Experiments were performed using male Sprague-Dawley rats (body weight, 250–300 g) with the approval of the Bioethics Committee of Beijing Friendship Hospital, Capital Medical University. Rats were housed three per cage in a room with an environment temperature of 21–24°C with a 12:12-hour light–dark cycle and were fed standard rat chow with ad libitum access to water. Rats were fasted for 12 hours before surgery.
Establishment of the septic rat model and treatment
Sepsis was induced by cecal ligation and puncture (CLP), as described previously to establish a rat model of abdominal infection after surgery. An intraperitoneal injection with 10% chloral hydrate at 0.03 mL/100 g for anesthesia was used to maintain spontaneous breathing, and the animal was placed supine on the heated platform that can maintain the temperature at 37°C. A longitudinal skin midline incision was made with a scalpel, being careful not to penetrate the peritoneal cavity, and then small scissors were used to extend the incision and to gain entry into the peritoneal cavity (3–4 cm). The linea alba was identified, and we made an intermuscular incision. When locating the cecum and exteriorizing it, it is critical not to breach or damage the mesenterial blood vessels. The membrane was dissected at the mesenteric site of the cecum, and the cecum was ligated at half the distance between the distal pole and the base of the cecum (mid-grade sepsis). The cecal contents were gently pushed toward the distal cecum, and at the time of cecal puncture using 18-gauge needles, we gently aspirated any trapped air or gases to avoid puncturing blood vessels. After removing the needle, we extruded a droplet of feces from both the mesenteric and antimesenteric penetration holes to ensure patency. We moved the cecum into the abdominal cavity, and closed the peritoneum, fasciae, and abdominal musculature with wax-coated braided silk nonabsorbable surgical sutures 4-0. Animals were resuscitated by injecting prewarmed normal saline (37°C, 5 mL per 100 g body weight) subcutaneously to demonstrate the early, hyperdynamic phase of sepsis. Rats were returned to their cages immediately at the end of the surgical procedures where access to water and food was available.
Rats were divided into three groups: sham group, the cecum was squeezed gently (no perforation) and was injected with the same volume of 0.9% saline via the tail vein; sepsis group, the steps are described above; and ulinastatin group, rats were treated with 50 000 U/kg/24 h of ulinastatin via the tail vein immediately after the operation.
We measured the blood pressure and heart rate of rats with a rat tail blood pressure monitor (BP-98A, Softron, Tokyo, Japan). Blood lactic acid was measured using a portable blood lactate meter (YK-Scout, Germany).
Assessment of renal microcirculation perfusion
All rats were anesthetized with 3% isoflurane in oxygen before imaging. The rats were placed on the ventral side of the heated imaging platform after anesthesia was induced. Anesthesia was maintained by delivering 1%–2% isoflurane in oxygen through a nose cone during imaging. We used the Vevo 2100 ultrasound system (FujiFilm VisualSonics Inc., Toronto, Canada) in B-mode. We placed the MS-400 probe in the renal artery to measure the width of the renal artery, and used the M-mode to measure the arterial flow rate. The analysis indices were the renal artery resistance index (RI) and average blood flow velocity. We set the relevant parameters as follows: frequency, 30 MHz; power, 100%; frame rate, 12; gain, 28.0 dB; dynamic range, 60 dB; and depth, 7.07 mm.
The unit was equipped with an MS250 transducer, and we used a wide beam width setting to ensure a low, uniform transmission pressure for all depths. We set the relevant parameters as follows: frequency, 18 MHz; energy, 10%; frame rate, 36; contrast gain, 37.0 dB; two-dimensional gain: 18.0 dB; depth, 20 mm; and dynamic range, 35 dB.
