The mortality of acute lung injury and acute respiratory distress syndrome (ALI/ARDS) remains high and efforts for prevention and treatments have shown little improvement over the past decades. The present study investigated the efficacy and mechanism of leukocytapheresis (LCAP) to partially eliminate peripheral neutrophils and attenuate lipopolysaccharide (LPS)-induced lung injury in dogs. A total of 24 healthy male mongrel dogs were enrolled and randomly divided into LPS, LCAP and LCAP-sham groups. All animals were injected with LPS to induce endotoxemia. The serum levels of leucocytes, neutrophil elastase, arterial blood gas, nuclear factor-kappa B (NF-
Acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) are characterized by increased permeability of the alveolar-capillary barrier, resulting in an influx of protein-rich edematous fluid and a consequent impairment in arterial oxygenation. Mortality remains high in spite of sophisticated intensive care [
It is generally accepted that most forms of ALI involve lung neutrophil entrapment and activation, as well as neutrophil-mediated pulmonary injury [
Leukocytosis, a condition characterized by an elevated number of leukocytes in the blood, is the prominent feature of bacteritic sepsis. In patients with sepsis-related ALI, significantly higher neutrophil counts (10-fold) in the bronchoalveolar lavage fluid (BALF) were consistently noted in non-survivors as compared to survivors, 7–14 d after intubation [
Several
The Experimental Animal Care and Use Committees, Third Military Medical University, Chongqing, China, reviewed and approved the animal use and care protocols of this study. Twenty-four dogs (15–20 kg) from the Animal Center of the university were equally divided into LPS, LCAP, and LCAP-sham groups. Animals were housed at room temperature with a 12-h/12-h light/dark cycle for one week and deprived of food for 12 h before surgery.
Animals received general anesthesia induced by intravenous injection of ketamine (25 mg/kg body weight) and xylazine (7 mg/kg). Sodium pentobarbital (1 mg/kg) was intermittently administrated to maintain anesthesia. All surgery procedures were performed in an animal surgery facility under sterile conditions.
After tracheotomy and intubation, mechanical ventilation was conducted using SIMV with Tv 10 mL/kg, 30 breathes/min, inspiratory oxygen fraction 29%, and PEEP 3 cm H2O (Newport 200, USA). The distal end piece of the infusion set was inserted into the left jugular and right femoral veins to prepare for connection to a blood cell separator (COM.TEC, Fresenius, Germany). A heparin-filled catheter was inserted into the right femoral artery for monitoring blood pressure through a pressure transducer and collecting blood samples. The parameters were continuously recorded with PowerLab/16SP (AD Instruments) for data acquisition.
Following anesthesia, monitoring, ventilation, and vessel preparation, all animals were intravenously injected with LPS (2 mg/kg
LCAP was performed by an automated continuous-flow blood cell separator after stable hemodynamics were attained. The mononuclear cell program was selected to separate peripheral leucocytes. The blood flow rate and total number of separation cycles were set according to the animal’s gender, body weight, height, hematocrit level and targeted peripheral leucocyte count, and the total separation cycles were adjusted as needed. The counts of leukocytes and neutrophils in the periphery and collected storage bag were sampled to estimate the efficiency of separation during the LCAP process.
The sham-LCAP group did not undergo removal of the leucocytes, carried out by continuous reinfusion of separated leucocytes (Figure
Schematic summary of LCAP and sham-LCAP procedures. LCAP was performed by an automated continuous-flow blood cell separator. The parameters were set based on the targets. The total separated cycles were adjusted according to the efficiency of separation during the LCAP process. The sham-LCAP group underwent all the same procedures as the LCAP group except for removal of the leucocytes, carried out by continuous reinfusion of the separated leucocytes in the storage bag.
After anesthesia, as a basal value or at other time points, 5 mL of blood was collected, a portion of which was used for the leucocyte and neutrophil counts and arterial blood gas analysis (I-STAT, Abbott, USA). The remaining was centrifuged at 1500 ×g for 10 min; the supernatant was collected and stored at −80°C for other uses. Each animal received a chest X-ray when the oxygenation index was <300 mmHg.
