Endothelial cell overexpression of plasminogen activator inhibitor - 1 (PAI-1) inhibits smooth muscle cell migration

S2

Surgical Forum Abstracts

J Am Coll Surg

but not NF␬B. Activation of NFAT was inhibited by PD 98059, curcumin and CsA but not SB203580. Conclusions: NFAT plays a key role in VEGF induced gene expression in EC that is regulated by p42/44 MAPK, JNK and is independent of p38 MAPK. This suggests a possible role for selective inhibition of signal transduction pathways in EC for anti-angiogenic and antiinflammatory therapy.

Inhibition of vascular smooth muscle and endothelial cell migration by apolipoprotein J Nayan Sivamurthy MD, Darren I Rohan MD, David H Stone MD, William C Quist MD PhD, Frank W LoGerfo MD. Beth Israel Deaconess Medical Center, Division of Vascular Surgery; Harvard Medical School. Address Correspondence to: Nayan Sivamurthy MD; BIDMC Vascular Surgery Research; Harvard Institute of Medicine; 4 Blackfan Circle, Room 130; Boston, MA 02115, USA; Phone # 617667-0094 Introduction: Intimal hyperplasia (IH) is a significant cause of delayed prosthetic arterial graft failure. Recently, we identified several genes with altered expression at the distal anastomosis of prosthetic grafts. One of these up regulated genes codes for a glycoprotein known as apolipoprotein J (apo J). The purpose of this study was to elucidate the role of apo J in human aortic vascular smooth muscle cell (SMC) and human umbilical vein endothelial cell (HUVEC) migration and to localize apo J in IH with immunohistochemistry. Methods: To study cell migration, a 48 well microchemotaxis chamber with a semi permeable polycarbonate membrane was used. Assays were performed with apo J alone or in combination with PDGF-bb, ␤FGF, or VEGF. The number of cells per high power field was used to assess migration. Immunohistochemistry was performed on the distal carotid polytetrafluoroethylene graft anastomosis harvested at 5, 14, and 30 days from canines using a monoclonal antibody to the alpha chain of Apo J. Results: Apo J did not cause migration in vitro for SMC or HUVEC from concentrations of 1 – 50 ␮g/ml. In fact, apo J decreased migration under the conditions studied. Cell

Media ⴙ

ApoJ1 ⴙ

ApoJ

ApoJ

ApoJ

25␮g/ml ⴙ

5␮g/ml ⴙ

10␮g/ml ⴙ

VEGF3

␤-FGF

␤-FGF4

6.4 ⫾ 2.3

28.2 ⫾ 7.7

4.8 ⫾ 1.6

Type

2% FBS

2% FBS

PDGF-bb

PDGFbb

VSMC

4.2 ⫾ 1.4

1.7 ⫾ 0.5

38.5 ⫾ 8.3

16.3 ⫾ 3.8

(p⬍0.02) HUVEC 338 ⫾ 41

6.8 ⫾ 3.6 (p⬍0.001)

VEGF

(p⬍0.04) 36 ⫾ 6

(p⬍0.001)

(p⬍0.02)

1. Apo J at 10 ␮g/ml for VSMC and 1 ␮g/ml for HUVEC in media ⫹ 2% FBS

Furthermore, immunohistochemistry localized apo J to SMC in the neointima of IH at 5, 14, and 30 days. Conclusions: The data suggest that apo J may play an important role in regulating the development or progression of IH by inhibiting the migration of SMC and endothelial cells. Elucidating the role of apo J in the development of hyperplastic lesions will have broad applications to the field of vascular wall biology.

acute arterial injury and has been implicated in cell migration. The purpose of this study is to define the functional role of JNK in TSP-1induced migration by transfection with dominant negative mutants. Methods: Quiescent bovine aortic VSMC were exposed to TSP-1 (20␮g/mL) for 0, 15, 30, and 120 minutes. Western blot analysis for JNK activation was performed on cell lysates. Then VSMC were transfected with JNK(k-r) (dominant negative mutant, 1.5␮g/mL) or pcDNA (vector control, 1.5␮g/mL), and ␤-Gal (transfection efficiency marker, 0.5␮g/mL) plasmids in SuperFect Transfection Reagent (transfection vehicle, 24␮g/mL, ⬃40% transfection rate). Migration studies were performed using a modified Boyden chamber with TSP-1 (20␮g/mL) or serum free medium (SFM) in the bottom chamber wells and transfected quiescent VSMC (50,000) in the top wells. Cells were stained with hematoxylin for total cell counts and X-gal for transfected cell counts. Results were recorded as VSMC migrated/5 fields (400X) and analyzed by paired t-test. Results: JNK was minimally activated (densitometry n ⫽ 3, @30 min. p54 x ⫽ 1.42 ⫾ 0.18; p46 x ⫽ 1.40 ⫾ 0.37) by TSP-1 at all time points (Fig 1). JNK(k-r) transfected VSMC had no significant difference in migration than pcDNA transfected VSMC when exposed to TSP-1 and both had more migration than the negative control (p ⬍ 0.05, Fig 2).

