment and were spun and aliquoted. Serum was stored at −80°C until testing.
ELISA
Serum Egfl7 level was quantitatively measured by a sandwich ELISA system. First, a monoclonal antibody (Abnova, Taiwan, China) specific to Egfl7 was added to a 96-well microplate (Greiner Bio-One, Germany) as a capture antibody and incubated for 2 hours at room temperature. The 96-well microplate was then coated at 4°C overnight. The coated plate was washed in 0.05% phosphate-buffered saline/tween (PBST). After washing away any unbound antibody, 4% bovine serum albumin (BSA) was added to the wells and incubated for 2 hours at 4°C for blocking. After washing with 0.05% PBST, 250-fold diluted sera were added to the wells and incubated for 2 hours at room temperature. After washing away any unbound substances, a polyclonal antibody (Santa Cruz Biotechnology) specific for Egfl7 was added to the wells as a detection antibody and incubated for 2 hours at room temperature. After a wash to remove any unbound antibody, goat anti-rabbit IgG antibody was added to the wells and incubated for 2 hours. After a wash, TMB Soluble Reagent (Tiangen Biotech Co., Ltd., Beijing, China) was added to the wells and allowed to react for 30 minutes at room temperature. The reaction was stopped by adding 1 M H2SO4 (Jingmei Biotech Co., Ltd.). Color intensity was determined by a photometer at a wavelength of 450 nm, with a reference wavelength of 570 nm. Recombinant protein Egfl7 (Abnova, Taiwan, China) was used as a standard sample in each assay.
Follow-up study
Follow-up data were obtained by reviewing the hospital records or through direct communication with the patients after surgical resection (132 HCC patients and 102 breast cancer patients). The follow-up period was defined from the date of surgical excision of the tumor to the date of death or the last follow up. Deaths from other causes were treated as censored cases. Recurrence and metastasis were diagnosed by clinical examination, serial AFP level mensuration, and ultrasonography or computed tomography. To determine factors influencing survival after operation, 8 conventional variables together with Egfl7 expression were tested in all 132 HCC patients: age, gender, cirrhosis, Edmondson-Steiner grade, capsule, size of the tumor, the number of tumor nodes and vein invasion; 9 conventional variables together with Egfl7 expression were also tested in all 102 breast cancer patients: age, size of the tumor, pathological types, histological grade, TNM stage, lymphatic metastasis, estrogen receptor (ER), progesterone receptor (PR) and human epidermal growth factor receptor 2 (Her2).
Statistical analysis
Experimental results were expressed as mean values ± standard deviation (mean±SD). Statistical analyses for RT-PCR and Western blotting were performed with the independent t-test for comparing all pairs of groups (SPSS software, version 13.0). Spearman correlation analysis was used to analyze the relationship between Egfl7 expression levels and clinicopathological variables of HCC as well as breast cancer patients. Kaplan-Meier curves were calculated for Egfl7 expression and differences in survival times among patient subgroups were analyzed using the log-rank test. Univariate and multivariate analyses were performed with the Cox proportional hazard regression model to determine associations between clinicopathological variables and cancer-related mortality. All tests were 2-tailed and p<0.05 was considered statistically significant.
Results
Egfl7 expressed differentially in normal adult human tissues
To determine the Egfl7 expression in normal adult human tissues, we used semiquantitative RT-PCR as well as Western blot to measure Egfl7 mRNA and protein in 19 adult human normal tissues from 4 postmortem cases and found that Egfl7 expressed differentially in normal human adult tissues. High Egfl7 expression was found in lung, heart, brain and retina tissues, and moderate Egfl7 expression was observed in thyroid, pancreas, spleen, ureter, bladder, prostate, testicle, muscle, portal vein and aorta tissues. The lowest Egfl7 levels were detected in digestive system tissues including liver, stomach, duodenum, intestine and colon (Fig. 1 and Tab. I).