The animals were euthanized at 36 h under anesthesia followed by BALF, tracheostomy, and
Myeloperoxidase and NE levels in serum and BALF were assayed with ELISA kits (R&D, USA) according to the manufacturer’s instructions, to determine neutrophil accumulation and NE release.
The total protein in BALF was measured by the method of Bradford according to the manufacturer’s protocol (Bio-Rad, USA). Tissue samples from lung were homogenized in PBS containing a protease inhibitor cocktail (Applygen, China) and nuclear fractionation was performed using an EZ nuclei isolation kit (Applygen, China) according to the manufacturer’s protocol. The supernatant was collected and mixed. The production of p65 and TNF-
The detection of MDA in parenchyma was performed based on the manufacturer’s instructions (Nanjing KeyGen, China). Optical density was read at 532 nm with the spectrophotometer. The values of MDA were expressed as nM.
The incised section of lung tissue was weighed immediately and dried at 96°C in an oven for 24 h and weighed again. The volume of lung tissue edema was measured by calculating the wet/dry weight ratio.
Tissues fixed in 4% paraformaldehyde were dehydrated in an ascending series of ethyl alcohol, cleared in xylene, and embedded in paraffin. Paraffin sections were cut at 15
The terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay was performed using an In Situ Cell Death Detection Kit, Peroxidase (Nanjing KeyGen, Nanjing, China) according to the instructions from the manufacturer. Ten fields of each section were randomly selected to count the numbers of positive apoptotic cells.
All values are expressed as the mean ± standard error (SE). The data were analyzed with Statistical Products and Service Solutions (SPSS) 13.0 for Windows software. Comparisons among the groups at the corresponding time points were performed using Student’s
Diagnoses for ALI and ARDS followed the criteria set by the American-European Consensus Conference of 1994 [
The counts of leucocytes and neutrophils in the periphery decreased rapidly following the injection of LPS, then increased slowly and returned to basal values about 16 h after injection, and afterward gradually increased further (Figures
The effect of LPS and LCAP on WBC and PMN counts and NE level in periphery. The numbers of WBCs and PMNs collected from the right femoral artery at indicated time periods after LPS injection were detected using the Digicell 500 cell counter and enumerated with Giemsa-Wright Stain. The animals in the LPS group underwent only LPS injection at the zero time point. Except for LPS administration, the animals in the sham-LCAP group underwent a sham-LCAP procedure (carried out by reinfusion of the end-products collected in a storage bag) at time point 16 h, for up to 4 h. and the animals in the LCAP group underwent both LPS injection and LCAP at corresponding time points. (a) WBC counts following the LPS injection and LCAP (/sham-LCAP). (b) PMN counts following LPS injection and LCAP (/sham-LCAP). (c) The level of NE following LPS injection and LCAP (/sham-LCAP). *indicates significant difference among the LCAP, LPS, and sham-LCAP groups at the indicated time (*
The average count of neutrophils in the LCAP group was statistically lower than in the LPS or the sham-LCAP groups (
The effect of LCAP on the average levels of PMN and NE in three groups. The numbers of PMN in the right femoral artery were counted at the indicated time period using the Digicell 500 cell counter (a), and the mean levels of NE in sera at the same time periods were determined by ELISA kits (R&D, USA) according to the manufacturer’s instructions (b). *indicates significant difference between the LCAP and the sham-LCAP groups (
The effect of LCAP on p65 protein levels in lung parenchyma, total protein concentration in BALF, wet/dry ratio in lung, and apoptotic changes in lung tissues. The p65 was examined by ELISA kits, and optical density was read at 450 nm (a). The water content of the tissues from left inferior lung was shown as wet/dry weight ratio (b). The total protein concentration in BALF assayed by methods of Bradford (c); apoptotic cells (arrow) in left inferior lung parenchyma were detected by TUNEL
In addition, the average levels of PMN and NE in peripheral blood were inversely correlated with oxygenation indices at 36 h (Figures
The correlation between oxygenation indices and the average levels of the PMN count in periphery and NE in serum. We calculated the mean levels of PMN in the peripheral blood at indicated time period and showed the correlation between the mean levels of PMN in the periphery and the oxygenation indices (PaO2/FiO2) at 36 h (
Lung interstitial edema, hyperemia, and neutrophil detention in microvascular exudation in lung interstitial and alveolar spaces were found in all groups after the LPS challenge. However, relatively minor abnormal lesions were observed in the LCAP group (Figures
The pathologic changes of lung tissues. Lung tissues from the left-lower lobe underwent routine fixation, dehydration, sectioning, and staining with H&E. General pathological changes of lung tissues were observed by optical microscope. The average neutrophil (arrows) counts were calculated in 10 random high-power fields per section under photomicroscope (10 × 40). (a) LPS group; (b) sham-LCAP group; (c) LCAP group.
