Pulmonary endothelial cell activation during experimental acute kidney injury

Abstract

Acute kidney injury (AKI) leads to increased lung microvascular permeability, leukocyte infiltration, and upregulation of soluble inflammatory proteins in rodents. Most work investigating connections between AKI and pulmonary dysfunction, however, has focused on characterizing whole lung tissue changes associated with AKI. Studies at the cellular level are essential to understanding the molecular basis of lung changes during AKI. Given that the pulmonary microvascular barrier is functionally abnormal during AKI, we hypothesized that AKI induces a specific proinflammatory and proapoptotic lung endothelial cell (EC) response. Four and 24 h after kidney ischemia/reperfusion injury or bilateral nephrectomy, murine pulmonary ECs were isolated via tissue digestion followed by magnetic bead sorting. Purified lung ECs were analyzed for changes in mRNA expression using real-time SuperArray polymerase chain reaction analysis of genes related to EC function. In parallel experiments, confluent rat pulmonary microvascular ECs were treated with AKI or control serum to evaluate functional cellular alterations. Immunocytochemistry and FACS analysis of Annexin V/propidium iodide staining were used to evaluate cytoskeletal changes and promotion of apoptosis. Isolated murine pulmonary ECs exhibited significant changes in the expression of gene products related to inflammation, vascular reactivity, and programmed cell death. Further experiments using an in vitro rat pulmonary microvascular EC system revealed that AKI serum induced functional cellular changes related to apoptosis, including structural actin alterations and phosphatidylserine translocation. Analysis and segregation of both upregulated and downregulated genes into functional roles suggest that these transcriptional events likely participate in the transition to an activated proinflammatory and proapoptotic EC phenotype during AKI. Further mechanistic analysis of EC-specific events in the lung during AKI might reveal potential novel therapeutic targets for the deleterious kidney-lung crosstalk in the critically ill patient.

Figure 1. Experimental AKI Leads to Renal Injury

Prior to endothelial cell isolation, mice underwent 60 minutes of bilateral IRI, bilateral nephrectomy or sham surgery, and were then sacrificed at either 4 or 24 hours post-procedure. Normal mice (without any surgical intervention) were also sacrificed at the final time point as internal controls. Serum creatinines were measured prior to surgery (0h) and at time of sacrifice (4h or 24h). Both bilateral IRI and nephrectomy result in significant renal injury (*) while sham surgery does not increase creatinine levels above normal baseline. There was no significant difference between the rise in serum creatinine induced by IRI sersus bilateral nephrectomy. Results shown are average serum creatinine values for a representative experiment with error bar signifying standard deviation (n=5 mice per group).

Figure 2. Magnetic Isolation of Pulmonary Endothelial Cells

Whole lung tissue from mice (n = 5 mice per surgical group) was pooled and homogenized into a single cell suspension and pulmonary endothelial cells isolated using a microbead sorting system. (A) Total cells prior to sorting, without fluorescent antibody label. (B) Cells stained with CD45-FITC and CD31-PE (C) Pure population of CD31(+), CD45(−) cells after a negative sort for CD 45 and a second, positive selection for CD31.

Figure 3. Renal Injury Leads to Alterations in Pulmonary Vascular Endothelial Gene Expression

Vascular endothelial cell gene expression was assessed via RT-PCR array 4 and 24 hours after injury. Fold changes in gene expression were clustered by average linkage analysis using the Cluster and Treeview software and illustrated in heatmap format. Red indicates increased gene expression while green indicates decreased gene expression. All fold changes are relative to appropriate sham surgeries. Clustering reveals similar gene expression profiles for several functional groupings, including apoptosis and stress response, vascular reactivity as well as angiogenesis.

Figure 4. AKI induces Actin Reorganization

Rat Lung Microvascular Endothelial Cells (RLMVECs) were treated for 24 hours with 10% serum from animals that underwent either sham surgery or bilateral IRI. Cells were stained with phalloiden and DAPI to evaluate actin organization and nuclear structure. In cells exposed to IRI serum, actin stress fiber formation and nuclear disorganization are apparent compared with cells grown in sham serum. All pictures are at 20X magnification

Figure 5. AKI Induces Apoptosis in Vascular Endothelial Cells

Rat Lung Microvascular Endothelial Cells (RLMVECs) were cultured in the presence of serum from animals that underwent either bilateral IRI or sham surgery. Serum was isolated 24 hours after injury; cells were then cultured in either 5%, 10% or 25% serum for 24 hours. After serum exposure, cells were stained with a FITC-conjugated antibody against Annexin V as well as PI and then evaluated by FACS. Cells exposed to IRI serum showed increased evidence of apoptosis in a dose-dependent manner (A) compared to cells exposed to sham serum. There was a statistically significant difference in apoptosis between IRI and Sham in the 10% and 25% serum groups. Notably there was no significant difference in apoptosis amongst the sham groups. (B) An RLMVEC stained with Annexin V and Hoechst for nuclear visualization reveals membrane localization of Annexin V as well as nuclear changes consistent with apoptosis.

Figure 6. AKI Induces ICAM-1 protein expression in Vascular Endothelial Cells

Representative Western blot of homogenates obtained from Rat Lung Microvascular Endothelial Cells (RLMVECs) treated for 24 hours with 10% serum from animals that underwent either sham surgery or bilateral IRI. ICAM-1 protein expression was increased in RLMVECs treated with serum from IRI-treated mice when compared to those treated with sham serum.