Answers are grounded in this paper and cite their sources.
In the 1980s, Shiga-toxin (Stx)-producing Escherichia coli O157:H7 were identified as a cause of bloody diarrhea in the United States and were associated with hemolytic uremic syndrome (HUS) characterized by microangiopathic hemolytic anemia, thrombocytopenia, and renal failure [1]. One of the pathologic hallmarks of HUS is the presence of hyaline thrombi composed of fibrin and platelets in the microvasculature of involved tissues [2, 3]. How platelets become activated in diarrhea-associated HUS remains unknown.Multiple clinical studies have demonstrated systemic platelet activation in HUS. Most have focused on measurements of platelet α-granule contents released into plasma. Levels of β-thromboglobulin and platelet factor 4 were increased in the plasma of children with HUS [4]. Soluble P-selectin, an integral membrane glycoprotein stored in both α-granulase of platelets and Weibel-Palade bodies of endothelial cells, was increased in the plasma of patients with HUS [5, 6]. One study has demonstrated increased urinary levels of platelet-activating factor in children with HUS [7]. Two studies have assessed platelet responses to known agonists in patients with HUS and have determined that these responses are impaired [8, 9]. However, there is very little information about possible direct effects of Stx on platelet function [10, 11].The implicated bacterial etiologic agents produce one or more members of a protein toxin family, the Stxs. Stxs are heterodimers containing an enzymatically active A subunit and a pentamer of identical noncovalently linked B subunits. The A subunit is known to have RNA glycohydrolase activity, inhibiting protein synthesis by cleavage of a specific adenine residue of 28S ribosomal RNA. The B subunit confers binding specificity for a Stx glycolipid receptor, globotriaosylceramide (Gb3), which is present on the platelet membrane [12]suggesting that Stx binding could occur. Although our preliminary data [11]suggested that treatment of human platelets with Stxs enhanced ADP-associated platelet aggregation, we now report that Stxs from E. coli do not directly stimulate platelet activation, do not enhance aggregation with common platelet agonists, and do not bind to the platelet membrane.Human blood was collected from 5 healthy donors, 1 female age 32 and 4 males ages 24–42 who had not taken aspirin during the previous 10 days. Platelet-rich plasma (PRP) was prepared by centrifugation of citrated blood at 200 g for 20 min, and removing the platelet fraction using care not to disturb the buffy coat. Stxs were prepared as previously described [13]. Aggregometry was performed on test and control specimens prepared as follows. Test specimens were prepared by preincubating PRP for 1–2 h at room temperature with various concentrations of Stx 1 and Stx 2; control specimens contained PRP and distilled water, the toxin diluent. Each test specimen was pipetted into a cuvette and allowed to incubate at 37°C for 2–3 min, along with a control specimen in a separate cuvette. Stir bars were added, and baseline aggregation was measured for approximately 1 min. Agonists were then added, and aggregation patterns were allowed to develop. Agonists used were either adenosine-5′-diphosphate (ADP, Bio/Data Corp.) or a 14-amino-acid thrombin receptor activator peptide, or TRAP (TRA, Sigma Corp.). Slope of aggregation and percent total aggregation were calculated for each pair of test and control specimens.Figure 1 shows the slope and degree of aggregation in specimens preincubated with increasing concentrations of Stx 1 and Stx 2 compared to simultaneous control specimens averaged over 2–8 separate experiments using different platelet donors. Representative tracings are shown from the 100 ng/ml toxin concentration specimen and simultaneous control specimen for each agonist and toxin group. PRP preincubated with Stx 1 and Stx 2 did not enhance either the slope of aggregation or the total percent of aggregation in the presence of either ADP or TRAP as agonists. There was no difference in either slope or total percent of aggregation, even at very high concentrations of toxin (1 µg/ml). Stx 1 and Stx 2 alone do not independently stimulate aggregation of PRP (data not shown).In order to determine whether Stx stimulate the secretion of α-granules from platelets, we exposed gel-filtered platelets to various concentrations of Stx 1 and Stx 2 and assayed for surface expression of P-selectin using a phycoerythrin-conjugated monoclonal antibody to P-selectin (AC1.2, Becton Dickinson). P-selectin is a transmembrane α-granule constituent that is expressed on the platelet surface in an activation-dependent manner [14, 15]. Figure 2a shows FACS analysis of P-selectin