There's a multi-billion-dollar drug that's been on the market since 1935 whose composition and mode of action leave the FDA largely in the dark, along with the pharmaceutical industry and the surgeons who inject it every day.
This deep-cover compound is heparin – the across-the-board anticlotting agent of choice. "It's extensively used when you have people undergoing coronary bypass surgery," said self-styled biotechnologist and heparin researcher Ram Sasisekharan, of the Massachusetts Institute of Technology (Cambridge, Massachusetts), "and putting them in extracorporeal devices such as heart-lung and kidney dialysis machines, where you don't want the blood to clot. And heparin is used post-surgically as well, to prevent thrombus formation – obstruction of blood vessels."
Heparin – more correctly heparan sulfate – occurs in every tissue throughout the bodies of all mammals. It's expressed on the surface of cells and in the extracellular matrix between them. It's produced most abundantly in liver, lungs and the intestinal mucosa. The pharmaceutical industry extracts it from a porcine source.
But what it extracts is like no other FDA-approved compound.
"For instance," said Ram [as Sasisekharan is known], "if you had a protein drug like EPO [erythropoietin], the first thing the FDA asks for is the precise molecular composition of what's in a vial of EPO. Imagine that all one could do was extract it from serum and say, 'Contained in that vial is EPO with such-and-such a biological activity – but there are a lot of other things in it besides.' So with heparin, you're talking only in terms of fractions in the range of function instead of precise, controlled amounts of material content that correlates with its anticlotting activity.
"The simple way to picture it," he said, "is just imagine putting heparin into a food blender and getting smaller fragments of polysaccharides – sugar molecules. You'd end up with a mixture of one-sugar oligosaccharides, dual dodeca, quadruple tetradeca levels. But still each is a fragment of the larger fraction."
Until a decade or so ago, normal commercial heparin weighed in at 12,000 to 20,000 Daltons molecular weight. Then, in recent years, low-molecular-weight versions emerged, tending to fall in the category of around 6,000 D. "That's where we are with low-molecular-weight heparin [LMWH]," Ram noted. "Still dealing with it in an analogous fashion in terms of extract purity and consistency."
Ram is senior author of two back-to-back papers in the Sept. 12, 2000, issue of the Proceedings of the National Academy of Sciences (PNAS), released Sept. 12, 2000. Their respective titles: "Sequencing of 3-0 sulfate containing heparin decasaccharides with a partial antithrombin III binding site," and "Cleavage of the antithrombin III binding site in heparin by heparinases and its implication in the generation of low molecular weight heparin."
A deficiency of antithrombin-III (AT-III) prevents inhibiting the clotting factor, thrombin, causing recurrent thromboses. When heparin binds to AT-III, its action increases several-fold. Ram's first PNAS paper revealed that analysis of heparin fragments generated in a fashion similar to the first American LMWH, DuPont-Merck's Tinzaparin, released last July, points to the fact that AT-III is cleaved off. It does not contain the intact AT-III binding decasaccharide (10-sugar unit). "These serious quality-control issues," Ram said, "arise because there has been no way to put a finger on the exact composition-activity relationship of either the high or low MW heparins."
He and his co-authors have activated that analytical finger. Their novel tool is based, he said, "on the fact that the heparin field has lagged far behind the mainstream work on DNA and proteins." That's because the complex sugars have many more building blocks than their better-known cousins. DNA relies on four nucleotides – adenine, cytosine, guanine and thymidine (A-C-G-T). Proteins depend on the 20 essential amino acids. Heparin's sugars consist of 32 building blocks.
"Our MIT analytical tool," Ram said, "is a quick, easy way to determine the structure, or order of building blocks, in these sugars. Once you have their sequence for a given polysaccharide," he added, "you can start cracking its function in the body.
"So we developed what we called a binary or alphanumeric code, which is the sort of language the computer uses. One particular code would contain additional characteristic information – as for instance the building blocks that make heparin are sulfated in different positions. So the tool we fashioned was a way to translate that data in a single code so the computer could very easily handle that information. Essentially packaging as much information as possible in a code, rather than just one alphabetical character, as a way to visualize a heparin building block."
Given its highly variable and unpredictable content, clinical heparin is not without its side effects, mainly hemorrhage and bleeding.
"There also is a very specific autoimmune indication known as heparin-induced thrombocytopenia," Ram said. "In this whole mixture of heparin, only between 20% and 30% – depending on how you prepare it and who prepares it – has an active anticoagulant ingredient. The rest of the mix has nothing to do with its clot-curbing function."
But, he added, "we're now beginning to understand that heparin also plays a role in modulating numerous growth factors and cytokines found in blood. One such molecule is platelet factor 4 – one of the original products that Repligen has brought to market. Heparin forms a complex in the blood with PF4, to which the body raises autoantibodies – the makings of an autoimmune disease."
Another big problem associated with heparin, according to Ram, is that its fragments that are not involved in anticoagulation precipitate osteoporosis in patients. "Moreover," he said, "it has been demonstrated in the last year or two that in the brain, proteins get deposited on the surface of neurons, and form a big complex with these heparin-like sugars. That leads to the amorphous amyloid texture of amylodosis – which leads to the beta-amyloid plaques of Alzheimer's disease."
He said the new MIT analytical tool "will enable us now to tackle the composition and sequences of these sugars associated with amyloid formation. Once we know what they are, we can start designing drugs to target those sequences."
Ram said the current focus is on angiogenesis and cancer biology. "What you see in amyloid formation also appears in tumor cells. They rapidly change the sugar composition so that they become extremely metastatic. So we're now trying to identify composition and sequences of how normal cells become transformed cells in sugar-associated transformation. Then we can hopefully come up with newer ways to develop anticancer and antimetastatic drugs."