| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online September 19, 2005 as doi:10.1096/fj.05-4037fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
The Center for Vascular Biology Research and Division of Molecular and Vascular Medicine, Beth Israel Deaconess Medical Center,
Department of Pathology, Beth Israel Deaconess Medical Center, and
The Research Center for Advanced Science and Technology, the University of Tokyo, Tokyo, Japan.
1 Correspondence: Beth Israel Deaconess Medical Center, Molecular and Vascular Medicine; RW-663, 330 Brookline Ave., Boston MA 02215, USA. E-mail: waird{at}bidmc.harvard.edu
SPECIFIC AIMS
Endothelial cells are highly sensitive to changes in the extracellular milieu. Sepsis (defined as the systemic inflammatory response to infection) results in activation of inflammatory and coagulation pathways. We hypothesized that sepsis-associated mediators may alter the signal transduction capacity (termed "set point") of endothelial cells. The aim of this study was to elucidate the effect of preconditioning on thrombin-mediated gene expression in human umbilical vein endothelial cells (HUVEC)
PRINCIPAL FINDINGS
1. Net signal input is crucial in modulating subsequent thrombin signaling in endothelial cells
To determine the effect of different sepsis-like preconditioning regimes on thrombin-mediated gene expression, HUVEC were pretreated for 16 h in the presence or absence of high glucose, tumor necrosis factor 
(TNF-), and/or lipopolysaccharide (LPS), then treated with or without 1.5 U/mL thrombin for 4 h. Alternatively, HUVEC were grown under hypoxic or hyperthermic conditions prior to thrombin treatment. Cells were harvested for RNA and assayed for gene expression by RNase protection assay (Fig. 1
). In the absence of preconditioning, thrombin induced the expression of tissue factor (TF), platelet-derived growth factor (PDGF) -A, intercellular adhesion molecule (ICAM) -1, urokinase-type plasminogen activator (u-PA), cyclo-oxygenase (COX) -2, E-selectin, vascular cell adhesion molecule (VCAM) -1, but not CD44. The various preconditioning regimens had different effects on thrombin-mediated gene expression. Pretreatment with TNF- or LPS significantly reduced thrombin stimulation of TF, ICAM-1, u-PA, E-selectin, and VCAM-1 but had little or no effect on PDGF-A. The data suggest that TNF- or LPS sensitizes CD44 and COX-2 to the effect of thrombin. In contrast to TNF- or LPS, preconditioning with high glucose, hyperthermia, or hypoxia alone had little or no effect on thrombin response. In cells exposed to combination of high glucose and hypoxia, there was a trend toward superinduction of thrombin-responsive genes. In DNA microarray studies, preconditioning with TNF- or LPS displayed overlapping yet distinct effects on thrombin-mediated gene expression. Results were validated using TaqMan PCR.
|
2. Preconditioning with TNF- or LPS inhibits thrombin-p65 NF-
B signaling in human endothelial cells
In EMSAs, preconditioning with TNF- and, to a lesser extent, LPS significantly inhibited thrombin-mediated induction of p65 NF-B binding (Fig. 2
). These data were confirmed using ELISA-based assays for p65 NF-B binding activity. In transient transfection assays of HUVEC, TNF- and LPS each inhibited thrombin stimulation of the NF-B-responsive ICAM-1 promoter. In Western blot assays, pretreatment with TNF- and LPS blocked thrombin-mediated depletion of cytoplasmic IB and induction of phospho-IB. In immunofluorescent studies, preconditioning with TNF- or LPS significantly inhibited the capacity of thrombin to induce nuclear translocation of p65 NF-B.
|
3. The effect of TNF- on thrombin signaling is delayed for up to 24 h after its removal from the culture medium
After preincubation with TNF- or LPS, endothelial cells were allowed to recover in fresh medium for 24 h prior to thrombin treatment. After the 24 h "washout" period, TNF-- and LPS-primed endothelial cells demonstrated ongoing desensitization to the effects of thrombin on gene induction, p65 NF-B binding, and IB phosphorylation. These effects were still observed at 48 h but were lost when cells were propagated.