| Egfl7 expression levels (mean±SD) | |||
|---|---|---|---|
| Normal adult human tissues | N | mRNA levels | Protein levels |
| Brain | 4 | 0.28±0.02 | 0.33±0.02 |
| Retina | 4 | 0.27±0.02 | 0.31±0.03 |
| Heart | 4 | 0.30±0.02 | 0.35±0.02 |
| Lung | 4 | 0.33±0.03 | 0.39±0.03 |
| Thyroid | 4 | 0.18±0.01 | 0.21±0.02 |
| Liver | 4 | 0.07±0.02 | 0.08±0.02 |
| Pancreas | 4 | 0.19±0.02 | 0.23±0.01 |
| Spleen | 4 | 0.20±0.01 | 0.23±0.02 |
| Stomach | 4 | 0.04±0.02 | 0.04±0.02 |
| Duodenum | 4 | 0.05±0.01 | 0.06±0.01 |
| Intestine | 4 | 0.05±0.02 | 0.06±0.02 |
| Colon | 4 | 0.04±0.02 | 0.04±0.02 |
| Ureter | 4 | 0.15±0.02 | 0.18±0.03 |
| Bladder | 4 | 0.15±0.01 | 0.18±0.02 |
| Prostate | 4 | 0.14±0.02 | 0.17±0.02 |
| Testicle | 4 | 0.13±0.01 | 0.15±0.01 |
| Muscle | 4 | 0.16±0.02 | 0.18±0.03 |
| Portal vein | 4 | 0.16±0.02 | 0.18±0.03 |
| Aorta | 4 | 0.14±0.02 | 0.17±0.02 |
Egfl7 overexpressed in human epithelial tumors
To examine the expression of Egfl7 in 10 human epithelial tumor tissue samples, we employed semiquantitative RT-PCR and Western blot to measure its mRNA and protein levels in tumor tissues and corresponding ANTs. Our results showed that the Egfl7 mRNA and protein levels were significantly upregulated in all tumor tissues we tested, especially in HCC and breast cancer (Fig. 2 and Tab. II).
| Malignant tumor | Egfl7 mRNA levels (mean±SD) | Egfl7 protein levels (mean±SD) | ||||||
|---|---|---|---|---|---|---|---|---|
| N | T | ANT | p value | T | ANT | p value | ||
| HCC | 36 | 0.69±0.13 | 0.12±0.05 | <0.001 | 0.83±0.16 | 0.16±0.06 | <0.001 | |
| Breast cancer | 30 | 0.58±0.09 | 0.21±0.05 | <0.001 | 0.71±0.10 | 0.26±0.05 | <0.001 | |
| Lung cancer | 30 | 0.64±0.20 | 0.46±0.18 | 0.001 | 0.81±0.26 | 0.59±0.23 | 0.001 | |
| Malignant glioma | 3 | 0.61±0.14 | 0.30±0.04 | 0.023 | 0.78±0.12 | 0.35±0.04 | 0.004 | |
| Ovarian cancer | 3 | 0.48±0.11 | 0.16±0.04 | 0.009 | 0.59±0.15 | 0.19±0.05 | 0.012 | |
| Prostate cancer | 4 | 0.48±0.09 | 0.17±0.04 | 0.003 | 0.62±0.11 | 0.22±0.03 | 0.002 | |
| Renal cancer | 6 | 0.55±0.11 | 0.25±0.04 | 0.001 | 0.69±0.14 | 0.32±0.05 | 0.001 | |
| Esophageal cancer | 5 | 0.45±0.07 | 0.09±0.04 | 0.001 | 0.56±0.08 | 0.12±0.04 | 0.001 | |
| Gastric cancer | 30 | 0.39±0.06 | 0.08±0.03 | 0.001 | 0.48±0.07 | 0.10±0.04 | <0.001 | |
| Colorectal cancer | 30 | 0.42±0.05 | 0.10±0.02 | 0.001 | 0.52±0.06 | 0.12±0.02 | <0.001 | |
N, sample size; T, tumor; ANT, adjacent nontumorous tissues; HCC, hepatocellular carcinoma
Localization of Egfl7 protein in human epithelial tumor tissues and cell lines
Immunohistochemistry was subsequently employed to explore the localization of Egfl7 protein in human epithelial tumor tissues. Our results showed that Egfl7 protein mainly expressed in tumor cells, although weak Egfl7 expression could also be found in ECs (Fig. 3A, 5A and 6A). To further examine the subcellular localization of Egfl7 protein in cell lines of these epithelial tumors, cytoimmunofluorescence was also carried out and our data showed that Egfl7 is mainly located in the cytoplasm of these tumor cells, suggesting that tumor cells could express Egfl7 by themselves (Fig. 3B).
