Thrombomodulin activity on human saphenous vein grafts prepared for ,coronary artery bypass James M. Cook, M D , Christine D. Cook, CT(ASCP), Richard Marlar, PhD, Maurice M. Solis, M D , L o u Fink, M D , and J o h n F. Eidt, M D , Denver, Colo.,

and Little Rock, Ark. Thrombomodulin, a membrane glycoprotein present on normal vascular endothelium, binds circulating thrombin and is important in protein C activation. These functions contribute to the nonthrombogenic nature of endothelium. Damage during harvest and ex vivo storage of vein grafts may result in dysfunction of this endothelial anticoagulant barrier and possibly contribute to early graft thrombosis. We studied the functional activity and antigenic expression of thrombomodulin on saphenous veins before (initial) and after (h~awested) harvest and storage for coronary artery bypass grafting in 15 patients. Also, fresh saphenous vein was studied after mechanical endothelial stripping. After storage for 2.7 -+. 0.6 hours at room temperature in heparinized saline, thrombomodulin functional activity in harvested vein segments was 28% less than initial segments (p = 0.08). Endothelial stripping resulted in a 79% reduction in thrombomodulin activity compared with initial segments (p = 0.04). Immunohistochemical staining confirmed thrombomodulin antigen on vein grafts after harvest and storage, but not on segments stripped of endothelium. Thrombomodulin functional activity and antigenic expression on human saphenous vein grafts is not significantly changed by harvest and relatively short periods of storage at room temperature in heparinized saline. (J VAse SuRG 1991;14:147-51.)

Normal vascular endothelium presents a nonthrombogenic surface to circulating blood. At least four substances produced by endothelial cells contribute to this property: prostacyclin, i tissue type plasminogen activator, 2 heparin-containing proteoglycans, 3 and thrombomodulin. 4'5 Thrombomodulin, is an integral membrane glycoprotein present on the luminal surface of endothelial cells. It serves a dual role in preventing thrombosis by acting as an essential cofactor for protein C activation and also by inhibiting the procoagulant properties of circulating thrombin. 4'5 Although much information has been gained by in vitro experiments, no report of thrombomodulin activity in intact human blood vessels or vascular grafts has appeared. Harvest and ex vivo storage of vein grafts during From the Departments of Surgery and Pathology, John L. McClellan Veterans Administration Hospital and University Hospital of Arkansas, Little Rock, and Department of Pathology, Denver Veterans AdministrationHospital, Denver. Reprint requests: James M. Cook, MD, John L. McClellan Veterans AdministrationHospital, 112-PV, 4300 West 7th St., Little Rock, AR 72205. 24/1/28730

vascular bypass procedures is known to result in endothelial cell loss and dysfunction. We hypothesized that this trauma could adversely affect thrombomodulin activity on the vein lumen. To address this question wc studied the functional activity and antigenic expression of thrombomodulin on human saphenous veins (HSV) before and after harvest for coronary artery bypass grafting. METHODS In 15 patients undergoing cardiac surgery, a 2 cm segment of greater saphenous vein was obtained before dissection of the entire graft length ("initial" segment). These segments were assayed immediately for thrombomodulin activity as described below. The remainder of the vein was harvested by means of a standard "no touch" technique and gently distended with heparinized saline (1 unit/ml) without a pressure limiting syringe. It was then stored in heparinized saline at room temperature and used as needed for coronary artery bypass grafting. At the completion of the procedure, unused remnants of vein ("harvested" segments) were assayed and compared with initial segments. 147

,[ournai o f VASCULAR SURGERY

148 Cook et al.

Table I. Thrombomodulin activity in human saphenous vein grafts

Initial (n = 15) H a r v e s t e d (n = 15) D e n u d e d (n = 5)

Thrombomodulin activity (mean OD +- SD)

Confidence interval (95% level)

