Genetic disruption of protein kinase Cδ reduces endotoxin-induced lung injury

Abstract

The pathogenesis of acute lung injury and acute respiratory distress syndrome is characterized by sequestration of leukocytes in lung tissue, disruption of capillary integrity, and pulmonary edema. PKCδ plays a critical role in RhoA-mediated endothelial barrier function and inflammatory responses. We used mice with genetic deletion of PKCδ (PKCδ^−/−) to assess the role of PKCδ in susceptibility to LPS-induced lung injury and pulmonary edema. Under baseline conditions or in settings of increased capillary hydrostatic pressures, no differences were noted in the filtration coefficients (k_f) or wet-to-dry weight ratios between PKCδ^+/+ and PKCδ^−/− mice. However, at 24 h after exposure to LPS, the k_f values were significantly higher in lungs isolated from PKCδ^+/+ than PKCδ^−/− mice. In addition, bronchoalveolar lavage fluid obtained from LPS-exposed PKCδ^+/+ mice displayed increased protein and cell content compared with LPS-exposed PKCδ^−/− mice, but similar changes in inflammatory cytokines were measured. Histology indicated elevated LPS-induced cellularity and inflammation within PKCδ^+/+ mouse lung parenchyma relative to PKCδ^−/− mouse lungs. Transient overexpression of catalytically inactive PKCδ cDNA in the endothelium significantly attenuated LPS-induced endothelial barrier dysfunction in vitro and increased k_f lung values in PKCδ^+/+ mice. However, transient overexpression of wild-type PKCδ cDNA in PKCδ^−/− mouse lung vasculature did not alter the protective effects of PKCδ deficiency against LPS-induced acute lung injury. We conclude that PKCδ plays a role in the pathological progression of endotoxin-induced lung injury, likely mediated through modulation of inflammatory signaling and pulmonary vascular barrier function.

acute respiratory distress syndrome (ARDS) is a major cause of morbidity and mortality in sepsis and severe pneumonia (26). Transudation of vascular fluid and protein from the pulmonary capillary lumen into the interstitium and alveolar air spaces impairs gas exchange, decreases lung compliance, and initiates a cascade of inflammatory events that leads to respiratory failure. This cascade is often characterized by sequestration of leukocytes in lung tissues, intravascular coagulation, and disruption of capillary integrity, leading to pulmonary edema (4).

The vascular endothelium is a critical target for an endotoxin such as LPS, an established inflammatory and edemagenic agent. Circulating levels of LPS are elevated in patients with severe pneumonia or sepsis and can be used as a predictor of the development of ARDS (14).

PKCδ plays an important role in the maintenance of the lung vascular barrier. We observed enhanced endothelial barrier function following PKCδ overexpression in vitro, effects that were mediated through the RhoA-GTPase pathway (21). We further noted that chemical and molecular inhibition in vivo and in vitro results in pulmonary edema formation and endothelial barrier dysfunction through reduced RhoA-GTPase-mediated disruption of stress fibers and focal contacts (15, 21), respectively. Similar to our studies, PKCδ inhibition promoted endothelial barrier dysfunction in coronary artery and brain-derived endothelial monolayers (9, 20). Conversely, PKCδ inhibition has been correlated with a barrier-protective role. Tinsley et al. (38) noted that PKCδ chemical and molecular inhibition attenuated phorbol ester-induced pulmonary endothelial monolayer hyperpermeability. A recent study showed that intratracheal administration of a cell-permeant peptide that inhibits PKCδ subcellular translocation attenuated the degree of protein and cell infiltration into the alveolar space of septic rodent lungs (19). Thus PKCδ represents a potential therapeutic target in the treatment of acute lung injury (ALI); however, whether the protection offered by inhibition of PKCδ in the development of ARDS is mediated through inflammatory or vascular cellular systems has yet to be fully elucidated.