on resting platelet membranes. As indicated in figure 2c, e, exposure of platelets to Stx did not stimulate the surface expression of P-selectin. In contrast, SFLLR, a peptide agonist which acts through the thrombin receptor, elicited marked P-selectin surface expression (fig. 2b). Preincubation of platelets with Stx at various concentrations for 30 min did not affect the surface expression of P-selectin in response to SFLLR (fig. 2d, f). Thus, Stx neither stimulated platelet α-granule secretion nor inhibited SFLLR-stimulated granule secretion. Recalcification of gel-filtered platelets did not affect these results (data not shown).The Stx B subunit pentamer mediates toxin binding to the glycolipid receptor Gb3. To determine if Stx 1 B subunit bound directly to Gb3 present on platelets, we exposed gel-filtered platelets to increasing concentrations of Stx 1 B subunit (0.5–500 µg/ml) labeled with Oregon Green or to increasing concentrations of ovalbumin (0.5–500 µg/ml) labeled with Oregon Green, and assayed for binding by fluorescence. Stx 1 B subunit was prepared as previously described [16]. Stx 1 B subunit and ovalbumin were conjugated with Oregon Green 488 according to the manufacturer’s instructions (FluoReporter Protein Labeling Kit, Molecular Probes). Concentrations of protein conjugates were determined by the modified Lowry method. Degree of fluorescent labeling of protein conjugates was determined by measuring the absorbance at 496 nm and calculating dye per molecule. The integrity of protein conjugates was verified by polyacrylamide gel electrophoresis, showing single fluorescent bands at migration distances appropriate for molecular weight. The Stx 1 B subunit protein conjugate retained ability to bind specific monoclonal antibody based on ELISA, bound to Vero cells which are known to express its receptor, Gb3, and did not bind to cells that do not contain Gb3.The protein concentration of the Oregon Green labeled Stx-1 B subunit was adjusted for efficiency of protein labeling so that equivalent amounts of fluorophore were added (fig. 3a, b). The final concentration of Oregon Green-labeled Stx 1 B subunit was approximately 40% that of Oregon Green-labeled ovalbumin when this was done. As shown in figure 3a, b, Oregon Green-labeled Stx 1 B subunit bound no better than Oregon Green-labeled ovalbumin to gel-filtered platelets, and neither protein showed any significant degree of binding until extremely high concentrations were added (the open histograms in figure 3 correspond to 500 µg/ml protein). Figure 3c shows Oregon Green-labeled Stx 1 B subunit at the same protein concentrations as ovalbumin used in figure 3b. The open histogram represents 0.5 µg/ml; the solid histogram represents 500 µg/ml protein. Increasing the binding time to 4 h and changing the membrane composition by activating the platelets with SFLLR did not affect binding (data not shown).In conclusion, we have carried out a careful study to determine whether the pathogenesis of HUS involves a direct effect of Stx on platelets. After preincubation of human platelets with Stx, we did not observe an enhanced aggregation of human platelets in response to ADP or TRAP. A very recent publication supports our observations [17]. Stx added simultaneously with the agonists ADP, collagen, or epinephrine did not enhance platelet aggregation.We have no evidence that Stxs from E. coli activate human platelets in these donors. Moreover, E. coli Stxs do not inhibit SFLLR-induced α-granule secretion, even using an extremely sensitive method of determining α-granule secretion, namely measuring P-selectin expression on platelet membrane surfaces. These negative findings may be due to the fact that platelets, either resting or activated, do not appear to bind the Stx 1 B subunit efficiently. Stx 1 B subunit is known to bind to the Stx receptor Gb3. The labeled Stx 1 B subunit bound well to other cells expressing GB3 on their surface, and there is no reason to think that the absence of the A subunit affected binding. In summary, the thrombotic microangiopathy seen in diarrhea-associated HUS does not appear to occur via a mechanism involving direct effects of Stx on circulating platelets.Research support for this study includes the following grants from the National Institute of Health, Bethesda, Md.: HL-55660, A1-16242, A1-39067, and P 30 DK-34928 for the Center for Gastroenterology Research on Absorptive and Secretory Processes. C.M.T. was supported by NIH Training Grant T32-AI-07329. R. F. is a Howard Hughes Medical Institute Physician Post-Doctoral Fellow. We thank Mr. James Acheson for his expert assistance with figure 1.
No takes yet. Share an insight, caveat, or question.
Thorpe et al. (1999) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: