This article is mentioned in:

Introduction

The extracellular matrix (ECM) is a complex three-dimensional network structure composed of macromolecules, including fibrous protein, proteoglycans and glycoproteins. The ECM mediates cellular adhesion and is involved in the intracellular signal transduction pathway, which, in turn, affects the occurrence, development and metastasis of malignancies (1). The adhesive cell-cell and cell-matrix attraction may be affected and modified when the cell undergoes malignant transformation.

A demonstrated that signal transduction of the urokinase plasminogen activator receptor (uPAR) activates the Ras-mitogen-activated protein kinase (MAPK) signaling pathway through the activity of a series of proteases (2). High expression levels of uPA and uPAR may result in the activation of integrin α5β1 and initiate a signaling cascade by aggregating the epidermal growth factor receptor. The signaling cascade subsequently causes a sustained and high level of extracellular signal-regulated kinase (ERK) activity and tumorigenicity (3). Integrin α5β1 is the most important transmembrane receptor in the uPAR signaling pathway, and may promote the development of malignant lesions, invasion and metastasis. uPAR, a heavily glycosylated single-chain glycoprotein, has three homologous domains: D1, D2 and D3. uPA dissociates these three regional proteins of uPAR into: uPAR (D1), comprised of the D1 domain, uPAR (D2D3), comprised of the D2 and D3 domains, uPAR (D1D2), comprised of the D1 and D2 domains, and the alternative splicing isomer of the uPAR gene, comprised of the D3 domain (4). In a previous study, the concentration and intensity of uPAR (D1D2) mRNA were found to be significantly increased in para-carcinoma and liver cancer tissue in primary culture as compared with normal tissue (5). The result indicated that uPAR (D1D2) mRNA overexpression may be due to uPAR isomer conformational changes, resulting in cell signal conduction abnormalities. In addition, the overexpression was closely associated with liver cell differentiation, abnormal clonal proliferation and increases in the degree of malignant transformation. Therefore, in the present study, the expression levels of uPAR (D1D2) and integrin α5β1 in normal, para-carcinoma and hepatic tissues were detected by polymerase chain reaction (PCR) and in situ hybridization.

Materials and methods

Liver tissue sample collection

The collection of 60 hepatocellular carcinoma (HCC) tissue samples (from 42 males and 18 females) and 25 hemangioma tissue samples (from 18 males and 7 females), from patients with a pathological diagnosis and surgical resection, was conducted at the First Affiliated Hospital of Guangxi Medical University (Nanning, China) between December 2011 and April 2013. Prior to specimen collection, ethical informed consent was obtained from each patient. The study was approved by the Ethics Commitee of The First Affiliated Hospital of Guangxi Medical University (Nanning, China)

Liver cancer tissues were removed from the cancer tissues that were not yet necrotic, para-carcinoma tissues were extracted from areas surrounding the cancer tissues ~2 cm and normal liver tissues were obtained from hepatic hemangioma patients undergoing surgical resection. One portion of each tissue sample was placed in a liquid nitrogen environment until mRNA extraction was performed, and another portion was fixed using neutral formalin solution, paraffin-embedded and sectioned.

Instruments and reagents

The following instruments and reagents were used: Ultraviolet Spectrophotometer (Nanodrap 2000; Thermo Fisher Scientific, Rockford, IL, USA), ultra-low temperature freezer (Thermo Forma 984; Thermo Fisher Scientific), Gel Doc™ XP + Gel imaging system, voltage steady flow electrophoresis apparatus (DYY-8B), thermal cycling machine (ABI Veriti™) and an inverted microscope (Nikon Corporation, Tokyo, Japan). Human uPAR (D1D2), integrin α5, integrin β1, the in situ hybridization kit and 3% hydrogen peroxide were purchased from Wuhan Boster Biological Techology, Ltd. (Wuhan, China). TRIzol, the RT-PCR kit and the nucleic acid dye were obtained from Treasure Biological Company (Dalian, China). The PCR kit and agarose were purchased from Thermo Fisher Scientific. Primers 1 and 4 were designed by Shanghai Sangon Biological Engineering Technology & Services Co., Ltd. (Shanghai, China) and synthesized by Invitrogen Trading Co., Ltd. (Shanghai, China). Primers 2 (6) and 3 (7) were synthesized by Takara Bio, Inc. (Shiga, Japan).

