Adult male Sprague-Dawley (SD) rats (provided by the department of experimental animals, China Medical University) weighing 300-400 g were used in the present study. They were housed in stainless steel cages, three rats per cage, under controlled temperature (23°C) and humidity conditions, with 12-h dark/light cycles, and maintained on standard laboratory diet with tap water ad libitum throughout the experiment, except for an overnight fast before surgery.

The experiments were carried out according to the guidelines set forth by the Ethics Committee of China Medical University, Shenyang, China.

Experimental protocol: Thirty rats were randomly divided into two groups: Group S (n = 10) served as sham-operation and group C (n = 20) served as CPB. The rats of group C underwent 60 min of CPB and were sacrificed at the end of CPB (n = 10) or at 2 h after CPB (n = 10) randomly.

Surgical procedure for CPB[20-22]: Rats were anaesthetized with intraperitoneal administration of chloral hydrate (350 mg/kg). Once the anesthesia was achieved for the surgical level, the rats were placed in supine position. A 14G catheter was inserted into the trachea and the lungs were mechanically ventilated with a small animal ventilator (DH2800, Zhejiang University Medical Apparatus Co, China). The tidal volume was approximately 10 mL/kg, the respiratory rate was 60/min, peak airway pressure was 20-30 mmH₂O and the respiration concentration of O₂ was 100%. Anesthesia was maintained throughout the experiment with additional doses of chloral hydrate (100-150 mg/kg). All subsequent procedures were performed under aseptic conditions.

The left femoral artery was cannulated with a 24-gauge catheter to monitor systemic arterial pressure (SAP) and to collect arterial blood (0.2 mL) for arterial blood gas analysis before CPB, 30 min, 1 h and 2 h after CPB (blood gasanalyzer, GEM Premier 3000, USA). The homolateral femoral vein was cannulated with a 20-gauge catheter for fluid replacement (0.9% NS 0.5-1.0 mL/h). Following administration of heparin (300 U/kg), a 16-gauge catheter, modified to a multiside-orifices cannula in the forepart, was inserted into the right jugular vein and advanced to the right atrium for blood drained. A 22-gauge catheter was cannulated to the right carotid artery which served as the arterial infusion line for the CPB circuit. The electrocardiogram (ECG) was performed and rectal temperature (RT) was monitored.

The minute CPB circuit comprised a venous reservoir, a specially designed membrane oxygenator (Ke Wei Co. Ltd, Guandong, China ), a roller pump(Longer Precision Pump Co. Ltd, Baoding, China), and sterile tubing with an inner diameter of 4 mm for the venous line and of 1.6 mm for the arterial line. The main priming fluid of the circuit constituted Hespander (synthetic colloid, 7 mL), Ringer’s lactate solution (7 mL), 15% mannitol (1 mL) and 8% sodium bicarbonate (1 mL). After careful check, the clamps were released and the bypass was started. With an inspired oxygen fraction of 100%, 50 mL/min gas flow was sufficient to achieve adequate oxygenation and to maintain the PaCO₂ within 35-45 mmHg. At the initiation of perfusion, the flow-rate was gradually adjusted to a level that would sustain the arterial pressure near 60 mmHg. At this point, mechanical ventilation was terminated and CPB was stably performed at 90-100 mL/kg per min throughout the experiment. Body central temperature was monitored with a rectal probe and kept at about 33°C. After 60 min, the bypass was terminated when the temperature gradually elevated to 37°C by a heat lamp placed above the animal and the infusion warmer (BIGGER, CBW68, USA). The mechanical ventilation was resumed until the rats gained steady active breath. The remaining priming solution was infused gradually after the termination of CPB and dopamine (3 μg/kg per min) was infused when the main arterial pressure was less than 60 mmHg.

The anesthetic and surgical procedure performed in the sham-group (n = 10) were the same as in the CPB group. Stable Mbp was maintained, but without CPB, for the same duration of surgery and anesthesia as in the CPB 1 h or 2 h group.

Sample collection: The rats from each group were sacrificed by decapitation. The arterial blood (0.2 mL) was sampled before CPB, at the end of CPB and 2 h after CPB for determination of serum DAO and d-lactate levels. At end of CPB and 2 h after CPB, intestinal (ileum) tissue samples were collected for immunohisto-chemical and Western blotting analysis. The animals in the sham-group were sacrificed at the end of the time period equivalent to the CPB-2 h group. All samples were determined in a blinded fashion; results were not available until completion of the study.