In the current study, we assessed how PKCδ influenced susceptibility of the endothelium to LPS-induced lung injury. Our studies demonstrate that, in contrast to our previous studies (21), somatic deletion of PKCδ protected against LPS-mediated edema formation assessed by fluid accumulation within the lung, as well as protein and cell infiltration across the endothelial barrier. In addition, acute transient overexpression of dominant-negative PKCδ cDNA within the lung endothelium attenuated the effect of LPS-induced lung injury in PKCδ^+/+ mice and endothelial barrier dysfunction in vitro. Surprisingly, transient overexpression of wild-type PKCδ cDNA in the pulmonary vasculature of mice with genetic deletion of PKCδ (PKCδ^−/−) did not reverse the protective effect of PKCδ deletion in these animals. Thus the data suggest that PKCδ inhibition effectively attenuates endotoxin-induced lung injury and, thus, plays a facilitatory role in endothelial barrier dysfunction associated with ALI.

MATERIALS AND METHODS

PKCδ knockout mice.

Two breeding pairs of PKCδ^+/− mice bred into the C57BL/6 background, originally derived by the laboratory of Dr. Keiichi I. Nakayama (32), were a kind gift from Dr. Brooke Mossman (University of Vermont, Burlington, VT) (37). The mice were subsequently maintained and bred into the C57BL/6 background at the Providence Veterans Affairs Medical Center animal facility. The genotype of the mice was confirmed by PCR using the following primers (Integrated DNA Technologies): common primer 1 (5′GGAAGAATAAGAAACTGCATCACC3′), amplification of wild-type product (240 bp), primer 2 (5′GAAGGAGCCAGAACCGAAAG3′), and amplification of disrupted PKCδ product (150 bp), primer 3 (5′TGGGGTGGGATTAGATAAATG3′), as described by Dr. Keiichi I. Nakayama (RBRC00457, Riken BioResource Center). In parallel, ear tissue from PKCδ^+/+ and PKCδ^−/− mice was examined by Western blot analysis using antibodies directed against the COOH terminus of PKCδ to confirm absence or presence of PKCδ protein expression.

Cell lines and reagents.

Rat lung microvascular endothelial cells (LMVEC; Vec Technologies, Rensselaer, NY) were maintained in MCDB-131 (Vec Technologies) and used between passages 5 and 11. LMVEC used throughout the study maintained the traditional endothelial cell characteristics of von Willebrand factor and vascular endothelial (VE)-cadherin expression, uptake of acetylated LDL, and positive staining for the lectin Griffonia simplicifolia, as we previously described (10, 24).

LPS (endotoxin) from Escherichia coli serotype 011:B4 was obtained from Enzo Life Sciences (Plymouth Meeting, PA), and naphthol AS-D chloroacetate esterase granulocyte stain was purchased from Sigma-Aldrich (St. Louis, MO). Mouse monocyte chemoattractant protein (MCP-1), IL-6, and TNF-α levels were measured using the cytometric bead array technique (BD Cytometric Bead Array Mouse Inflammation Kit, BD Biosciences) according to the manufacturer's instruction and analyzed on a bioanalyzer (FACSArray, BD Biosciences).

The vectors encoding wild-type PKCδ (PKCδ^wt) and catalytically inactive PKCδ (PKCδ^K376R) were purchased from Addgene (Cambridge, MA), phosphorylated green fluorescent protein (pGFP-C1) was obtained from Clontech (Mountain View, CA), and antibodies directed against PKCδ, VE-cadherin, and β-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA).

Liposome preparation.

For production of liposomes, dimethyldioctadecyl-ammonium bromide and cholesterol were mixed in chloroform, as described elsewhere (22, 43), and the mixture was rotated under constant vacuum pressure and temperature in a rotary evaporator. Liposomes dissolved in 5% glucose solution were combined with cDNA (50 μg) and injected into the retrobulbar sinus of the orbit of 8- to 10-wk-old anesthetized PKCδ^+/+ and PKCδ^−/− mice.

In vivo model of LPS-induced ALI.

LPS was administered to adult PKCδ^+/+ and PKCδ^−/− mice via a single intraperitoneal injection (5 mg/kg) or intratracheally (2.5 mg/kg). At ∼24 h, mice were given a lethal dose of anesthesia (pentobarbital sodium, 120 mg/kg ip). Bronchoalveolar lavage (BAL) was performed using 1 ml of sterile normal (0.9%) saline. BAL cell counts were performed using a standard hemocytometer, and the BAL protein concentration was determined using a Folin-Lowry assay. Additionally, the right lung was frozen and homogenized; the left lung was fixed in formalin, processed for hematoxylin staining and for the presence of the granulocyte marker chloroacetate esterase using the substrate naphthol AS-D chloroacetate, and viewed using standard light microscopy.

In additional experiments, at 24 h following intraperitoneal injection of LPS or vehicle into PKCδ^+/+ and PKCδ^−/− mice, lungs were isolated and perfused as previously described (10, 11). Filtration coefficient (k_f) was determined using the rate of weight gain during the final 2 min following an ∼8-cmH₂O increase in venous pressure (high hydrostatic challenge) for 15 min divided by the change in capillary pressures induced by the double-occlusion technique. The k_f value was normalized to 100 g wet lung mass, empirically derived as 0.00472 body mass (40).

All animal experimental protocols were approved by the Institutional Animal Care and Use Committees of the Providence Veterans Affairs Medical Center and Brown University and comply with the Health Research Extension Act and US Public Health Service policy.

Endothelial cell isolation from mouse lungs.

Lung homogenates were incubated with platelet-endothelial cell adhesion molecule (PECAM, CD31)-conjugated magnetic beads, and immunocomplexes were separated from homogenate with a magnetic column (25) and seeded on culture dishes in medium for 24 h. Cells were collected, and lysates were used for immunoblot analysis.

Endothelial monolayer permeability.

Changes in endothelial monolayer permeability were assessed using the electrical cell impedance sensor technique (Applied Biophysics, Troy, NY), as previously described (11, 13, 15, 21). For analysis of monolayer permeability following transfection of LMVEC with cDNA encoding PKCδ^K376R, Polyjet reagent (SignaGen, Gaithersburg, MD) was used to seed cells to confluence onto collagen-coated electric cell-substrate impedance-sensing arrays. Monolayers were then treated with LPS (1 μg/ml), and resistance was measured over time. cDNA overexpression was confirmed at 48 h posttransfection via immunoblot analysis.

Statistical analysis.

For three or more groups, differences among the means were tested for significance in all experiments by ANOVA with Fisher's least significance difference test. Significance was reached when P < 0.05. Values are means ± SE.

RESULTS

Genetic deletion of PKCδ is protective against LPS-induced lung edema.

We previously demonstrated that inhibition of PKCδ caused monolayer barrier dysfunction in cultured pulmonary endothelial cells in vitro (15) and promoted lung edema in vivo (21). We now examined if genetic deletion of PKCδ altered the response of lung vasculature at baseline or in response to pathophysiological challenges. Values of k_f in isolated, perfused lungs not significantly different between PKCδ^+/+ and PKCδ^−/− mice (Fig. 1A). Next, we sought to determine if an increase in the intravascular volume, as noted in cardiogenic pulmonary edema, differentially caused pulmonary edema in PKCδ^+/+ and PKCδ^−/− mice. Hydrostatic challenge promoted a similar degree of edema formation, as measured by the wet-to-dry lung weight ratio, in PKCδ^+/+ and PKCδ^−/− mice (Fig. 1B). While k_f was higher in lungs harvested from LPS-exposed PKCδ^+/+ and PKCδ^−/− mice than in saline-injected animals (Fig. 1C), indicative of edema formation, k_f values were significantly lower in LPS-exposed PKCδ^−/− than LPS-exposed PKCδ^+/+ mice.

Fig. 1.

PKCδ-deficient (PKCδ^−/−) mice demonstrate attenuated hydrostatic lung edema formation following LPS exposure. A: lung vascular permeability was assessed by measuring capillary filtration coefficient (k_f) in ex vivo lungs isolated from PKCδ^+/+ and PKCδ^−/− mice. B: after jugular vein catheterization, sedated PKCδ^+/+ and PKCδ^−/− mice were subjected to a hydrostatic challenge, i.e., infusion of NaCl at 40 μl/g mouse over a 2-min period. Lungs were harvested after 30 min, and lung edema was determined as change in wet-to-dry lung weight ratio. C: lungs of PKCδ^+/+ and PKCδ^−/− mice that had been injected intraperitoneally with saline or LPS (5 mg/kg) were collected at 24 h postinjection, and k_f was determined. Values are means ± SE; n = 8–12 (A), 6–14 (B), and 3–8 (C). *P < 0.001 vs. vehicle. #P < 0.001 vs. LPS-treated PKCδ^+/+ mice.

Parallel experiments measured the protein content and number of cells infiltrated in the BAL fluid of the lungs of PKCδ^+/+ and PKCδ^−/− mice following 4 and 24 h of exposure to LPS. After 4 h of LPS treatment, PKCδ^+/+ and PKCδ^−/− mice displayed a similar increase in protein content within the BAL fluid compared with saline-injected animals (Fig. 2A, panel i). Interestingly, after 24 h of exposure to LPS, PKCδ^−/− mice exhibited an attenuated level of protein within the BAL fluid, relative to PKCδ^+/+ mice (Fig. 2A, panel ii). At 4 h after exposure to LPS, no significant increase in the number of cells infiltrated into the lungs of PKCδ^+/+ or PKCδ^−/− mice was noted (Fig. 2B, panel i). However, at 24 h following exposure to LPS, the number of cells within the BAL fluid was significantly greater in PKCδ^+/+ mice. Interestingly, the number of cells that infiltrated into the lungs of PKCδ^−/− mice exposed to LPS for 24 h was not different from that in BAL fluid isolated from the saline-exposed PKCδ^+/+ and PKCδ^−/− mice (Fig. 2B, panel ii). We further show reduced thickening of alveolar septae, vascular congestion, and cellularity in lungs from PKCδ^−/− mice exposed to LPS for 24 h compared with lungs from PKCδ^+/+ mice (Fig. 2C).

Fig. 2.

Diminished cell infiltration, protein accumulation, and cellularity in PKCδ^−/− mouse lungs upon LPS instillation. A and B: bronchoalveolar lavage (BAL) fluid was collected from PKCδ^+/+ and PKCδ^−/− mice treated with LPS (2.5 mg/kg) intratracheally for 4 h (i) or 24 h (ii) and analyzed for protein concentration and cell count. Values are means ± SE; n = 4–17. *P < 0.01 vs. vehicle. C: representative images of lungs from PKCδ^+/+ and PKCδ^−/− mice treated with LPS (5 mg/kg ip) for 24 h and inflation-fixed and immunohistologically stained with Leder stain and hematoxylin. Scale bars, 50 μm.

We next examined if the LPS-mediated response of proinflammatory cytokines in BAL fluid collected from PKCδ^+/+ mice was different from that in BAL fluid collected from PKCδ^−/− mice. After 24 h of LPS exposure, we observed no real differences in IL-12, TNF-α, IFN-γ, IL-10, and IL-6 levels in BAL fluid from PKCδ^+/+ and PKCδ^−/− mice (Table 1). Additionally, the low basal level of MCP-1 was elevated to the same extent in PKCδ^+/+ and PKCδ^−/− mice after 24 h of LPS exposure.

Table 1.

Induction of pulmonary cytokines by LPS is unaffected by PKCδ deficiency

	PKCδ+/+		PKCδ−/−
	Vehicle	LPS	Vehicle	LPS
IL-12	ND	1.3 ± 1.0	ND	0.28 ± 0.28
TNF-α	ND	210.2 ± 82.3	ND	1,449 ± 1,184
IFN-γ	ND	3.08 ± 2.08	ND	1.18 ± 0.68
MCP-1	25.6 ± 3.6	921.7 ± 286.7*	50.6 ± 15.7	1,277 ± 430.8†
IL-10	ND	3.24 ± 1.4	ND	1.55 ± 1.55
IL-6	ND	270.7 ± 130.4	ND	212.7 ± 40.6

Values (means ± SE) are pg/ml; n = 2–8. Cytokines [IL-12, TNF-α, IFN-γ, monocyte chemoattractant protein (MCP-1), IL-10, and IL-6] were measured in bronchoalveolar lavage fluid collected from PKCδ^+/+ and PKCδ^−/− mice treated with LPS (2.5 mg/kg it) for 24 h.

