The following tissue specimens were enrolled in this study: 40 specimens of colorectal adenocarcinoma (including 16 specimens of moderately differentiated colorectal adenocarcinoma, 24 poorly differentiated adenocarcinoma, 14 lymph node metastatic carcinoma, and 17 invasive carcinoma) and 22 specimens of normal colorectal tissue, 40 specimens of gastric adenocarcinoma (including 20 specimens of moderately differentiated gastric adenocarcinoma, 20 poorly differentiated adenocarcinoma, 15 lymph node metastatic carcinoma, and 23 invasive carcinoma) and 22 specimens of normal gastric tissue, 120 specimens of intraductal carcinoma of breast (including 86 specimens of moderately differentiated intraductal carcinoma, 34 poorly differentiated intraductal carcinoma, 30 lymph node metastatic carcinoma and 58 invasive carcinoma) and 20 specimens of normal breast tissue, 72 specimens of lung cancer (including 13 specimens of moderately differentiated carcinoma, 19 poorly differentiated carcinoma, 24 lymph node metastatic carcinoma) and 23 specimens of normal lung tissue. All patients underwent surgery at Xijing Hospital of the Fourth Military University, China in 2003. The resected specimens were fixed in 10% formaldehyde.

CD44s and CD44v6 monoclonal antibodies were obtained from MBI. SABC kit was bought from Sina-America Biotechnology Company. nm23 oligonucleotide probe and CSA test kit were purchased from Boshide Biotechnology Company. Biotin labeled anti-digitoxin (DIG) antibody was obtained from Sigma.

Samples were fixed in buffered formaldehyde and subsequently paraffin-embedded. Histological sections (5 μm) were prepared from the specimens, and then diagnosed and labeled by the pathologist. Tissue microarray was performed on the instrument of microarray from Beecher, USA. A hole (diameter 0.5 mm) was made on the recipient block (blank paraffin block), then the tissue core was taken on the supply block (labeled tissue block) and put in the hole. The former procedure was repeated and tissue microarray was completed. Histological sections (3-5 μm) were prepared with the instrument of Leica, Germany, and then rediagnosed by the pathologist.

The sections for tissue microarray were dewaxed in water by normal technique, and then the antigen was restored in high pressure (at 121 °C for 10 min). Immunohistochemistry was performed with the SABC kit. The primary antibodies were CD44s and CD44v6 monoclonal antibodies. Staining was performed following the instructions of SABC kit.

All reagents and containers were treated by diethy-lprocarbonate (DEPC). In brief, the sections for tissue microarray were dewaxed in water by normal technique, immersed into 3% H₂O₂ at room temperature for 10 min, and then washed twice with distilled water. Twenty microgram per milliliter of freshly diluted protease K was added and digested for 20 min at 37 °C to expose mRNA nucleic acid segments. Twenty milliliter of glycerine (20%) was added to the dry bottom of the test kits to keep humidity, 20 μL of the pre-hybridization solution was added to each section and kept at 37 °C for 4 h. Then the supernumerary liquid was absorbed without washing, hybridization solution was added as mentioned above. The sections were then covered with protective membrane and put in homeothermia at 40 °C overnight. After the coverglass was removed, the sections were washed twice at 37 °C in 2×SSC for 5 min, once in 0.5×SSC for 15 min, once in 0.2×SSC for 15 min, and kept in 3% BSA at 37 °C for 30 min and in seal solution at 37 °C for 30 min. After being washed with 0.5 mol/L PBS, the biotin labeled anti-digitoxin antibody was added for 60 min at 37 °C, followed by biotin-peroxidase for 20 min at 37 °C and then washed with PBS. Color was showed by DAB. Finally, the sections were restained, dehydrated, pellucidated, and sealed. Negative controls were designed.

Specimens were considered positive when >50% of the tissue components were immunohistochemically stained brown-yellow in appropriate cellular compartment. Specimens were considered positive for ISH when >50% of the tissue components were stained blue in appropriate cellular compartment.

Statistical analysis was performed with χ² test. P<0.05 was considered statistically significant.

The expression of CD44s in tissues of colorectal adeno-carcinoma and normal colorectal mucosa was 42.0% (17/40) and 13.6% (3/22), respectively (Figures 1A and B), and there was a significant difference between them (χ² = 5.18, P<0.05). In addition, there was a statistical significance between colorectal adenocarcinoma of with and without invasion (χ² = 10.52, P<0.01), as well as with and without lymph node metastasis (χ² = 12.48, P<0.01). However, there was no statistical significance between moderately and poorly differentiated colorectal adenocarcinoma (P>0.05), indicating that the expression of CD44s was related to the oncogenesis and development of colorectal adenocarcinoma. The expression of CD44v6 in tissues of colorectal adenocarcinoma and normal colorectal mucosa was 55.8% (22/40) and 0% (0/22), respectively (Figure 1C). There was a statistical significance between moderately and poorly differentiated colorectal adenocarcinoma (P<0.05) with and without invasion (P<0.05), as well as with and without lymph node metastasis (χ² = 9.22, P<0.01). The results showed that the expression of CD44v6 was related not only to the invasion of colorectal carcinoma, but also to the metastasis. The expression of nm23 mRNA in tissues of colorectal adenocarcinoma and normal colorectal mucosa was 45.0% (18/40) and 40.9% (9/22), respectively (Figure 1D), and there was no statistically significant difference (P>0.05). However, the expression of nm23 mRNA in gastric adenocarcinoma with/without lymph node metastasis was 28.5% (4/14) and 53.8% (14/26), respectively. The difference was statistically significant (χ² = 8.47, P<0.05). In addition, the expression of CD44v6 in colorectal adenocarcinoma was related to the invasion (P<0.05), but not to the pathological grade (P>0.05, Table 1).