Plasmad-lactate measurement: Plasma d-lactate levels were measured by an enzymatic spectrophotometric assay using a centrifugal analyzer at 30°C (Hoffmann-LaRoche, Basle, Switzerland) as described earlier[23]. d-lactate, d-lactate dehydro-genase and NAD⁺ were purchased from Sigma Chemical Company (Milan, Italy).

Plasma DAO measurement: Plasma DAO activities were determined using an enzymatic spectrophotometry as described earlier[24]. O-dianisidine, Gadaverine Dihydrochloride and DAO were purchased from Sigma Chemical Company (Milan, Italy).

Immunohistochemistry for occludin and ZO-1: Four to 5 μm thick histologic sections, mounted on gelatinized glass slides, were dewaxed in xylene and hydrated through a series of graded concentrations of alcohol. Endogenous peroxidase activity was blocked with 1% hydrogen peroxide in methanol for 15 min. Sections were then processed in a microwave oven twice, 5 min each time at high power for antigen retrieval. Subsequently, a standard streptavidin-biotin-peroxidase technique was applied to detect occludin and ZO-1. After incubation with 1% bovine serum albumin to block nonspecific binding, sections were incubated with rabbit polyclonal antioccludin or antiZO-1 antibody (1:100, Zymed Laboratories, San Francisco, CA) for 1 h, and with biotinylated antirabbit immunoglobulin-G for 30 min. Color was developed with diaminobenzidine (Sigma Fast DAB tablets, D-4293) and counterstained with Harris hematoxylin. All procedures took place at room temperature and sections were washed in tris-buffered saline. The specimens were examined under microscope (Olympus BX51, Japan). The integrated optical density of ZO-1 or occludin in intestinal sections was analyzed using an image analysis system (UIC, MetaMorph, USA).

Western blotting analysis: After rats were killed as described above, terminal ileums were isolated, excised, flushed with ice-cold PBS to remove the fecal debris, and then opened longitudinally. Intestinal mucosa was harvested with a razor blade. Tissues were homogenized by mechanical disruption in lysis buffer (50 mmol/L Tris-Cl, pH 8.0, 150 mmmol/L NaCl, 0.1% dodecyl sodium sulfate, 1% Nonidet P-40, 0.5% sodium deoxycholate, 100 mg/mL phenylmethylsulfonyl fluoride, 1 mg/mL Aprotinin) and incubated on ice for 30 min. Homogenate was then pelleted at 6000 g for 10 min and supernatant was collected. Using Lowry method, the quantity of total protein was determined in the supernatant. The protein extracts were incubated with primary antisera (rabbit anti-occludin and anti-ZO-1, 1:500; Zymed Laboratories, San Francisco, CA) at 4°C overnight and then the mixture was incubated with secondary antibody (horseradish peroxidase conjugated goat anti-rabbit IgG, 1:1000; Beijing Zhongshan Biotech,Co.) at 4°C for 6-8 h. Immunoprecipitates were subjected to SDS-PAGE and electrotransferred to nitrocellulose membrane. After a final wash in TBST, using the enhanced chemiluminescence (ECL) kit (Shang hai Pufei Biotech, Co.), chemluminescent signals were collected on autoradiographic film. The quantity of band intensity was carried out using electrophoresis Gel Imaging Analysis System (ChemiImager 5500, ALPHA INNOTECH, USA). Results were expressed as ratio of relative intensity of target protein to that of the internal standard, β-actin.

Electron microscopy: Samples were immediately fixed in 2.5% glutaraldehyde in 0.1 mol/L sodium cacodylate buffer (pH 7.4) for 2 h at 22°C, rinsed for 18 h (4°C) with 0.05 Tris buffer (pH 7.6), and washed three times， for 5 min each time. An H-2000 transmission electron microscope was used for the examination.

Data were analyzed using a sigma stat statistical software (SPSS, Chicago, IL) and expressed as mean ± SD. Statistical significance of differences between groups were determined using one way analysis of variance (ANOVA) followed by LDS (for multiple comparisons. Linear regression analysis was used to assess associations between TJ protein relative expression levels and DAO or d-lactate levels at each period. P < 0.05 was considered as significant in difference.

