Methods for assessing the burden of parasitic zoonoses: echinococcosis and cysticercosis

Page 1

Update

|Research Focus

299 Malaria parasite transporters as a drug-deliverystrategyGiancarlo A. Biagini, Stephen A. Ward and

Patrick G. Bray

302 Model-organism genomics in veterinaryparasite drug-discoveryJohn S. Gilleard, Debra J. Woods and Julian A.T. Dow

305 The occurrence of strongylid nematodes in theepididymides of wood miceMaurizio Casiraghi and Marco Ferraguti

| Letter

307 Preventing confusion about side effects in acampaign to eliminate lymphatic filariasisKapa D. Ramaiah, Rengachari Ravi and Pradeep K. Das

|Book Reviews

308 Avian Malaria Parasites and OtherHaemosporidia (by Gediminas Valkiunas)Glenn A. McConkey

309 Nematology: Advances and Perspectives(Vols 1 and 2) (edited by Zhongxiaio Chen,Senyu Chen and Donald W. Dickson)Marc Pilon

310 Malaria Immunology (2nd edn)(edited by Peter Perlmann andMarita Troye-Blomberg)Christian Engwerda

311 The Trypanosomiases (edited by Ian Maudlin,Peter Holmes and Michael Miles)Keith Matthews

312 GIS and Spatial Analysis in Veterinary Science(edited by Peter A. Durr and Anthony C. Gatrell)Madelaine Norstrom

313 Livestock Trypanosomoses and their Vectors inLatin America (by Marc Desquesnes)Geoff Hide

314 Parasite Genomics Protocols(edited by Sara E. Melville)Jane M. Carlton

315 Oxford Handbook of Tropical Medicine(2nd edn) (by Michael Eddleston,Robert Davidson, Robert Wilkinson andStephen Pierini)Felissa R. Lashley

Opinion

316 Trypanosoma equiperdum: master of disguise or historical mistake?Filip Claes, Philippe Buscher, Louis Touratier and Bruno Maria Goddeeris

322 SAGE and the quantitative analysis of gene expression in parasitesDavid P. Knox and Philip J. Skuce

Review

327 Methods for assessing the burden of parasitic zoonoses: echinococcosis andcysticercosisHelene Carabin, Christine M. Budke, Linda D. Cowan, A. Lee Willingham III and

Paul R. Torgerson

334 Comparative folate metabolism in humans and malaria parasites (part II): activitiesas yet untargeted or specific to PlasmodiumAlexis Nzila, Steve A. Ward, Kevin Marsh, Paul F.G. Sims and John E. Hyde

340 Emerging technologies for the detection and genetic characterization of protozoanparasitesPaul T. Monis, Steven Giglio, Alexandra R. Keegan and R.C. Andrew Thompson

TRENDSin

July 2005

Vol. 21, No. 7

pp. 299–346

ParasitologyFORMERLY PARASITOLOGY TODAY

Editor Sally I. Hirst

Assistant Editor Peter Chapman

Editorial Coordinator Freddie Scott

Illustrations The Studio

Publishing Manager O. Claire Moulton

Editorial Enquiries

Trends in Parasitology

Elsevier,

84 Theobald’s Road,

London, UK WC1X 8RR

Tel: +44 (0)20 7611 4400

Fax: +44 (0)20 7611 4470

E-mail: [email protected]

Subscription Enquiries

E-mail: [email protected]

Advisory Editorial Board

J. Alvar, Majadahonda, Spain

Y. Aoki, Nagasaki, Japan

J. Behnke, Nottingham, UK

J.C. Boothroyd, Stanford, CA, USA

C. Burri, Basel, Switzerland

J. Carlton, Rockville, MD, USA

D.G. Colley, Atlanta, GA, USA

B.M. Cooke, Melbourne, Australia

G.H. Coombs, Glasgow, UK

F.E.G. Cox, London, UK

H. Ginsburg, Jerusalem, Israel

B. Gottstein, Berne, Switzerland

E. Handman, Melbourne, Australia

L. Hviid, Copenhagen, Denmark

M. Kennedy, Glasgow, UK

E. Ottesen, Atlanta, GA, USA

D.W. Severson, Notre Dame, IN, USA

C. Sibley, Seattle, WA, USA

W. Sirawaraporn, Bangkok, Thailand

J. Stevens, Exeter, UK

R.C.A. Thompson, Murdoch, Australia

A.J. Trees, Liverpool, UK

S. Upton, Manhattan, KS, USA

J. Vinetz, Galveston, TX, USA

A.P. Waters, Leiden, The Netherlands

M.E.J. Woolhouse, Edinburgh, UK

Y. Yuthavong, Bangkok, Thailand

Forthcoming articles

What distinguishes malaria parasites from other pigmented haemosporidians?G. Valkiunas, A.M. Anwar, C.T. Atkinson, E.C. Greiner, I. Paperna and M.A. Peirce

Worms can worsen malaria: towards a new means to roll back malaria?P. Druilhe, A. Tall and C. Sokhna

Cover: Female pig farmer in Mbulu District (Tanzania), where cysticercosis poses an obstacle to the marketing of pigs and pork:

reducing farmers’ incomes substantially and affecting thehealth, social life andproductivity of thepeople infected (see pp. 327–333).

Image supplied by A. Lee Willingham III. Design by Geraldine Woods.

Page 2

Research Focus

Malaria parasite transporters as a drug-deliverystrategy

Giancarlo A. Biagini, Stephen A. Ward and Patrick G. Bray

Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK, L35 QA

The recent characterization of the choline carrier of the

malaria parasite and its role in the selective delivery of

novel antimalarial drugs has reignited interest in para-

site transporters as a drug-delivery strategy. In this

article, we discuss these findings in relation to the wider

context of developing a sustainable antimalarial-drug-

development portfolio.

Why study malaria parasite transporters?

Levels of death and morbidity from malaria are increas-ing, largely because of parasite drug resistance [1]. Thegrim determination of the malaria parasite means thatthe limited armamentarium of antimalarial drugs willsoon be unable to cope. Recognition of this problem by theinternational community and the engagement of thepharmaceutical industry have resulted in the recommen-dation of several strategies [2], one of which is thedevelopment of new drugs with novel targets. Whendeveloping new antimalarial compounds, one must keepin mind that, because the parasite resides in red bloodcells (RBCs), there are multiple membranes that must betraversed to access most intraparasitic targets, includingthe host (red) cell membrane (HCM), the parasitophorousvacuolar membrane (PVM), the parasite plasma mem-brane (PPM) and, possibly, a further organelle membrane[e.g. a food vacuole membrane (FVM) or a mitochondrialmembrane (MM)], depending on the site of action of thedrug. Many promising antimalarial compounds arecharged and, therefore, require transportation into theparasite; a thorough understanding of parasite trans-porter function and specificity offers huge potential for thefuture of parasite chemotherapy.

The malaria parasite protects itself from the hostimmune system by residing in RBCs. By doing so,however, it places itself in an intracellular milieu of low[NaC] and high [KC], which is the inverse of what isexperienced by most eukaryotic cells [3]. For this reason(for the first 15 h of RBC infection, at least), the malariaparasite cannot use the NaC gradient to drive the uptakeand efflux of charged molecules. Instead, it seems that theparasite uses its large membrane potential (Jm), gener-ated by the energy-dependent pumping of protons [4,5], as‘currency’ for the transport of charged molecules. Proton-dependent transport across the PPM has been describedfor several substrates, including lactate, pyruvate [6] andthe vitamin pantothenate [7].

Corresponding author: Biagini, G.A. ([email protected]).Available online 26 May 2005

www.sciencedirect.com

In addition, it has been known for 30 years thatmalaria-infected human RBCs show an enhanced perme-ability to a range of molecules [8]. Seminal physiologicalstudies [9–11] have since shown that, upon infection,RBCs demonstrate a new permeation pathway (NPP), orpathways, exhibited as an inwardly rectified channel(or channels) showing a preference for anions rather than,but not to the exclusion of, cations. It is still debatedwhether the NPP consists of single or multiple channelsand whether the channel activity is parasite derived orhost derived [12,13]; however, it is clear that the presenceof the NPP is symptomatic of parasite residence withinRBCs.

Translation of malaria physiology to chemotherapy

So, how can the newfound knowledge of parasite transportphysiology be turned into a strategy for chemotherapy?The first strategy has been to target the NPP and PPMtransporters directly. Several compounds have beenidentified that target the NPP, as demonstrated by theirinhibition of NPP-mediated transport of small solutes.These compounds include phlorizin [14], sulfonyl ureas[15], arylaminobenzoates [16,17], cinnamic acid deriva-tives [18] and chalcones [19]. Unfortunately, no currentcompounds are active at a low enough concentration to bepharmacologically relevant (setting aside other issuessuch as in vivo activity, bioavailability, toxicology, medi-cinal chemistry and cost), and a defining proof-of-conceptstudy demonstrating that the NPP is a legitimate target isstill needed. At present, a more promising alternative is totarget PPM transporters such as the hexose transporterPfHT with O-3-hexose derivatives [20], which have beenshown to block transport, leading to a reduction of ATPlevels, a loss of pH homeostasis [21] and, ultimately, death.It remains to be seen whether this approach can be used todevelop a viable drug candidate in the near future.

The second strategy is more subtle in its approach butno less devastating for the parasite. During the earlystudies characterizing the NPP, it was suggested thatthe unique properties of this parasite-induced channelcould be used to target into infected RBCs drugs that,otherwise, cannot penetrate (i.e. are safe to) host cells[22]. Since then, major advances have been made withthis approach in targeting parasite purine metabolism[23]. It was shown that several nucleoside analogues,including L-enantiomers, that cannot penetrate normalRBCs could enter not only infected RBCs through the NPP[24] but also the PPM through the nucleoside transporter

Update TRENDS in Parasitology Vol.21 No.7 July 2005

Page 3

FV

NPP

HCM

PPM PVM

++ +

+

+

+

+

+

− − −−−−−−

Organic-cation transporter

CholinePentamidineT16

Parasite

(i)(ii)

(iii)

(iv)

Erythrocyte

TRENDS in Parasitology

Figure 1. Hypothesized accumulation and selective targeting of pentamidine and

T16. (i) Choline, pentamidine and the bisquaternary ammonium drug T16 enter

infected RBCs through parasite-induced NPPs. Choline and the drugs are

transported across the PVM (ii) and the PPM (iii) by the choline carrier, which is

energized by the transmembrane HC potential. (iv) The binding of choline and the

drugs to haem (ferriprotoporphyrin IX) drives their accumulation up to millimolar

levels in the digestive food vacuole (FV): a process that is vital for antimalarial

activity.