% Decreasefrom initial

0 . 2 8 6 _+ 0 . 1 3 0 0 . 2 0 5 +- 0 . 1 1 6 0.061 + 0.105

0.214 - 0.358 0.141 - 0.270 -0.069 - 0.192

0 28 79

Thrombomodulin functional activity assay Vessel segments were cut longitudinally and placed in an acrylic resin template with eight reaction wells 6 mm in diameter in contact with the luminal vein surface (modified from Stern et al.6). The wells were rinsed three times with Tris-buffered saline, and bovine thrombin (0.05 units), calcium chloride (18 eg), 0 1% gelatin in Tris-buffered saline (65 Ixl), then human recombinant protein C (2txg; generously provided by Robert Wydro, PhD; Genzyme Corp.; Framingham, Mass.) were added. After incubation at 37 ° C for 45 minutes, the reaction was stopped by a 10-minute incubation with hirudin (60 ~g). A 100 mm 3 aliquot was removed, placed in a multiwell reading plate, and chromogenic substrate for activated protein C was added ($2366, Helena Laboratories, Beaumont, Texas; 25 mm 3 in 125 mm ~ of 0.5 mol/L Tris and 0.01 mol/L NaC1 buffer, pH 8.2). Thrombomodulin activity on the intimal surface, as evidenced by generation of activated protein C, was determined spectrophotometrically as change in optical density (OD) at 480 um 2 hours later (Autoreader II, Ortho Diagnostic Systems Inc., Raritan, N.J.). Background OD from wells without vein were subtracted from those in which intima was exposed. Control wells not containing thrombin and/or protein C were run on each vein segment. To further assess the importance of endothelium for protein C activation, five additional flesh veins were assayed after mechanical intimal stripping with a scalpel. Histology Immunoperoxidase staining of formalin-fixed, paraffin embedded HSV was performed with a rabbit polyclonal antithrombomodulin antibody (Codon, Inc., South San Francisco, Calif.) and a rabbit antifactor VIII von Willebrand factor antibody (VIII-vWF; Signet Laboratories, Inc., Dedham, Mass.). Immunostains were obtained with peroxidase staining kits (Signet Laboratories) with a 3-amino-9-ethylcarbazole substrate chromogen in formamide and appropriate blocking reagents. Histologic sections were examined with light microscopy

to determine the presence and distribution of thrombomodulin and VIII-vWF antigen on initial and harvested vein segments. Statistical analysis A two-tailed Student's t test was used to compare mean values of OD obtained from each vein segment, and statistical significance was assumed at the 0.05 level. With these parameters, the power (1-[3) of the study could be determined. 7 Also, 95% confidence intervals of OD values were calculated and compared. RESULTS Harvested veins were stored at room temperature in heparinized saline for 2.7 + 0.6 hours (mean + SD) before the thrombomodulin assay. As seen in Table I, thrombomodulin functional activity in harvested HSV segments was 28% less than in initial segments (95% confidence interval, 0.214-0.358 initial, 0.141-0.270 harvested). This difference did not reach statistical significance (i0 = 0.08). Mechanical endothelial denudation resulted in a 79% reduction in thrombomodulin activity compared to initial segments (95% confidence interval, 0.069-0.192; p = 0.04) (Fig. 1). The statistical power to detect a 50% decrease in thrombomodulin activity between initial and harvested segments was 86%. Immunohistochemical staining confirmed the presence of thrombomodulin and VIII-vWF in both initial and harvested samples (Fig. 2). The distribution of both antigens was limited to endothelial cells in all sections. No staining for either thrombomodulin or VIII-vWF was present on veins that had been stripped of endothelium (Fig. 3). DISCUSSION

Since the discovepy of thrombomodulh~ by Esmon and Owens 8 in 1981, attention has focused on its isolation and role in protein C activation in vitro. Recently the molecule has been successfully sequenced and cloned. 9-~1 There are 559 amino acid residues with a molecular weight of 75,000 to 105,000, and the structure of the molecule shares

Volume 14 Number 2 August 1991

Thrombomuodulis activity on human saphenous vein grafts

149

.45 ,4

.35

ILl

.25

..,I

.2

p=0.08

~__ .15 Q. 0 .1 ,05

INITIAL

HARVESTED

DENUDED

Fig. 1. Thrombomodulin activity in initial, harvested, and denuded saphenous vein segments.

