In total, 128 male Wistar rats (10–12 weeks old, weight: 230–280g) were provided by the Institute of Zoology, Chinese Academy of Sciences. The rats were anaesthetized with intraperitoneal injection of 3% sodium pentobarbital (30–50mg/kg). The MAP was monitored by placing a catheter in separated right common carotid artery and measured by connecting a pressure transducer. Arterial blood was sampled by placing a catheter in the carotid artery. HSD and 0.9% NaCl solution (NS) were administrated by placing a catheter in the separated left jugular vein. A rat sepsis model was established using the CLP approach. We performed traditional disinfection of the abdomen and a median skin incision of approximately 2–3cm to expose the abdominal cavity. The cecum was freed from the mesentery; the base of the cecum was ligated with a 3-0 suture, and punctured at two sites 3mm apart by using a No. 9 needle. The intestinal canal was then restored to ensure smooth intestinal passage, the abdominal layers were sutured, the lost fluid was replenished, and the wounds were bandaged after disinfection with tamed iodine. Operation procedures for the sham operative (SOP) group were the same as-mentioned above, except for the CLP approach.

The rats were randomly divided into four groups: the SOP group (n=21) received a subcutaneous injection of 30mL/kg 0.9% NaCl after the operation; the CLP group (n=45) received a subcutaneous injection of 30mL/kg 0.9% NaCl after the operation; the CLP+NS group (n=45) received a subcutaneous injection of 30mL/kg 0.9% NaCl and a jugular vein infusion of 5mL/kg 0.9% NaCl at 0.4mL/(kgmin) for 3h after the operation; and the CLP+HSD (n=28) group received subcutaneous injection of 30mL/kg 0.9% NaCl and a jugular vein infusion of 5mL/kg 7.5% NaCl/6% dextran at 0.4mL/(kgmin) for 3h after the operation.

The experiments were performed in adherence with the National Institutes of Health guidelines for the treatment of animals and ethical animal research. In addition, animals were killed at 0, 9, and 18h after the operation. Blood samples and lung tissues were collected for examination. Equal volumes of 0.9% NaCl were immediately injected upon each blood sampling.

Plasma TNF-α, IL-1β, and IL-6 levels were determined using enzyme-linked immunosorbent assay (ELISA). Blood samples were collected from the common carotid artery at 0, 9, and 18h after CLP. Blood aliquots of 0.8mL were added to Eppendorf tubes containing sterile pyrogen-free ethylenediaminetetraacetic acid (EDTA) anticoagulant and centrifuged (3000rpm) for 15min at 4°C. The plasma samples (100μL) obtained at 0, 9, and 18h after the operation were used, and the levels of TNF-α, IL-1β, and IL-6 were measured in duplicate using an ELISA kit (R&D Systems, Minneapolis, MN), according to the manufacturer's instructions.

The carotid artery blood of rats in each group was sampled (0.2mL) at 0, 9, and 18h after the operation, and the pH, PaO2, and PaCO2 in the blood samples were assayed using a portable blood gas analyzer (i-STAT, Abbott, Chicago, USA).

BALF collection and cell sorting methods were performed as reported by Reis et al.7 and Callol et al.8 The trachea and lung of the sacrificed rats were exposed at 18h after the operation. The right bronchus was ligated, the trachea was fixed on the annular cartilage with forceps, and a transverse incision was made under it. By removing the No. 7 needle, an approximately 10-cm-long plastic infusion tube was kept remaining. The tube was inserted into the tracheal cavity, sutured at one end, and fixed on a 5-mL syringe at the other end. Sterile saline (2mL, 37°C) was slowly injected into the lung, and suctioned back after 30s. The above procedures were repeated thrice and a total of 5mL of BALF was collected. The BALF was centrifuged at 3000rpm for 10min and the cells in the pellet were stained using Wright's staining. One hundred cells were counted and classified under a microscope equipped with an oil-immersion lens according to leukocyte features. The percentage of neutrophils was calculated as follows: percentage of neutrophils (%)=number of neutrophils counted/number of total cells counted×100%.

All the rats were sacrificed 18h after the operation. The frozen right lower lobe lung samples were then weighed and homogenized in 2mL of 50mmol/L phosphate buffer at pH 6.0 containing 0.5% hexadecyltrimethylammonium bromide (Sigma Chemical Co., St. Louis, MO). The samples were then frozen on dry ice and thawed at room temperature thrice, after which they were sonicated. The suspensions were then centrifuged at 4000rpm for 15min. MPO activity in the supernatant was assessed by measuring the hydrogen peroxide (H2O2)-dependent oxidation of o-dianisidine dihydrochloride (Sigma Chemical Co.), according to a previous report.9 We mixed 0.1mL of the supernatant with 2.9mL of phosphate buffer (50mmol/L, pH 6.0) containing 0.0005% hydrogen peroxide (Sigma) and 0.167mg/mL o-dianisidine dihydrochloride (Sigma). The change in absorbance was measured at 460nm using a spectrophotometer (U-2000; Hitachi Instruments, Webster, NY). MPO activity was derived from the observed change in absorbance per minute. The MPO activity was normalized further to the total protein content of the supernatant, which was measured by the micro-Lowry technique.10 Activity is expressed as units of MPO activity per milligram of protein.

