RNA as drug and antidote

Article metrics

The biggest cause of serious illness and premature death in Western societies today is the blockage of blood vessels by abnormal blood clots1,2,3. On page 90 of this issue Rusconi and colleagues4 describe how they discovered a new drug that prevents blood clotting, and how they designed an antidote. The drug is an RNA 'aptamer', the sequence of which was found by iterative in vitro selection from a large combinatorial library of RNAs, screened for binding to a protein that promotes blood coagulation. Aptamers can be made of RNA or DNA, and they achieve their target-binding selectivity by virtue of the unique shapes they adopt through the formation of internal base pairs. Coagulation proteins are popular targets for both traditional and aptamer-based drugs, because they are available pure in soluble form, and their activity can be readily measured in assays of blood-clotting time or by their cleavage of small, colour-generating substrates5.

Coagulation proteins — or 'factors' — circulate in the bloodstream as inactive precursors, which sequentially activate each other after an initial trigger from tissue factor, exposed by damaged or diseased blood vessels (Fig. 1). An active coagulation factor is given the suffix 'a'. So a complex of tissue factor and factor VIIa activates factor IX (ref. 6) — the target of Rusconi and colleagues' aptamer4. Factor IXa, together with factor VIIIa, then activates factor X. Next, factor Xa, with help from factor Va, activates factor II (also called prothrombin), producing thrombin. Finally, thrombin converts soluble fibrinogen to fibrin, the protein that polymerizes to form a solid blood clot. Thrombin also activates and promotes the aggregation of platelets, which are blood cells that help to form a plug in the wall of the wounded vessel.

Figure 1: How blood clots.

From the top: clotting is triggered when factor VII, circulating in the blood, encounters tissue factor, exposed by blood-vessel damage or disease. The complex of tissue factor and factor VIIa (activated factor VII) activates factor IX, which in turn, with factor VIIIa, activates factor X. With factor Va, factor Xa converts prothrombin to thrombin; thrombin then converts fibrinogen to fibrin; and fibrin forms blood clots. There are various checks and balances — for example, factors VIII and V are back-activated by thrombin and can be degraded by activated protein C (APC), produced by thrombin in the presence of thrombomodulin (found on intact blood-vessel walls). Also, the complex of tissue factor and factor VIIa can activate factor X both indirectly (via factor IX) and directly: the direct pathway is rapidly shut off by tissue-factor-pathway inhibitor, whereas factor IX can be further activated by factor XI (top right), which is itself activated by thrombin. Many proteins and reactions have been omitted for clarity. Rusconi et al.4 have developed an RNA-based drug (an aptamer) that inhibits factor IXa, and its antidote. vWF, von Willebrand factor.

Normally, this process represents a vital first line of defence against blood loss. But a clot or platelet plug in the wrong place can cause a heart attack or stroke, by preventing blood flow and therefore oxygen supply to a particular area of tissue — hence the importance of drugs that inhibit blood clotting. Drugs that block the production of thrombin are termed anticoagulants, whereas those that inhibit platelet function are called antiplatelet agents. Remarkably, until quite recently there was only one class of antiplatelet agent, aspirin and its relatives, and only two classes of anticoagulant, heparins and antagonists of vitamin K, in widespread clinical use. Even more surprisingly, these were discovered (or introduced to the clinic) as long ago as 1900, 1916 and 1940, respectively. So the anti-clotting armamentarium remained sparsely stocked for most of the past century. Some innovative approaches have, however, started to fill it.

Beginning with leeches (in medicinal use for at least 1,000 years), researchers have screened a series of blood-feeding invertebrates and even vampire bats for the presence of anticoagulants in their digestive tracts — obviously a requirement for successfully eating blood. Hirudin is a thrombin inhibitor in leech saliva, and genetically engineered hirudin, as well as a second-generation derivative7, is now in clinical use for a few conditions. An anticoagulant peptide from the soft tick is a highly selective inhibitor of activated factor X, and a peptide from hookworms inhibits factor VII. The giant Amazon leech has a unique platelet-disaggregating enzyme in its saliva8, and vampire bats keep the blood flowing with a powerful clot-dissolving enzyme. None of these peptides or enzymes has reached the clinic yet. But several small synthetic molecules that inhibit thrombin or factor Xa directly are in clinical trials9. Moreover, a mouse antibody, modified to resemble human antibodies, that blocks a fibrinogen receptor on the platelet membrane is in widespread clinical use10.