ND, not detectable.

^*P < 0.05,

^†P < 0.01 vs. vehicle.

Taken together, these data show that the genetic deletion of PKCδ confers strong protection against LPS-induced ALI via attenuated capillary permeability and edema formation.

Acute inhibition of PKCδ in the pulmonary endothelium attenuates LPS-induced increases in monolayer permeability and edema formation.

Because of the similar inflammatory responses via cytokine production in PKCδ^+/+ and PKCδ^−/− mice exposed to LPS, we next sought to determine if the attenuation of LPS-induced lung edema formation in PKCδ^−/− mice was mediated at the pulmonary endothelial barrier. Thus we transiently overexpressed catalytically inactive PKCδ (PKCδ^K376R) cDNA in lung endothelium in vivo and in vitro.

Equivalent numbers of LMVEC were transiently transfected with eukaryotic vectors encoding PKCδ^K376R or GFP. Overexpression of PKCδ in LMVEC transfected with PKCδ^K376R cDNA was confirmed by immunoblot analysis of PKCδ protein (Fig. 3A, inset), demonstrating an ∼76% increase in PKCδ protein in these cells relative to GFP-overexpressing LMVEC. As previously described using adenoviral particles expressing PKCδ dominant-negative protein (15), we observed a significant reduction in endothelial monolayer baseline resistance with PKCδ^K376R overexpression relative to GFP-overexpressing endothelial cells (Fig. 3A). Interestingly, the LPS-induced barrier dysfunction noted in GFP-overexpressing LMVEC was completely attenuated in PKCδ^K376R-overexpressing LMVEC (Fig. 3, B and C).

Fig. 3.

Disruption of PKCδ attenuates LPS-induced increases in pulmonary vascular damage. A: equal numbers of lung microvascular endothelial cells (LMVEC) were transfected with eukaryotic vectors encoding catalytically inactive PKCδ (PKCδ^K376R) or green fluorescent protein (GFP) cDNA. Resistance across monolayers was measured at 48 h posttransfection. PKCδ protein overexpression was confirmed by immunoblot (IB) analysis of lysates of transiently transfected LMVEC (inset), with relative levels of PKCδ determined via densitometry and expressed as means ± SE. ru, Resistance units. B: normalized resistance of LMVEC overexpressing PKCδ^K376R or GFP cDNA in the presence and absence of LPS (1 μg/ml). Arrow indicates point of addition. A representative trace is shown. C: percent drop in transendothelial resistance (TER) of LMVEC overexpressing PKCδ^K376R or GFP cDNA relative to addition of LPS. Values are means ± SE; n = 8–11. *P < 0.05 vs. GFP vehicle.

To correlate the in vitro observations with pulmonary vascular function in vivo, we next examined the effect of acute modulation of PKCδ within the intact lung on pulmonary edema formation by transiently overexpressing PKCδ^K376R and PKCδ^wt cDNA in the vasculature of PKCδ^+/+ and PKCδ^−/− mice, respectively, with GFP cDNA used as control. Mice were injected with cationic liposomes encapsulating cDNA, and 24 h postinjection, LPS was administered. A total of 48 h after liposome injection, k_f of the ex vivo isolated perfused lung was determined. Transient overexpression of cDNA within the lung endothelium of these mice was confirmed by precipitation of endothelial cells from lung homogenates using PECAM-1 antibody-coated magnetic beads and immunoblotting of endothelial cell lysates for PKCδ overexpression (Fig. 4A). For confirmation that the cells precipitated were endothelium, the blots were stripped and reprobed for VE-cadherin (Fig. 4A). Similar to untransfected PKCδ^+/+ mice exposed to LPS (Fig. 1C), PKCδ^+/+ mice transiently overexpressing GFP exhibited a significant increase in k_f following LPS exposure relative to those exposed to saline vehicle (Fig. 4B). Furthermore, Fig. 4B shows that acute inhibition of PKCδ, via transient overexpression of PKCδ^K376R into the pulmonary vasculature of PKCδ^+/+ mice, led to complete attenuation of the LPS-induced increase in k_f (Fig. 4C). These results were similar to k_f values in PKCδ^−/− mice (Fig. 1C). Finally, the reverse experiments were performed to see if transient overexpression of PKCδ^wt cDNA would affect the phenotype generated by somatic deletion of PKCδ in the PKCδ^−/− mice. Despite overexpression of PKCδ^wt cDNA in the lung endothelium of PKCδ^−/− mice, LPS-induced lung edema formation remained attenuated, similar to LPS-exposed PKCδ^−/− mice transiently overexpressing GFP (Fig. 4C).