Update TRENDS in Parasitology Vol.21 No.7 July 2005300

PfNT1 [25]. However, although several promising com-pounds have been synthesised that are selectivelypermeable and that could target purine metabolism [23],this strategy remains at an early stage of evaluation andrequires more rigorous testing.

The idea of using parasite transporters as a selectivedrug-delivery route has gained further momentum in arecent study of antimalarial drugs that are designed totarget parasite phospholipid (PL) metabolism [26]. Ithad been shown that bisquaternary ammonium anddiamidine are selectively taken up by infected RBCsthrough the parasite-induced NPP [27,28]. These com-pounds represent a large number of compounds that hasbeen synthesized to mimic the structure of choline. Leadbisquaternary ammonium or diamidine compounds havebeen shown to exhibit potent in vitro activity againstPlasmodium falciparum and Plasmodium vivax [28–30],and good in vivo activity against P. falciparum andPlasmodium cynomolgi in Aotus and Rhesus monkeys,respectively [31]. PL metabolism is absent from matureerythrocytes but is rampant in infected RBCs (the PLcontent increases by 500% [32,33]), with much of itderiving from the de novo pathway for phosphatidyl-choline (PC), in which choline is actively salvaged from thehost serum [34]. The antimalarial activity of cholineanalogues has been correlated to the inhibition of the denovo pathway for PC synthesis [29] and it was speculatedthat these choline analogues exert their antimalarialactivity by inhibiting the PPM choline transporter [31].Because all of these drugs are positively charged, it islikely that they also need to be transported into theparasite (because a positive charge would prevent simplediffusion), although it was not clear which transporter wasresponsible. This work led to the investigation of howthese compounds can be further transported across thePPM and how they might interfere with de novo PCmetabolism in the parasite.

Little was known about choline transport across thePPM and, therefore, one of the first objectives was tocharacterize the parasite choline transporter. To do this,studies were performed with saponin-freed parasites toremove interference from the choline transporters on theHCM. Choline uptake across the PPM was shown to becarrier mediated (Km of choline was 25 mM) and NaC

independent but energized by the transmembrane Jm

[26]. An independent study of the P. falciparum cholinePPM transporter that was published at the same time asRef. [26] gave similar results and confirmed the electro-genic uptake of choline across the PPM [35]. A comparisonof choline-uptake rates in infected RBCs and freedparasites revealed that the PPM carrier transportscholine at a rate 60 times that of the combined uptakeby the endogenous and induced infected-RBC transporters[26]. This suggests that, at steady state, the kinetics ofcholine transport measured using intact infected RBCswould relate to the rate-limiting processes on the HCMrather than those on the PPM. This signifies that there is areduced choline concentration in the host cell relative tothe serum, likening the PPM carrier to a choline ‘vacuumcleaner’ for the host cell. Also of note is that thebiochemical properties of the P. falciparum PPM choline

www.sciencedirect.com

transporter that were presented in the two studies [26,35]distinguish it from the high-affinity choline transporterobserved in neurons [36]. In many respects, including itsdependence on a proton-motive force, it is more akin to thecholine transporter characterized in Leishmania major[37]. This functional divergence from host orthologuesmakes the P. falciparum PPM choline transporter parti-cularly suitable for chemotherapeutic attack.

It was further demonstrated that choline uptake couldbe inhibited by titration of either pentamidine or T16(a lead bisquaternary ammonium compound). Radiolabelflux experiments of both drugs and choline demonstratedthat this inhibition was a result of competition for thesame PPM transporter, which demonstrates that thesecompounds gain access to the intracellular parasitethrough the PPM choline transporter (Figure 1). Inter-estingly, although both pentamidine and T16 inhibit thetransport of choline, it is unlikely that this inhibitionforms the basis of antimalarial action. Pentamidine has anIC50 of w70 nM but, against the same parasite strain, itsKi for inhibiting the choline PPM carrier is 3 mM (a similardisparity is observed with T16). Therefore, the datasuggest that the observed inhibition of PL metabolismby these compounds [29] is not due to the inhibition of thePPM choline carrier per se, instead indicating that latersteps in the Kennedy pathway are targeted. It has beenshown that the antimalarial activity of these drugs reliesprimarily on their high intracellular accumulation (theyaccumulate 500-fold inside the parasite, relative to theexternal concentration) as a result of binding to ferriproto-porphyrin IX [27–28]. With the intraparasitic concen-tration of these drugs reaching the millimolar range, it

Page 4

Update TRENDS in Parasitology Vol.21 No.7 July 2005 301

would not be surprising if several cellular functions weredisrupted before parasite death.

Chemotherapeutic implications for the PPM choline

carrier

The parasite-induced NPP and the parasite-encoded PPMcholine carrier form a cooperative transport network thatoffers great potential for the selective targeting of potentantimalarial drugs. In addition, because of its importanceto PL metabolism, the PPM choline transporter is likelyto be an excellent pharmacological target. A project isunderway in our laboratories to characterize further thePPM choline carrier at the molecular level for subsequentheterologous expression. Concurrently, we have initiateda medicinal chemistry programme that focuses ondetermining the chemical features that are required forboth the recognition and the inhibition of cholinetransport by the NPP and the PPM carrier. Thisinformation will feed into a further structure–activityrelationship programme to help improve the antimalarialactivity of the pharmacophore. It is hoped that this newantimalarial strategy will help to create a sustainableantimalarial-drug-development portfolio for the treat-ment of malaria in this current climate of drug resistance.

AcknowledgementsG.A.B. is a Leverhulme Trust Fellow. S.A.W. and P.G.B. are supported byfunding from the BBSRC, the MRC and the Wellcome Trust.

References

1 Snow, R.W. et al. (2001) The past, present and future of childhoodmalaria mortality in Africa. Trends Parasitol. 17, 593–597

2 Biagini, G.A. et al. (2003) Antimalarial chemotherapy: young guns orback to the future? Trends Parasitol. 19, 479–487

3 Kirk, K. (2001) Membrane transport in the malaria-infected erythro-cyte. Physiol. Rev. 81, 495–537

4 Saliba, K.J. and Kirk, K. (1999) pH regulation in the intracellularmalaria parasite, Plasmodium falciparum. HC extrusion via a v-typeHC-ATPase. J. Biol. Chem. 274, 33213–33219

5 Allen, R.J.W. and Kirk, K. (2004) The membrane potential of theintraerythrocytic malaria parasite Plasmodium falciparum. J. Biol.Chem. 279, 11264–11272

6 Elliott, J.L. et al. (2001) Transport of lactate and pyruvate in theintraerythrocytic malaria parasite, Plasmodium falciparum. Bio-chem. J. 355, 733–739

7 Saliba, K.J. and Kirk, K. (2001) HC-coupled pantothenate transport inthe intracellular malaria parasite. J. Biol. Chem. 276, 18115–18121

8 Homewood, C.A. and Neame, K.D. (1974) Malaria and the perme-ability of the host erythrocyte. Nature 252, 718–719

9 Kutner, S. et al. (1982) Alterations in membrane permeability ofmalaria-infected human erythrocytes are related to the growth stageof the parasite. Biochim. Biophys. Acta 687, 113–117

10 Kirk, K. et al. (1994) Transport of diverse substrates into malaria-infected erythrocytes via a pathway showing functional character-istics of a chloride channel. J. Biol. Chem. 269, 3339–3347

11 Desai, S.A. et al. (2000) A voltage-dependent channel involved innutrient uptake by red blood cells infected with the malaria parasite.Nature 406, 1001–1005

12 Alkhalil, A. et al. (2004) Plasmodium falciparum likely encodes theprincipal anion channel on infected human erythrocytes. Blood 104,4279–4286

13 Decherf, G. et al. (2004) Anionic channels in malaria-infected humanred blood cells. Blood Cells Mol. Dis. 32, 366–371

14 Kutner, S. et al. (1987) On the mode of action of phlorizin as anantimalarial agent in in vitro cultures of Plasmodium falciparum.Biochem. Pharmacol. 36, 123–129

www.sciencedirect.com

15 Kirk, K. et al. (1993) Glibenclamide and meglitinide block thetransport of low molecular weight solutes into malaria-infectederythrocytes. FEBS Lett. 323, 123–128

16 Kirk, K. and Horner, H.A. (1995) In search of a selective inhibitor ofthe induced transport of small solutes in Plasmodium falciparum-infected erythrocytes: effects of arylaminobenzoates. Biochem. J. 311,761–768

17 Staines, H.M. et al. (2004) Furosemide analogues as potent inhibitorsof the new permeability pathways of Plasmodium falciparum-infectedhuman erythrocytes. Mol. Biochem. Parasitol. 133, 315–318

18 Kanaani, J. and Ginsburg, H. (1992) Effects of cinnamic acidderivatives on in vitro growth of Plasmodium falciparum and on thepermeability of the membrane of malaria-infected erythrocytes.Antimicrob. Agents Chemother. 36, 1102–1108

19 Go, M.L. et al. (2004) Antiplasmodial chalcones inhibit sorbitol-induced hemolysis of Plasmodium falciparum-infected erythrocytes.Antimicrob. Agents Chemother. 48, 3241–3245

20 Joet, T. et al. (2003) Validation of the hexose transporter ofPlasmodium falciparum as a novel drug target. Proc. Natl. Acad.Sci. U. S. A. 100, 7476–7479

21 Saliba, K.J. et al. (2004) Inhibition of hexose transport and abrogationof pH homeostasis in the intraerythrocytic malaria parasite by anO-3-hexose derivative. FEBS Lett. 570, 93–96

22 Ginsburg, H. and Stein, W.D. (1987) New permeability pathwaysinduced by the malarial parasite in the membrane of its hosterythrocyte: potential routes for targeting of drugs into infectedcells. Biosci. Rep. 7, 455–463

23 Gero, A.M. et al. (2003) New malaria chemotherapy developed byutilization of a unique parasite transport system. Curr. Pharm. Des. 9,867–877

24 Upston, J.M. and Gero, A.M. (1995) Parasite-induced permeation ofnucleosides in Plasmodium falciparum malaria. Biochim. Biophys.Acta 1236, 249–258

25 Carter, N.S. et al. (2000) Isolation and functional characterization ofthe PfNT1 nucleoside transporter gene from Plasmodium falciparum.J. Biol. Chem. 275, 10683–10691

26 Biagini, G.A. et al. (2004) Characterization of the choline carrier ofPlasmodium falciparum: a route for the selective delivery of novelantimalarial drugs. Blood 104, 3372–3377