After the rats were sacrificed, the right lung tissues were harvested and the wet weight was measured first. Then, the tissue samples were dried at 80°C for 72h and weighed again to obtain the dry weight. The ratio of wet to dry weight was calculated.

The lung tissues were harvested at 18h for histopathological analyses. The harvested lung tissues were fixed in buffered formaldehyde (10% in PBS, pH 7.4) for 8h. The fixed organs were dehydrated in graded ethanol and embedded in paraffin (Tissue-processor, Japan). Then, 4-μm sections were cut using a sliding microtome (Leica Jung SM 2000, Germany) and the paraffin was removed using xylene. The tissue sections were then stained with hematoxylin and eosin and visualized under a light microscope.

The rats were sacrificed at 9 and 18h after the operation. Total RNA was extracted using the one-step approach by adding 1mL Trizol to 100mg of right lung tissue. cDNA was synthesized according to the manufacturer's instructions, and 1μL cDNA was used as the template for PCR amplification using the following primer sequences: upstream: 5′-CGGTAGACACAAGCAAGAGA-3′; downstream: 5′-GCAGGGATTGACCATAATT-3′; β-actin (207bp) upstream: 5′-CCAACTGGGACGATATGGAG-3′; downstream: 5′-CAGAGGCATA CAGGGACAAC-3′. The PCR reaction conditions were as follows: predenaturation at 94°C for 5min until start of the cycle; followed by 35 cycles at 94°C for 30s, 56°C for 30s, and 72°C for 1min; and a final extension step at 72°C for 7min. For β-actin amplification, the following conditions were used: predenaturation at 94°C for 5min until start of the cycle; followed by 35 cycles at 94°C for 30s, 56°C for 30s, 72°C for 1min; and a final extension step at 72°C for 7min. After agarose gel electrophoresis, the amplified products were scanned using a gel imaging system. The mRNA expression level was expressed as the ratio of ICAM-1 to β-actin band luminosity.

The survival rate of experimental animals in each group was given by point statistics at 9 and 18h postoperatively.

Measurement data were statistically processed using SPSS 11.5 software and expressed as means±standard deviation (x¯±s). Means among multiple groups were compared using ANOVA, two groups were compared using t-test, and p<0.05 was deemed statistically significant.

The baseline MAP in each group was not significantly different (Table 1). During the entire research period, the MAP was stable in the SOP group, but decreased significantly between 9 and 18h in the CLP and CLP+NS groups. Hemodynamic disorders were detected but the MAP was significantly better in the CLP+HSD group than in the CLP and CLP+NS groups at 9h. The difference was statistically significant and the HSD showed a continued improvement in blood pressure from 9h onwards.

The baseline TNF-α, IL-1β, and IL-6 levels in each group were not significantly different (Table 1). The plasma TNF-α, IL-1β, and IL-6 levels in the SOP group were not significantly different at 9 and 18h, suggesting that the laparotomy did not significantly influence the expression of these markers. The plasma TNF-α levels in the CLP and CLP+NS groups were significantly increased at 9h and gradually increased thereafter. During the entire experiment phase, the plasma TNF-α level in the CLP+HSD group was significantly lower than the corresponding values in the CLP and CLP+NS groups (Fig. 1A). Furthermore, in the CLP and CLP+NS groups, the IL-1β expression levels increased sharply at 9h and 18h. The plasma IL-1β level in the CLP+HSD group showed biphasic changes and although it was significantly lower than that in the CLP, the IL-1β level in this group was significantly different from that in the SOP group at 9 and 18h (Fig. 1B). The plasma IL-6 levels in the CLP and the CLP+NS groups increased sharply until 9h and decreased slowly thereafter. The plasma IL-6 level in the CLP+HSD group was initially lower than that in the CLP and CLP+NS groups; however, the levels were similar among these groups at 18h (Fig. 1C).

Fig. 1.

Effect of HSD on (A) TNF-α, (B) IL-1β, (C) IL-6, (C) pH, (D) PaO2, and (E) PaCO2 in rats with sepsis induced by peritonitis. The graphs illustrate the changes during the experimental period in animals that received laparotomy plus vehicle (SOP; n=10), cecal ligation and puncture (CLP; n=45), CLP plus normal saline (CLP+NS, jugular vein injection of 5mL/kg 0.9% NaCl at an infusion speed of 0.4mL/(kgmin) for 3h after the operation; n=45), and CLP plus HSD (CLP+HSD, jugular vein infusion of 5mL/kg 7.5% NaCl/6% dextran at an infusion speed of 0.4mL/(kgmin) for 3h after the operation; n=28). Data are expressed as means±SEM. *p<0.05 for all groups vs. sham. #p<0.05 for HSD vs. NS.