Fig. 4.

Acute expression of PKCδ^K376R confers protection against LPS-induced acute lung injury (ALI) in PKCδ^+/+ mice, but acute expression of PKCδ^wt does not influence onset of ALI in LPS-treated PKCδ^−/− mice. PKCδ^+/+ (B) and PKCδ^−/− (A and C) mice were injected with liposomes encapsulating plasmid cDNA encoding PKCδ^K376R, PKCδ^wt, or GFP. At 24 h after liposome injection, LPS was administered (5 mg/kg ip). After an additional 24 h, k_f was assessed. A: to prove overexpression of cDNA in lung endothelium, lungs were harvested from PKCδ^−/− mice that had been injected with liposomes encapsulating GFP or PKCδ^wt cDNA for 24 h, and endothelial cells were isolated from lung homogenate using platelet-endothelial cell adhesion molecule (PECAM-1) antibodies conjugated to magnetic beads and plated for an additional 24 h. Lysates of harvested lung endothelial cells were immunoblotted for PKCδ. Membranes were stripped and reprobed for vascular endothelial (VE)-cadherin expression to confirm that harvested cells were endothelium-derived. Values are means ± SE; n = 3–8. *P < 0.05 vs. GFP vehicle.

Taken together, these data show that chronic deletion or acute transient inhibition of PKCδ leads to protection against lung injury in response to LPS exposure. However, acute transient overexpression of PKCδ^wt cDNA in PKCδ^−/− mice was insufficient to reverse the protective effects of PKCδ deletion on lung edema formation in response to LPS. These results suggest that targeted PKCδ inhibition may provide protection against ARDS in settings of endotoxemia and that PKCδ inhibition acts by affecting the systemic immune responses and the pulmonary endothelial barrier.

DISCUSSION

In this study, we demonstrated that somatic deletion of PKCδ attenuated ALI in mice exposed to LPS; however, there appeared to be no effect on protein levels of select cytokines released into the lungs in response to LPS. In addition, acute transient overexpression of dominant-negative PKCδ cDNA within the lung endothelium attenuated the effect of LPS-induced ALI in PKCδ^+/+ mice and endothelial barrier dysfunction in vitro. Surprisingly, transient overexpression of PKCδ^wt cDNA in the pulmonary vasculature of PKCδ^−/− mice did not reverse the protective effect of PKCδ deletion in these animals, suggesting a coordinated response of the immune and vascular systems. Thus our data support a role for PKCδ in regulating the response of the endothelium to LPS-induced lung injury and suggest that PKCδ inhibition effectively attenuates endotoxin-induced injury. Our previous studies suggested that PKCδ protein expression plays a protective role in pulmonary endothelial barrier function (21); however, the findings in the present study imply that PKCδ is a facilitator of endothelial barrier dysfunction. Discrepancy in the role of PKCδ at the endothelial monolayer may be due to the mechanisms by which PKCδ modulates endotoxin-induced ALI. There is evidence that PKCδ has an important role as a regulator of the inflammatory response. For example, PKCδ^−/− mice were originally shown by two independent groups to have altered B cell production and increased levels of autoimmune antibodies (29, 32). Furthermore, neutrophils isolated from PKCδ^−/− mice are dysfunctional in their adhesive and migratory properties, as well as the respiratory burst and degranulation responses in vitro (7). In addition, PKCδ has been shown to regulate monocyte and neutrophil functions, such as reactive oxygen species release and phagocytosis, through direct phosphorylation and activation of NADPH oxidase subunits, p47^phox and p67^phox (3, 41, 42). While we did not detect significant differences in the release of select inflammatory cytokines into the lungs of PKCδ^−/− mice relative to wild-type mice, we did note a significant decrease in the number of cells infiltrating into the lung parenchyma, evidence suggestive of a muted inflammatory response. The attenuated immune response may explain the LPS-induced reduction in lung neutrophil accumulation in PKCδ^−/− mice and the subsequent decrease in ALI onset, as assessed by k_f and BAL protein content.