27 Biagini, G.A. et al. (2003) Heme binding contributes to antimalarialactivity of bis-quaternary ammoniums. Antimicrob. AgentsChemother. 47, 2584–2589

28 Stead, A.M. et al. (2001) Diamidine compounds: selective uptake andtargeting in Plasmodium falciparum. Mol. Pharmacol. 59, 1298–1306

29 Ancelin, M.L. et al. (1998) Antimalarial activity of 77 phospholipidpolar head analogs: close correlation between inhibition of phospho-lipid metabolism and in vitro Plasmodium falciparum growth. Blood91, 1426–1437

30 Ancelin, M.L. et al. (2003) Potent inhibitors of Plasmodiumphospholipid metabolism with a broad spectrum of in vitro anti-malarial activities. Antimicrob. Agents Chemother. 47, 2590–2597

31 Wengelnik, K. et al. (2002) A class of potent antimalarials and theirspecific accumulation in infected erythrocytes. Science 295, 1311–1314

32 Holz, G.G., Jr. (1977) Lipids and the malarial parasite. Bull. WHO. 55,237–248

33 Vial, H.J. and Ancelin, M.L. (1998) Malarial lipids. In Malaria:Parasite Biology, Pathogenesis and Protection (Sherman, I.W., ed.),pp. 159–175, ASM Press

34 Vial, H.J. et al. (1999) Transport of phospholipid synthesis precursorsand lipid trafficking into malaria-infected erythrocytes. NovartisFound. Symp. 226, 74–83

35 Lehane, A.M. et al. (2004) Choline uptake into the malaria parasite isenergized by the membrane potential. Biochem. Biophys. Res.Commun. 320, 311–317

36 Rabkin, S.W. (1990) Effect of amiodarone and desethylamiodarone oncholine uptake and phosphatidylcholine biosynthesis in cultured chickcardiac myocytes. J. Mol. Cell. Cardiol. 22, 965–974

37 Zufferey, R. and Mamoun, C.B. (2002) Choline transport in Leishma-nia major promastigotes and its inhibition by choline and phosphocho-line analogs. Mol. Biochem. Parasitol. 125, 127–134

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.013

Page 5

Update TRENDS in Parasitology Vol.21 No.7 July 2005302

Model-organism genomics in veterinary parasite drug-discovery

John S. Gilleard1, Debra J. Woods2 and Julian A.T. Dow3

1Division of Infection and Immunity, Institute of Comparative Medicine, Faculty of Veterinary Medicine, University of Glasgow,

Glasgow, UK, G61 1QH2Veterinary Medicine Research and Development, Pfizer Animal Health, Ramsgate Road, Sandwich, Kent, UK, CT13 9NJ3Division of Molecular Genetics, Institute of Biomedical and Life Sciences, University of Glasgow, Glasgow, UK, G11 6NU

A recent article about genomic filtering highlights

exciting new opportunities for antiparasitic drug dis-

covery resulting from major advances in genomic

technologies. In this article, we discuss several

approaches in which model-organism genomics and

proteomics could be applied to the identification and

validation of novel targets for antiparasitic drug

discovery in veterinary medicine.

The drug-target challenge

Antiparasitic drug discovery is estimated to cost $25million per year across the animal health industry, withonly one or two classes of compound commercialized perdecade and few mechanisms targeted by antiparasiticmolecules. All new drug–target interactions must inter-fere with parasite survival, be selective (i.e. safe) and notbe cross-resistant with known resistance mechanisms.The target must also be amenable to in vitro screening,ideally with a high-throughput screening (HTS)-compa-tible functional endpoint. Incorporate the need to build inspectrum, bioavailability, stability and persistence, inaddition to development costs of $50 million to 100 million(industry figures), and the low success rate is notsurprising. Recent advances in genomic technology offerthe opportunity to identify, validate and develop screeningconstructs for new antiparasitic drug targets, using amore empirical approach than was previously possible.

Model organisms are powerful systems for drug-targetdiscovery and, following completion of the Drosophilamelanogaster and Caenorhabditis elegans genomeprojects, one could consider all 13 000 predictedD. melanogaster genes or all 19 000 predicted C. elegansgenes to be prospective antiparasitic drug targets. Targetvalidation is an expensive and painstaking process,however, so a key challenge is to select a small numberof potential targets that are worthy of the investmentrequired to advance them as genuine targets. In thisarticle, we discuss some of the ways in which model-organism genomics can be used to help achieve thisobjective.

Corresponding author: Woods, D.J. ([email protected]).Available online 26 May 2005

www.sciencedirect.com

How closely do model organisms resemble target

species?

A central assumption of the model-organism approach isthat genetic model organisms produce information that isrelevant to target species. Clearly, some aspects ofparasite biology will not be highly conserved with free-living model organisms (e.g. mechanisms of immuneevasion, and aspects of feeding and digestion). For manyaspects of core biology, however, it is reasonable toanticipate a high degree of functional conservation [1];indeed, the evidence of this is excellent in the cases of drugmode-of-action and development of insecticide resistancein ectoparasites. D. melanogaster, for example, is knockeddown by all classical insecticides, and resistance innatural D. melanogaster populations seems to occur bythe same mechanisms as in related species. WorldwideD. melanogaster resistance to both dichlorodiphenyltri-chloroethane (DDT) and a range of unrelated insecticidesis caused by massive overexpression of a single gene –cyp6g1 – that encodes a cytochrome P450. InD. melanogaster, the overexpression is caused byupstream insertion of an Accord element [2]; in Drosophilasimulans, it is caused by insertion of a Doc element at asimilar location [2]. Broad resistance to major classes ofinsecticide is also associated with another class ofmetabolic enzyme, glutathione-S-transferase, across awide range of insects [3,4]. Similarly, resistance tocyclodiene insecticides is associated with substitution atalanine 302 in the Drosophila g-aminobutyric acid(GABA) receptor gene resistance to dieldrin (rdl); theequivalent residue is mutated in all other resistant insectspecies [5]. The major mechanism of benzimidazoleresistance in nematodes is the Phe-Tyr P200 mutation,which seems to be highly conserved in many parasiticnematode species and C. elegans [6,7]. The mechanisms ofresistance in the other anthelmintic classes, however, areyet to be resolved. More importantly, with respect to drugdiscovery, there is evidence that the major broad-spectrumanthelmintics – benzimidazoles, levamisole and ivermec-tin – function at similar targets in parasitic nematodesand C. elegans [8]. Indeed, most of the understanding ofthe mode of action of these compounds comes from detailedexperimentation in C. elegans followed by experiments intarget parasite species to validate relevance. Although theuse of model organisms is not the only approach to

Page 6

5118

393

30045010

362

5880

15 947

Drosophila melanogaster13 525

Anopheles14 364

Homo sapiens22 313

TRENDS in Parasitology

Figure 1. Venn diagram of the homology among the genomes of Drosophila

melanogaster, Anopheles and Homo sapiens. Figure was prepared using Ensmart

(http://www.ensembl.org/Multi/martview) – part of the Ensembl package – in which

in silico homologies between annotated genomes have been precomputed [24],

making such calculations relatively simple.

Update TRENDS in Parasitology Vol.21 No.7 July 2005 303

antiparasitic drug discovery, experience suggests thatthey offer a promising route to success.

Reverse genetics

A simple approach would be to exploit the genetic workthat has gone into model organisms. In D. melanogaster,for example, tens of thousands of P element (transposon)insertions were generated and screened for lethality [9].One could consider all genes flanking a recessive lethal Pelement insertion to be essential and, therefore, a validtarget. This would identify w20% of the genes in thegenome as targets; several start-up companies haveadopted this approach, although there are drawbacks.First, the gene disrupted by a transposon insertion is oftendifficult to identify. P elements can be lethal even wheninserted many kilobases away from the gene they disrupt.Second, insertion of P elements into the genome is notrandom. Genes will be missed, not because they arenonessential but because they are not ‘hot spots’ for Pelement insertion. For completeness, the D. melanogasterlethality screens must be repeated (which is currentlyunderway) using other transposons such as hobo orpiggyBac that have random (or different) insertion-sitespecificities [10].

The ease with which RNA interference (RNAi) can beapplied to C. elegans has enabled researchers to under-take genome-wide RNAi screens, and bacterial librariesexpressing dsRNA from 86% of the 19 427 predicted genesare now available [11,12]. Genome-wide RNAi screenshave identified 10% of genes as having essential functions.Although RNAi does not incur the problems of bias anddifficulty of gene identification that are described forinsertional mutagenesis in D. melanogaster, there aredisadvantages. Crucially, gene inactivation is often partialand fails to produce a full loss-of-function phenotype,leading to false negatives, particularly for neuronal genesor long-lived transcripts. Notably, 40% of genes withknown mutant phenotypes showed no phenotype in theoriginal genome-wide RNAi-feeding screen [11]. Never-theless, the approach represents a ‘first-pass’ filter forpotential drug targets.

So far, most D. melanogaster genetic screens havefocused on lethality and, similarly, RNAi produces partialor complete loss of function. Although loss-of-functionlethality is a useful validation of a target, manyantiparasitic drugs (e.g. avermectins and neonicotinoids)function by catastrophic activation of a target. Genetically,these drug–target interactions correspond to neitherhypomorphs nor hypermorphs (the latter can be generatedin overexpression screens [13]) but, instead, dominantlethal neomorphs, which are extremely difficult to identifyby screening. Consequently, although genetic screens foressential genes are likely to identify new targets, theapproach is not comprehensive. Furthermore, geneticscreens in isolation are insufficiently selective for identify-ing manageable numbers of candidate genes.

Bioinformatics

A second approach, using comparative genomics, wasrecently described as genomic filtering [14]. Searchingexpressed sequence tag (EST) datasets for genes that are

www.sciencedirect.com

represented widely across the nematode phylum but thatlack vertebrate homologues, a subset of 1200 genes wasselected from the C. elegans genome. This was reduced to100 potential targets using reverse genetic data. It isimpressive that such a small number of candidate genescan be chosen from a genome using simple criteria; this isequally powerful when applied to the D. melanogastergenome. One might predict that a good insecticide targetin the D. melanogaster genome would have close homol-ogues in other completed insect genomes (such as that ofAnopheles) but no close human homologue. How does thislook in practice?

