Image: Courtesy of Barry R. Lentz
The findings in an August paper by Barry R. Lentz and colleagues are so controversial, it took three years to get them into print. Lentz, director of molecular and cellular biophysics at the University of North Carolina (UNC), purports to present "compelling evidence" that the lipid molecule phosphatidylserine (PS) is the key regulator of an enzyme complex central to blood coagulation.1
If true, this discovery would overturn a dogma of decades concerning the principal role played by membranes in clotting. It could change not only the way researchers think about coagulation, but how they prepare experiments to manipulate it, and which binding sites they choose for drug development. It might even lead to a new, targeted class of medications that prevent or treat the clots of strokes and heart attacks, hopefully without the unwanted bleeding that accompanies most such drugs. Lentz's main conclusion is that by binding to two key proteins, PS alters their functions and turns on the final phase of the clotting process, a job ascribed until now to the membrane surface.
He seems sanguine about his stance, chuckling at the memory of how he started investigating coagulation. As a young biochemist in the mid-1970s, he wanted to work at the interface between basic science and clinical health. UNC offered that potential and he was approached by the then-director of the clinical coagulation lab, Frederick A. Dombrose. He recalls Dombrose saying, "You know that membranes are really important in blood coagulation, and you and I are going to find out why." The two collaborated until Dombrose abandoned academia in 1982 for a career in industry. "I was left holding the bag," Lentz quips. He has peered into that bag ever since, mostly focusing on PS.
The fatty molecule has long been known to be an important part of the coagulation process, but just how important is the question. In a resting platelet, PS contributes to membrane structure by assembling in two layers. Most is on the inner layer or leaflet, due in part to an energy-promoting cellular pump that moves the molecules from the outside inward. When the disk-shaped platelet is activated to participate in repair of vascular damage, it undergoes "horrendous reorganization," Lentz says. "It shoots out protrusions, pieces of membrane break off, it spills part of its guts; it's a very dramatic process." The bilayer flips over, exposing the lipids and creating a negatively charged membrane surface.
THEORIES DIVERGE Here, Lentz's conclusions differ from mainstream thought. The negative charge has been considered for many years to be essential to assembly of the enzyme complex that cleaves the prothrombin molecule to produce high volumes of thrombin, the primary protein in coagulation. Thrombin cuts the big fibrinogen molecule into fibrin, which reinforces the spongy plug at the damage site that has been created by platelets sticking to each other. Without a negatively charged membrane surface, thrombin will be generated too slowly for proper clotting, according to the prevailing theory. But Lentz's team says it has proven that PS can regulate thrombin production at a rate comparable to that of a membrane surface, even in the absence of such a surface. It does this by binding to two proteins that are needed for assembly of the enzyme complex that cleaves prothrombin. These proteins are the clotting factor Xa (or 10 activated) and its cofactor, Va.
Of course, he can only demonstrate this effect in a solution that doesn't include a membrane surface. That's a problem for his critics, but not for his supporters. Lentz explains that when the thought occurred to him about a decade ago that PS has a special role in coagulation, he knew that the technical hurdle was an inability to characterize membrane binding sites. He needed to study individual molecules. That's why his excitement rose in the mid-1990s, when he supervised then-postdoc Vishwanath Koppaka in demonstrating that a soluble form of PS would bind to Xa and enhance its ability to activate prothrombin in solution.2 A series of experiments and publications followed, culminating in the August paper. But some people are unimpressed.
"Those papers of Lentz's have been kicking around now for about six years. That's because nobody believes it," says Kenneth G. Mann, the University of Vermont's biochemistry chairman and a leading researcher in coagulation. "You can replicate what a cell does except for one thing: cell activity."
Mann points out that after preparing a soluble PS solution, Lentz performs a measurement that shows it does not have a critical miceller concentration (CMC) at which macromolecular complexes form. Next, Lentz combines the soluble PS with factors Xa and Va in another solution, which changes the CMC. He then claims that the lipid, rather than the lipid-protein complex, is the cause of this change, Mann charges. In short, he misinterprets the data.
