The present study comprised 28 SACC patients (16 males and 12 females; mean age, 50 years; range, 26–67) and 10 ACA patients (6 males and 4 females; mean age, 51.4 years; range, 31–65) treated at the Department of Oral and Maxillofacial Surgery, School of Stomatology, Fourth Military Medical University between 2004 and 2008. The present study was approved by the Ethics Committee of the Fourth Military Medical University. Informed consent was obtained from all subjects prior to obtaining the tissue samples during surgery.

Samples were divided into two sections. For immunohistochemical staining and hematoxylin and eosin (H&E) staining, one section was fixed immediately in 10% formalin for 24 h and was embedded in paraffin. Subsequently, 4-μm serial sections were prepared with a microtome. For pre-embedding immunogold-silver cytochemistry, the tissue was incised into a 2×2 mm slice and fixed into a mixture of 4% paraformaldehyde, 0.05% glutaraldehyde and 15% saturated picric acid in 0.1 M phosphate buffer (pH 7.4) at room temperature.

Slides were prepared from each sample using 4-μm H&E-stained serial sections and reviewed by two independent oral pathologists who were blind to the clinical findings. The presence or absence of microscopic PNI was recorded.

Mouse anti-human Leu-7 was used as the primary antibody and biotinylated goat anti-mouse IgG was used as the secondary antibody. The two antibodies were reacted with a streptavidin-biotin-peroxidase complex. Briefly, sections were mounted on glass slides and air-dried. Dewaxing and quenching of endogenous peroxidase was carried out using 3% H₂O₂/methanol and subsequently protease digestion was carried out for antigen retrieval. The samples were then incubated in goat serum (Invitrogen, Carlsbad, CA, USA) at 1:10 dilution for 20 min to block non-specific binding. Mouse anti-human monoclonal antibody Leu-7 (Abcam, Cambridge, MA, USA) was diluted in phosphate-buffered saline (PBS)/0.3% bovine serum albumin (Sigma-Aldrich, St. Louis, MO, USA) and used at a predetermined optimal dilution of 1:25. Following overnight incubation at 4°C, the sections were incubated for 30 min at 37°C in goat anti-mouse IgG (Invitrogen). The peroxidase-labeled streptavidin (Invitrogen) was then added for 30 min at 37°C. The substrate color reaction was developed by incubation in liquid diaminobenzamine (DAB; Dako, Glostrup, Denmark). For the negative control sections, the primary antibody was replaced by non-immune normal goat serum. The results of immunohistochemical staining were evaluated by two independent pathologists blind to the clinical findings.

A total of 22 samples that expressed Leu-7 were included in immunofluorescence double-staining carried out. Briefly, sections were stored and then exposed to antigen retrieval. Mouse anti-human monoclonal antibody Leu-7 (Abcam, 1:50) and rabbit anti-human polyclonal antibody α-SMA (Abcam, 1:200) were used. Sections were incubated in a mixture of the two primary antibodies. Subsequent to overnight incubation at 4°C, the sections were rinsed three times for 15 min in PBS (0.01 mol/l, pH 7.4) and incubated for 60 min at 37°C in a biotinylation goat anti-mouse IgG (Invitrogen, 1:8,000). Sections were then incubated for 3 h at room temperature in a mixture of goat anti-rabbit FITC-conjugated IgG (Invitrogen, 1:30) and avidin-conjugated Cy3 (Sigma, 1:100).

The labeled tissues were then rinsed in PBS (0.01 mol/l, pH 7.4), mounted on gelatin-coated slides and coverslipped. The distribution and intensity of labeling were assessed using a Zeiss epi-illumination fluorescence microscope and a Zeiss LSM 510 confocal laser scanning microscopy system. The software of the CLSM exhibited images acquired with 495-nm laser illumination (FITC) color-coded in green and those acquired at 543 nm (Cy3) color-coded in red. In each immunostaining run, the two primary antibodies were replaced by non-immune normal goat serum as negative controls.