Serum levels of Egfl7 in healthy donors and cancer patients
Because Egfl7 encodes a secreted protein, we established a sandwich ELISA system to clarify whether Egfl7 could be detected in human serum samples. We tested the serum specimens of 286 healthy donors and 390 patients with 10 types of malignant tumors. Our results showed that Egfl7 could be found in all healthy individuals and the median and 95% confidence interval (CI) of serum Egfl7 levels in healthy individuals were 1082.12 (7.50-4660.58) ng/mL. On the other hand, Egfl7 was also detectable in serum samples from the cancer patients and the serum Egfl7 levels in cancer patients were significantly higher than in healthy donors (Fig. 4 and Tab. III). Interestingly, the serum Egfl7 levels in HCC patients, the median and 95% CI of which were 4028.85 (539.47-9510.97) ng/mL, were the highest among all the cancer patients we measured.
| Group | N | Serum Egfl7 levels (ng/mL, median and 95% CI) | p value* |
|---|---|---|---|
| Malignant glioma | 11 | 2718.16 (489.02-4109.91) | 0.004 |
| Breast cancer | 47 | 3265.04 (12.62-9954.90) | <0.001 |
| Esophageal cancer | 18 | 2388.46 (183.84-5386.36) | 0.002 |
| Lung cancer | 46 | 3558.53 (17.09-11505.05) | <0.001 |
| Gastric cancer | 65 | 2158.84 (23.09-8878.84) | <0.001 |
| Colorectal cancer | 64 | 2038.53 (65.49-6086.73) | <0.001 |
| HCC | 78 | 4028.85 (539.47-9510.97) | <0.001 |
| Renal cancer | 29 | 3052.40 (14.31-8838.46) | <0.001 |
| Ovarian cancer | 15 | 2803.51 (104.17-3513.57) | 0.007 |
| Prostate cancer | 17 | 2065.37 (110.16-8844.56) | 0.011 |
| Healthy donors | 286 | 1082.12 (7.50-4660.58) |
*
Compared with healthy donors
Egf17 expression levels were correlated with clinicopathological characteristics and prognosis of HCC
To further explore the role of Egfl7 in HCC, we analyzed the correlations between Egfl7 expression in 132 cases of HCC and the clinicopathological features as well as prognosis of these HCC patients. Our results showed that Egfl7 expression was correlated closely with vein invasion (p=0.020), multiple nodes (p=0.036) and capsule formation (p=0.026) (Tab. IV). Based on the immunohistochemistry results, HCC patients were divided into a low Egfl7 expression group (immunohistochemistry score 0 and 1+; n=57) and a high Egfl7 expression group (immunohistochemistry score 2+ and 3+; n=75). Patients in the high Egfl7 expression group had significantly shorter disease-free survival and overall survival than those in the low Egfl7 expression group (median disease-free survival time 315 vs 560 days, p=0.007; median overall survival time 410 vs 640 days, p=0.001; Fig. 5B and 5C). By multivariable Cox regression analysis, high Egfl7 expression (relative risk [RR] 2.256; p=0.001), venous invasion (RR 1.939; p=0.004) and multiple nodes (RR 1.828; p=0.018) were found to be independent prognostic factors for overall survival of HCC (Tab. V).
| Egfl7 expression levels | ||||||
|---|---|---|---|---|---|---|
| Clinicopathological parameter | N | 0 | 1+ | 2+ | 3+ | p value |
| Gender | ||||||
| Male | 116 | 11 | 39 | 51 | 15 | |
| Female | 16 | 4 | 3 | 6 | 3 | 0.791 |
| Age | ||||||
| ≤60 years | 107 | 12 | 35 | 46 | 14 | |
| >60 years | 25 | 3 | 7 | 11 | 4 | 0.726 |
| Liver cirrhosis | ||||||
| Presence | 89 | 9 | 32 | 34 | 14 | |
| Absence | 43 | 6 | 10 | 23 | 4 | 0.896 |
| Tumor size | ||||||
| ≤5 cm | 46 | 6 | 17 | 20 | 3 | |
| >5 cm | 86 | 9 | 25 | 37 | 15 | 0.128 |