All of these newer anti-clotting weapons have a drawback, however: they have no rapidly acting antidote. Consequently, overdosage may result in serious bleeding, with no effective treatment. The older drugs — heparin, warfarin and aspirin — are not wholly immune to this criticism either, although protamine, a protein from fish sperm, can be used to neutralize heparin by virtue of its positive charge (heparin is negatively charged). An overdose of warfarin severely reduces levels of all the vitamin-K-dependent coagulation factors, and can be reversed in an emergency by plasma infusion — but this carries a risk of virus transmission. The use of aspirin is associated with a high risk of stomach bleeding; it also leads to bleeding in the brain in a few long-term users. Aspirin is an irreversible inhibitor of a platelet enzyme, cyclo-oxygenase, and no antidote is available, or indeed possible. Ticlopidine, which also inhibits platelets, similarly suffers the 'no antidote' problem.

So Rusconi et al.4 wanted to see whether they could design both an anticoagulant and its matching antidote by taking advantage of the properties of RNA. The authors' choice of factor IX as their target is significant: previous animal studies11 involving the infusion of factor IXa that had an inhibited active site revealed a favourable therapeutic margin compared with heparin in preventing experimentally induced clotting. ('Therapeutic margin' refers to the dosage difference between preventing clotting and inducing blood-vessel failure through overt bleeding.) Until now, however, searches of natural products and screens of chemical libraries have not revealed a selective inhibitor of factor IXa.

Rusconi et al. started with eight rounds of selection for factor-IXa-binding aptamers from a 1014-fold combinatorial library of 80-nucleotide RNAs. They thereby generated an aptamer that is 5,000 times more specific for factor IXa than for the structurally related factors VIIa, Xa, XIa and activated protein C. This last point is important, because a protein C inhibitor would promote blood clotting (see Fig. 1). By comparing the effective sequences, the authors generated a truncated form to which polyethylene glycol could be attached without affecting the aptamer's affinity for factor IXa or its anticoagulant activity. (Polyethylene glycol increases the time that RNA aptamers can spend in the blood circulation.) Finally, although Rusconi et al. conducted all their studies in vitro, they did detect a significantly longer clotting time, compared with controls, after adding their aptamer to blood plasma.

What of the antidote? Aptamers achieve selectivity by adopting unique three-dimensional folds, as a result of internal Watson–Crick base pairing. To disrupt these folds it is simply necessary to design a nucleotide sequence that will compete effectively for the critical base pairs. Rusconi et al. have done this convincingly, using a complementary sequence to reverse the anticoagulant effects of the selected aptamer completely and rapidly.

How soon can we expect to see this class of drug and its antidote in clinical use? First we have to wait for in vivo studies in animals, which are now in progress. There is no reason to suppose that the aptamer and antidote would be toxic, or that they would affect clotting in vivo differently from that in vitro. Nonetheless, it is impossible to predict whether other enzymes or physiological systems will be adversely affected until the animal — and, ultimately, human — studies have been carried out.

It has been proposed that life on Earth had its origins in RNA, which had both a passive role as an information store and an active role as an enzyme that propagates this information. Later, these tasks became split between DNA and proteins. But perhaps we have now come full circle: the work of Rusconi et al.4 shows how one RNA molecule can have an active role as a drug, and can store the information required to make its own perfect antidote.


  1. 1

    Hirsh, J. & Lee, A. Y. Blood 99, 3102–3110 (2002).

  2. 2

    Turpie, A. G. & Antman, E. M. Arch. Intern. Med. 161, 1484–1490 (2001).

  3. 3

    Mannucci, P. M. & Poller, L. Br. J. Haematol. 114, 258–270 (2001).

  4. 4

    Rusconi, C. P. et al. Nature 419, 90–94 (2002).

  5. 5

    Bock, L. C. et al. Nature 355, 564–566 (1992).

  6. 6

    Eigenbrot, C. & Kirchhofer, D. Trends Cardiovasc. Med. 12, 19–26 (2002).

  7. 7

    Carswell, C. I. & Plosker, G. L. Drugs 62, 841–870 (2002).

  8. 8

    Sawyer, R. T., Jones, C. P. & Munro, R. Blood Coag. Fibrinol. 2, 153–159 (1991).

  9. 9

    Pauls, H. W. & Ewing, W. R. Curr. Top. Med. Chem. 1, 83–100 (2001).

  10. 10

    Harnick, D. J. & Vorchheimer, D. A. Curr. Cardiol. Rep. 3, 355–361 (2001).

  11. 11

    Himber, J. et al. Thromb. Haemost. 85, 475–481 (2001).

Download references

Author information

Correspondence to Edward Tuddenham.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tuddenham, E. RNA as drug and antidote. Nature 419, 23–24 (2002) doi:10.1038/419023a

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.