Liver tissue RNA extraction

The liver tissues were removed from the liquid nitrogen. The required quantity of tissue was added to 1 ml TRIzol and was then placed on ice for 5–10 min. The tissues were mechanically homogenized (JB50-S; Shanghai Heying Instrument Manufacturing Co., Ltd., Shanghai, China) and chloroform was added at a ratio of 5:1 lysate:chloroform (Shanghai Heying Instrument Manufacturing Co., Ltd.). The samples were left to rest for 5 min subsequent to being completely agitated and mixed, then centrifuged for 15 min (6,720 × g) at 4°C. DNA, protein and RNA were separated into sublayers, a white middle layer and an upper layer one after another.

The colorless upper layer of liquid was transferred to a new Eppendorf tube and an equal volume isopycnic isopropanol (Shanghai Heying Instrument Manufacturing Co., Ltd.) was added. The suspension was mixed, incubated at room temperature for 10 min and centrifuged for 15 min (5,600 × g) at 4°C.

The supernatant was discarded and the white sediment was retained. A volume of 1 ml 75% absolute ethyl alcohol was added, the solution was pipetted gently and then centrifuged for 10 min (6,720 × g) at 4°C. The alcohol was carefully discarded and the above procedure was repeated. The alcohol was added to an appropriate volume of diethylpyrocarbonate (DEPC; D100T; Sigma) to dissolve the RNA subsequent to drying in the air for 10 min.

Measurement of RNA purity and PCR

The ultraviolet spectrophotometer was used to measure the concentration of RNA and the optical density (OD) value. The RNA purity was deemed to be at an adequate level if the OD260/OD280 value was between 1.8 and 2.0. The quantity of RNA was measured using the RT-PCR kit, according to the manufacturer’s instructions. Reverse transcription was performed using an ABI Veriti Thermal Cycler instrument (Applied Biosystems, Foster City, CA, USA). The primers used are listed in Table I.

Table I PCR primers.

Table I

PCR primers.

Gene	Primer type	Primer sequence	No. of bases	Length (bp)
uPAR (D1D2)	Forward	5′-GACCTCTGCAGGACCACGAT-3′	20	367
	Reverse	5′-GGTGGCGGTCATCCTTTG-3′	18
Integrin α5	Forward	5′-AAAAACGGGAAGCTCCAAGCCGCA-3′	24	337
	Reverse	5′-AGGATGATGATCCACAGTGGGACG-3′	24
Integrin β1	Forward	5′-GTGGAGAATGTATACAAGCAGGGC-3′	24	511
	Reverse	5′-TTCCTGAGCTTAGCTGGTGTTGTG-3′	24
GAPDH	Forward	5′-GGTGCTGAGTATGTCGTGGAG-3′	21	292
	Reverse	5′-CAGTCTTCTGAGTGGCAGTGAT-3′	22

[i] PCR, polymerase chain reaction; bp, base pairs; uPAR, urokinase-type plasminogen activator receptor.

In situ hybridization

The in situ hybridization kit used a polyphase oligonucleotide probe corresponding to the target gene labeled with digoxigenin. Table II shows the mRNA probe sequences of the target genes.

Table II mRNA probe sequences of the target genes.

Table II

mRNA probe sequences of the target genes.

Target gene	mRNA probe sequence
uPAR (D1D2)	5′-TGTAA GACCA ACGGG GATTG CCGTG TGGAA GAGTG-3′ 5′-CACTC AGAGA AGACC AACAG GACCC TGAGC TATCG-3′ 5′-AGGAT GACCG CCACC TCCGT GGCTG TGGCT ACCTT-3′
Integrin α5	5′-CAGTG CACCC CCATT GAATT TGACA GCAAA GGCTC-3′ 5′-CAGGA GCAGA TTGCA GAATC TTATT ACCCC GAGTA-3′ 5′-ACCAT CTTCC CCGCC ATGTT CAACC CAGAG GAGCG-3′
Integrin β1	5′-AAATC TTGTG GAGAA TGTAT ACAAG CAGGG-3′ 5′-CAGTC ACTGA AGAGT TCCAG CCTGT TTACA-3′ 5′-GTGCC GTGAC AACTG TGGTC AATCC GAAGT-3′

[i] uPAR, urokinase-type plasminogen activator receptor.

Cover slips and other vessels used in the in situ hybridization experiment were treated with water containing 1% DEPC to remove RNA enzymes prior to use. The experiment was conducted according the manufacturer’s instructions of the kit and the results were observed under a microscope (Nikon, Tokyo, Japan).