| Tumor nodule number | ||||||
| Multiple (≥2) | 49 | 6 | 9 | 24 | 10 | |
| Solitary | 83 | 9 | 33 | 33 | 8 | 0.036 |
| Capsule formation | ||||||
| Presence | 54 | 9 | 20 | 20 | 5 | |
| Absence | 78 | 6 | 22 | 37 | 13 | 0.026 |
| Edmondson Steiner grade | ||||||
| I-II | 77 | 6 | 30 | 31 | 10 | |
| III-IV | 55 | 9 | 12 | 26 | 8 | 0.709 |
| Vein invasion | ||||||
| Presence | 71 | 6 | 18 | 35 | 12 | |
| Absence | 61 | 9 | 24 | 22 | 6 | 0.020 |
| Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|
| Variables | N | RR (95% CI) | p value | RR (95% CI) | p value |
| Gender | |||||
| Male | 116 | 1 | |||
| Female | 16 | 0.873(0.464-1.641) | 0.673 | 0.806 (0.405-1.603) | 0.539 |
| Age | |||||
| ≤60 years | 107 | 1 | 1 | ||
| >60 years | 25 | 1.516(0.911-2.522) | 0.110 | 1.184 (0.699-2.005) | 0.529 |
| Liver cirrhosis | |||||
| Absence | 43 | 1 | 1 | ||
| Presence | 89 | 1.272(0.822-1.968) | 0.280 | 1.263 (0.759-2.103) | 0.369 |
| Tumor size | |||||
| ≤5 cm | 46 | 1 | 1 | ||
| >5 cm | 86 | 1.042(0.674-1.610) | 0.853 | 0.818 (0.507-1.321) | 0.412 |
| Capsule formation | |||||
| Presence | 54 | 1 | 1 | ||
| Absence | 78 | 1.601(1.049-2.444) | 0.029 | 1.519(0.934-2.472) | 0.092 |
| Edmondson-Steiner grade | |||||
| I-II | 77 | 1 | 1 | ||
| III-IV | 55 | 1.185(0.779-1.803) | 0.427 | 1.339 (0.845-2.121) | 0.214 |
| Vein invasion | |||||
| Absence | 61 | 1 | 1 | ||
| Presence | 71 | 2.032(1.345-3.068) | 0.001 | 1.939 (1.232-3.053) | 0.004 |
| Tumor nodule number | |||||
| Solitary | 83 | 1 | 1 | ||
| Multiple (≥2) | 49 | 1.835(1.209-2.785) | 0.004 | 1.828 (1.109-3.013) | 0.018 |
| Egfl7 expression | |||||
| Low | 57 | 1 | 1 | ||
| High | 75 | 2.060(1.345-3.154) | 0.001 | 2.256 (1.397-3.645) | 0.001 |
Egf17 expression levels were correlated with clinicopathological characteristics and prognosis of breast cancer
The immunohistochemical staining results showed that upregulation of Egfl7 protein expression in breast cancer was strongly correlated with TNM stage (p=0.011), lymphatic metastasis (p=0.006), ER (p=0.033) and Her2 (p=0.025) (Tab. VI). To examine the correlation between the Egfl7 expression level and prognosis, breast cancer patients were divided into a low Egfl7 expression group (immunohistochemistry score 0 and 1+; n=45) and a high Egfl7 expression group (immunohistochemistry score 2+ and 3+; n=57). Patients in the high Egfl7 expression group had shorter disease-free survival and overall survival than those in the low Egfl7 expression group (median disease-free survival time 1450 vs 2110 days, p=0.018; median overall survival time 1900 vs 2710 days, p=0.001; Fig. 6 B, C). By multivariable Cox regression analysis, high Egfl7 expression (RR 7.096; p=0.001), lymphatic metastasis (RR 6.243; p=0.002) and ER negative status (RR 8.047; p=0.003) were found to be independent prognostic factors for overall survival of patients with breast cancer (Tab. VII).
| Egfl7 expression levels | ||||||
|---|---|---|---|---|---|---|
| Clinicopathological parameters | N | 0 | 1+ | 2+ | 3+ | p value |
| Age | ||||||
| ≤ 50 years | 60 | 7 | 20 | 26 | 7 | |
| > 50 years | 42 | 3 | 15 | 17 | 7 | 0.560 |
| Tumor size | ||||||
| ≤ 2 cm | 22 | 5 | 5 | 6 | 6 | |
| 2-5 cm | 54 | 2 | 20 | 27 | 5 | |
| > 5 cm | 26 | 3 | 10 | 10 | 3 | 0.615 |
| Pathological types | ||||||
| Invasive ductal carcinoma | 84 | 7 | 28 | 37 | 12 | |
| Other types | 18 | 3 | 7 | 6 | 2 | 0.258 |
| Histological grade | ||||||
