Treatment of experimental animals
Adult male Wistar rats (n = 80) weighing 200 - 250 g were used and obtained from the Laboratory Animal Center of the National Taiwan University. We followed the Guide for the Care and Use of Laboratory Animals (1985) as stated in the United States NIH guidelines (NIH publication no.86-23) to treat animals for this study including the care, housing, handling and experimental procedures. All the experiments were approved by our Laboratory Animal Center, China Medical University, Taiwan. All efforts were made to minimize both the animal suffering and the number of animals used for this investigation.
The experimental animals were anesthetized with an intramuscular (i.m.) injection of mixtures of zoletil (30 mg/kg) and xylazine (10 mg/kg). After anesthesia, the left vagus and the hypoglossal nerves were subjected to crush injury by clamping with a small hemostatic forceps for 30 s; for the former, the level of clamping was mid-cervical, whereas for the latter, right below the tendon of the digastric muscle. In sham-operated animals (controls), similar procedures were carried out except both of the left side nerves remained intact. The experimental animals were divided into four groups: I, II, III, and IV (10, 25, 50 mg/kg EGCG-pretreatment + PNCI and PNCI only, respectively). Previous in vivo studies have demonstrated that EGCG can pass the blood-brain barrier and reach the brain parenchyma [34, 38, 41, 42]. Animals in each group were therefore received daily intraperitoneal injections of EGCG or normal saline for successive six days with the last injection at 30 min before PNCI. Different doses (10, 25, or 50 mg/kg body weight dissolved in normal saline) of EGCG (cat No.E4143, Sigma-Aldrich, St Louis, MO, USA) were applied in the present study. Each experimental group was subdivided into four groups (n = 5) and sacrificed at 3, 7, 14, and 28 days. All the experimental animals were housed under the same conditions (controlled temperature [22°C] and 12-h light/dark cycle) and given free access to food and water ad libitum.
Perfusion and tissue preparation
At each time point, rats were deeply anesthetized with mixtures of zoletil (30 mg/kg) and xylazine (10 mg/kg) and perfused transcardially with 100 ml of Ringer's solution followed by 300 ml of 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4. After perfusion, the brainstem and its nuclei were quickly removed, rinsed in 0.1 M PB, and transferred through a series of increasing concentrations of sucrose buffer (10 - 30%) for cryoprotection at 4°C overnight. Serial 30-μm thick sections of the brainstem were cut transversely with a cryostat on the following day and were alternately placed at a distance of 120 μm apart into four-well dishes. Sections collected in the first and second wells were processed for NADPH-d histochemistry and counterstained with neutral red; sections in the third well were processed for nNOS immunohistochemistry and those in the fourth well were processed for nNOS immunofluorescence labeling combined with NADPH-d histochemistry.
NADPH-d used as a selective marker for nNOS as described previously [21, 27, 43, 44]. Briefly, sections in the first two wells were incubated with NADPH-d medium (0.1 mg/ml nitroblue tetrazolium, 1 mg/ml β-NADPH, and 0.3% Triton X-100 in 0.1 M PB [pH 7.4]) for 1 h at 37°C, washed several times in 0.1 M PB to terminate the reaction. For counting neuronal numbers, sections from the second well were further counterstained with neutral red, dehydrated through a graded series of alcohols, cleared with xylene, and coverslipped with Permount.
Sections in the third well were rinsed in 0.01 M phosphate buffered saline (PBS), pH 7.4, treated with 0.01 M PBS containing 10% methanol and 3% hydrogen peroxide for 1 h to abolish the endogenous peroxidase activity, rinsed 3 times with PBS, and incubated with medium containing 3% normal horse serum, 2% bovine serum albumin, and 0.1% Triton X-100 for 1 h. The reacted sections were then washed several times with PBS, incubated with antibody to nNOS (1:100; Santa Cruz Biotechnology, Burlingame, CA, USA) for 24 h at 4°C, treated with biotinylated secondary antibody (1:200; Vector Laboratories, Burlingame, CA, USA) for 1 h at room temperature, and incubated with Streptavidin/HRP (DAKO A/S, Glostrup, Denmark). The signal was developed with diaminobenzidine (a peroxidase substrate).