Not so, counters UNC medicine and pathology professor Harold R. Roberts, a distinguished researcher in the field. He and UNC kineticist Dougald ("Mac") Monroe are currently preparing an editorial for a journal on this topic. Roberts acknowledges that the proteins are necessary to clotting, just as the negative membrane charge clearly plays a role in vivo. But Lentz has found that PS orients the complex and turns it on like a light switch. This reaction is allosteric, meaning that Xa's activity is altered via a conformational change caused by lipid binding to a site other than the enzyme's active site.
"We think PS is necessary but not sufficient to account for coagulation," Monroe says. He adds that he understands why Lentz's paper took so long to get published. "He had to jump through an incredible number of hoops to show reviewers that the CMC was below active levels at all points."
Photo: Courtesy of Barry R. Lentz
QUESTION OF CATALYSIS "Lentz is getting catalysis, he just says that his catalysis is optimal. I'm not convinced," says Charles T. Esmon, head of cardiovascular biology research at the Oklahoma Medical Research Foundation. "The affinity of Xa and Va are enhanced immensely by a membrane surface." But he allows, "There probably are allosteric effects above and beyond that." Esmon's own work suggests that, after oxidation and with a small amount of PS, another phospholipid called phosphatidylethanolamine becomes a potent stimulator of anticoagulation.3
At the Scripps Research Institute, experimental hemostasis and thrombosis professor John H. Griffin has also shown stimulation of anticoagulation via several other lipids.4,5 All these findings are controversial, and Griffin is careful to point out that the procoagulant and anticoagulant cascades are much different, but he believes that evidence is mounting of important signaling roles played by lipids in these pathways. That, too, is controversial: Lipids are well-known messengers inside the cell, but not in clotting or bleeding processes. Yet, Griffin reveals that he has discovered another lipid messenger in anticoagulation, to be described in a paper in preparation. "The membrane surface paradigm is dead as a simple model," he pronounces.
Gary L. Nelsestuen disagrees. The University of Minnesota professor of biochemistry, molecular biology, and biophysics has collaborated with Roberts at UNC but voices a serious concern about Lentz's work. Nelsestuen says he thought for several days about how to summarize his skepticism and finally realized that the problem is that Lentz's experiments are based on negative results. Lentz claims that no bilayers are present, but zero is not measurable. The membrane might be transient or small; just because it isn't found doesn't prove its nonexistence. "If you can't prove it right and you can't prove it wrong, I'd sort of ignore it, because it's really not a scientific experiment. It's that fundamental," says Nelsestuen
He offers the cautionary example of protein kinase C, which binds phospholipids and was "the hottest regulatory protein of the 1980s." Many potential therapeutic uses for it were explored, but the enzymology that required the addition of phospholipids turned out to be influenced by artifacts. No useful compound ever was produced, Nelsestuen says. "Sometimes revolutionary ideas are right. I'd say more often, they're not."
Another coagulation expert, biochemistry professor Michael E. Nesheim at Queens University in Canada, thinks that Lentz's work "could have some fundamental mechanistic implications," but he's withholding judgment. Similarly, Nesheim thinks the jury is still out on recent data concerning the extent of allosteric influences exerted by phospholipids on coagulation and anticoagulation.
More important to him is a new direction in which he believes coagulation research should head. "We're going to have to understand the system itself," he says. "Just understanding how the individual reactions work won't be enough, because the system begins to assume properties that aren't explained by understanding individual components." Such answers won't be found in cultures, he suggests, but in the data streams of computer models.
Steve Bunk (firstname.lastname@example.org) is a contributing editor.
1. R. Majumder et al., "Soluble phosphatidylserine triggers assembly in solution of a prothrombin-activating complex in the absence of a membrane surface," Journal of Biological Chemistry, 277:29765-73, Aug. 16, 2002.
2. V. Koppaka et al., "Soluble phospholipids enhance factor Xa-catalyzed prothrombin activation in solution," Biochemistry, 35:7582-91, 1996.
3. O. Sofa et al., "Lipid oxidation enhances the function of activated protein C," Journal of Biological Chemistry, 276:1829-36, 2001.
4. H. Deguchi et al., "Neutral glycosphingolipid-dependent inactivation of coagulation factor Va by activated protein C and protein S," Journal of Biological Chemistry, 277:8861-5, March 15, 2002.
5. J.A. Fernandez et al, "Cardiolipin enhances protein C pathway anticoagulant activity," Blood Cells, Molecules and Diseases, 26:115-23, 2000