The expression of Leu-7/α-SMA was explored using pre-embedding immunogold-silver cytochemistry. Sections of 50-μm were cut with a vibratome (VT 1000S; Leica, Mannheim, Germany) and placed in PBS containing 25% sucrose and 10% glycerol for 1 h for cryo-protection. Following a freeze-thaw treatment, sections were blocked with 5% BSA and 5% NGS for 4 h to block non-specific immunoreactivity. Subsequent immunohistochemical procedures were performed at room temperature.

For Leu-7/α-SMA double labeling, Leu-7 was detected by immunoperoxidase and α-SMA by an immunogold-silver staining method. The sections were incubated overnight in primary antibodies against Leu-7 with those against α-SMA, diluted to working concentrations as described above, in PBS containing 1% BSA and 1% NGS. Following rinsing in PBS, sections were incubated overnight in a mixture of secondary antibodies of biotinylated goat anti-mouse IgG and goat anti-rabbit IgG conjugated to 1.4-nm gold particles (1:100, Nanoprobes, Yaphank, NY, USA). In addition, following rinsing in PBS, sections were post-fixed in 2% glutaraldehyde in PBS for 45 min. Silver enhancement was carried out in the dark with an HQ Silver kit (Nanoprobes). Prior to and following the silver enhancement, sections were rinsed several times with deionized water. The sections were then incubated in ABC solution (Sigma, 1:300) for 4 h and visualized using the glucose oxidase-3,3′-diaminobenzidine method. Immuno-labeled sections were fixed with 0.5% osmium tetroxide in 0.1 M phosphate buffer for 1 h, dehydrated in graded ethanol series, then in propylene oxide and finally flat-embedded in Epon 812. Following polymerization, the sections were examined under the light microscope. Sections containing Leu-7/α-SMA immunoreactivity were selected, trimmed under a stereomicroscope and mounted onto blank resin stubs. Ultra-thin sections were cut and mounted on mesh grids. The sections were then counterstained with Uranyl Acetate and lead citrate and observed under a JEM-1230 electron microscope (JEOL Ltd., Tokyo, Japan). Electron micrographs were developed in the dark room.

Statistical analysis was performed using SPSS 18.0 software (Chicago, IL, USA). To determine the difference of the ability of the levels of PNI and Leu-7 expression between SACC and ACA, statistical analysis was performed using Fisher’s exact test. To find the correlation between PNI and Leu-7 expression, κ analysis was used. P<0.05 was considered to indicate a statistically significant difference.

PNI appeared in 16 out of 28 SACCs (57.1%). In addition, 1 case of PNI was observed microscopically out of the 10 specimens of ACA (10%). The H&E section findings of PNI are summarized in Table I. The differences in the rate of PNI between SACC and ACA were significant (Fisher’s exact test, P=0.009; Table II).

Positive staining for Leu-7 was revealed in 22 out of 28 SACCs (78.6%; Fig. 1a), while none of the 10 ACAs were positive for Leu-7 (Fig. 1b). The differences between SACC and ACA in the expression of Leu-7 were significant (Fisher’s exact test, P=0.000; Table II).

SACCs (16/28) were observed with PNI and these were all positive for Leu-7. The κ test revealed that there was a significant correlation between Leu-7 expression and PNI (κ=0.533, P=0.01; Table III).

In the SACC, myoepithelial cells were detectable by α-SMA staining. In the CLSM images, green fluorescence revealed positive staining of α-SMA, red fluorescence revealed positive staining of Leu-7 and yellow fluorescence revealed green and red fluorescence in the same location (Fig. 1d-f). All 22 SACC cases were double-stained for Leu-7/α-SMA, which were generally localized diffusely in the cytoplasm of myoepithelial cells.

Under the electron microscope, myoepithelial cells were recognizable by α-SMA staining. Immunogold-silver-enhanced particles indicative of α-SMA immunoreactivity were present diffusely in the cytoplasm (arrows, Fig. 1c) in SACC. The immunoperoxidase reaction product indicative of Leu-7 immunoreactivity was localized mainly in the cytoplasm (arrowheads, Fig. 1c). Co-expression of Leu-7 and α-SMA was identified in the same myoepithelial cells (Fig. 1c).