Figure 1 shows that only 3000 genes (20% of theD. melanogaster genome) are common to D. melanogasterand Anopheles but not Homo sapiens. Application of thelethality criterion would probably reduce the number ofcandidates to w600, or 4% of the genome. Additionalcriteria could further refine the list of candidate genes forinvestigation. One might, for example, focus on the centralnervous system (CNS), in which most targets are found.The gene list could be merged with a microarray analysisof the CNS transcriptome, selecting insect-specific genesthat are enriched in the brain. Such sets of candidatescould then be sifted, based on more-empirical, experiment-based criteria. Although this approach would reject someknown targets (nicotinic acetylcholine receptors andb-tubulins are well conserved across phyla), it highlightsa set of potentially exciting genes, O70% of which remainto be named and examined.

Proteomics

Advances in bioinformatics, 2D electrophoresis [including2D difference gel electrophoresis (DIGE)] and massspectrometry enable more-accurate identification of pro-teins from femtomolar quantities of tissue. The

Page 7

Update TRENDS in Parasitology Vol.21 No.7 July 2005304

measurement of conditional expression, abundance andposttranslational modification of proteins is now accessibleto drug-discovery scientists. Does this information fit intothe search for antiparasitic drug targets? Differentialexpression among tissues (CNS, muscle and gut) mighthelp to identifyproteinsenrichedor upregulated inessentialorgans.Measuringchanges in targetproteins,detoxificationpathways and/or compensatory pathways (e.g. redundantparallel pathways with similar function to the targetprotein) after drug treatment has the potential to aid targetidentification [15]. Extremely powerful proteomic resourcesare being developed in model organisms; more than 12 000C. elegans open reading frames have been cloned into theGateway system (http://worfdb.dfci.harvard.edu/) to pro-vide a platform for proteomic studies [16]. More than10 000 of these have now been cloned into the yeast two-hybrid system to enable global ‘interactome mapping’, andthere is active development of genome-wide protein chipsand reverse transfection resources [17]. Many of thesetools have massive potential for the HTS of compoundsthat disrupt particular aspects of protein function and forthe study of lead-compound mode of action.

Chemical genetics

The approaches discussed can be described as ‘biologyfirst’ because target identification is based on biologicalcriteria. One disadvantage is the difficulty in determiningwhether gene function can be disrupted or altered bysmall molecules. An alternative approach is defined as‘chemical first’, whereby antiparasitic molecules are usedas probes to identify drug targets. Targets identified inthis way are, by nature, susceptible to disruption by smallmolecules. One strategy is to isolate model-organismmutants that are resistant to a compound and then tomap the locus genetically. This method has a good trackrecord when used in C. elegans and D. melanogaster,identifying the targets of many important antiparasiticcompounds (e.g. avermectins, benzimidazoles, levamisole,organophosphates and DDT). An example of the power ofthis chemical genetics approach is the identification of twoindependent activities of the drug fluoexetine (Prozace) inC. elegans, which led to the identification of a novel class oftransmembrane receptor [18]. These forward geneticstrategies are powerful but often extremely labourintensive. The availability of full-genome sequences,sophisticated bioinformatic resources and new genetic-mapping strategies makes the chemical genetics approachmuch more rapid and accessible. ‘Snip–SNP’ mapping inC. elegans, for example, involves PCR genotyping ofmapped single nucleotide polymorphisms (SNPs) in theF2 progeny of a cross with bulk segregant analysis [19].This technique enables mapping of a genetic locus byanalysing a single genetic cross with a relatively smallnumber of PCR reactions. This process takes days orweeks compared with months of multiple genetic crossesusing more-traditional strategies.

Functional biology and integrative physiology

It is, perhaps, in functional biology and integrativephysiology that genetic-model organisms have the mostto offer. After identifying a manageable subset of genes,

www.sciencedirect.com

reverse genetics can be used, not as a simple screen forlethality but, instead, to provide the substrate for aphenotypic, ‘functional’ analysis. Following recent callsfor a return to mechanism-based screening [20], it will becrucial to increase interactions between drug-discoveryscientists and biologists who are interested in themechanisms implicated in the action of new genes.These biologists might be interested in tissues(e.g. nervous system and midgut) or processes(e.g. synaptic transmission, detoxification and homeosta-sis), and it is this expertise that is essential to theenterprise. Genetic-model organisms are famous for theirrelative lack of physiological investigation: what has beentermed the ‘phenotype gap’ [21]. As comparative physiol-ogists become aware that the use of functional genomicsrequires function (and, thus, the skills that they have),this imbalance should be addressed rapidly.

The model-organism functional genomics approachoffers a tangible way to identify and validate novelantiparasitic targets. The sophisticated tools andresources available for model organisms will ensurethat they have an important role for many years,maximizing the benefits that will come from sequen-cing parasite genomes {e.g. sequencing of the Brugiamalayi genome is advanced [22], and sequencing of theIxodes scapularis [23] and Haemonchus contortus genomes(http://www.sanger.ac.uk/Projects/H_contortus/) is under-way}. Collaborations between industry and academiawill be crucial for fully exploiting and developing thesenew technologies and resources.

References

1 Couthier, A. et al. (2004) Ectopic expression of a Haemonchuscontortus GATA transcription factor in Caenorhabditis elegans revealsconserved function in spite of extensive sequence divergence. Mol.Biochem. Parasitol. 133, 241–253

2 Schlenke, T.A. and Begun, D.J. (2004) Strong selective sweepassociated with a transposon insertion in Drosophila simulans. Proc.Natl. Acad. Sci. U. S. A. 101, 1626–1631

3 Hemingway, J. et al. (2004) The molecular basis of insecticideresistance in mosquitoes. Insect Biochem. Mol. Biol. 34, 653–655

4 Enayati, A.A. et al. (2005) Insect glutathione transferases andinsecticide resistance. Insect Mol. Biol. 14, 3–8

5 Ffrench-Constant, R.H. et al. (2000) Cyclodiene insecticide resistance:from molecular to population genetics. Annu. Rev. Entomol. 45,449–466

6 Driscoll, M. et al. (1989) Genetic and molecular analysis of aCaenorhabditis elegans b-tubulin that conveys benzimidazole sensi-tivity. J. Cell Biol. 109, 2993–3003

7 Wolstenholme, A.J. et al. (2004) Drug resistance in veterinaryhelminths. Trends Parasitol. 20, 469–476

8 Geary, T.G. and Thompson, D.P. (2001) Caenorhabditis elegans: howgood a model for veterinary parasites? Parasitology 101, 371–386

9 Bellen, H.J. et al. (2004) The BDGP gene disruption project: singletransposon insertions associated with 40% of Drosophila genes.Genetics 167, 761–781

10 Singer, M.A. et al. (2004) A complementary transposon tool kit forDrosophila melanogaster using P and piggyBac. Nat. Genet. 36,283–287

11 Kamath, R.S. et al. (2003) Systematic functional analysis of theCaenorhabditis elegans genome using RNAi. Nature 421, 231–237

12 Vastenhouw, N.L. et al. (2003) A genome-wide screen identifies 27genes involved in transposon silencing in C. elegans. Curr. Biol. 13,1311–1316

13 Rorth, P. (1996) A modular misexpression screen in D. melanogasterdetecting tissue-specific phenotypes. Proc. Natl. Acad. Sci. U. S. A. 93,12418–12422

Page 8

Update TRENDS in Parasitology Vol.21 No.7 July 2005 305

14 McCarter, J.P. (2004) Genomic filtering: an approach to discoveringnovel antiparasitics. Trends Parasitol. 20, 462–468

15 Barrett, J. et al. (2000) Parasite Proteomics. Parasitol. Today 16,400–403

16 Reboul, J. et al. (2003) C. elegans ORFeome version 1.1: experimentalverification of the genome annotation and resource for proteome-scaleprotein expression. Nat. Genet. 34, 35–41

17 Rual, J-F. et al. (2004) ORFeome projects: gateway between genomicsand omics. Curr. Opin. Chem. Biol. 8, 20–25

18 Choy, R.K.M. and Thomas, J.H. (1999) Fluoxetine-resistant mutantsin C. elegans define a novel family of transmembrane proteins. Mol.Cell 4, 143–152

19 Wicks, S.R. et al. (2001) Rapid gene mapping in Caenorhabditiselegans using a high density polymorphism map. Nat. Genet. 28,160–164

Corresponding author: Casiraghi, M. ([email protected]).

www.sciencedirect.com

20 Geary, T.G. et al. (2004) The changing landscape of antiparasitic drugdiscovery for veterinary medicine. Trends Parasitol. 20, 449–455

21 Dow, J.A.T. and Davies, S.A. (2003) Integrative physiology andfunctional genomics of epithelial function in a genetic modelorganism. Physiol. Rev. 83, 687–729

22 Ghedin, E. et al. (2004) First sequenced genome of a parasiticnematode. Trends Parasitol. 20, 151–153

23 Hill, C.A. and Wikel, S.K. The Ixodes scapularis Genome Project: anopportunity for advancing tick research. Trends Parasitol. 21,151–153

24 Kasprzyk, A. et al. (2004) EnsMart: a generic system for fast andflexible access to biological data. Genome Res. 14, 160–169

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.007

The occurrence of strongylid nematodes in theepididymides of wood mice

Maurizio Casiraghi1,2 and Marco Ferraguti3

1DIPAV, Sezione di Patologia Generale e Parassitologia, Universita degli Studi di Milano, via Celoria 10, 20133 Milano, Italy2BTBS, Dipartimento di Biotecnologie e Bioscienze, Universita degli Studi di Milano Bicocca, Piazza della Scienza 2,

20126 Milano, Italy3Dipartimento di Biologia, Sezione di Zoologia e Citologia, Universita degli Studi di Milano, via Celoria 26, 20133 Milano, Italy

The recent discovery of a larval nematode in the

epididymides of free-living wood mice (Apodemus

sylvaticus) suggests a sexual transmission of these

parasites. They have been placed within the bursate

nematodes (order Strongylida) through 18S rDNA

analysis, suggesting that they are undetermined meta-

strongyloid nematodes. The possibility that these

parasites are transmitted sexually opens an intriguing

field of research because sexually transmitted metazoan

parasites are known to occur mainly in invertebrates,

whereas in vertebrates sexually transmitted parasites

are usually microparasites such as viruses, bacteria

and protozoa.

State of the art

At the end of the 18th century, Lazzaro Spallanzani – oneof the founders of experimental biology – observed inmammalian semen what he called ‘little spermaticanimals’. Spallanzani proposed that they were a particu-lar kind of parasite that lived and reproduced within theirhosts and were capable of transmission to offspring [1].Later, the ‘little spermatic animals’ were recognized tobe spermatozoa.

