Introduction

Endothelial dysfunction in sepsis is characterized by an inflammatory response, able to activate endothelial cells (EC).  ("Schouten M  - 2008" [1] (full text)

Endothelial-to-mesenchymal transition-derived cells are believed to function as fibroblasts in damaged tissue. (Potenta S - 2008) [2] (full text)

Aim of the study was to investigate the response of EC after stimulation with LPS and the role of Rapamycin in the modulation of Endothelial dysfunction.

Methods

MTT test and FACS analysis were performed on ECs (EA.hy 926) stimulated with LPS (2 µg/ml and 4 µg/ml ) and LPS (2 µg/ml and 4 µg/ml ) + Rapamycin 5nM  for 24 hours.  

Results

By MTT cell viability assay,  we observed a significantly proliferation of ECs after LPS 2μg/ml (0.720 ± 0.03 vs basal 24 h 0.469 ± 0.004, p= 0.03) and LPS 4μg/ml (0.597 ± 0.03 vs basal 24 h 0.469 ± 0.004, p= 0.04) stimulation. Moreover, the presence of Rapamycin 5 nM reported the values of cell growth to basal condition by inhibiting LPS-induced activation (Rapamycin   + LPS 2μg/ml: 0.511 ± 0.008 vs. LPS 2μg/ml, p = 0.05; Rapamycin + LPS 4μg/ml: 0.446 ± 0.01 vs LPS 4μg/ml, p = 0.02) (Figure 1).

Flow cytometry analysis (AnnV / IP) showed that only a small percentage of ECs underwent to apoptosis after LPS 2μg/ml (8.9% ±1.6  vs basal 5, 9% ± 1.2 ) and LPS 4μg/ml (7.8% ± 1.6) stimulation (Figure 2 A). In the same way the addition of Rapamycin 5nM did not induce a damage in ECs (Rapamycin 5nM: 4,4 ±1,1; Rapamycin   + LPS 2μg/ml: 5,2± 1,8; Rapamycin   + LPS 4μg/ml: 4,5± 1,1) (Figure 2 B).

In particular, after LPS 2μg/ml stimulation, ECs showed a reduction of endothelial markers CD31 (70.2% ± 1.3 vs basal: 95.1% ± 2.1) and VE-cadherin (23% ± 1.8 vs basal: 56.4% ± 1.45) and an increased expression of fibroblast markers N-cadherin (66% ± 2.15 vs basal: 9.6% ± 1.3) Vimentin (65,12% ± 1.43 vs basal: 43,28% ± 2.3) and FSP-1 (47,13% ± 1.24 vs basal: 24,33% ± 1.3) (Figure 3 E).

These data showed a phenotypic transition of ECs, called EndMT. The stimulation of EC with LPS at higher concentration (4μg/ml) caused a greater decrease of CD31 (44% ± 1.08) and VE-cadherin (19.5% ± 1.62), and a further increase of N-cadherin (78.45% ± 2.5), Vimentin (72,62% ± 1.25) and FSP-1 (57,69% ± 1.68) (Figure 3 E). Moreover, we performed a double-staining for CD31 and intracellular FSP-1 and analyzed  transitioning EC by confocal microscopy (Figure 3 B-D).

Interestingly,the addition of Rapamycin abrogated the LPS-induced EndMT by restoring the expression of CD31 (Rapamycin + LPS2μg/ml: 84.67% ± 1.7; Rapamycin + LPS4μg/ml: 81.16% ± 1.62 ), VE-cadherin (Rapamycin+ LPS2μg/ml: 63% ± 1.02; Rapamycin+ LPS4μg/ml: 64,16% ± 2,01 )  N-cadherin (Rapamycin + LPS2μg/ml: 7.3% ± 0.8; Rapamycin + LPS4μg/ml: 8,26% ± 1,2), Vimentin (Rapamycin + LPS2μg/ml: 47,61% ± 0.6; Rapamycin + LPS4μg/ml: 44,28% ± 2,3) and FSP-1 (Rapamycin + LPS2μg/ml: 11,4% ± 1.8; Rapamycin + LPS4μg/ml: 10,6% ± 1,2) to basal condition (Figure 4).  

Rapamycin inhibition of EndMT process was also showed by confocal microscopy (Figure 5).

Conclusions

Our data demonstrate that LPS acts directly on ECs by blocking apoptosis, activating cellular proliferation and promoting EndMT.

The inhibition of this pro-fibrotic pathway via Rapamycin may represent a possible therapeutic strategy to prevent the development of fibrosis during sepsis.