| I | 22 | 4 | 7 | 6 | 5 | |
| II | 45 | 4 | 18 | 20 | 3 | |
| III | 35 | 2 | 10 | 17 | 6 | 0.232 |
| TNM stage | ||||||
| I | 8 | 2 | 2 | 3 | 1 | |
| II | 43 | 6 | 18 | 16 | 3 | |
| III | 51 | 2 | 15 | 24 | 10 | 0.011 |
| Lymphatic metastasis | ||||||
| Absence | 35 | 4 | 19 | 9 | 3 | |
| Presence | 67 | 6 | 16 | 34 | 11 | 0.006 |
| ER | ||||||
| Positive | 69 | 4 | 22 | 32 | 11 | |
| Negative | 33 | 6 | 13 | 11 | 3 | 0.033 |
| PR | ||||||
| Positive | 67 | 5 | 25 | 26 | 11 | |
| Negative | 35 | 5 | 10 | 17 | 3 | 0.604 |
| Her2 | ||||||
| Positive | 59 | 4 | 18 | 25 | 12 | |
| Negative | 43 | 6 | 17 | 18 | 2 | 0.025 |
| Univariable analysis | Multivariable analysis | ||||
|---|---|---|---|---|---|
| Variables | N | RR (95% CI) | p value | RR (95% CI) | p value |
| Age | |||||
| <50 years | 60 | 1 | |||
| >50 years | 42 | 1.733 | 0.179 | 2.276 | 0.123 |
| Tumor size | (0.777-3.867) | (0.801-6.466) | |||
| <2 cm | 22 | 1 | 1 | ||
| >2 cm | 80 | 1.206 | 0.763 | 0.343 | 0.141 |
| Pathological types | (0.357-4.074) | (0.082-1.425) | |||
| Invasive ductal carcinoma | 84 | 1 | 1 | ||
| Other types | 18 | 0.586 | 0.388 | 0.618 | 0.490 |
| Histological grade | (0.175-1.969) | (0.158-2.418) | |||
| I | 22 | 1 | 1 | ||
| II-III | 80 | 0.384 | 0.123 | 0.774 | 0.732 |
| TNM stage | (0.114-1.294) | (0.178-3.355) | |||
| I-II | 51 | 1 | 1 | ||
| III | 51 | 2.347 | 0.039 | 1.711 | 0.275 |
| Lymphatic metastasis | (1.043-5.282) | (0.652-4.491) | |||
| Absence | 35 | 1 | 1 | ||
| Presence | 67 | 2.929 | 0.024 | 6.243 | 0.002 |
| ER | (1.156-7.422) | (1.976-19.730) | |||
| Positive | 69 | 1 | 1 | ||
| Negative | 33 | 2.829 | 0.012 | 1.939 | 0.004 |
| PR | (1.261-6.345) | (1.232-3.053) | |||
| Positive | 67 | 1 | 1 | ||
| Negative | 35 | 2.018 | 0.085 | 0.508 | 0.312 |
| Her2 | (0.907-4.489) | (0.137-1.886) | |||
| Negative | 43 | 1 | 1 | ||
| Positive | 59 | 0.983 | 0.968 | 1.026 | 0.958 |
| Egfl7 expression | (0.433-2.231) | (0.389-2.705) | |||
| Low | 45 | 1 | 1 | ||
| High | 57 | 4.006 (1.641-9.777) | 0.002 | 7.096 (2.208-22.805) | 0.001 |
Discussion
Fitch et al have reported that Egfl7 was highly expressed during early mouse embryogenesis, while low levels were detected in adult lung, heart, ovary, uterus, and renal tissues (11). In the present study, we first examined the Egfl7 expression profiles in normal adult human tissues using RT-PCR and Western blot. Among 19 adult normal human tissues, obvious Egfl7 expression was found in lung, heart, brain and retina tissues but only very low Egfl7 expression could be detected in liver, duodenum, intestine and colon tissues, which is similar to the expression patterns of Eglf7 in adult mice (9-11). We subsequently examined the Egfl7 expression profiles in 10 common human epithelial tumors using RT-PCR and Western blot and found that Egfl7 expression was significantly increased in all tumor tissues compared to the corresponding ANTs, especially in digestive system malignancies and breast cancer, indicating that upregulation of Egfl7 might be a universal event during the development of human epithelial tumors. Recently, Azhikina et al found that methylation of Egfl7 promoter was heterogeneous in nonsmall cell lung cancer (22), suggesting that the epigenetic modification might be one of the mechanisms underlying the overexpression of Egfl7 in human epithelial tumors. However, further research is required to verify this hypothesis.