Calculation of expression level scores

Positive expression of uPAR (D1D2) and integrin α5β1 was identified in the cytoplasm, exhibiting a yellow granular appearance. A total of 10 high power fields (HPFs) for each slice were randomly selected to determine the result and 100 tumor cells were counted to calculate the percentage of positive cells in each HPF. To comprehensively determine the expression levels of the two molecules, positive cell number and stained intensity scores were used. The positive cell score was calculated as follows: 0 points, number of positive cells <10%; 1 point, number of positive cells <25%; 2 points, number of positive cells <50%; 3 points, number of positive cells >50%. The positive staining intensities were determined as follows: 0 points indicate negativity, consistent with the negative control; 1 point indicates low positive staining, appearing faint yellow (slightly more intense than the negative control); 2 points indicate positive staining, appearing medium yellow (more intense than the negative control); 3 points indicated marked positive staining, appearing tan or brown. The total score of each slide was calculated using the positive cell number score plus the stained intensity score: 0–2 points signified low expression levels (−/+), 3–4 points signified medium expression levels (++) and 5–6 points signified high expression levels (+++).

Statistical analysis

The PCR results are expressed as gray values (mean ± standard deviation). The in situ hybridization result was determined by positive cell number and staining intensity scores. SPSS version 17.0 (SPSS, Inc., Chicago, IL, USA) was used to process the data. Pairwise comparisons were calculated using one-way analysis of variance, categorical variable comparisons were performed with the χ2 test and correlation analysis was conducted using Spearman’s rank correlation. P<0.05 was considered to indicate a statistically significant difference.

Results

uPAR (D1D2) and integrin α5β1 mRNA expression levels

In the liver cancer and para-carcinoma HCC tissues and normal liver tissues, uPAR (D1D2; Fig. 1A), integrin α5 (Fig. 1C) and integrin β1 (Fig. 1E) mRNA expression levels were determined using electrophoresis. The signal intensities in para-carcinoma and cancer tissues were markedly increased as compared with those in normal tissue. Water served as a blank control to exclude specific bands. uPAR (D1D2; Fig. 1B), integrin α5 (Fig. 1D) and integrin β1 (Fig. 1F) samples served as internal reference bands corresponding to the three target genes. The function of these bands was to reduce the effect of differences in the doses of sample injection and the experimental results.

Figure 1 Electrophoresis results of uPAR isomer D1D2, integrin α5 and integrin β1 detection in cancer, paracarcinoma and normal tissue. (A) uPAR (D1D2), (C) integrin α5 and (E) integrin β1. Lane 1, marker; lane 2, cancer tissue; lane 3, para-carcinoma tissue; lane 4, normal tissue; lane 5, blank control. (B, D and F) indicate the internal reference bands corresponding to the three target genes: (B) uPAR (D1D2), (D) integrin α5 and (F) integrin β1. uPAR, urokinase-type plasminogen activator receptor.

IMAGE J software analysis of each sample and the grey value of the corresponding internal reference

The expression levels of uPAR (D1D2) and integrin α5β1 mRNA, as determined by PCR, were calculated according to the following equation: Grey level ratio = (grey value of target band - grey value of water)/(grey value of internal reference - grey value of water) (Table III). mRNA levels of uPar (D1D2), integrin α5 and integrin β1 were increased in para-carcinoma as compared with those in normal tissues, and even more increased in carcinoma tissues.

Table III Expression levels of uPAR (D1D2) and integrin α5β1.

Table III

Expression levels of uPAR (D1D2) and integrin α5β1.

		Grey value ratio
Group	n	uPAR (D1D2)	Integrin α5	Integrin β1
Normal tissue	25	0.542±0.048	0.509±0.070	0.329±0.401
Adjacent tissue	60	0.772±0.063b	0.735±0.039b	0.596±0.08b
Cancer tissue	60	1.222±0.144a	1.316±0.108a	1.168±0.106a

a P<0.05 compared with normal tissue and para-carcinoma tissue.

b P<0.05 compared with para-carcinoma tissue and cancer tissue.

{ label (or @symbol) needed for fn[@id='tfn5-mmr-10-05-2568'] } uPAR, urokinase-type plasminogen activator receptor.