Conclusions: JNK is minimally activated by TSP-1; however, blocking JNK with its dominant negative mutant, JNK(k-r), does not inhibit migration of VSMC. These results suggest that TSP-1-induced migration is functionally independent of JNK.

Endothelial cell overexpression of plasminogen activator inhibitor - 1 (PAI-1) inhibits smooth muscle cell migration Richard R Proia, MD, Peter R Nelson, MD, Mary-Jo Mulligan-Kehoe, PhD, Arthur J Kehas, Robert J Wagner, Richard J Powell, MD, FACS. Dartmouth-Hitchcock Medical Center, One Medical Center Drive, Lebanon, NH 03756, USA. Phone # (603) 650-8190. Introduction: PAI-1, a known inhibitor of plasminogen activators, may regulate smooth muscle cell migration (SMC) through alteration in matrix metalloproteinase (MMP) activity. Methods: To study the effect of endothelial cell (EC) PAI-1 overexpression on SMC migration, RT-PCR was used to clone the full length PAI-1 gene, which was then ligated into the pCMV/myc/ER expression vector. Using electroporation, bovine aortic EC were transfected with either the PAI-1 construct or an empty vector control. EC PAI-1 overexpression was demonstrated using a PAI-1 activity assay and ELISA. The effect of EC PAI-1 overexpression on SMC migration was measured using a Boyden-chamber assay. SMC MMP expression was measured using zymography.

TSP-1-Induced VSMC migration is independent of JNK Shoichi Fuse, MD, Xiu-Jie Wang, MD, Alliric I Willis, MD, Eric Olson, BS, Susan Nesselroth, MD, Bauer E Sumpio, MD, PhD, FACS, Vivian Gahtan, MD, FACS. Yale University School of Medicine. Contact: V. Gahtan, MD, Yale University School of Medicine, 333 Cedar St., FMB 137, New Haven, CT 06520, USA (203) 737-4036, Fax (203) 785-7556 Introduction: Thrombospondin-1 (TSP-1), a transient extracellular matrix protein acutely elevated with vascular injury, stimulates vascular smooth muscle cell (VSMC) migration, a process important for the development of arterial lesions. The mitogen activated protein kinase pathway enzyme, c-jun N-terminal kinase (JNK), is activated after

Results: Selected clones (EC9, EC21) had a 2-5 fold increase in PAI-1 activity compared to untransfected EC and empty vector EC (ECC; Fig.

Vol. 191, No. 4S, October 2000

Surgical Forum Abstracts

1). Similarly, ELISA results showed a 2-4 fold increase in PAI-1 levels in EC9 and EC21 vs. ECC. Untransfected EC and ECC had similar effects on SMC migratory patterns. Migration of SMC exposed to PAI-1 overexpressing EC was inhibited by 35–57% compared to ECC (Fig. 2). This inhibititory effect was reversed by addition of exogenous uPA. Zymography showed downregulation of MMP 2 and 9 in SMC exposed to PAI-1 overexpressing EC.

S3

in coculture with EC resulted in a 47% reduction in EC-induced SMC migration, which under these conditions was only 10-fold greater than control (Figure 2).

Conclusions: EC PAI-1 overexpression inhibits SMC migration. This effect may be mediated by decreased SMC MMP activity.

Endothelial cells exposed to nicotine act as a chemoattractant for vascular smooth cell migration Gabriele Di Luozzo, M.D., Ajay K Dhadwal, M.D., Spiros G Frangos, M.D., Alan H Chen, M.D., Brian W Jeffries, B.S., Stanley J Dudrick, M.D. Bauer E Sumpio, M.D.,Ph.D. Yale University School of Medicine, Section of Vascular Surgery, 333 Cedar St. New Haven, 137 FMB, CT 06520-8602, (203) 785-2561 and St. Mary’s Hospital, Waterbury, CT 06706, USA Introduction: The long-term effect of nicotine on the vascular endothelium has clinically been expressed as advanced, diffuse atherosclerosis. However, the mechanism to date has not been elucidated. The aim of this study was to characterize the effect of nicotine exposed endothelial cells (EC) on vascular smooth muscle cell (VSMC) migration. ⫺8

Methods: Quiescent bovine ECs were exposed to either 10 M nicotine (preconditioned EC, PEC) or serum free media (SFM) for 48 hours. Bovine VSMCs were quiesced in SFM for 48 hours. Chemotaxis assays were performed using a modified Boyden microchemotaxis chamber. SFM, 10% FBS, Nicotine (10⫺8M), unconditioned EC media (EC-media), or PEC-media were placed in the respective lower wells, followed by a porous (8␮m) membrane. VSMCs were placed in the top wells (5 ⫻ 10⫺4 cells). The chamber was incubated for 4 hours at 37°C. The membrane was then removed, and stained in hematoxylin. Migrated cells were counted 5 fields per well at 400⫻. The average of 3 wells was expressed as mean ⫾ SEM. Results: There is a 14-fold increase in VSMC migration in the 10% FBS group, a positive chemoattractant, versus control (SFM). Nicotine alone did not stimulate VSMC migration. EC-media augmented VSMC migration by 3-fold. There was a 19-fold increase in VSMC migration in response to PEC-media. In other experiments, heating PEC media completely eliminates the chemoattractant activity. Conclusions: PECs release a transferable, heat-labile compound into the conditioned media that enhances VSMC migration significantly. Nicotine exposure appears to disturb the endothelial-smooth muscle cell interaction that could facilitate the development of atherosclerosis.