Our immunohistochemistry result confirmed the Egfl7 expression in the same human epithelial tumors and showed predominant expression of Egfl7 in tumor cells despite weak Egfl7 expression in ECs. Cytoimmunofluorescence experiments were then carried out in 10 tumor cell lines, which were derived from the same human epithelial tumors we analyzed, to explore the subcellular localization of endogenous Egfl7 protein. Our results showed that Egfl7 is mostly located in the cytoplasm of these cells, which was consistent with our previous findings in HCC (18). Recently, Campagnolo et al also provided evidence for the expression of Egfl7 in non-ECs, such as primordial germ cells (12). These results therefore suggested that human epithelial tumor cells could express Egfl7 protein by themselves. Altogether, our data showed that Egfl7 was expressed differently in normal adult human tissues but significantly upregulated in common human epithelial tumors, suggesting Egfl7 to be a potential biomarker for human epithelial tumors. However, further studies are required to elucidate how and why Egfl7 expression is upregulated in these tumors.
In light of the fact that Egfl7 is identified as a secreted protein (13), we also hypothesized that Egfl7 could be detected in the serum of cancer patients. We determined Egfl7 levels in serum of healthy donors and patients with 10 epithelial tumors by a sandwich ELISA system. Interestingly, we found that Egfl7 levels were significantly elevated in serum samples from cancer patients compared to healthy donors, with the highest levels being found in HCC. These results were in agreement with our RT-PCR and Western blot results, which have indicated that HCC also had the highest Egfl7 expression among the 10 epithelial tumor tissues we detected, suggesting that a high serum Egfl7 level in HCC patients might derive from the high Egfl7 expression in HCC tissue. Our results presented here sufficiently showed a potential clinical usefulness of Egfl7 as a serological biomarker for epithelial tumors, especially HCC. In any case, further studies are essential to validate the value of Egfl7 as a novel serological marker in early detection, metastasis/recurrence monitoring and high-risk population screening of HCC (23).
To understand the role of Egfl7 in the development of human malignancies, we further analyzed the association of Egfl7 expression with the clinicopathological characteristics as well as prognosis of patients with HCC and breast cancer, malignancies which have a high incidence in Chinese men and women, respectively. The results showed that Egfl7 overexpression is closely related to capsule formation, multiple nodes, and vein invasion of HCC, which are all widely accepted factors associated with metastasis and poor prognosis of HCC (24, 25). HCC patients with high Egfl7 expression in general had a worse prognosis than those with low Egfl7 expression. Similarly, we also found that the overexpression of Egfl7 was significantly associated with clinicopathological features related to metastasis and poor prognosis of breast cancer, such as TNM stage, lymphatic metastasis, ER and Her2 (26). Breast cancer patients with high Egfl7 expression also had a worse prognosis than those with low Egfl7 expression. Importantly, the multivariable Cox regression analyses indicated that high Egfl7 expression is an independent risk factor for the prognosis of patients with HCC as well as breast cancer, suggesting that Egfl7 may be a useful prognostic biomarker for these 2 malignancies. Previously, we have revealed the important role of Egfl7 in the metastasis of HCC through EGFR-dependent FAK pho-sphorylation (18). Recently, Delfortrie et al demonstrated in a mouse model that Egfl7 in breast cancer could promote tumor immunity escape and progression by reducing the expression of endothelial molecules that mediate immune cell infiltration, such as IFN-γ, interleukin-12 (IL-12), intercellular adhesion molecule 1 (ICAM-1) and vascular cell adhesion molecule 1 (VCAM-1) (27). However, we believe the biological function of Egfl7 in breast cancer remains valuable to explore, for example, in mediation of cell motility or angiogenesis (18, 21).
In conclusion, the present study has exhibited the expression patterns of Egfl7 in normal adult human tissues and the significant increase in Egfl7 in human epithelial tumor tissues and sera of cancer patients. We have also documented the close correlation between Egfl7 overexpression and the clinicopathological features as well as poor prognosis of human HCC and breast cancer, providing novel evidence for Egfl7 to be a potential tumor biomarker, especially for HCC and breast cancer.
Footnotes
Financial Support: This study was supported by National Key Technologies R and D Program of China (No. 2001BA703B04 and No. 2004BA703B02), National Keystone Basic Research Program of China (No. 2004CB720303), National High Technology Research and Development Program of China (No.2006AA02Z4B2), National Science Fund for Distinguished Young Scholars of China (No. 30328028), National Natural Science Foundation of China (No. 30571826), Clinical Subjects' Key Project of Ministry of Health (2007-2009), The Science and Technology Significantly Special - The Significant New Medicine Formulation (2009ZX09103- 681).
Conflicts of Interest Statement: The authors have no potentially conflicting financial relationships relevant for this paper to disclose.
References
1. Jemal A., Bray F., Center M.M., Ferlay J., Ward E., Forman D. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90.
2. Andre F., McShane L.M., Michiels S.et al. Biomarker studies: a call for a comprehensive biomarker study registry. Nat Rev Clin Oncol 2011; 8: 171–6.