(0.26MB).

The baseline levels of arterial blood pH, partial pressure of oxygen, partial pressure of carbon dioxide in each group were not significantly different (Table 1). The arterial blood pH in each group was decreased at 9h, but it further decreased at 18h in the CLP and CLP+NS groups, whereas the blood pH showed the same gradual increasing trend in the SOP and CLP+HSD groups (Fig. 1E).

The arterial partial pressure of oxygen did not significantly change in the SOP group, but it showed a sharp decrease at 9h followed by a gradual decrease until 18h in the CLP and CLP+NS groups. The arterial partial oxygen pressure in the CLP+HSD group decreased at 9h, but it was significantly higher than the corresponding values in the CLP and CLP+NS groups. Such effects continued until 18h (Fig. 1F).

The arterial partial pressure of carbon dioxide increased slightly at 9h in the CLP, CLP+NS, and CLP+HSD groups, but showed a continuous rising trend even at 18h in the CLP and CLP+NS groups. The arterial partial pressure of carbon dioxide in these two groups showed a higher increasing trend than that in the CLP+HSD group, which was slightly increased during the study duration but did not show significant change compared to the baseline value at 0h in the SOP group (Fig. 1G).

The percentage of BALF neutrophils was higher in the CLP and CLP+NS groups than in the SOP group. There were more BALF neutrophils in the CLP+HSD group than in the SOP group; however, this value was still significantly lower than the corresponding values in the CLP and CLP+NS groups (Fig. 2A). The MPO activities in the CLP and CLP+NS groups were significantly increased as compared to the SOP group; however, the MPO activity in the lung tissue in the CLP+HSD group was lower than that in the CLP and CLP+NS groups (Fig. 2B). The CLP and CLP+NS groups had a higher wet to dry weight ratio of lung tissues as compared to the SOP and CLP+HSD groups (Fig. 2C).

Fig. 2.

Effect of HSD on (A) BALF neutrophil number, (B) lung tissue MPO and (C) wet/dry weight ratio of lung tissue activities in rats with sepsis induced by peritonitis. The graph illustrates changes during the experimental period in animals that received laparotomy plus vehicle (SOP; n=6), cecal ligation and puncture (CLP; n=6), CLP plus NS (CLP+NS; n=6), and CLP plus HSD (CLP+HSD; n=6). Data are expressed as means±SEM. *p<0.05 for all groups vs. sham. #p<0.05 for HSD vs. NS.

(0.16MB).

There were no significant pathological changes in rat lung tissue sections in the SOP group at 18h (Fig. 3A). The lung tissues in the CLP and CLP+NS groups showed significant inflammatory pathological changes such as hemorrhage, alveolar congestion, thickening of alveolar walls/hyaline membrane formations, and infiltration and aggregation of neutrophils in airspaces or vessel walls (Fig. 3B and C). The CLP+HSD group had significantly fewer inflammatory changes as compared to the CLP and CLP+NS groups (Fig. 3D).

Fig. 3.

Histopathological studies of lung tissue. Light microscopy showed lung sections of rats in the (A) sham-operated (SOP) group, (B) cecal ligation and puncture (CLP) group, (C) CLP+NS group, and (D) CLP+HSD group. Sections were stained with hematoxylin and eosin (original magnification, 200×).

(0.35MB).

The ICAM-1 levels in the CLP and CLP+NS groups were increased at 9h and further increased at 18h. The ICAM-1 levels in the CLP+HSD group were significantly lower than those in the CLP and CLP+NS groups (Fig. 4).

Fig. 4.

Expression of ICAM-1 mRNA in lung tissues. The expression of ICAM-1 was the lowest in the SOP group. The expression level in the CLP and CLP+NS groups increased at 9h and 18h. The ICAM-1 level in the CLP+HSD group was significantly lower than that in the CLP and CLP+NS groups (p<0.05).

(0.14MB).

The rats in the SOP group did not die within 18h. The survival rates of animals in the CLP group at 9 and 18h were 62.2% and 31.1%, respectively. The survival rates of rats in the CLP+NS group at 9 and 18h were 57.8% and 35.6%, respectively. The survival rates of the CLP+HSD group at 9 and 18h were 85.7% and 64.3%, respectively. Thus, intravenous infusion of HSD can improve the early survival rate (within 18h) in CLP rats. However, the survival rate of the CLP+HSD group was decreased at 9h. Therefore, additional therapeutic means must be implemented in order to improve the survival rate of CLP rats using HSD.