Rho is a small GTPase and reported to be the molecular switch for intracellular signalling (10). Numerous effector molecules of rho have been identified, among which two serine/threonine kinases, rock-I and rock-II, are frequently reported (11). Rho kinases are composed of NH2-terminal catalytic, coiled coil, rho binding, and COOH-terminal pleckstrin-homology domains (12). These kinases phosphorylate various substances, including myosin light chain phosphatase and mediate the formation of actin stress fibers and focal adhesions in various cell types. These molecules are involved in many aspects of cell motility which include smooth muscle cell contraction and cell migration (13,14). Pharmacological manipulation of this pathway can be achieved with agents such as clostridium botulinum toxin C3, which inhibits small GTPase rho, or fasudil or Y-27632 that specifically inhibit its effector rho kinase (13). Reorganisation of the endothelial cytoskeleton, which include actin filaments, microtubules and intermediated filaments, leads to alteration in cell shape and provides a structural basis for increase of vascular permeability. This process has been implicated in the pathogenesis of pulmonary endothelial barrier dysfunction in acute lung injury. Tasaka et al. (14) has suggested an important role of rho GTPase-mediated signalling in the endotoxin-induced acute lung injury. However, an in vivo model and investigation of detailed mechanism of action are required to define the role of rho/rho kinase in sepsis-induced lung injury.

It has been shown that the rho/rho kinase pathway is involved in the mechanisms of apoptosis (15,16). Shiotani et al. (17) have demonstrated that rho kinase-mediated production of reactive oxygen species (ROS) and inflammatory cytokines are substantially involved in the pathogenesis of ischemia reperfusion injury. Although it has been demonstated that rho kinase regulates the production of ROS through activation of an NADPH oxidase in neutrophils, it is unknown whether Rho kinase is involved in sepsis-induced ROS and reactive nitrogen species (RNS) production and tissue injury in vivo (18,19). It has been shown that rho effector protein ROCK 1 is cleaved during apoptosis to generate a truncated active form (20,21). Furthermore, direct cleavage of ROCK 2 by granzyme B defined ROCK 2 as the inducer of apoptotic membrane blebbing in a caspase-independent manner in several cell lines (22). In a parallel manner, we have also shown the fragmentation of ROCK 2 in human placentas from preeclamptic patients (23). However, there has been no report documenting this pathway in an animal model of sepsis.

Experimental sepsis by CLP: Anesthesia was induced by intramuscular (i.m.) administration of ketamine 50 mgkg-1, and xylazine 7 mgkg-1. After shaving the abdomen and application of a topical disinfectant, a two cm midline incision was made below the diaphragm to expose the abdominal organs. After the identification of the cecum, it was ligated below the ileocecal valve without occluding the bowel passage. The cecum was then subjected to a single “through and through” perforation with an 18-gauge needle distal to the point of ligation. The needle was removed and a small amount of stool was extruded from both punctures to ensure potency. After repositioning the bowel, the abdominal incision was closed with 4/0 sterile synthetic absorbable suture (Polyglactin 910, Vicryl, Ethicon Ltd., Edinburg) and the skin clips (Ethicon, Somerville, NJ). Sham-operated animals underwent the same procedure except for ligation and puncture of the cecum.

Experimental Protocol: After fasting overnight, 60 rats were randomly divided into four groups. The first group (sham group, n=15) served as sham-operated and the third group (CLP group, n=15), was subjected to cecal ligation and puncture (CLP). The second group (sham + Y27632 group, n=15) and the fourth group (CLP + Y-27632, n=15) were given (+)-(R)-trans-4-1-aminoethyl-N-4-pyridyl cyclohexane carboxamide dihydrochloride monohydrate (Y-27632, a selective ROCK inhibitor) 1.5 mgkg-1 intraperitoneally (i.p.) 20 min prior to sham or CLP operations. All animals received fluid resuscitation. Twenty four hours later, rats were anesthetized with i.m. ketamine 80 mgkg-1. After mid-line sternotomy, blood samples were taken with cardiac puncture and the right bronchus was clamped. Bronchoalveolar lavage of the left lung was performed with 2 ml of saline and then both lungs were harvested. Right lung was divided into four equal parts. To evaluate the CLP-induced lung injury one part was fixed in 10% formaldehyde and the other was preserved for W/D weight ratios. The final two parts were taken for biochemical assay and Western blot. Lung specimens were kept frozen at -70°C until analysis. BALF was used for measurement of infiltrating cells and protein concentrations.