3. Fang F., Chang R., Yang L. Heat shock factor 1 promotes invasion and metastasis of hepatocellular carcinoma in vitro and in vivo. Cancer 2012; 118: 1782–94.
4. Servais E.L., Suzuki K., Colovos C.et al. An in vivo platform for tumor biomarker assessment. PLoS One 2011; 6: e26722.
5. Fang F., Yang L., Tao Y., Qin W. FBI-1 promotes cell proliferation and enhances resistance to chemotherapy of hepatocellular carcinoma in vitro and in vivo. Cancer 2012; 118: 134–46.
6. Kalinina J., Peng J., Ritchie J.C., Van Meir E.G. Proteomics of gliomas: initial biomarker discovery and evolution of technology. Neuro Oncol 2011; 13: 926–42.
7. Tolles J., Bai Y., Baquero M., Harris L.N., Rimm D.L., Molinaro A.M. Optimal tumor sampling for immunostaining of biomarkers in breast carcinoma. Breast Cancer Res 2011; 13: R51.
8. El-Serag H.B., Rudolph K.L. Hepatocellular carcinoma: epidemiology and molecular carcinogenesis. Gastroenterology 2007; 132: 2557–76.
9. Nichol D., Stuhlmann H. EGFL7: a unique angiogenic signaling factor in vascular development and disease. Blood 2012; 119: 1345–52.
10. Parker L.H., Schmidt M., Jin S.W.et al. The endothelial-cell-derived secreted factor Egfl7 regulates vascular tube formation. Nature 2004; 428: 754–8.
11. Fitch M.J., Campagnolo L., Kuhnert F., Stuhlmann H. Egfl7, a novel epidermal growth factor-domain gene expressed in endothelial cells. Dev Dyn 2004; 230: 316–24.
12. Campagnolo L., Moscatelli I., Pellegrini M., Siracusa G., Stuhlmann H. Expression of EGFL7 in primordial germ cells and in adult ovaries and testes. Gene Expr Patterns 2008; 8: 389–96.
13. Campagnolo L., Leahy A., Chitnis S.et al. EGFL7 is a chemoattractant for endothelial cells and is up-regulated in angiogenesis and arterial injury. Am J Pathol 2005; 167: 275–84.
14. Schmidt M., Paes K., De Mazière A.et al. EGFL7 regulates the collective migration of endothelial cells by restricting their spatial distribution. Development 2007; 134: 2913–23.
15. Ruoslahti E., Seppälä M. alpha-Fetoprotein in cancer and fetal development. Adv Cancer Res 1979; 29: 275–346.
16. Bacolod M.D., Barany F. Molecular profiling of colon tumors: the search for clinically relevant biomarkers of progression, prognosis, therapeutics, and predisposition. Ann Surg Oncol 2011; 18: 3694–700.
17. Malaguarnera G., Giordano M., Paladina I., Berretta M., Cappellani A., Malaguarnera M. Serum markers of hepatocellular carcinoma. Dig Dis Sci 2010; 55: 2744–55.
18. Wu F., Yang L.Y., Li Y.F., Ou D.P., Chen D.P., Fan C. Novel role for epidermal growth factor-like domain 7 in metastasis of human hepatocellular carcinoma. Hepatology 2009; 50: 1839–50.
19. Díaz R., Silva J., García J.M.et al. Deregulated expression of miR-106a predicts survival in human colon cancer patients. Genes Chromosomes Cancer 2008; 47: 794–802.
20. Huang C.H., Li X.J., Zhou Y.Z., Luo Y., Li C., Yuan X.R. Expression and clinical significance of EGFL7 in malignant glioma. J Cancer Res Clin Oncol 2010; 136: 1737–43.
21. Li J.J., Yang X.M., Wang S.H., Tang Q.L. Prognostic role of epidermal growth factor-like domain 7 protein expression in laryngeal squamous cell carcinoma. J Laryngol Otol 2011; 125: 1152–7.
22. Azhikina T., Kozlova A., Skvortsov T., Sverdlov E. Heterogeneity and degree of TIMP4, GATA4, SOX18, and EGFL7 gene promoter methylation in non-small cell lung cancer and surrounding tissues. Cancer Genet 2011; 204: 492–500.
23. Wright L.M., Kreikemeier J.T., Fimmel C.J. A concise review of serum markers for hepatocellular cancer. Cancer Detect Prev 2007; 31: 35–44.