Unexpectedly, more than 200 years later, Clarke et al.[2] observed the larval stages of an unknown nematode inthe epididymal fluid of wood mice (Apodemus sylvaticus)from a population collected in Oxfordshire (UK), providing

a potential case of sexual transmission of a nematode in amammalian host. It is unsurprising that these nematodeswere not reported previously in A. sylvaticus because,despite being a well-studied mammal, the prevalence ofthe nematodes in A. sylvaticus was less than 20% (ninehosts infected out of 46 examined) [2].

Phylogenetic positioning using 18S rDNA

It was not possible to identify the larval nematodes;however, they were placed within the bursate nematodes(order Strongylida) through 18S rDNA sequencing andcomparison with available 18S rDNA nematode sequences[2,3]. Distance and parsimony analyses indicated that thenematodes nested among members of the superfamilyMetastrongyloidea and clustered in a clade with twospecies of the genera Angiostrongylus and Filaroidesmartis, but with low bootstrap support. Analyses alsoindicated that the nematode was unrelated to Heligmoso-moides polygyrus, a trichostrongyloid nematode ofA. sylvaticus, and was distinct from but clearly relatedto the metastrongyloid nematode of wood mice Angio-strongylus dujardini. These results support the possibilitythat the unidentified nematode is a metastrongyloidnematode parasite of wood mice.

Possible routes of transmission

The presence of nematode larval stages in the epididy-mides of wood mice suggests that the nematodes might betransmitted to females at ejaculation. Are they parasitesand is this their mode of transmission? It is not clear, andfour possible transmission routes were presented by

Page 9

Update TRENDS in Parasitology Vol.21 No.7 July 2005306

Clarke et al. [2]: (i) sexual transmission of larvae whoseadult forms live in other tissues in wood mice;(ii) supplementary sexual transmission of larvae that, asfree-living forms, enter male hosts by oral or percutaneousroute and migrate to the reproductive tract for venerealtransmission; (iii) a paratenic cycle in which a predator ofwood mice is the definitive host and in which larvalmigration to the epididymis occurs, possibly for evasion ofthe host immune response; and (iv) a dead-end cycle inwhich wood mice are not the natural host of the parasite.

In addition, it must be emphasized that the orderStrongylida encompasses parasites that are known forcomplex biological cycles, showing both free-living andparasitic behaviours, such as Strongyloides stercoralis[4,5]. These complex life cycles include the nematodeentering the host, and one possible route is penetrationthrough the skin of the host. One might speculate that theunidentified nematodes found in wood mice penetrateddirectly through the genital opening, eventually reachingthe epididymis. The large number of nematodes found inthe epididymides of wood mice by Clark et al. [2] arguesagainst penetration ‘by chance’ through the genitalopening and suggests an extremely efficient strategy tocolonize this part of the genital tract. However, based oncurrent knowledge of the life cycles of bursate nematodes[5], efficient penetration through the genital openingseems unlikely. A ‘descendent’ route of infection(i.e. from internal body compartments to the epididymis)seems more likely than an ‘ascendant’ route (i.e. from theexternal environment to the epididymis).

Sexually transmitted parasites

Lockhart et al. [6] reviewed in detail the sexualtransmission of microparasites and macroparasites.Most sexually transmitted diseases are caused bymicroparasites such as viruses, bacteria and protozoa.Sexual transmission of metazoan parasites is far lesscommon and occurs primarily in invertebrates. Sexualtransmission of nematodes has been investigated in detailin insects [7]; for example, Noctuidonema guyanense is anectoparasitic nematode that is transmitted during themating of a moth of the Noctuidae family [8]. Thesenematodes are ectoparasites that reside on the interseg-mental membranes of the terminal part of the abdomen,and their transmission is analogous to that of pubic liceand scabies in humans.

There is only a small number of cases of sexualtransmission of endoparasitic nematodes. The infectivestages of Mehdinema alii, a parasite of the alimentarycanal of the cricket Gryllus sigillatus, occur only in thegenital chamber and are transmitted venereally duringcopulation [9]. Similarly, sexual transmission of a para-sitic nematode has been described in molluscs: forexample, Nemhelix bakeri, a nematode that reproducesin the genital apparatus of the snail Helix aspera [10].

Only three nematodes have been reported to betransmitted sexually in vertebrate hosts, two of which –the pinworm Enterobious vermicularis and S. stercoralis –are transmitted as a consequence of heterosexual or, moreobviously, homosexual encounters [11–13]. However, thesecases are probably not the rule and, instead, can be

www.sciencedirect.com

considered ‘evolutionary accidents’. Other nematodeshave been found in the mammalian urogenital system,such as lymphatic filarial nematodes (Wuchereria ban-crofti and Brugia malayi in humans) and Trichomosoidescrassicauda (Trichuridae) of rats. These nematodes havenot been reported as being transmitted sexually, althoughthe microfilariae of filarial nematodes and the eggs ofT. crassicauda have been found in the urine of the host [5].

Because sexually transmitted diseases affect thehuman imagination in an extremely personal way, it isnot surprising that they have become the subject ofimportant medical and veterinary research. Furthermore,parasites that are transmitted during copulation mighthave a key role in the biology, ecology and evolution oftheir hosts, affecting, for example, mating behaviour andchoice of mate, in addition to adapting to the reproductivephysiology of the host. In other words, sexually trans-mitted parasites, which rely on host mating and reproduc-tion for their transmission, must be intimately adapted tothe reproductive physiology and behaviour of their host,and are likely to influence host reproductive physiologyand behaviour [14–18], even affecting the evolution ofcomplex traits such as social behaviour and eusocialityin insects [19].

Although sexual transmission is typically linked withdisease and parasitism, it has also been reported fornon-parasitic symbioses: for example, an unidentifiednematode of the genus Bursaphelenchus that is sexuallytransmitted during copulation among beetles of the familyNitidulidae [20]. Here, the relationship between beetleand nematode is phoretic because Bursaphelenchus is afree-living mycophagous organism.

The results reported by Clarke et al. [2] could representthe first step towards the description of another intriguingcase of sexual transmission of a parasitic nematode in ananimal group in which this phenomenon seems, thus far,to occur only by accident. It is hoped that these results willlead to further investigation of the sexual transmission ofparasitic nematodes in mammals.

References

1 Rostand, J., ed. (1951) Les origines de la biologie experimentale etl’Abbe Spallanzani, Fasquelle Editeurs, Paris

2 Clarke, J.R. et al. (2004) Sexual transmission of a nematode parasite ofwood mice (Apodemus sylvaticus)? Parasitology 128, 561–568

3 Blaxter, M.L. et al. (1998) A molecular evolutionary framework for thephylum Nematoda. Nature 392, 71–75

4 Georgi, J.R. (1982) Strongyloidiasis. In CRC Handbook Series onZoonoses: Parasitic Zoonoses (Vol. 2), pp. 257–267, CRC PressReprints

5 Anderson, R.C., ed. (2000) Nematode Parasites of Vertebrates: TheirDevelopment and Transmission (2nd edn), CABI Publishing

6 Lockhart, A.B. et al. (1996) Sexually transmitted diseases in animals:ecological and evolutionary implications. Biol. Rev. Camb. Philos. Soc.71, 415–471

7 Knell, R.J. and Webberley, K.M. (2004) Sexually transmitted diseasesof insects: distribution, evolution, ecology and host behaviour. Biol.Rev. Camb. Philos. Soc. 79, 557–581

8 Marti, O.G. et al. (1990) Pathological effects of an ectoparasiticnematode Noctuidonema guyanense (Nematoda: Aphelenchoididae)on adults of the fall armyworm (Lepidoptera: Noctuidae). Ann.Entomol. Soc. Am. 83, 956–960

9 Luong, L.T. et al. (2000) Venereal worms: sexually transmittednematodes in the decorated cricket. J. Parasitol. 86, 471–477

Page 10

Update TRENDS in Parasitology Vol.21 No.7 July 2005 307

10 Morand, S. (1993) Sexual transmission of a nematode: study of amodel. Oikos 66, 48–54

11 Waugh, M.A. (1972) Threadworm infestation in homosexuals. Trans.St Johns Hosp. Dermatol. Soc. 58, 224–225

12 Waugh, M.A. (1974) Sexual transmission of intestinal parasites. Brit.J. Venereal Dis. 50, 157–158

13 Sorvillo, F. et al. (1983) Sexual transmission of Strongyloidesstercoralis among homosexual men. Br. J. Vener. Dis. 59, 342

14 Abbot, P. and Dill, L.M. (2001) Sexually transmitted parasites andsexual selection in the milkweed leaf beetle, Labidomera clivicollis.Oikos 92, 91–100

15 Hurst, G.D.D. et al. (1995) Sexually transmitted disease in apromiscuous insect, Adalia bipunctata. Ecol. Entomol. 20, 230–236

16 Webberley, K.M. and Hurst, G.D.D. (2002) The effect of aggregativeoverwintering on an insect sexually transmitted disease system.J. Parasitol. 88, 707–712

Corresponding author: Ramaiah, K.D. ([email protected]).

www.sciencedirect.com

17 Webberley, K.M. et al. (2004) Host reproduction and a sexuallytransmitted disease: causes and consequences of Coccipolipushippodamiae distribution on coccinellid beetles. J. Anim. Ecol. 73,1–10

18 Lien, T. et al. (2005) Sexually transmitted nematodes affectspermatophylax production in the cricket, Gryllodes sigillatus.Behav. Ecol. 16, 153–158

19 O’Donnell, S. (1997) How parasites can promote the expression ofsocial behaviour in their hosts. Proc. R. Soc. Lond. B. Biol. Sci. 264,689–694

20 Giblin, R.M. (1985) Association of Bursaphelenchus sp. (Nematoda:Aphelenchoididae) with nitidulid beetles (Coleoptera: Nitidulidae).Rev. Nematol. 8, 369–375

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.014

Letter

Preventing confusion about side effects in a campaignto eliminate lymphatic filariasis

Kapa D. Ramaiah, Rengachari Ravi and Pradeep K. Das

Vector Control Research Centre (Indian Council of Medical Research), Medical Complex, Indira Nagar, Pondicherry 605 006, India

Of the 83 countries that are endemic for lymphaticfilariasis (LF), 32 have already initiated mass adminis-tration of antifilarial drugs to eliminate the disease [1]. Asmany as 1.1 billion people throughout the countriesendemic for LF are expected to receive antifilarial drugs.In India alone, w450 million are expected to receive 1.1billion to 1.5 billion diethylcarbamazine (DEC) tablets [2],making it the largest ever drug-distribution program.