t

Fig. 2. Antithrombomodulin immunoperoxidase stain of harvested human saphenous vein (arrow on endothelial cell staining for thrombomodulin) (counterstained with hematoxylin). some features with the low-density lipoprotein receptor. As seen in Fig. 4, thrombomodulin is composed of severa][ discrete molecular regions. The extracellular portion includes a hydrophobic aminoterminal domain and a region composed of six epidermal growth factor (EGF)-like repeats. A domain within the EGF-like region is thought to be responsible for both thrombin binding and protein C activation.12'1 In vivo, thrombomodulin is expressed on the luminal surface of endothelium in arteries, veins, capillaries, lymphatics, mad syncytiotrophoblast of placentas. 4 Circulating thrombin binds to a specific receptor site on the exposed thrombomodulin molecule (Fig. 5). The thrombin-thrombomodulin corn-

plex, in the presence of calcium, converts inactive protein C (synthesized in the liver) to activated protein C. Activated protein C then combines with protein S on the surface of either platelets or endothelium, and this complex catalyzes the proteolyric inactivation of coagulation factors Va and VIIIa. Activated protein C is neutralized by a protein C inhibitor and alphal-antitrypsin. 4,5 The clinical importance of systemic protein C deficiency is well recognized. The homozygous form of congenital protein C deficiency (neonatal purpura fulminans) is uniformly fatal. 14 Heterozygotes and sporadic cases usually have recurrent episodes of venous thrombosis beginning after puberty, although arterial thromboembolism has been

150

Journalof VASCULAR SURGERY

Cook et al.

Fig. 3. Antithrombomodulin immunoperoxidase stain of human saphenous vein stripped of endothelium (counterstained with hematoxylin).

Membrane

PROTEIN C

Extracellular

Cytoplasm

ACTIVATED PROTEIN C

® COOH--

NH 6 5 4 3 2 1

~ F THROMBIN ] TM

---------I~-~

L

TB

I

APC

I

Fig. 4. Thrombomodulin molecule (TB, Thrombin binding site; APC, site for protein C activation).

reported.IS 17 These patients characteristically have less than 50% of normal protein C antigen levels and functional activity. TM An acquired form of protein C deficiency is thought to play an important role in the subcutaneous thrombosis seen in warfarin-associated skin necrosis.19 In light of clinical observations, it has been suggested that thrombomodulin activation of protein C may be physiologically most important in vessels with low flow rates and in the microcirculation. 5 Recently it has been shown that various humoral agents, such as interleukin-1, 2° endotoxin, 21 and rumor necrosis factor, = decrease thrombomodulin activity in vitro. This may contribute to the so-called hypercoagulable state often seen in patients with sepsis and carcinoma.

ENDOTHELIAL MEMBRANE

Fig. 5. Mechanism of thrombomodulin-dependent protein C activation. Trauma from vein harvest has been shown to cause small areas of deendothelialization seen on electron microscopy, which may favor platelet deposition on these denuded surfaces. 2a Our data demonstrate that areas of complete endothelial cell loss have a markedly decreased ability to activate protein C. This might contribute to early graft thrombosis, particularly in bypasses with low flow rates (e.g., distal lower extremity bypass) despite treatment with antiplatelet agents. A strong trend toward decreased thrombomodulin activity in H S V grafts after harvest and storage was seen but did not reach statistical significance. Since patients with systemic protein C activities greater than 50% of normal do not have increased in vivo thromboses, the relatively modest 28% decrease found in this study suggests that protein C activation is functionally intact in these vein grafts at

Volume 14 Number 2 August 1991

the time of implantation. Further research is needed to determine whether the slight decrease in thrombomodulin activity is a result of endothelial cell loss or receptor dysfunction during ex vivo storage. If the former explanation is true, thrombomodulin activity may be a quantitarve measure of endothelial integrity. In the latter circumstance, methods used to increase thrombomodulin activity in vitro (i.e., incubation with forskolin or cyclic adenosine monophosphate 24 may have clinical applications during vein harvest and storage. CONCLUSIONS Mechanical endothelial denudation results in markedly diminished protein C activation on the lumen of saphenous vein grafts. Thrombomodulin antigen is correspondingly absent in denuded segments, but is present in endothelial cells of vein grafts after harvest and short periods of storage at room temperature in heparinized saline. Although thrombomodulin functional activity may be somewhat decreased in these grafts, protein C activation is not depressed to a clinically important degree in routine cases.

