Contribution of different types of phospholipids in blood coagulation

Tavoosi, Narjes

Permalink

https://hdl.handle.net/2142/44746 Copy

Description

Title

Contribution of different types of phospholipids in blood coagulation

Author(s)

Tavoosi, Narjes

Issue Date

2013-05-28T19:17:52Z

Director of Research (if dissertation) or Advisor (if thesis)

Morrissey, James H.

Doctoral Committee Chair(s)

Morrissey, James H.

Committee Member(s)

• Sligar, Stephen G.
• Fratti, Rutilio A.
• Tajkhorshid, Emad

Department of Study

Biochemistry

Discipline

Biochemistry

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• blood clotting
• γ-carboxyglutamate-rich (GLA) domain
• phospholipid membrane

Abstract

In case of an injury platelets become activated and proper interaction of blood clotting proteins with membranes of activated platelets is essential for maintaining a balanced hemostasis. Hemophilia and thrombosis are the two extremes of an imbalanced hemostasis where patients with hemophilia suffer from severe bleeding and those with thrombosis from unnecessary blood clots that block normal flow of blood. There are several therapeutics available for treating patients suffering from different types of hemophilia and thrombosis, but lack of specific and efficient medications requires scientist to look for new and more specific activators or inhibitors of the blood clotting cascade. This will not be possible unless we have a detailed knowledge of the nature of the interactions between blood clotting proteins and phospholipid membranes. In this study, we benefitted from combining the biochemical and biophysical techniques with Nanodisc technology (Sligar lab), molecular dynamics (MD) simulations (Tajkhorshid lab) and solid state-NMR (SSNMR) (Rienstra lab), as a part of a collaborative project at university of Illinois. This approach enabled us to look into the protein-phospholipid interactions in blood clotting at a molecular scale. The first part of my thesis focuses mainly on γ-carboxyglutamate-rich (GLA) domains of factor X (fX) and prothrombin as representative GLA domains of clotting factors, which have been the subject of several protein-phospholipid interaction studies. Based on the results of this project, we proposed a new mechanism for the binding of GLA domains of clotting proteins to phospholipid membranes. The blood clotting reactions take place on membranes with exposed phosphatidylserine (PS). GLA domains are the most common PS-binding motifs in blood coagulation. A further intriguing aspect of the membrane’s contributions to clotting is that phosphatidylethanolamine (PE) supports little or no clotting activity in the absence of PS, but strongly synergizes with small amounts of PS to enhance many clotting reactions. The mechanism of PE/PS synergy is poorly understood, although several hypotheses have been proposed which rely on specific properties of PE’s phosphoethanolamine headgroup. We proposed a novel, general model for GLA domain binding to membranes; the ABC (Anything But Choline) hypothesis, supported by biochemical studies, solid-state NMR analyses and molecular dynamics simulations. This hypothesis invokes two types of protein-phospholipid interactions: “phospho-L-serine-specific” and “phosphate-specific.” In the latter, the accessible phosphate groups in phospholipids interact with tightly bound Ca2+ in GLA domains. We proposed that based on this model, phospholipids that can satisfy the phosphate-specific interactions should be able to synergize with PS to support fX activation. We showed that almost any glycerophospholipid other than phosphatidylcholine (PC) synergizes strongly with PS to enhance factor X activation by factor VIIa/tissue factor. We proposed that PC and sphingomyelin (the major external phospholipids of healthy cells) provide an anticoagulant surface to healthy cells because their bulky, highly hydrated choline headgroups sterically hinder access to their phosphates. Following cell activation, lysis, or damage, PE and PS are exposed on the outer leaflet where they can collaborate to create binding sites for GLA domains, by providing phosphate-specific and phospho-L-serine-specific interactions, respectively. The ABC hypothesis study is mainly focused on the GLA domains of fX and prothrombin. There are seven proteins in blood clotting that bind reversibly to phospholipid membranes through γ-carboxyglutamate-rich (GLA) domains. Although the GLA domains of these blood clotting proteins are very similar structurally, their membrane-binding affinities vary by almost three orders of magnitude. To our knowledge, no study has compared the membrane binding affinities of all seven human GLA domain-containing blood clotting proteins side by side under the same conditions. Furthermore, the PS stereospecificity of GLA domain binding (i.e., whether they preferentially recognize PS containing L-serine versus D-serine) has not been evaluated for all of these clotting proteins. As the second project of my thesis, we employed surface plasmon resonance binding studies and enzymatic assays to systematically investigate the phospholipid specificity of these seven GLA domain-containing proteins. It has long been thought that GLA domains of blood clotting proteins bind preferentially to bilayers containing PS; but we found, surprisingly, that two of the GLA domain-containing blood clotting proteins (factor VII and protein C) actually bound preferentially to membranes containing phosphatidic acid (PA) or phosphatidylinositol phosphate (PIP), compared to membranes containing PS. Furthermore, PA and PIP strongly enhanced the enzymatic activities of factor VIIa and activated protein C. Incidentally, of the seven blood clotting proteins with GLA domains, factor VII and protein C are known to have the lowest binding affinities for PS-containing membranes. The results of our experiments provide new insights into the membrane binding mechanism for these two GLA domain-containing clotting proteins, through PA- and PIP-specific binding interactions. Beside these projects, there are two other ongoing projects that were studied briefly as part my thesis. The first one was to investigate the effect of mixtures of Ca2+ and Mg2+ on blood clotting reactions. We usually use 2.5 mM calcium in our experiments which is an optimal concentration for supporting blood clotting reactions. Physiologically there are 1.25 mM free calcium and 0.5 mM magnesium in blood, and the crystal structures of GLA domains of fVIIa and fIX prepared in mixtures of these ions show occupation of two of these calcium binding sites with magnesium ions. Comparing rates of fX activation by cell surface complexes of tissue factor (TF) and factor VIIa (fVIIa) (or TF:fVIIa enzyme complex) on liposomes prepared with mixtures of PS/PE/PC showed that at lower concentrations of calcium, and on membranes with lower PS contents, the effect of magnesium ions is higher. Increasing the calcium concentration diminishes the effects of magnesium but does not overcome it. These results might be due to the effect of magnesium on phospholipid clustering in phospholipid membranes and also to changes in GLA domain structure and binding affinity for phospholipid membranes. These could be further tested by SS-NMR and MD simulations. The second ongoing project focuses on testing the ABC hypothesis in the prothrombinase complex (the complex of factors Va and Xa, which are responsible for converting prothrombin to thrombin). For fX activation by TF:fVIIa, it was shown that effects of PE and phosphatidylglycerol (PG) are basically similar and that there is a linear enhancement in rates of fX activation when PE is added to PG. Similar experiments with the prothrombinase complex showed that PE is not able to synergize with PG and PA as well as it does with the PS, but that it still shows significant synergy with non-PS phospholipids. This is mainly due to the fact that prothrombinase binds to anionic phospholipids well, but binds to PS better than the other anionic phospholipids. The phospho-L-seine specific interactions for the prothrombinase complex might not be as stringent as the tenase complex(the complex of TF and fVIIa, which are responsible for converting fX to fXa). In conclusion, we proposed a new hypothesis for binding of GLA domains of clotting factors to phospholipid membranes and showed a new phospholipid specificity for two blood clotting proteins.

Graduation Semester

2013-05

Permalink

http://hdl.handle.net/2142/44746

Copyright and License Information

Copyright 2013 Narjes Tavoosi

Owning Collections