Selectin-mucin interactions as a probable molecular explanation for the association of Trousseau syndrome with mucinous adenocarcinomas

Abstract

Trousseau described spontaneous, recurrent superficial migratory thrombophlebitis associated with occult cancers, and this was later correlated with disseminated microangiopathy (platelet-rich clots in small blood vessels). Trousseau syndrome often occurs with mucinous adenocarcinomas, which secrete abnormally glycosylated mucins and mucin fragments into the bloodstream. Since carcinoma mucins can have binding sites for selectins, we hypothesized that selectin-mucin interactions might trigger this syndrome. When highly purified, tissue-factor free carcinoma mucin preparations were intravenously injected into mice, platelet-rich microthrombi were rapidly generated. This pathology was markedly diminished in P- or L-selectin-deficient mice. Heparin (an antithrombin-potentiating agent that can also block P- and L-selectin recognition of ligands) ameliorated this platelet aggregation, but had no additional effect in P- or L-selectin-deficient mice. Inhibition of endogenous thrombin by recombinant hirudin also did not block platelet aggregation. Mucins generated platelet aggregation in vitro in hirudinized whole blood, but not in platelet-rich leukocyte-free plasma nor in whole blood from L-selectin-deficient mice. Thus, Trousseau syndrome is likely triggered by interactions of circulating carcinoma mucins with leukocyte L-selectin and platelet P-selectin without requiring accompanying thrombin generation. These data may also explain why heparin ameliorates Trousseau syndrome, while vitamin K antagonists that merely depress thrombin production do not.

Figure 1

Purified mucin preparations are large and retain calcium-dependent binding sites for all three selectins. (a) A 4% SDS-PAGE gel with detection of carbohydrates by PAS reagent. (b) ELISA in which immobilized mucin preparations are probed with recombinant soluble murine selectins. Values represent mean ± SEM; n = 5 individual mucin preparations. P, P-selectin; L, L-selectin; E, E-selectin.

Figure 2

Time-course of platelet-rich microthrombi in wild-type mouse lungs following a single injection of mucin. (a) Typical representations of in vivo platelet-rich microthrombi in lung sections observed using fluorescently labeled Ab against the platelet marker, CD41. (b) Quantitative histologic determination, quantitating average number of CD41-positive pixels per 200× view field (mean value of 20 random fields). Each bar represents the mean ± SEM. *P < 0.001 compared with PBS-injected mice, n = 7; **P < 0.05, n = 5.

Figure 3

Selectin deficiency attenuates platelet-rich microthrombus formation following mucin injection. (a) Typical representations of platelet-rich microthrombi in lung sections observed as in Figure 2, at 5 minutes after injection. (b) Quantitative determination of CD41 in lung sections as in Figure 2. Values shown represent the mean ± SEM. *P < 0.001 compared with mucin-injected wild-type mice, n = 7; **P < 0.05, n = 7. White bars, PBS-injected mice; black bars, mucin-injected mice.

Figure 4

Relative importance of fluid-phase coagulation versus selectins in mucin-induced microthrombus formation. (a and b) Quantitative determination of CD41 in lung sections as in Figure 2, at 5 minutes after injection. (a) Ability of heparin or hirudin to block the effects of mucin in wild-type mice. *P < 0.05 compared with mucin alone, n = 5; **P < 0.001 compared with PBS-injected mice, n = 5. (b) Effects of coinjecting subthreshold levels of thrombin with mucin. (c) Three-color immunofluorescence microscopic analysis of platelet-rich microthrombi in wild-type mice injected with mucin (with or without hirudin) to demonstrate fibrin, an indicator of thrombin activity. Bright green indicates deposited fibrin; pale green indicates autofluorescence of luminal surface of longitudinally sectioned venule; red indicates platelets; and blue indicates nuclei. Examples of occluding microthrombi and diffuse platelet aggregates are shown.

Figure 5

In vitro demonstration of the essential role of leukocyte L-selectin in mediating platelet activation. Flow-cytometric quantification of platelet activation based on P-selectin expression in stirred, hirudin-anticoagulated whole blood. Black bars, wild-type mouse blood; gray bars, L-sel^–/– mouse blood. Values represent the mean ± SEM (n = 3); *P = 0.007 relative to PBS-treated controls; **P = 0.003 relative to mucin-treated wild-type blood.

Figure 6

Proposed model for L- and P-selectin–mediated mucin-induced activation and aggregation of platelets. Yellow circle labeled with L indicates L-selectin ligands on mucins. Blue boxes labeled with P indicate P-selectin ligands, including carcinoma mucins and PSGL-1 expressed on leukocytes. L-sel, L-selectin; P-sel, P-selectin.