The determination of thiobarbituric acid reactive substances: Tissue was homogenized in 10 parts 15 mmol/L KCL for thiobarbituric acid reactive substances (TBARS) assay. The TBARS levels, an index for lipid peroxidation, was determined by thiobarbituric acid reaction described by Yagi (24). The principle of the method is based on spectrophotometric determination of the intensity of the pink color produced by interaction of the barbituric acid with malondialdehyde liberated as a result of lipid peroxidation. We used 1,1,3,3 tetraetoxypropane as the primary standard.

The determination of 3-nitrotyrosine/total tyrosine ratio: For tyrosine assay lung tissue was homogenized in ice-cold phosphate buffered saline (pH=7.4). Equivalent amounts of each sample were hydrolyzed in 6N HCI at 100ºC for 18-24h, samples were then analyzed on a HP 1049 HPLC apparatus. The analytical column was a 5 µm pore size Spherisorb ODS-2 C18 reverse-phase column (4, 6-250 mm; Alltech, Deerfield, IL, USA). The guard column was a C18 cartridge (Alltech). The mobile phase was 50 mmoL l-1 sodium acetate/50 mmoL l-1 citrate/8% methanol, pH=3.1. HPLC analysis was performed under isocratic conditions at a flow rate of 1 ml min-1 and the UV detector was set at 274 nm. 3-nitrotyrosine (3-NT) determination was made by comparison of the sample’s peak area with the peak area produced by the external standard solution of 10 µmolL-1 3-NT (25). The results were expressed as 3-NT / total tyrosine ratio.

Lung wet-to-dry weight ratio: Lung wet-to-dry (W/D) weight ratios were used as a measure of pulmonary edema. The W/D weight ratio of the left lung was calculated by weighing the freshly harvested organ, and then heating it at 90°C in a gravity convection oven for 72 hours to attain its dry weight (26).

Number of Infiltrating Cells and Protein Concentration in BALF: The number of infiltrating cells and the protein concentration in BALF were used as indicators of the degree of lung inflammation. BALF samples were stored in ice water until testing. Cell counts and total protein concentrations were measured on the day of sample collection. For cell counting, 1µl of bronchoalveolar aspirate was placed on a glass slide, air-dried, and then stained by modified Giemsa’s method. The total number of inflammatory cells (polymorphonuclear leukocytes-[PMNs]) in each 1 µl sample was then counted under the light microscope by a pathologist unaware of the groups. The remainder of the bronchoalveolar fluid was preserved for analysis of protein content. The total protein concentration in a bronchoalveolar fluid sample was measured using the method of Lowry et al (27).

Histopathological Examination: The specimens were fixed in 10% formalin for 24h, and standard dehydration and paraffin-wax embedding procedures were used. H&E-stained slides were prepared by using standard methods. Light microscopic analyses of lung specimens were done by blinded observation to evaluate pulmonary architecture, tissue edema formation and infiltration of the inflammatory cells as previously defined (9). The results were classified into four grades where Grade 1 represented normal histopathology; Grade 2 indicated minimal neutrophil leukocyte infiltration; Grade 3 represented moderate neutrophil leukocyte infiltration, perivascular edema formation and partial destruction of pulmonary architecture and finally Grade 4 included dense neutrophil leukocyte infiltration, abcess formation and complete destruction of pulmonary architecture.

TUNEL Assay: TUNEL assay was performed on paraffin embedded lung tissues from Sham, CLP and CLP + Y-27632 groups. Sections 5 µm thick were mounted on subbed slides, deparaffinized at 57 ºC for 5 minutes on a slide warmer then immersed in 2 changes of mixed xylenes 5 minutes each. The sections were hydrated by immersing them in an alcohol series ranging from 100% to 70% ethanol. This was followed by 2 washes in PBS 5 minutes each. The staining of the sections was performed according to R&D systems Tacs TdT In Situ Apoptosis Detection kit (TA4626). Briefly sections were permeabilized with proteinase K at 37 ºC for 22 min. This was followed by quenching of endogenous peroxidases by treating the sections with a 3% hydrogen peroxide solution for 5 minutes. The labelling reaction with biotinylated dNTPs was incubated for 2 hours at 37ºC in a humidified chamber. Detection was performed by using streptavidin conjugated to HRP at double the concentration for 10 minutes at room temperature. The addition of Tacs blue for 5 minutes to each section was followed by 3 washes in water. The sections were counterstained with nuclear fast red, had coverslips mounted with Permount® (FisherScientific) and documented with a Ziess Axiovert 200M deconvolution microscope. Five fields of view for each section were counted using the 40X objective and the number of blue cells expressed as a percent of the total cell number was averaged over the 5 fields to give the percent of apoptotic cells per treatment.

Western Blot: Western blot experiments were performed as previously described (28). In brief, lung tissues were homogenized in cold buffer containing 50 mM Tris-HCl (pH 7.5), 400 mM NaCl, 2 mM EGTA, 1 mM EDTA, 1 mM DTT, 10 µM PMSF,10 µg ml-1 leupeptine, 1 µg ml-1 pepstatin and 1mM benzamidine. Nuclei and unlysed cells were removed by low speed centrifugation at 900 x g, 4ºC for 10 min. Protein concentration of supernatant was determined by Lowry method. The supernatant (200 µg of protein) was mixed with an equal volume of 2x SDS sample buffer and boiled for 5 min. Proteins were separated by SDS polyacrylamide gel electrophoresis (8 % acrylamide) and blotted onto a PVDF membrane. Membranes were blocked for 1 h with 5 % (w/v) dry nonfat milk in TBS-tween. Blots were then incubated in mouse monoclonal anti-Rock 1 and anti-Rock 2 antibody which detects both active and inactive forms of Rocks (1:500, Transduction Laboratories) actin (1:500, LabVision) and caspase 3 which only detects 19 and 17 kDa fragment of caspase 3 (1:1000, Cell Signalling) for 3 h. An HRP-conjugated secondary antibody was used in conjunction with an enhanced chemiluminescence detection kit (ECL Plus) from Amersham Pharmacia Biotech to visualize the immunopositive bands on X-ray film. Equal protein loading was verified using actin antibody and coomassie brillant blue (cbb) staining of membranes. The intensities of the bands were quantified by densitometry using Scion image computer program (Scion Corp. Beta 4.0.2). Percent fractionation of ROCK 1 and 2 was calculated by the following equation;

% Fraction of rock 1 or 2 = [ODact x 100] / [ODact + ODinact]

ODact : Optic Density of active fragment.

ODinact : Optic Density of inactive fragment.

The transendothelial migration of neutrophils is a critical step in inflammation and the role of infiltrating PMNs, ROS and/or RNS in sepsis induced organ damage is well established. Recent studies suggest that rho and rho kinase are key mediators of myosin light chain (MLC) phosphorylation and have important roles in neutrophil migration (29,30). Endothelial rho and rho kinase regulate transendothelial neutrophil migration by modulating the cytoskeletal events that mediate such migration (14,30). In our study, the increased levels of inflammatory cell counts in BALF and increased levels of lung tissue TBARS and 3-NT/total tyrosine ratio (a marker of peroxynitrite formation) indicate that leukocyte recruitment and oxidative and/or nitrosative stress are induced after CLP. Histopatologic data showing edema and leukocyte infiltration in lung tissues obtained from the CLP group support these biochemical changes. The demonstration of rho kinase-mediated leukocyte infiltration in endotoxemic liver injury and ROS production in I/R injury are also in concordance with our results (17,31). Another finding of this study is the simultaneous suppression of TBARS and 3-NT/total tyrosine ratio by Rho kinase inhibitor, Y-27632, preventing not only oxygen centered free radical damage but also peroxynitrite mediated lung injury. No evidence of Y-27632 antioxidant activity in vitro has been reported, thus the fact that Y-27632 can protect cells from lipid peroxidation in this study may result from its antiinflammatory activity rather than from its direct antioxidant activity.