24. Huang G.L., Li B.K., Zhang M.Y.et al. Allele loss and down-regulation of heparanase gene are associated with the progression and poor prognosis of hepatocellular carcinoma. PLoS One 2012; 7: e44061.
25. Yang L.Y., Fang F., Ou D.P., Wu W., Zeng Z.J., Wu F. Solitary large hepatocellular carcinoma: a specific subtype of hepatocellular carcinoma with good outcome after hepatic resection. Ann Surg 2009; 249: 118–23.
26. Ha S.A., Lee Y.S., Kim H.K.et al. The prognostic potential of keratin 18 in breast cancer associated with tumor dediffer-entiation, and the loss of estrogen and progesterone receptors. Cancer Biomark 2011; 10: 219–31.
27. Delfortrie S., Pinte S., Mattot V.et al. Egfl7 promotes tumor escape from immunity by repressing endothelial cell activation. Cancer Res 2011; 71: 7176–86.
Cite article
Cite article
Cite article
OR
Download to reference manager
If you have citation software installed, you can download article citation data to the citation manager of your choice
Information, rights and permissions
Information
Published In
Article first published online: March 2, 2018
Issue published: January-March 2013
Keywords
PubMed: 23558933
Authors
Metrics and citations
Metrics
Journals metrics
This article was published in The International Journal of Biological Markers.
VIEW ALL JOURNAL METRICSArticle usage*
Total views and downloads: 749
*Article usage tracking started in December 2016
Articles citing this one
Receive email alerts when this article is cited
Web of Science: 42 view articles Opens in new tab
Crossref: 32
- Targeting SMYD2 inhibits angiogenesis and increases the efficiency of ...
- Paeoniflorin Inhibits the Proliferation and Metastasis of Ulcerative C...
- Copper-containing titanium alloys promote angiogenesis in irradiated b...
- EGFL7 expression profile in IDH-wildtype glioblastomas is associated w...
- Impact of Activation of EGFL7 within Microenvironment of High Grade Ov...
- The Polymorphism and Expression of EGFL7 and miR-126 Are Associated Wi...
- Serum epidermal growth factor-like domain 7 serves as a novel diagnost...
- A pan-cancer study of spalt-like transcription factors 1/2/3/4 as ther...
- Nanomedicine for the Treatment of Advanced Prostate Cancer
- Prostate cancer: Delivery aspect and prospect
- A Pan-Cancer Study of Epidermal Growth Factor-Like Domains 6/7/8 as Th...
- EGFL7 Antagonizes NOTCH Signaling and Represents a Novel Therapeutic T...
- Epidermal Growth Factor Like-domain 7 and miR-126 are abnormally expre...
- High EGFL7 expression may predict poor pro...
- Increased circulating levels of Epidermal Growth Factor-like Domain 7 ...
- EGFL7 promotes hepatocellular carcinoma cell proliferation and inhibit...
- EGFL7: Master regulator of cancer pathogenesis, angiogenesis and an em...
- EGFL7 enhances surface expression of integrin α 5 ...
- Egfl7 Represses the Vasculogenic Potential of Human Endothelial Progen...
- Prognostic importance of circulating epidermal growth factor-like doma...
- Prognostic and biological significance of the proangiogenic factor EGF...
- EGFL7 is an intercellular EGFR signal messenger that plays an oncogeni...
- Viral nanoparticles decorated with novel EGFL7 ligands enable intravit...
- Expression of epidermal growth factor-like domain 7 may be a predictiv...
- Different expression of VEGF and EGFL7 in human hepatocellular carcino...
- Suppression of the Epidermal Growth Factor-like Domain 7 and Inhibitio...
- Integrative Analysis of Normal Long Intergenic Non-Coding RNAs in Pros...
- Expression of Epidermal Growth Factor-like Domain 7 is Increased by Tr...
- A Distinct Mechanism of Vascular Lumen Formation in Xenopus Requires E...
- Prognostic significance of epidermal growth factor-like domain 7 in pa...
- Prostate cancer relevant antigens and enzymes for targeted drug delive...
- High expression of epidermal growth factor-like domain 7 is correlated...
Figures and tables
Figures & Media
Tables
View Options
View options
PDF/ePub
View PDF/ePubGet access
Access options
If you have access to journal content via a personal subscription, university, library, employer or society, select from the options below:
loading institutional access options
Alternatively, view purchase options below:
Purchase 24 hour online access to view and download content.
Access journal content via a DeepDyve subscription or find out more about this option.