Studies in animals and humans have shown that,relatively speaking, DEC is an extremely safe drug [3].Although DEC is known to cause some transient sideeffects such as drowsiness, nausea, vomiting, gastricupset, dizziness, fever, headache and body pains [4,5] insome patients, these reactions are easily manageable.In 1973, Hawking stated that ‘although hundreds ofthousands of people have been treated, no case of deathproved to be due to DEC has been reported. The untowardreactions in man are never dangerous [3]’. Unfortunately,however, severe side effects and some deaths that occurredin treated populations in two Indian states in 2004 wereattributed to administration of DEC (6 mg kgK1 body-weight). Following mass drug administration (MDA) forthe first time in June 2004, many people in the Gulbargaand Bidar districts of Karnataka state (India) reportedside effects and sought treatment in rural hospitals. In thetown of Thiruchirapally in Tamil Nadu (India) – where theseventh annual MDA was implemented in September2004 – 65 people aged 60–90, most of them with otherserious illnesses, sought treatment in government andprivate hospitals for symptoms of giddiness, vomitingand diarrhea that were thought to be caused by DEC.Four patients in Thiruchirapally and one each in the

Thiruvannamalai and Pudukottai districts (India) diedwithin two to seven days of consuming DEC. On the basisof its safety record [3], DEC was not thought by the healthauthorities to be the cause of mortality. However, becausethe deaths occurred within a few days of consuming DECand after side effects, some community members attrib-uted the deaths to DEC. These incidents coincided withthe extension of the program to 170 more districts(a population of 353 million) in 2004, from only 31 districtsin 2003. Such a massive expansion led to several otherlogistic and operational problems [6].

The deaths in Thiruvannamalai and Pudukottai wereinvestigated by the health authorities and were attributedto fits and encephalitis, respectively. However, the exactreasons for the deaths that occurred following MDA inThiruchirapally are not yet clear; postmortem reports forthose who died there are being examined by state healthauthorities to ascertain the causes of death. Samples ofDEC tablets – supplied mostly by small pharmaceuticalfirms – distributed in Thiruchirapally are also beinganalyzed for quality. Nevertheless, in September 2004, theevents in Thiruchirapally were reported extensively bythe media, leading to adverse public opinion about theMDA program.

It has been established that post-MDA adversereactions in treated communities are an impediment toachieving higher compliance with treatment and theLF-elimination campaign [2,7]. Incidents such as those inThiruchirapally will only hamper the campaign further.Therefore, remedial measures – particularly the trainingof drug distributors and the health education of com-munities – that highlight the beneficial effects of DECare necessary. To counter the shortage of personnel andto overcome the poor compliance with treatment [7],

Page 11

Update TRENDS in Parasitology Vol.21 No.7 July 2005308

social-welfare staff and village volunteers, alongsidehealth personnel, were also involved in drug distributionin India. Tamil Nadu alone deploys 120 000 drugdistributors. The involvement of a large number of drugdistributors must be matched by high-quality training –through competent trainers and adequate fund allocation– about exclusion criteria, pharmacology of the drug,dosage and side effects. More importantly, MDA should bepreceded by extensive health education because people areapprehensive about the side effects and the need fortreatment of the entire population [7]. At present, theeducation material is limited and often reaches healthcenters only a few days before MDA is carried out,resulting in poor use and effect of the material. Achallenging task in health education is to make peopleaware of not only the benefits but also the adverse effectsof a drug. Currently, under the MDA program in India,only US$0.005 per capita is allocated for health education,which is grossly inadequate.

Upscaling and sustenance of MDA for the requiredduration of more than six years [8,9] necessitates someadditional safety measures that include: (i) drug procure-ment only from companies that follow good manufacturingpractices and that are approved by the World HealthOrganization (WHO; http://www.who.int/en/) to thwartthe entry of substandard drugs; (ii) confidence buildingin communities through follow-up visits and promptmanagement of side effects by medical officers and drugdistributors; and (iii) exclusion from MDA for extremelyold and seriously ill people. Inadequate preparation and

Corresponding author: McConkey, G.A. ([email protected]).

www.sciencedirect.com

implementation of MDA might earn a good strategy a badname and endanger the future of MDA.

References

1 World Health Organization (2002) Global Programme to EliminateLymphatic Filariasis: Annual Report on Lymphatic Filariasis, WHO

2 Das, P.K. et al. (2001) Towards elimination of lymphatic filariasis inIndia. Trends Parasitol. 10, 457–460

3 Hawking, F. (1978) Diethylcarbamazine. A Review of the Literaturewith Special Reference to its Pharmacodynamics, Toxicity, and Use inthe Therapy of Onchocerciasis and Other Filarial Infections, WorldHealth Organization (WHO/Oncho/78.142)

4 Sundaram, R.M. et al. (1974) Studies on bancroftian filariasis controlwith diethylcarbamazine. Frequency and nature of drug reactions.J. Com. Dis 6, 290–300

5 Dreyer, G. et al. (1994) Tolerance of diethylcarbamazine by micro-filaraemic and amicrofilaraemic individuals in an endemic area ofbancroftian filariasis, Recife, Bazil. Trans. R. Soc. Trop. Med. Hyg. 88,232–236

6 Sabesan, S. et al. (2005) Elimination of lymphatic filariasis in India.Lancet Infect. Dis. 5, 4–5

7 Ramaiah, K.D. et al. (2000) A programme to eliminate lymphaticfilariasis in Tamil Nadu state, India: compliance with annual singledose DEC mass treatment and some related operational aspects. Trop.Med. Int. Health 5, 842–847

8 Ramaiah, K.D. et al. (2003) The impact of six rounds of single dose massadministration of diethylcarbamazine or ivermectin on the trans-mission of Wuchereria bancrofti by Culex quinque fasciatus and itsimplications for lymphatic filariasis elimination. Trop. Med. Int. Health8, 1082–1092

9 Stolk, W.A. et al. (2003) Prospects for elimination of bancroftianfilariasis by mass drug treatment in Pondicherry, India: a simulationstudy. J. Infect. Dis 188, 1371–1381

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.015

Book Reviews

Not just for the birdsAvian Malaria Parasites and Other Haemosporidia by Gediminas Valkiunas. CRC Press, 2004. $238.00 (932 pages) ISBN 0415300975

Glenn A. McConkey

School of Biology, Faculty of Biological Sciences, Miall Building, University of Leeds, Leeds, UK, LS2 9JT

Many people are surprised when theydiscover that there is malaria in birds orthat it is found in all the different climatesthroughout the world, yet Avian MalariaParasites and Other Haemosporidia fills932 pages on this topic. Not sinceGarnham’s classic Malaria Parasites andOther Haemosporidia (1966) has such acomplete work about the description of

malaria species been written and illus-

trated. Avian malaria species provided much of the basis ofthe understanding and treatment of malaria, from SirRonald Ross’ demonstration of parasite transmission toantimalarial discovery from testing compounds against themodel Plasmodium gallinaceum. This book not onlyencompasses the findings of these historical studies but

also provides a modern interpretation and includes manystudies from rare Russian sources that have not previouslybeen published in English.

The book begins with a general description of life cycles,ultrastructural characteristics, pathogenicity, distri-bution and methods of study, followed by a comprehensivecatalogue of species of Haemosporidia. It is not onlydescriptive but also filled with analysis and interpre-tation, although the wording is sometimes verbose. Thereis a helpful chapter entitled ‘Certain peculiarities of theecological study of bird parasites’ that notes importantdifficulties, including migratory behaviour, sampling biasand blood-film identification of infections. The effect ofmixed infections such as aspergillosis is one area that isnot considered; this could provide scope for the next volume.

Several applications of the study of avian malaria andother Haemosporidia are described in Avian MalariaParasites and Other Haemosporidia. In one chapter

Page 12

Update TRENDS in Parasitology Vol.21 No.7 July 2005 309

(certainly, the numbering of chapters would have helped),the use of Haemosporidia to test the Hamilton-Zukhypothesis (that plumage colour affects mate choice andindicates parasite resistance) is assessed. Another chapterdetails how evolutionary biology studies have focused onHaemosporidia. The author notes the advantages of usingHaemosporidia to study parasites in the environment andthe lack of investigations in this area. He discusses brieflythe evolution of avian malaria parasites that might havepassed from reptiles to birds and served as a source ofmammalian species of malaria parasite. Examples of thesevere ecological ramifications of this group of parasites,including the introduction of avian malaria into an area(e.g. the Hawaiian Islands) and the introduction ofsusceptible birds into endemic areas (e.g. penguins inzoos), are topics covered in another chapter.

Following the chapters about generalized characters,there are individual descriptions of 206 species of avianHaemosporidia, including details of vertebrate hosts,vectors, geographical distribution, pathogenicity anddevelopmental stages. This exhaustive list of knownavian Haemosporidia is comprehensive. There are evenspecies identified as recently as 2002. The book is packedwith hundreds of black and white illustrations of parasitesat different stages. These would have benefited fromcolour: a point that is accentuated by the three colourplates of stained parasites at the end of the book. The

Corresponding author: Pilon, M. ([email protected]).

www.sciencedirect.com

descriptions are pithy and well considered. There are alsoinsights into the etymology (e.g. Leucocytozoon lovatibeing named in honour of Lord Lovat, who was chairmanof the Scottish Grouse Commission in the early 20thcentury). Each description is followed by author commentsthat propose experiments to clarify systematics, identifygaps in experimental investigations or describe difficultiesin performing investigations.

There is, however, a notable lack of molecular data. Withthe recent description of molecular markers and ongoingP. gallinaceumgenome sequencing, this book might serve asa basis for molecular phylogeny. As Valkiunas suggests, itmight be time to reinvestigate avian malaria as a model ofthe ecology and evolution of parasites; indeed, it could be amodel for vaccine trials. This book should serve as a helpfultextbook for researchers, from the experimental parasitol-ogist to the ecologist interested in host–parasite inter-actions. Many specifications for a model of the effects ofparasites on populations are fulfilled by avian Haemospor-idia, and Valkiunas hopes that his book will encouragefuture investigations in this area. Avian Malaria Parasitesand Other Haemosporidia should benefit anyone who isstudying malaria or who is interested in the systematics ofthese parasitic protists.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.006

The amazing world of nematodesNematology: Advances and Perspectives (Vols 1 and 2) edited by Zhongxiaio Chen, Senyu Chen and Donald W. Dickson.