Detection of uPAR (D1D2), integrin α5 and integrin β1 expression using in situ hybridization

In situ hybridization was observed under a microscope; positive signaling of the target genes was located in the cytoplasm and appeared as a tan color. The positive signal percentages for uPAR (D1D2), integrin α5 and integrin β1 were highest in cancer tissues, followed by para-carcinoma tissues and were lowest in the normal liver tissues (Figs. 2–4). Subsequently, the expression levels of uPAR (D1D2), integrin α5 and integrin β1 in the liver tissues were calculated from the microscopy results. The adopted positive cell and staining intensity scores were used to comprehensively determine the expression levels of the respective molecules (Table IV).

Figure 2 Expression of urokinase-type plasminogen activator receptor isomer D1D2 mRNA (tan color) in cancer, para-carcinoma and normal liver tissues (magnification, ×400). (A) Cancer tissue; (B) paracarcinoma tissue; (C) normal tissue.

Figure 4 Expression of integrin β1 mRNA (tan color) in cancer tissue, adjacent tissue and normal tissue. (A–C) Integrin β1 mRNA expression in liver tissue (magnification, ×400). (A) Cancer tissue; (B) adjacent tissue; (C) normal tissue.

Table IV Expression levels of uPAR (D1D2), integrin α5 and integrin β1 in liver tissue.

Table IV

Expression levels of uPAR (D1D2), integrin α5 and integrin β1 in liver tissue.

			Staining results
Gene	Group	n	−	+	++	+++	Positive rate (%)
uPAR (D1D2)	Normal tissue	25	16	6	3	0	36.0a
	Adjacent tissue	60	24	12	16	8	60.0b
	Cancer tissue	60	10	4	25	21	83.3
Integrin α5	Normal tissue	25	22	3	0	0	12.0a
	Adjacent tissue	60	38	6	12	4	36.7b
	Cancer tissue	60	21	7	18	14	65.0
Integrin β1	Normal tissue	25	22	2	1	0	12.0b
	Adjacent tissue	60	37	8	10	5	38.3b
	Cancer tissue	60	23	6	15	16	61.7

a P<0.05 compared with normal and para-carcinoma tissues.

b P<0.05 compared with para-carcinoma and cancer tissues.

{ label (or @symbol) needed for fn[@id='tfn8-mmr-10-05-2568'] } uPAR, urokinase-type plasminogen activator receptor.

The correlations among the expression levels of uPAR (D1D2) and integrin α5β1 in HCC, and the clinical pathological features were determined to be as follows: The positive rate of uPAR (D1D2) in HCC was not correlated with α fetal protein (AFP), ferritin (SF), pathologic stage or tumor size, but was correlated with DNA copy number. When the DNA copy number was >1.0×103, the positive percentage of uPAR (D1D2) was 90.7%; When the DNA copy number was <1.0×103, the positive percentage was 64.7% and a statistically significant difference was detected between the two groups (P<0.05). The integrin α5β1 expression levels did not correlate with AFP, SF, hepatitis B virus (HBV) DNA copy number or tumor size (P>0.05), but correlated with the pathologic stage. At different pathological stages, the percentages of positive integrin α5β1 in HCC were significantly different (P<0.05).

The correlation between the expression intensities of uPAR (D1D2) and integrin α5β1 in HCC tissues was positive. Thus, a synergistic effect between the expression levels of the two molecules may be considered (rs1=0.257, P<0.05; rs2=0.261, P<0.05; Tables V and VI).

Table V Correlation between uPAR (D1D2) expression and integrin α5 expression in HCC.

Table V

Correlation between uPAR (D1D2) expression and integrin α5 expression in HCC.

Integrin α5 expression	uPAR (D1D2) expression				Total
	−	+	++	+++
−	7	1	8	5	21
+	1	0	4	2	7
++	0	2	9	7	18
+++	2	1	4	7	14
Total	10	4	25	21	60

[i] uPAR, urokinase-type plasminogen activator receptor; HCC, hepatocellular carcinoma.

Table VI Correlation between uPAR (D1D2) expression and integrin β1 expression in HCC.

Table VI

Correlation between uPAR (D1D2) expression and integrin β1 expression in HCC.

Integrin β1 expression	uPAR (D1D2) expression				Total
	−	+	++	+++
−	5	1	13	4	23
+	2	1	1	2	6
++	1	1	6	7	15
+++	2	1	5	8	16
Total	10	4	25	21	60

[i] uPAR, urokinase-type plasminogen activator receptor; HCC, hepatocellular carcinoma.