REFERENCES

1. Bush HL, Graber JN, Jakubowski JA, et al. Favorable balance ofprostacyclin and thromboxane A2 improves early patency of human in situ vein grafts. J VASe SURG 1984;2:149-59. 2. Gerard RD, Meidell RS. Regulation of tissue plasminogen activator expression. Ann Rev Physiol 19'89;51:245-62. 3. Fransson LA, Carlstedt I, Coster L, Malmstrom A. The fimctions of the heparin sulphate proteoglycans. Ciba Foundation Symposium 1986;124:125-42, 4. Dittman WA, Majerus PW. Structure and function of thrombomodulin: a natural anticoagulant. Blood 1990;75: 329-36. 5. Esmon CT. The roles of protein C and thrombomodulin in the regulation of blood coagulation. }"Biol Chem 1989;264: 4743-6. 6. Stern DM, Naworth PP, Kiesiel W, et al. A coagulation pathway on bovine aortic segments leading to generation of factor Xa and thrombin. I Clin Invest 1984;74:1910-21. 7. Rosner B. Fundamentals of biostatistics, 3rd ed. Boston: PWS-Kent Publishing Co, 1986:275-6. 8. Esmon CT, Owen WG. Identification of an endothelial cell cofactor for the thrombin.-catalyzed activation of protein C. Proc Natl Acad Sci USA 1981;78:2249-52. 9. Wen D, Dittman WA, Ye RD, et at. Human thrombomodulin: complete cDNA sequence and chromosome localization of the gene. Biochemistry 1987;26:4350-7.

Thrombomuodulin activity on human saphenous vein grafts

151

10. Jackman RW, Beeler DL, Fritze L, Soft G, Rosenberg RD. Human thrombomodulin gene is intron depleted: nucleic acid sequences of the cDNA and gene predict protein structure and suggest sites of regulatory control. Proc Nail Acad Sci USA

1987;84:6425-9. 11. Suzuki K, Kusumoto H, Deyashiki Y, et al. Structure and expression of human thrombomodulin, a thrombin receptor on cndothelium acting as a cofactor for protein C activation. EMBO J 1987;6:1891-7'. 12. Kurosawa S, Stearns DJ, Jackson KW, Esmon CT. A 10-kDa cyanogen bromide fragment from the epidermal growth factor homology domain of rabbit thrombomodulin contains the primary thrombin binding site. J Biol Chem 1988;263: 5993-6. 13. Suzuki K, Hayashi T, Nishioka J, et al. A domain composed of epidermal growth factor-like structures of human thrombomodulin is essential for thrombin binding and for protein C activation. J Biol Chem 1989;264:4872-6. 14. Seligsohn U, Berger A, Abend M, et al. Homozygous protein C deficiency manifested by massive venous thrombosis in the newborn. N Engl J Med 1984;310:559-62. 15. Griffin JH, Evatt B, Zimmerman TS, Kleiss AJ, Wideman C. Deficiency of protein C in congenital thrombotic disease. J Clin Invest 1981;68:1370-3. 16. Miletich J, Sherman L, Broze GJ. Absence of thrombosis in subjects with protein C deficiency. N EngI J Med 1987;317: 991-6. 17. Horellou MH, Conard J, Bertina RM, Samama M. Congenital protein C deficiency and thrombotic disease in nine French families. Br Med J 1984;289:1288-7. 18. D'Angelo SV, Comp PC, Esmon CT, D'Angelo A. Relationship between protein C antigen and anticoagulant activity during oral anticoagulation and in selected disease states. I Clin Invest 1986;77:416-25. 19. Naworth PP, Handiey DA, Esmon CT, Stern DM. Interleukin 1 induces endothelial cell procoagulant while suppressing cell-surface anticoagulant activity. Proc Natl Acad Sci USA 1986;83:3460-4. 20. Moore KL, Andreoli SP, Esmon NL, Esmon CT, Bang NU. Endotoxin enhances tissue factor and suppresses thrombomodulin expression of human vascular endothelium in vitro. J Clin Invest 1987;79:124-30. 21. Naworth PP, Stern DM. Modulation of endothelial cell hemostatic properties by tumor necrosis factor. J Exp Med 1986;163:740-5. 22. Gundry SR, lones M, Ishihara T, Ferrans VJ. Optimal preparation techniques for human saphenous vein grafts. Surgery 1980;88:785-94. 23. Green D. Protein C and ]protein S. In: Kwaan HC, Samama MM, eds. Clinical thrombosis, 1st ed. Boca Raton, Fla: CRC Press, Inc, 1989:243-53. 24. 1to T, Ogura M, Motishita Y, et al. Enhanced expression of thrombomodulin by cyclic AMP in human megakaryoblastic leukemia cell lines. Blood 2989; 74(Suppl 1):130a. Submitted Dec. 17, 1990; accepted Feb. 16, 1991.