Smooth muscle cells alter endothelial cell regulation of smooth muscle cell migration Peter R Nelson, MD, Arthur J Kehas, Robert J Wagner, Richard R Proia, MD, Jack L Cronenwett, MD, FACS, Richard J Powell, MD, FACS. Dartmouth-Hitchcock Medical Center, Dartmouth Medical School, One Medical Center Drive, Lebanon, NH 03756, USA. (603) 650-8190. Introduction: Endothelial cells (EC) stimulate vascular cell migration in angiogenesis, but may have a suppressive effect on smooth muscle cell (SMC) migration in vascular injury. Using a coculture model, we examined the influence of SMC on EC-induced SMC migration. Methods: Bovine aortic EC and SMC were grown in bilayer coculture on opposite sides of a porous membrane. A modified Boyden chamber assay (Figure 1) was then developed to allow for the study of SMC migration either unstimulated (Control), or stimulated by EC alone (A), SMC alone (B), or EC grown in coculture with SMC (C). Results: Unstimulated SMC demonstrated low levels of random basal migration which served as control. Exposure to EC alone increased SMC migration by 19-fold compared with control. Exposure to SMC alone also stimulated SMC migration by 16-fold. The addition of SMC

Conclusions: EC by themselves exert a stimulatory effect on SMC migration, such as seen during angiogenesis. EC grown in coculture with SMC, however, display an attenuated ability to induce SMC movement. This model may better represent the interaction of these cell types at the site of vascular injury. This cellular interaction is likely the result of a diffusable molecular signal, since it is observed over a short time period in the absence of cell-cell interaction.

Human vascular smooth muscle cell (HVSMC) injury alters mitochondrial distribution Thomas N Robinson MD, Stephanie A Miller MD, Benjamin J Pomerantz MD, Anirban Banerjee PhD, Alden H Harken MD FACS. Univ. of Colorado Health Sciences Center, Dept. of Surgery, Campus Box C-320, 4200 East Ninth Ave., Denver, CO 80262, USA (303) 315-7060 Introduction: HVSMC migration is a critical step in the early stages of wound healing and atherosclerosis. Migration requires energy. Mitochondria (mitos) produce energy by generating an electrochemical proton gradient across the inner-membrane. Microtubules distribute mitochondria within a cell. We hypothesized that mitos with a high transmembrane potential would localize to the migrating area of the HVSMC after injury and that this process is microtubule dependent. Methods: Injure HVSMC by dividing a confluent layer in cell culture with a needle. Treat cells with Vinblastine 1 ␮M (depolymerizes microtubules) or vehicle for 2 hours. Migrating area defined as cell edge advancement over injured area. Migration evaluated by live digital imaging microscopy after JC-1 (measures mitos transmembrane potential) dye loading. Three separate donors used for each condition. T-test analysis performed. Results: Vinblastine reduces the area of mitos present in the migrating portion of the HVSMC from 9 ⫾ 2% in control to 0.2 ⫾ 0.2% (p ⬍ 0.05). In the migrating area of the HVSMC, 41 ⫾ 5% of all mitos are charged compared to 18 ⫾ 4% (p ⬍ 0.05) in the non-migrating area. Conclusions: Energized mitos preferentially distribute to the migrating area of the HVSMC. Mitochondrial redistribution requires microtubule integrity. Microtubule function appears important for energy distribution during migration in the injured HVSMC.

3-Aminopropanal in the pathogenesis of ischemia Jennifer R Syrek, MD, Svetlana Ivanova, PhD, Steven Friedman, MD FACS, Larry Scher, MD FACS, Kevin J Tracey, MD FACS. North Shore University Hospital, 300 Community Drive, Manhasset, NY 11030, USA. (Contact: Kevin J. Tracey, MD 516-562-2416) Introduction: The pathogenesis of ischemic tissue damage involves many factors including cytokines, arachidonic acid metabolites, and free radicals. 3-Aminopropanal (3-AP), a cytotoxic product of polyamine-oxidation, has recently been implicated in the pathogenesis of cerebral ischemia. However, the role of 3-AP in peripheral muscle ischemia was previously unknown. Methods: Male Sprague-Dawley rats were subjected to complete left hindlimb ischemia (HLI) by ligation of all branches on the left side of the infrarenal aorta and of the left iliac artery, followed by left common femoral artery ligation. After 24 hours, intracardiac (IC) (n ⫽ 4) and left common femoral vein (IV) (n ⫽ 4) blood samples, and left and right