CABI Publishing, 2003 and 2004. £85.00 and £75.00 (656 and 608 pages) ISBN 0851996450/0851996469

Marc Pilon

Chalmers University, Lundberg Laboratory, Medicinaregatan 9C, Box 462, Goteborg S-405 30, Sweden

Nematodesare the mostabundantmetazo-ans on earth. They crawl through the soil,preying on bacteria and each other. Theylive in the bodies of vertebrates and inthe roots, shoots or seeds of plants. Theylive in Antarctic ice, in high-temperatureenvironments and at the bottom of theoceans. The nematode Caenorhabditiselegans is one of the best-understoodorganisms and is a favorite model fordevelopmental geneticists and scientistsstudying human disease. Unfortunately,the diversity of nematodes and theirbehaviors, life cycles and physiology arerarely part of biology programs, and itseems that fewer books are publishedabout the subject each decade. I was glad

of the opportunity to read the two volumes

ofNematology:AdvancesandPerspectives. With 25chapterswritten authoritatively, there are probably no other twobooks that, together, cover such a breadth of topics onnematodes. Some chapters are judiciously well illustrated,such as the introductory chapter about the history ofnematology, the comprehensive chapter about C. elegansdevelopmental biology and the chapters about fungal andbacterial enemies of nematodes. However, other chapterscould have been illustrated better; in particular, the chapterabout plant diseases caused by nematodes has no images,and it is difficult for the noninitiated to visualize theappearance of the nematodes at various stages of their lifecycles and what the disease symptoms actually look like.

Furthermore, despite the subtitle Advances and Per-spectives, it is disappointing that no chapter covers themolecular biology of nematodes, leaving many questionsunanswered. To what degree are nematodes similar toeach other in their gene composition? Why does the onlynematode for which a genome sequence is available(C. elegans) have 20 000 genes, whereas fruit flies havemerely 14 000 genes and humans have 25 000 to 30 000

Page 13

Update TRENDS in Parasitology Vol.21 No.7 July 2005310

genes?What do the genome sequences of nematodes, partialor complete, say about their evolutionary relationshipswithother phyla? Other serious deficiencies include: the absenceof a proper chapter about nematode physiology, the absenceof a chapter about the molecular biology of plant resistanceto nematodes and the absence of a chapter about nematodesthat are parasites of animals. Advances in these areasmusthave occurred during the past 50 years.

Most chapters provide an excellent introduction totheir topics, often with an historical account of the field.Especially interesting chapters concern the interactions ofnematodes with other organisms: nematodes as virusvectors (chapter 14) and biological control of nematodes byfungal (chapter 20) or bacterial antagonists (chapter 21).The sense that nematodes belong to a complex ecologicalweb is convincingly and instructively conveyed in thesechapters. Viruses take advantage of nematode proteins toadhere to the pharyngeal cuticle and await opportunities tomove to new plant sites and hosts. Fungi trap theirnematode prey using sticky knobs, nets or lassoes, andevenpenetrate themusingcannon-likeguncells that launchcuticle-penetrating missiles. Although they have enemies,nematodes that feast on insects and other invertebrates alsoexist. Thus, in chapter 22 (Biological control of insects andother invertebrates), one learns of efforts in the 1970s tocommercialize the use of nematodes (Romanomermisculicivorax) to combat the malaria vector Anopheles, inaddition to the possible use of other nematode species asbiocontrol agents against the wood wasp and pest slugs.

The behavior of nematodes in their natural environ-ment is addressed thoroughly. In particular, the chapterabout nematode behavior and migration through soil andhost tissue provides an excellent review of the field fromboth a theoretical and an experimental point of view.Apparently, the ‘crawling on its side’ behavior observed inPetri dishes for many nematodes is an artifact of theculture condition.Althoughsomenematodescanapparently

Corresponding author: Engwerda, C. ([email protected]).

www.sciencedirect.com

jump (dauer larva of Steinernema spp.), most speciesnavigate by crawling and pushing themselves throughwater-filled interstices. They can also navigate alongthermal gradients with an awareness of temperatures aslow as 0.0018C, as documented for Meloidogyne incognita.Root-seeking, dispersal, repulsion from possible predators(cockroach feces are a repellent) and attraction to theopposite sex are behaviors that illustrate the acuteawareness of nematodes of their chemical surroundings,and this chapter does a good job of describing theexperiments that are gradually increasing the understand-ing of these behaviors.

Although no chapter is devoted to nematode evolutionor diversity in general (a well-illustrated chapter aboutthe taxonomy of insect parasitic nematodes seems out ofplace within this collection), there is an excellent sectionabout marine nematode biodiversity (chapter nine). In it,we learn that nematodes are everywhere (apparently, theywere thawed alive from Antarctic ice by the Shackletonexpedition!), that 80–90% of all metazoans are nematodes(it is estimated that there are currently 1!1019 nema-todes alive on earth) and that there are between 1!106

and 1!108 distinct species of nematode.A concluding note is that the broad nature of the

overviews contained within these books means that theyare not a very good starting point for anyone interested in aparticular plant disease or nematode species. Instead, theywill be useful to those interested in primers to individualresearch areas or topics, ranging from nematode manage-ment (including crop rotation, biocontrol, pesticide andirradiation) to anatomy, ecology and even the history ofnematology itself. Weighing in at more than 1200 pages,the two volumes of Nematology: Advances and Perspec-tives provided a hearty meal for this bookworm.

1471-4922/$ - see front matter Q 2005 Published by Elsevier Ltd.

doi:10.1016/j.pt.2005.05.002

Malaria immunology: still much more to understandMalaria Immunology (2nd edn) edited by Peter Perlmann and Marita Troye-Blomberg. Karger, 2002. £130.77 (412 pages)

ISBN 3805573766

Christian Engwerda

Immunology and Infection Laboratory, Queensland Institute of Medical Research, 300 Herston Road, Herston 4006, Australia

Malaria remains a major health problemin many of the poorest nations on earth.Although it has been eliminated frommost developed countries by variouscontrol measures and improvements inhealth and hygiene, the political, econ-omic and moral will of the global commu-nity to make such changes in the poorest

parts of the world is lacking. Furthermore, naturaldisasters such as the recent tsunamis in Southeast Asia,in addition to manmade calamities, can lead to a rapidescalation in the incidence of malaria. Therefore, the besthopes for controlling or eliminating this disease remainthe development and implementation of cheap, yet safeand effective vaccines and drugs. Of course, after atreatment becomes available, further challenges forfinancing and applying new products in the field remain.

During the past decade, many governments, in additionto charitable and philanthropic organizations, have

Page 14

Update TRENDS in Parasitology Vol.21 No.7 July 2005 311

invested hundreds of millions of dollars into researchaimed at improving the understanding of malaria and howthis disease might be controlled. One of the most dynamicand rapidly advancing areas of medical research is thestudy of immunology. Important recent discoveriesinclude the way that pathogens are first recognized byprimitive pattern-recognition receptors on host cells andhow signals from these molecules link innate and adaptiveimmune responses. Furthermore, the key cells and mol-ecules necessary for the generation of effective adaptiveimmunity by vaccines have been identified, as have thespecific tissue sites at which key cellular and molecularinteractions occur. This knowledge has had amajor impacton malaria research.

The second (revised and enlarged) edition of MalariaImmunology reviews the progress made in recent yearsin the understanding of host immune responses to thePlasmodium spp. parasites that cause malaria. The bookis divided into four sections. The first comprises threechapters about the interactions among malaria parasites,vertebrate hosts and mosquito hosts; the second includessix chapters about parasite antigens that can stimulatehost immune responses; the third contains three chaptersabout mechanisms of host immunity to malaria parasitesand how they are regulated by genetic and environmentalfactors; and the fourth section contains seven chaptersabout malaria vaccine development. All of the chapterswere written by leading researchers in their respectivefields. The final chapter – about malaria vaccine trials,written by Brian Greenwood and Pedro Alonso – is

Corresponding author: Matthews, K. ([email protected]).

www.sciencedirect.com

particularly interesting and highly relevant for anyoneinvolved in malaria vaccine research, especially thoseplanning clinical trials.

AlthoughMalaria Immunology (2nd edn) was publishedin 2002, most of the information that it contains is stillextremely relevant to scientists working in malariaresearch. Obviously, however, some of the recent clinical-trial results, advances inusingattenuatedparasite vaccinesand identification of Toll-like receptors that recognizeparasite molecules are not included. This book wascompleted before publication of the genome of Plasmodiumfalciparum (the major cause of severe malaria in humans).However, thisdoesnotdetract fromtherelevanceofMalariaImmunology (2nd edn). Instead, it could serve as a solidreference source for investigators involved in researchprogrammes that tap into the huge amount of datagenerated by annotating the genome of the parasite. Inparticular, it could help to focus genome research projectsinto areas that will foster the development of rational andeffective vaccines or drugs most effectively.

In summary,Malaria Immunology (2nd edn) would be avaluable addition to any laboratory involved in malariaresearch. In particular, students or researchers enteringthe field of malaria research will find the book extremelyuseful because it will provide them with a sound knowl-edge of where the field is now and where it could be withinthe next decade.

1471-4922/$ - see front matter Q 2005 Published by Elsevier Ltd.

doi:10.1016/j.pt.2005.05.005

Parasites revisitedThe Trypanosomiases edited by Ian Maudlin, Peter Holmes and Michael Miles. CABI Publishing, 2004. £99.50 (614 pages)

ISBN 085199475X

Keith Matthews

Institute of Immunology and Infection Research, School of Biological Sciences, Ashworth Laboratories, University of Edinburgh,

West Mains Road, Edinburgh, UK, EH9 3JT

The Trypanosomiases is an update of theoriginal classic text The African Trypa-nosomiases, which was edited by H.W.Mulligan and published more than 30years ago. It has broadened its remitsomewhat to include the South Americantrypanosomes and their vectors.

The coverage of topics in the book is

extremely wide ranging. In the first part, it focuses on thebiology and molecular biology of the parasites themselves.Thereafter, it examines the vector biology of both Africanand South American trypansomiases, combining thesetopics with a focus on diagnosis and epidemiology. Inthe third theme of the book, the diseases caused by the

parasites and their impact on both the health and theeconomy of afflicted regions are addressed. Finally,The Trypanosomiases examines treatment and controlstrategies for both the parasites (through drug treatment,analysis of disease-resistant reservoirs and transmissionmanagement) and the vector.

Overall, this is a worthwhile book. There are few placesin which almost all of the aspects of trypanosome biology,disease and transmission are gathered together. Thisis important, particularly at this time; as the genomesequences of parasites, vectors and hosts become known,the challenge is to place the rapid expansion of molecularinformation into the context of the biology of theorganisms concerned. In this way, vulnerable points inthe biology or transmission of these parasites might beidentified and targeted. As a reference source for

Page 15

Update TRENDS in Parasitology Vol.21 No.7 July 2005312

researchers of individual aspects of trypanosome biology,disease or transmission, this book, therefore, provides avaluable framework and presents the ‘big picture’ in aconvenient single text.