The numbers of apoptotic lung parenchyma cells in the LCAP group were significantly lower compared to that in the LPS and the sham-LCAP groups (
Apoptotic changes in alveolar epithelial and endothelial cells. Frozen tissues from the left inferior lung were routinely sectioned (20
LCAP has been reported to remove both neutrophils and other leucocytes [
As one of the most destructive proteases, NE can degrade key structural elements of connective tissues in endothelial and epithelial cells, thrombomodulin, and proteoglycans in basement membrane and lung interstitium [
By promoting expressions of many downstream inflammatory factors related to lung damage, activation of NF-
The protein content in BALF, the PMN counts and NE level in serum and BALF, the water content of lung tissue, MDA and apoptotic changes in lung parenchyma are important biomarkers relevant to lung injury occurrence. They act as inflammatory mediators in activation of PMNs and result in development of ALI/ARDS. Currently, blockade of these factors in an attempt to reduce ALI/ARDS has limited success. Also, no dependable therapies are currently available that can safely and reversibly interrupt PMN detention, recruitment, and infiltration in the lung or inhibit the production of inflammatory mediators, especially NE and reactive oxygen species. It seems that the appearance of ALI/ARDS is unavoidable in many cases, including serious sepsis or trauma. Although LCAP could not completely reverse the ascending trend of these inflammatory factors after the endotoxin challenge, partial removal of peripheral PMNs might be a potential therapeutic approach for this illness.
Moderate increase in PMN plays an important role in protecting the organism against various insults. ALI/ARDS is the result of continuous attacks to lung tissues by excessive inflammatory mediators, of which NE is the most important. We reasoned that ALI/ARDS might occur only when PMN and NE levels in periphery are raised persistently, rather than transiently. Our results showed that the mean levels of PMN and NE were significantly lower in the LCAP group as compared to the LPS and sham-LCAP groups, with the exception of at 20 h and 24 h. Also, the mean levels of PMNs and NE in animals without ALI were significantly lower than those in animals with ALI. This might explain that the decrease in PMNs and NE for a limited time could also prevent the animals from developing ALI and implied that the occurrence of ALI was due to the continuous and cumulative damages from excessive PMNs and NE in the periphery. Thus, the interruption of continuous damages from excessive neutrophils and NE by LCAP may be an alternative therapeutic strategy.
Although determination of “excessive” leucocytes can be easily made according to the diagnostic criteria for systemic inflammatory syndrome or sepsis, it seems impossible to make an acknowledged conclusion regarding the exact level of neutrophils when faced with a complicated clinical case. Previous reports suggested that LCAP was safe with few and minor adverse effects when used for some refractory autoimmune diseases [
The findings in our endotoxemia model suggest that the partial removal of leucocytes and neutrophils attenuates lung damage and prevents animals from developing ALI without significantly damaging innate immunity.
Although our finding is encouraging, it was obtained from preliminary research only. More studies that include a larger sample population are needed to confirm this result. An additional method of exclusion of neutrophils may be required. Furthermore, the preventive and therapeutic efficacies of LCAP in endotoxemia-associated ALI/ARDS need to be explored further.
Acute lung injury
Acute respiratory distress syndrome
Bronchoalveolar lavage fluid
Inspiratory oxygen fraction
Hematoxylin-eosin
Leukocytapheresis
Lipopolysaccharide
Malondialdehyde
Neutrophil elastase
Nuclear factor-kappa B
Positive end expiratory pressure
Polymorphonuclear neutrophil
Synchronous intermittent mandated ventilation
Tumor necrosis factor alpha (formerly tumor necrosis factor alpha)
Terminal deoxynucleotidyl transferase dUTP nick end labeling.
Zhi-Gao He and Jian Huang contributed equally to this work.