Concomitantly, we observed that selective inhibition of PKCδ within the pulmonary vasculature, via transient overexpression of PKCδ^K376R cDNA, attenuated the LPS-induced lung edema formation in PKCδ^+/+ mice and endothelial barrier dysfunction in vitro. While these findings suggest an important role for PKCδ in the lung endothelium, this effect may be due to the disruption of other functions of PKCδ. The kinase has been described to be critical in cell proliferation, cell cycle progression, adhesion, barrier function, and apoptosis. As such, PKCδ^−/− mice have been used to examine various pathological conditions.

We have demonstrated a significant protection against lung injury in PKCδ^−/− mice upon LPS-induced endotoxemia. Furthermore, we have demonstrated that selective inhibition of PKCδ within the pulmonary vasculature via acute transient overexpression of dominant-negative PKCδ cDNA attenuated the LPS-induced lung edema formation in PKCδ^+/+ mice and endothelial barrier dysfunction in vitro, results indicative of an important role of PKCδ in the lung endothelium. In keeping with our data, it was previously shown that intravascular administration of a cell-permeant PKCδ inhibitor, PKCδ-TAT, attenuated edema formation in the brains of hypertensive rats via maintenance of the blood-brain barrier (33). While it was shown previously that intratracheal instillation of PKCδ-TAT peptide blunted protein levels in BAL fluid, as well as cellular infiltration into the lungs of septic rats, the route of the peptide administration suggests that the PKCδ inhibitor was most likely affecting the alveolar epithelium and infiltrating inflammatory cells primarily (19). Thus the design and results of the current study support a critical role for PKCδ in the pulmonary endothelium and inflammatory response to LPS-induced lung edema formation. Our laboratory previously showed that acute inhibition of PKCδ caused endothelial barrier dysfunction in vitro and lung edema in vivo (15, 21). We further showed that basal endothelial monolayer formation required PKCδ-mediated maintenance of focal adhesion and stress fiber formation, as well as RhoA activity (15). In the current study, we have demonstrated endothelial barrier disruption upon PKCδ inhibition via transient overexpression of PKCδ^K376R cDNA. However, in the absence of LPS, these barrier-disruptive effects in endothelial cells were not observed in vivo, either upon transient overexpression of PKCδ^K376R or somatic deletion of PKCδ. This may be a result of PKC compensation within the intact lung or may highlight the sensitivity of measuring endothelial permeability with cell impedance in vitro as opposed to k_f measurements in vivo. Despite this discrepancy, both experimental systems showed that PKCδ^K376R overexpression conferred protection against LPS-induced endothelial barrier dysfunction in vitro and edema formation in vivo, consistent with observations in vivo in PKCδ^−/− mice.

LPS is known to activate the Toll-like receptor and cause apoptosis in the lungs. Thus it is possible that acute inhibition or ablation of PKCδ protein is protective against LPS-induced lung injury by the suppression of endothelial cell apoptosis. Indeed, we previously showed less endothelial cell apoptosis in lungs of PKCδ^−/− mice caged under hyperoxic conditions than in lungs of hyperoxic PKCδ^+/+ mice (16).

LPS activates the NF-κB pathway in endothelial cells (30, 34), resulting in cytokine release, enhanced surface expression of adhesion molecules (ICAM-1 and VCAM), and caveolin-mediated cytoskeletal rearrangement (31, 39). In most cells, NF-κB remains in an inactive complex with IκB proteins within the cytosol; upon pathway activation, the IκB protein is phosphorylated and targeted for degradation by the IκB kinase complex, freeing NF-κB to translocate into the nucleus and control gene target expression (36). PKCδ has been shown in endothelial cells to activate ICAM-1 gene transcription following thrombin treatment by activating and enhancing NF-κB binding to the ICAM-1 promoter (34). Thus it could be expected that acute introduction of PKCδ^wt cDNA into PKCδ^−/− mice would induce edema formation upon LPS stimulation via NF-κB activation; however, this was not observed. It is possible that the duration of transient overexpression of PKCδ^wt cDNA in PKCδ^−/− mice was not sufficient to promote adhesion molecule surface expression and/or cytoskeletal signaling and reversion to the edemagenic response to LPS.