Colocalization of nNOS and NADPH-d reactivity
Colocalization of NADPH-d and nNOS was detected by immunofluorescence labeling of nNOS followed by histochemical staining for NADPH-d [21, 27, 43, 44]. Sections in 4-well dishes were rinsed in 0.01 M PBS, incubated in medium containing 3% normal goat serum, 2% bovine serum albumin, and 0.1% Triton X-100 for 1 h to block nonspecific binding, rinsed with PBS, and incubated with mouse monoclonal antibody against nNOS (1:100; Santa Cruz Biotechnology) for 24 h at 4°C. The reacted sections were then treated with fluorescein isothiocyanate (FITC)-conjugated anti-rabbit secondary antibody (1:200; Vector Laboratories) for 2 h at room temperature, washed several times with PBS, and coverslipped with buffered glycerin to prevent fading. The tissue slides were photographed using a ZEISS fluorescence microscope equipped with an appropriate excitation filter (450-490 nm) for observing FITC-labeled nNOS-immunoreactive neurons. After photographing, the sections were again washed several times again in 0.1 M PB, processed for NADPH-d histochemistry (as described above), rapidly dehydrated through a graded series of alcohols, cleared with xylene, and coverslipped with Permount.
Quantitative study and image analysis
In each animal, 20-30 sections representing the entire length of the HN and DMN (extending from the obex 1.8 mm caudally and 1 mm rostrally) were collected for NADPH-d histochemistry and neutral red counterstaining and cell counting. To avoid bias for the counts, large neurons (25-50 μm) with clearly outlined nuclei but not small motor neurons were counted due to the latter could not be distinguished from interneurons being 10-18 μm in diameter. Both the NADPH-d positive and negative (neutral red-counterstained) cells in the HN and DMN, either ipsilateral or contralateral to the nerve crush site, were counted and summed to present the total number of cells in 20-30 sections (unit volume, u). The labeling percentage was calculated by dividing the number of NADPH-d(+) neurons on the lesion side by the total number of neurons on the same side. To confirm the coexistence of NADPH-d(+) and nNOS(+) neurons, double-labeled neurons in the HN and DMN of the same sections were counted on the photomicrographs (data not shown). Cells containing nNOS were counted in the sections processed for NADPH-d histochemistry since neuronal NADPH-d is a reliable marker of nNOS [21, 43]. Besides, the results of neuronal staining in sections stained for NADPH-d are usually better than nNOS, which in turn facilitates computer-assisted quantitative assessment. Labeling intensity of NADPH-d(+) neurons was quantified using a computer-based image analysis system (MGDS) and Image Pro-Plus software (Media Cybernetics, Silver Spring, MD, USA). Labeled sections were scanned at 100 × magnification in bright field using a digital camera mounted on a ZEISS microscope and the images were displayed on a high-resolution monitor. Up to 100 cells per section were measured along the entire length of the HN and DMN. At 100 × magnification, the cytoplasmic optical density (OD), which was used as an index of labeling intensity, in NADPH-d(+) neurons was measured by tracing the contour of the labeled soma in digitized images. The background OD in each section was measured by averaging five random polygons (area of polygon = 150 μm2) within the neuropil of the corresponding HN and DMN regions on the non-lesioned side. The staining intensity in a tissue section reflected the amount of enzyme activity. Thus, all parameters were controlled using methods to ensure consistent gray level adjustment, histogram stretch, and minimal optical density. To avoid introducing bias, two observers blinded to the animal treatment group counted cells in the NADPH-d stained sections and evaluated the data from image analysis of HN and DMN neurons.
All data of this study were expressed as mean ± SEM and statistical significance was determined with a commercially available software package (SPSS version 12; SPSS, Chicago, IL). Between-group differences in the percentage and OD of NADPH-d positive neurons at various time points, with or without EGCG treatment, were evaluated using one-way analysis of variance (ANOVA) followed by LSD post hoc test. The data collected in experimental groups within the same time point (i.e., 10 mg/kg EGCG pretreated versus control non-treated group) were further analyzed using Student's t test. Statistical difference was considered significant if p < 0.05.
Some negative controls were included to ensure the validity of the NADPH-d histochemical and nNOS immunohistochemical results. Thus, the medium without β-NADPH was used as the control in NADPH-d staining protocol, whereas media without primary or secondary antibodies were used as controls in nNOS immunochemical procedure.