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Expression of CD44s in human normal colorectal and aden-ocarcinoma tissue (A, B) and expression of CD44v6 (C) and nm23 mRNA (D) in human colorectal adenocarcinoma.

Expression of CD44s and CD44v6 was associated with invasion and lymph node metastasis of colorectal adenocarcinoma (r = 0.47, P<0.05). However, expression of nm23 mRNA was not associated with invasion and lymph node metastasis of colorectal adenocarcinoma (r = -0.49, P<0.05), suggesting that CD44s, CD44v6 and nm23 mRNA could regulate invasion and lymph node metastasis of colorectal adenocarcinoma. In addition, the expression level of nm23 mRNA, CD44s and CD44v6 was not associated with the age and sex of patients (P>0.05, Table 1).

The expression of CD44s in tissues of gastric adeno-carcinoma and normal gastric mucosa was 48% (19/40) and 13.6% (3/22), respectively (Figures 2A and B), and there was a significant difference between them (χ² = 10.29, P<0.01). However, the expression of CD44s was not significantly associated with lymph node metastasis. There was no statistical significance between moderately and poorly differentiated gastric adenocarcinoma with/without invasion (P>0.05). The expression of CD44v6 in tissues of gastric adeno-carcinoma and normal gastric mucosa was 63.3% (25/40) and 0% (0/22), respectively (Figure 2C). There was a statistical significance between moderately and poorly differentiated gastric adenocarcinoma (χ² = 9.19, P<0.01) with and without invasion (χ² = 22.22, P<0.01), as well as with and without lymph node metastasis (χ² = 10.36, P<0.01). The expression of nm23 mRNA in tissues of gastric adenocarcinoma and normal gastric mucosa was 47% (19/40) and 43% (9/22), respectively (Figure 2D) and there was no statistically significant difference between them (P>0.05). However, the expression of nm23 mRNA in gastric adenocarcinoma with/without lymph node metastasis was 26.7% (4/15) and 60% (15/25) respectively, the difference was significant (χ² = 18.47, P<0.01, Table 2).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Expression of CD44s in human normal gastric tissue (A, B) and expression CD44v6 (C) and nm23 mRNA (D) in human gastric adenocarcinoma.

The expression of CD44v6 was associated with lymph node metastasis of gastric adenocarcinoma (r = 5.04). However, the expression of nm23 mRNA was not associated with lymph node metastasis of gastric adenocarcinoma (r = -4.93, Table 2).

The expression of CD44s and CD44v6 in tissues of intraductal carcinoma of breast was 45.8% (55/120) and 53.3% (64/120), respectively (Figures 3A and B), but neither of them was expressed in normal breast tissue. The expression of CD44s and CD44v6 in intraductal breast carcinoma with and without invasion had a significant statistical difference (χ² = 9.52, P<0.01; χ² = 22.89, P<0.001), as well as with and without lymph node metastasis (χ² = 9.41, P<0.01; χ² = 8.75, P<0.01). Furthermore, the expression of CD44s was not significantly associated with moderately and poorly differentiated intraductal breast carcinoma (P>0.05). However, the expression of CD44v6 was not only related to invasion and lymph node metastasis of intraductal breast carcinoma, but also related to its pathological grade (χ² = 5.68, P<0.05, Table 3).

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Expression of CD44s (A) and CD44v6 (B) in human intraductal breast carcinoma.

The expression of nm23 mRNA in tissues of lung cancer and normal lung tissues was 55.2% (40/72) and 82.6% (19/23), respectively (Figure 4), and there was a significant difference between them (χ² = 5.42, P<0.05). The expression of nm23 mRNA in tissues of lung cancer with and without lymph node metastasis was 25.0% (6/24) and 70.8% (34/48), respectively and there was a significant difference between them (χ² = 13.61, P<0.001), as well as between moderately and poorly differentiated lung cancer (χ² = 9.61, P<0.01), indicating that the expression of nm23 mRNA was associated with the oncogenesis, lymph node metastasis and pathological grade of lung cancer. In addition, the expression of nm23 mRNA in lung cancer was not associated with the age and sex of patients and the pathologic feature of lung cancer (P>0.05, Table 4).

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Expression of nm23 mRNA in human lung cancer.