Many cytokines released during gram-negative sepsis play an important role in vascular injury associated with endotoxic shock. Kilpatrick et al. (19) noted that cecal ligation-and-puncture-induced polymicrobial sepsis caused an elevation in plasma and BAL fluid levels of cytokine-induced neutrophil chemoattractant (CINC-1) and macrophage inflammatory protein (MIP-2) that was attenuated upon introduction of the PKCδ TAT inhibitory peptide (19). In the present study, we observed no significant alteration, with genetic ablation of PKCδ, in the levels of the inflammatory cytokines IL-6, IL-10, IL-12, MCP-1, and TNF-α within the BAL fluid. It is possible that, unlike acute inhibition of the kinase, upon somatic deletion of PKCδ, alternate pathways, or PKC isoforms, compensate for the absence of PKCδ to ensure a functional immune system.

Previous studies indicate that LPS-induced signaling within the endothelium, rather than systemic effects of the endotoxin, are responsible for lung injury. Leukocyte adhesion to the endothelium has been observed following LPS administration directly into the pulmonary vasculature, mediated through endothelial cell ICAM-1 (17, 18). Upon leukocyte adhesion, endothelial cells are activated with increased intracellular calcium levels, myosin light chain kinase phosphorylation, and stress fiber formation through a p38-mediated pathway (2, 8). All three of these outcomes may directly or indirectly be mediated by PKCδ and result in dysfunction of the endothelial barrier function and pulmonary edema formation. Vascular congestion within the lung is a hallmark of ALI; accordingly, we also noted reduced cellularity within the PKCδ^−/− mouse lung, correlating with attenuated formation of pulmonary edema.

In summary, the present study supports the existing findings regarding the protective influence conferred by the inhibition of PKCδ on endothelial function (5, 6) and inflammatory response (7, 19) in settings of endotoxin challenge. We show for the first time that chronic deletion and acute inactivation of PKCδ attenuate LPS-mediated pulmonary endothelial barrier dysfunction and edema formation. Furthermore, we observe maintenance of PKCδ^−/− protection with the rescue of the PKCδ^+/+ phenotype. Thus we hypothesize that PKCδ plays a facilitatory role in endotoxin-induced barrier dysfunction and ALI, likely mediated via its role in activating proinflammatory cellular signaling pathways, in addition to its preservation of RhoA activity within the endothelium. Further elucidation of these protective mechanisms holds potential for the development of therapies targeting the PKCδ signaling pathway to restore pulmonary vascular function following endotoxin-induced edema formation.

GRANTS

This material is the result of work supported with resources and the use of facilities at the Providence Veterans Affairs Medical Center and supported by National Heart, Lung, and Blood Institute Grant R01 HL-67795 and American Heart Association Grant 10GRNT4160055 to E. O. Harrington. K. L. Grinnell was supported by National Heart, Lung, and Blood Institute Grant T32 HL-094300.

DISCLAIMER

The contents of this manuscript do not represent the views of the Department of Veterans Affairs or the US Government.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

H.C., K.L.G., B.C., C.-S.C., and J.B. performed the experiments; H.C., K.L.G., B.C., and C.-S.C. analyzed the data; H.C., K.L.G., B.C., C.-S.C., J.L.-N., A.A., and E.O.H. interpreted the results of the experiments; H.C. and E.O.H. prepared the figures; H.C. and E.O.H. drafted the manuscript; H.C., K.L.G., C.-S.C., A.A., S.R., J.R.K., and E.O.H. edited and revised the manuscript; H.C., K.L.G., B.C., C.-S.C., J.B., J.L.-N., A.A., S.R., J.R.K., and E.O.H. approved the final version of the manuscript; K.L.G., B.C., S.R., J.R.K., and E.O.H. are responsible for conception and design of the research.