The focus range was placed in the middle of the left kidney, and 5 mL of physiological saline was injected into the ultrasonographic contrast agent (SonoVue lyophilized powder; Bracco, Milan, Italy) according to the manufacturer’s instructions, and shaken to form a sulfur hexafluoride microbubble suspension. We withdrew 0.2 mL/150 g of rat body weight, and injected a rapid bolus into the tail vein of each rat using a dedicated syringe pump, followed by physiological saline (0.5 mL). Following the contrast injection, we recorded 500 frames and analyzed the stored dynamic images offline using Vevo LAB (VisualSonics, Inc.) and Vevo CQ software (nonlinear amplitude modulation contrast imaging, VisualSonics, Inc.).
The cortical and medullary regions of interest were evaluated at the same depths and positions, as much as possible, over a 0.4-mm2 area, and the time intensity curve was plotted. The analysis indicators were the peak enhancement (PE), representative of blood volume, mean transit time (mTT), renal resistance index (RI), pulse index (PI), and wash-in perfusion index (WiAUC/RT), representative of blood flow. All parameters were measured three times and averaged.
Enzyme-linked immunosorbent assay
Blood and urine samples were obtained. Renal function was monitored by measuring the concentration of creatinine (E02C0629, Bluegene) and blood urea nitrogen (E02C0697, Bluegene) in serum using enzyme-linked immunosorbent assay (ELISA) kits. We measured the neutrophil gelatinase-associated lipocalin (NGAL) level using rat NGAL ELISA kits (ab119602, Abcam, Cambridge, UK). Serum levels of tumor necrosis factor (TNF)-α (ab46070, Abcam), interleukin (IL)-1β (ab100767, Abcam), and IL-6 (ab119548, Abcam) were detected using a rat bioactive ELISA assay.
Contrast-enhanced ultrasonography of the kidney
We performed CEUS at four time points (3, 6, 12, and 24 hours). By injecting the contrast agent via the tail vein of the rats in the sham group, the contrast agent rapidly reached the renal cortex, and the medullary echo from the renal sinus along the renal arteries was visible. Eventually, the entire renal cortex and medulla formed a "fireball-like" enhancement and then began to subside, and the cortex subsided last.
Histological examination and kidney Paller score
The left kidney was isolated and fixed by immersion in 4% paraformaldehyde, embedded in paraffin, and sectioned at a thickness of 4 µm. Following hematoxylin and eosin staining, pathological changes in kidney cortical and medullary tissues were examined under a light microscope, and slides were reviewed blindly and scored on a semiquantitative scale to evaluate changes found in acute renal failure.
We used the Paller score to assess the severity of renal tubular injury. Slides were reviewed blindly and scored on a semiquantitative scale. Ten high-power fields were randomly selected under a light microscope, avoiding repeated scoring of different convolutions of the same tubule, and renal tubular injury was evaluated according to the Paller score. Higher scores represented more severe damage (maximum score per tubule was 10), with points given for the presence and extent of tubular epithelial cell flattening, dilated renal tubules (1 point), brush border injury, shedding (1 or 2 points), cytoplasmic vacuolization (1 point), interstitial edema (1 point), cell membrane bleb formation, cell necrosis, and tubular lumen obstruction (1 or 2 points).
The left kidney obtained from each group was frozen in liquid nitrogen and stored at -80°C. Tissue samples from various groups were homogenized using a protein extraction reagent containing protease inhibitors. In brief, protein samples were electrophoresed on 4%–12% SDS-PAGE polyacrylamide gels, incubated with the appropriate antibodies (VE-cadherin Rabbit mAb, ab231227, Abcam), LC3II (E5Q2K) Mouse mAb (#83506, CST), and measured using an enhanced chemiluminescence detection system. Densitometric analysis was performed using ImageJ software (National Institutes of Health, Bethesda, MD).
Statistical analyses were performed using SPSS 22.0. The data are expressed as the mean ± SD. After homogeneity test of variances, one-way analysis of variance (ANOVA) followed by multiple comparison of the Tukey test was used to determine the statistical significance of different groups. All experiments were conducted at least in triplicate from different cell or tissue samples. Differences were considered statistically significant at P < 0.05.