Microfilaments and cytoskeletal actin are the major structures involved in maintaining cell shape. Gaps between endothelial cells open in inflammation which may lead to extravasation of fluid and macromolecules. Involvement of rho kinase in neutrophil-stimulated endothelial hyperpermeability, microvascular leakage and lung microvascular permeability has been demonstrated (28,32,33). Alveolar epithelium can also contribute to inflammation by releasing inflammatory mediators which is governed by rho signalling (34). Zeng et al (35) studied the effect of recombinant human activated protein C on endothelial cell permeability and modulation of the intracellular cytoskeleton via rho kinase pathway to reveal clinically improved organ function. Our CLP model was associated with significant capillary leak and lung edema, as evidenced by increased protein in the BALF and increased W/D ratio. Pretreatment with Y-27632 significantly decreased BALF proteins and lung edema. The decrease in edema formation and amelioration of lung damage suggests that Rho kinase inhibition prevents the activation of neutrophil-dependent oxido-inflammatory pathways and thus contributes to the reduction of fluid extravasation and improved histopathology. Supporting our results, it has been shown that Rho and Rho kinase are involved in neutrophil stimulated increase in endothelial permeability and in cytokine-mediated barrier dysfunction in the pathogenesis of pulmonary edema associated with acute lung injury (28,32,33,36,37). The importance of rho kinase pathway in ROS, specifically H2O2-induced pulmonary edema has been described in one previous study (12). Our study suggests an additional role of RNS in the rho kinase related lung edema. In addition, a recent study demonstrated that rock inhibitor, Y-27632, prevented the TNF-arelated increased permeability in a lipopolysaccharide (LPS) model (14). Rho kinase inhibitors are also defined as a potent inhibitor of TNF-a and chemokines in bronchial epithelial cells and may be an additional therapeutic option in sepsis (38). Contrary to our findings, Lundblad et al (40) has reported the permeability-reducing effects of prostacyclin and stated that inhibition of Rho kinase did not counteract endotoxin-induced increase in permeability in cat skeletal muscle. The different findings between the two studies may be attributed to the selection of different animal models and organs such as cat skeletal muscle in LPS induced-endotoxemia versus rat lung in CLP-induced sepsis.

Peroxynitrite has been identified as a noxious stimulus for lung inflammation (25,41). Peroxynitrite is known to induce DNA laddering and caspase 3 activation in different cell types including human endothelial cells in cultures (42,43). On the other hand, it has been recently shown that rock 1 cleavage, which produces the 130 kDa active form of the enzyme, requires caspase 3 activation (44). There is increasing appreciation that the microfilamentous cytoskeleton may be intrinsically involved in the cell damage by regulating intracellular signaling or by transmitting death messages to downstream effectors. This is the first in vivo study which shows ROCK 1 & 2 fragmentation response to sepsis. It can be assumed that rho kinase may be required for membrane blebbing in apoptosis induced by peroxynitrite. In this study, rho kinase appears to have a critical role in indirect acute lung injury in sepsis. Inhibition of rho kinase pathway prevented the increase in peroxynitrite levels in lung tissue which was associated with a decrease in the number of apoptotic cells in the CLP + Y-27632 group. Of note, Thorlacius et al have demonstrated the protective effect of Y-27632 on apoptosis in LPS induced liver injury (31). It is possible to postulate the following vicious cycle in sepsis: rho kinase activation is associated with increased epithelial permeability and followed by leukocyte migration and neutrophil infiltration with the consequence of production of ROS and peroxynitrite which triggers caspase cleavage and so rho kinase activation again. Lending support to this hypothesis, one recent study has demonstrated the involvement of peroxynitrite in caspase-3 mediated apoptosis at the tissue level (45). In addition 3-NT, as a marker of peroxynitrite production, is also increased in the plasma of patients with acute lung injury (46). Recently, Walford et al (47) have stated that hypoxia associated with tissue inflammation can modulate the effects of RNS on endothelial function and promotes the apoptotic cell death via peroxynitrite. Our data, from a different hypoxic/dysoxic model support these results. Our present findings, which show that Y-27632 decreases leukocyte infiltration and ROS/RNS mediated damage, may help to explain the potent, antiapoptotic effect exerted by rho kinase inhibition in non-pulmonary sepsis-induced indirect acute lung injury.