4. Preconditioning with thrombin has little effect on LPS and TNF- signaling
Preconditioning with thrombin attenuated subsequent thrombin-mediated induction of TF, ICAM-1, E-selectin, and VCAM-1. However, no such effect was seen with PDGF-A, u-PA or COX-2, arguing against simple consumption or desensitization of protease-activated receptor (PAR) -1. In contrast to the effects on subsequent thrombin response, preconditioning with thrombin had no effect on TNF-- or LPS-mediated induction of TF, ICAM-1, E-selectin, or VCAM-1. These results suggest that a change in set point and its effect on subsequent signaling depends on the nature of the preconditioning agent, the agonist being tested, and the identity of the target gene.
CONCLUSIONS AND SIGNIFICANCE
We have shown that preconditioning of endothelial cells with endogenous signals commonly associated with sepsis and systemic inflammation results in altered thrombin signaling in endothelial cells.
Although our findings invoke a role for NF-B, they do not exclude a role for other signal intermediates and/or transcription factors in mediating the effects of preconditioning. For example, preconditioning with TNF- also inhibited thrombin induction of NF-AT and Down syndrome critical region (DSCR-1). Such an effect would be predicted to contribute to the attenuation of thrombin-induced expression of E-selectin, VCAM-1, ICAM-1, and TF. Still other mechanisms may be involved in mediating the positive effects of TNF- on thrombin-induced COX-2 and CD44, as well as the positive effects of high glucose combined with hypoxia on several thrombin-responsive target genes.
In recent years, other studies have demonstrated the potential for preconditioning regimens to sensitize or desensitize endothelial cells to other agonists. For example, the term ischemic preconditioning describes the cytoprotective effect of ischemia-reperfusion or hypoxia on tissues and/or endothelial cells. "Transplantation accommodation" occurs when xenoreactive or alloreactive antibodies induce resistance of the endothelium to humoral rejection. Repeated challenge of endothelial cells with LPS results in temporary insensitivity to subsequent LPS challenge, a process termed "endotoxin tolerance." Other preconditioning regimens that have been demonstrated to alter endothelial cell signaling include hypoxia, heat shock, and IGF-1. Together with our data, these studies support the notion that the history of signal input is a critical determinant of subsequent cell behavior. This is likely to be true under both normal and pathophysiological conditions. At any given time, each endothelial cell in the human body has been preconditioned by its past exposure to diverse biomechanical and biochemical forces. The term set point refers to the functional sequelae of that input history.
Collectively, these data suggest that the endothelium may be therapeutically primed in such a way as to reduce subsequent activation. A total of five phase 3 clinical trials have demonstrated improved survival in critically ill patients or patients with severe sepsis. These include the use of low tidal volume ventilation, activated protein C, low-dose glucocorticoids, intensive insulin therapy, and early goal-directed therapy. It is possible that each regimen exerts its benefit through a protective effect on the endothelium. For example, activated rhAPC may attenuate the response of endothelial cells to inflammatory mediators, early goal directed therapy may result in favorable hemodynamics at the level of the endothelium, and intensive insulin therapy reverses any deleterious effect of hyperglycemia on endothelial responses. The link between therapeutic efficacy and alteration of endothelial set point is hypothetical and requires further study.
Finally, our results have important implications for studying endothelial cells in vitro. In most instances, cultured cells are incubated with agonist at time0 and assayed at some later point for mRNA, protein, and/or function. Typically, the pretreatment conditions are standardized (e.g., serum-starved medium). We have shown that simple changes in the starting conditions result in complex shifts in subsequent behavior. Moreover, these changes may persist even after removal of the preconditioning regimen. Thus, when designing studies of cultured endothelial cells, it is important to consider (and, where necessary, modulate) the pretreatment conditions.
|
FOOTNOTES
To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-4037fje;
This article has been cited by other articles:
|
|
 |
|
  Md. R. Abid, K. C. Spokes, S.-C. Shih, and W. C. Aird NADPH Oxidase Activity Selectively Modulates Vascular Endothelial Growth Factor Signaling Pathways J. Biol. Chem., November 30, 2007; 282(48): 35373 - 35385. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Hawkins, L. A. Solt, I. Chowdhury, A. S. Kazi, M. R. Abid, W. C. Aird, M. J. May, J. K. Foskett, and M. Madesh G Protein-Coupled Receptor Ca2+-Linked Mitochondrial Reactive Oxygen Species Are Essential for Endothelial/Leukocyte Adherence Mol. Cell. Biol., November 1, 2007; 27(21): 7582 - 7593. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. C. Aird Mechanisms of Endothelial Cell Heterogeneity in Health and Disease Circ. Res., February 3, 2006; 98(2): 159 - 162. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