Discussion

Tumor transformation, growth, invasion and metastasis is a complex interaction process between tumor cells and the rest of the body. In this process, the various proteases adhered to the ECM (including serine proteases and matrix metalloproteinases) and growth factors (such as the integrin family and epidermal growth factor) are crucial (8).

Increasing evidence suggests that tumor-associated mRNA may induce tumor cells to produce special function proteins closely associated with cancer occurrence and development. uPAR is an important factor in the urokinase system; certain studies have shown that uPAR mRNA is a specific marker of malignant transformation and metastasis in stomach cancer, colon cancer and non-small cell lung cancer (9–11). mRNA splice variants are considered specific sequences of diagnostic malignancy (12,13), which may render certain protein areas missing, and subsequently result in loss or change of function, or the development of a novel function. There are multiple mRNA splice variants of uPAR; studies have demonstrated that these splice variants exert a marked effect in breast cancer, small cell lung cancer and other malignancies (14–17). In the present study, PCR and in situ hybridization techniques were used to detect the expression levels of uPAR (D1D2) mRNA in the liver cancer, para-neoplastic and normal tissues. The expression levels of the uPAR (D1D2) splice variant were observed to be significantly different in HCC, para-carcinoma and normal liver cells and in primary culture; the expression levels of uPAR (D1D2) exhibited an upward trend with liver cancer development. The expression intensity of uPAR (D1D2) in HCC was not associated with AFP, SF, pathological stage or tumor size, but was associated with the HBV DNA copy number (P<0.05). This is consistent with the previous results of primary cultured hepatoma cells. Continuous replication HBV activates particular proto-oncogenes, but renders certain tumor-suppressor genes inactive and mutational, thus promoting the occurrence of cancer. The integration of virus DNA may increase the occurrence of the HBV X antigen and induce malignant transformation of liver cells (18).

Integrins are a group of transmembrane glycoprotein receptors widely distributed on cell surfaces. In addition to mediating adhesion between the cell and the ECM, integrins affect ECM degradation and tumor cell chemotaxis, proliferation, apoptosis and metastasis (19). Integrin α5β1 is an important member of the integrin family. A number of studies have demonstrated the association between integrin α5β1 and numerous types of cancer, including liver, breast and colorectal cancer (20–21). Furthermore, uPA and uPAR, overexpressed in cancer cells, frequently combine with integrin α5β1 to form uPA-uPAR-α5β1 complexes, which induce the activation of signaling pathways to alter cell adhesion, proliferation and migration (22). Aguirre Aguirre Ghiso et al (23) considered the interaction between uPAR and integrin α5β1 on the cell surface to induce a series of reactions and this interaction may be involved in MAPK-ERK signaling pathway activation, eventually resulting in head and neck tumors, as determined in head and neck cancer models. Wei et al (24) deemed that the binding between uPAR and α5β1 requires maximum fibrin and tumor cell invasion, for which enhancement of the Src/Rac/ERK signaling pathway is necessary. In the present study, PCR and in situ hybridization were employed to detect the expression levels of integrin α5 and integrin β1 mRNA in HCC, para-carcinoma and normal liver tissues. Similar findings to those for uPAR (D1D2) were obtained; integrin α5β1 expression levels in HCC tissues were higher than those in para-carcinoma and normal liver tissues. Thus, the expression levels of integrin α5β1 in HCC were associated with the pathological stage, and the associations among integrin α5β1 expression levels, liver cancer differentiation and malignant transformation are inextricable. The results also revealed that uPAR (D1D2) and integrin α5β1 expression levels in HCC were positively correlated, and exerted a synergistic effect in cancer occurrence and development. The correlation between uPAR (D1D2) expression levels and liver cancer malignant transformation was initially observed in a previous study (5), but the correlation between integrin α5β1 expression levels and liver cancer, and the synergistic effect of uPAR (D1D2) and integrin α5β1 in HCC had not yet been reported, to the best of our knowledge.

In conclusion, the results of the present study demonstrated that the synergy of uPAR (D1D2) and integrin α5β1 expression in hepatocytes was closely associated with the occurrence of liver cancer and subsequent metastasis. The cell signal transduction pathways and gene therapy as anticancer strategies have become a predominant focus of cancer research. It has been found that the cell signal transduction pathways of different types of uPAR isomers and integrin-α5β1 are correlated with tumorigenesis but the mechanism has not been fully elucidated. Thus further studies will aim to determine the underlying mechanisms.

References