All experiments progressed without incident and all animals survived from the operative process. Table 1 displays the physiologic data from the CPB and Sham rats over the study period. In the CPB group, CPB and post-CPB MAP, PaO₂ values were significantly lower than baseline (P < 0.05), whereas PaCO₂ remained stable. In the Sham group, there were no changes in MAP, PaO₂ and PaCO₂during the whole experiment. Hematocrit (HCT) had a significant decrease in the CPB group (P < 0.05) during the perfusion process but had no change in the Sham group. The pH values in the CPB group decreased and this group showed a tendency to develop a metabolic acidosis.

There were no significantly differences between CPB groups and Sham group at baseline (P > 0.05, Table 2). In the CPB group, CPB and post-CPB DAO activity was significantly higher than baseline (P < 0.05), and DAO activity at 2 h after CPB was significantly higher than that at the end of CPB (P < 0.05). In the Sham group, there were no changes in plasma DAO activity during the whole experiment.

There were no significantly differences between CPB groups and Sham group at baseline (P > 0.05, Table 2). Plasma d-lactate levels increased significantly (P < 0.05) at the end of CPB compared with baseline or control group and remained higher at 2 h after CPB. In the Sham group, there were no changes in plasma d-lactate levels during the experiment.

All of the epithelial cells lining villi exhibited positive immunostaining for ZO-1 and occludin in control group (Figures 1-2A). In group C, there was loss of occludin and ZO-1 expressions in about 50% enterocytes lining villi (Figures 1-2B); this effect was more profound at 2 h after CPB and loss was about 80% (Figures 1 and 2).

Western blot analyses demonstrated alterations in expression of tight junctional proteins after CPB. Figure 3A shows the integral protein occludin with dual bands at 65-kDa .The 65-kDa band showed a significant (P < 0.05) decrease in group C compared with control(76.4% ± 22.5% vs 96.5% ± 28.5%) and reached lower value at 2 h after CPB (50.5% ± 10.5%). Figure 3B shows that ZO-1 expression significantly (P < 0.05) decreased after CPB compared with control (62.4% ± 10.1% vs 85.5% ± 25.6%) and became lower at 2 h after CPB (45.3% ± 9.5%).

To assess the relationship between changes of occludin and ZO-1 expressions and the damages of intestinal mucosa after CPB, we used regression analysis to assess associations between TJ protein relative expression levels, DAO and d-lactate levels at each period. Significant negative correlations were observed between occludin or ZO-1 expression and DAO (r² = 0.5629, r² = 0.5424, P < 0.05) or d-lactate levels (r² = 0.6512, r² = 0.7073, P < 0.05) both at the end of CPB and at 2 h after CPB.

To assess the relationship between changes of occludin and ZO-1 expressions and alterations of intestinal mucosal ultrastructure after CPB, we performed a morphological analysis of TJ (Figure 4), as occludin is an important component in TJ. In control group, regularly-aligned microvilli in intestinal epithelium, integral mitochondria and rough endoplasmic reticulum (RER) and distinct junction complex were observed. In experimental group, the amounts of microvilli were decreased with irregular length and arrangement. The mitochondria were swollen with cracked and vacuolated cristae. Some structures of the RER were destroyed. Intercellular space between epithelial cells was widened. The structure of TJ became short, and the dotted crystal structures were obscured or disappeared; other structures were disrupted and even disappeared. The damage was more severe at 2 h after CPB.

Some studies have shown that cardiopulmonary bypass (CPB) predisposes the intestines to inadequate perfusion, hypoxic injury, and increases in gut permeability. Tight junctions (TJs) are located at the apical part of lateral membranes of polar epithelial cells, forming a barrier that regulates the permeability of ions, macromolecules, and cells through the paracellular pathway. There are no previous studies investigating whether CPB disrupts intestinal mucosal barrier by altering the formation and function of TJs.

In general, changes in mucosal permeability and morphology during CPB are used to reflect the degree of damage of the barrier function. But studies on the molecular mechanism of the increased permeability of intestinal mucosa during CPB are rare.

In this study, the effects of CPB on occludin and ZO-1 expressions in the intestinal epithelium of rats were investigated. Authors chose the rat model because of the anatomical and physiologic similarity of the digestive tract and cardiovascular systems in rats and humans.

The results show for the first time that CPB induced gut barrier dysfunction is associated, at the cellular level, with regional loss of occludin and ZO-1 expressions in the intestinal epithelium. These findings offer further insight in the pathophysiology of gut barrier failure during and after CPB. Understanding the pathophysiology of barrier alterations during CPB at the molecular level will help find novel and more effective treatment approaches.

This is an interesting paper and it deals with a relevant clinical problem. The study appears well performed and the hypothesis is clear.