The focus of The Trypanosomiases is on the diseasesand their vectors, rather than the parasites themselves.Therefore, only a small proportion of the book deals withthe huge amount of information known about the mol-ecular cell biology of the trypanosomes. In some ways, thisis a shame; Mulligan’s book was written almost before theadvent of molecular technology, and there have been hugeadvances in understanding trypanosome biology in theinterim. For example, it would be valuable to haveinformation about the many advances in cell morphologyand division, the unusual mechanisms of gene expressionand the unique biochemistry of these organisms. Inaddition to their intrinsic interest, such areas mightpoint to new strategies for control or therapy and,therefore, their inclusion would have been worthwhile.Instead, the focus is more on the vectors and parasiteepidemiology, perhaps reflecting the editors’ interests butalso, rightly, emphasizing the areas that are most likely toachieve disease control in the short term. These areas arealso reviewed less frequently in the literature and,therefore, bringing them together in The Trypanosomiasesavoids the need for frustrating hours spent searchingPubMed (http://www.ncbi.nih.gov/entrez/query.fcgi) ororiginal research publications for a particular piece ofinformation.

The information is not as up to date as a recent reviewarticle (few of the cited references are from the 21st

Corresponding author: Norstrom, M. ([email protected]).

www.sciencedirect.com

century), although this is inevitable in a book of this kind.However, I was a little disappointed with the number orcontent of figures in each article – in many, there areeither no illustrations or very few, and often these arederived from source data. It might have been useful inseveral articles to provide illustrations to give an overviewof a particular area or concept. There is a series of colourplates within the book, but these are scarce and derivedfrom only a few of the topic areas. These are minor gripes,however, and probably reflect the need to balance the costof the book (at nearly £100, it is within the budget of mostlaboratories but well beyond that of research students).

I found most chapters to be well written and accessibleto the nonspecialist, and mistakes were infrequent. Thebreadth of topics covered and their collection into thissingle resource provide an easy route for finding infor-mation about many aspects of the trypansomiases. This isof value to postgraduate and postdoctoral workersinterested in these parasites in a broader context thantheir specific research project, and to heads of laboratoriesinterested in the wider application or direction of aresearch theme. Overall, this makes The Trypanosomiasesan invaluable work of reference that I hope will havethe same longevity as Mulligan’s original. However,the pace of discovery and the potential offered by thecurrent availability of genome information are breath-taking, and I expect that we will not have to wait 30 yearsfor a revision.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.003

Spatial veterinary epidemiology – an expanding scienceGIS and Spatial Analysis in Veterinary Science edited by Peter A. Durr and Anthony C. Gatrell. CABI Publishing, 2004. £60.00

(303 pages) ISBN 0 85 199 634 5

Madelaine Norstrom

National Veterinary Institute, Norwegian Zoonosis Centre, PO Box 8156 Dep., N-0033 Oslo, Norway

The field of geographical informationsystems (GIS) and spatial analysis isexpanding within the field of veterinaryscience. The new possibilities that thesemethods provide for exploratory analysisand advanced modelling of disease spreadwill probably have a large influence onthe development of veterinary science

and, especially, epidemiology in the

future. It is important to gather the existing knowledgeand use of these methods within veterinary science. GISand Spatial Analysis in Veterinary Science should, there-fore, be welcomed by everyone who has been looking for atextbook that addresses the use of GIS and spatial

analyses in this field. The book contains reviews of GISapplications in several different research and workingfields within veterinary science. The different chaptershave been written by well-known scientists in theirrespective working or research field, and several appli-cations are presented and discussed. This gives the readera good overview of the current GIS applications withinveterinary science. All examples of applications have beenpublished previously in scientific articles, a fact thatcontributes to increased quality.

GIS and Spatial Analysis in Veterinary Science isdivided into three parts: (i) introduction and overview; (ii)the wider context; and (iii) applications within differentareas of veterinary science. The introduction and overviewsection is comprehensive, and the first chapter is easy tounderstand and follow, even for someone new to this area

Page 16

Update TRENDS in Parasitology Vol.21 No.7 July 2005 313

of science. The reader is guided into the GIS world witheasy-to-understand examples. In the second chapter, anew science – spatial epidemiology – is introduced and theargument for it being a separate discipline is presented.This might not be obvious to all scientists, and furtherdiscussions about this definition should be expected.Chapters three and four might be difficult for someonewithout advanced statistical skills to understand. Theformer draws parallels with human health issues,whereas the latter deals with spatial statistical methodsand future directions. The most useful information is thatspatial analysis is a discipline in which other statisticalpackages carry out the desired analyses, whereas theresults might be linked to a GIS for visualization. I wouldsuggest that readers who are not familiar with advancedstatistical methods skip these chapters initially and readthe chapters in part three that are relevant to theirresearch or working fields of interest. This part discussesdifferent applications in veterinary science, including usein animal health, veterinary parasitology, modelling the

Corresponding author: Hide, G. ([email protected]).

www.sciencedirect.com

spread of animal diseases, use in companion-animalepidemiology, epidemic disease response and wildlifediseases. The final part of the book contains an appendixthat focuses on some basic ideas about, for example,purchasing a GIS package, where to find GIS resources,and web places.

In summary, GIS and Spatial Analysis in VeterinaryScience is a book that everyone with an interest in GIS andveterinary epidemiology should have on their bookshelf, notleast because of the comprehensive list of relevant refer-ences to the subject. Because there are many contributors,the style and quality vary but, overall, the content is of highquality and it is well worth reading. Some chapters requirethe reader to possess deeper statistical knowledge andunderstanding to get a full exchange of the content, whereasother chapters can be read by all veterinarians who areinterested in this field.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.017

New modes of disease transmission in Latin AmericaLivestock Trypanosomoses and their Vectors in Latin America by Marc Desquesnes. Office International des Epizooties, 2004. V40

(182 pages) ISBN 92-9044-634-X

Geoff Hide

School of Environment and Life Sciences, Peel Building, University of Salford, Salford, UK, M5 4WT

Over the years, there has been muchinterest in the livestock trypanosomosesof Africa, where many surveillance andcontrol studies have been conducted.However, these diseases have receivedmuch less attention in Latin America,and many of the etiological agents havebeen derived from their African counter-

parts; Trypanosoma vivax, Trypanosoma

evansi and Trypanosoma equiperdum were introduced tothe new world by humans importing their livestock. Thesespecies, unconstrained by the requirement for specialisttsetse fly vectors, have spread throughout Latin Americaand are on the verge of encroaching on North America.The other main livestock trypanosome – Trypanosomacruzi, perhaps more famous for its association with humandisease – is prevalent in fauna and threatens to progressinto the USA with the consequent risk of interactinghuman and animal transmission cycles. This is the settingfor Desquesnes’ new review of these diseases.

The core chapters of Livestock Trypanosomoses andtheir Vectors in Latin America are supported by goodintroductory chapters about the trypanosomatid species,their characteristics, hosts, pathogenicity and trans-mission cycles. The key adaptation that sets apart the

Latin American trypanosomes from those in Africa is theimportance of mechanical transmission by generalistbiting insects in the former, which releases them fromgeographical constraints imposed by specialist vectors.Desquesnes reviews much of the literature about themechanical transmission of T. vivax and T. evansi, andconcludes that the Tabanidae and the Stomoxyinae mighthave extremely important roles in transmission. Bycontrast, T. equiperdum adopts a different strategy –transmission during coitus – although mechanical trans-mission has not been ruled out.

A detailed chapter about the biology of vectors thatpromote mechanical transmission reviews their lifecyclesand considers the economics of the loss caused by thediseases spread by these vectors. For example, it isestimated that as many as 25 000 tons of meat are lostannually in the three Guyanas. Chapter four gives adetailed review of diagnostic techniques, and the nextchapter provides an in-depth review of the epidemiologicalstatus of the trypanosomoses, principally in FrenchGuyana but also across Latin America generally. Asurprising feature of these diseases that is revealedin this book is the sheer scale of the problem; forexample, there are 18 million head of cattle in Venezuela.Several different types of study of T. vivax prevalence(e.g. parasitological and serological) have been conductedbut all show results of 10–25%, with some studies

Page 17

Update TRENDS in Parasitology Vol.21 No.7 July 2005314

indicating that prevalence of 50–60% might not beuncommon. Similar stories are reported for othercountries and other parasites.

Following a discussion of vector-control strategies, thefinal chapter evaluates overall control strategies and looksto the future. Many of these diseases must be considered asemergent or re-emergent, and both T. vivax and T. evansithreaten to become major scourges of the huge areas ofanimal production in Central South America. Desquesnescalls for investment in the development of effectivediagnostic tools, the coordination of surveillance programs,and the development of control strategies and trypanocides.Finally, he takes us back to Africa. He speculates that, ifcontrol programs aimed at reducing tsetse fly populations in

Corresponding author: Carlton, J.M. ([email protected]).

www.sciencedirect.com

Africa are effective, we might see another effect occurring:the appearance of mechanical transmission of the nativetrypanosomoses to replace the cyclical transmissionthrough tsetse – a scary thought, indeed!

Overall, Livestock Trypanosomoses and their Vectors inLatin America, which was written in French originally,reads well in its English translation. Colour illustrationssupport the text, which is scientifically accurate and wellreferenced. This book is important because it is a timelyconsideration of a set of diseases that threatens to be ofconsiderable importance in the future.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.016

The year of the parasiteParasite Genomics Protocols edited by Sara E. Melville. Humana Press, 2004. £70.50 (452 pages) ISBN 158829062X

Jane M. Carlton

Institute for Genomic Research, Rockville, MD 20851, USA

Although 2005 has been designated the‘year of the rooster’ in the Chinese lunarcalendar, major contributions to thefield of parasitology might have led toit being renamed the ‘year of theparasite’. In particular, more completegenome sequences and comparativeanalyses of parasitic protist species

have been published this year than at

any other time; genomic and expression data from twoadditional species of rodent malaria parasite, the completesequence of Entamoeba histolytica, and sequence andcomparative analyses of two species of Theileria have beenpublished already, and the genome sequences of threespecies of trypanosomatid and the sexually transmittedhuman pathogen Trichomonas vaginalis are expected tobe published before the end of the year. So, it is extremelyexciting and timely to see the publication of ParasiteGenomics Protocols: a book in the Methods in MolecularBiology series.

The aim of the book, as outlined in the editorial preface,is to describe in detail the technologies that have beendeveloped for analyzing parasite genomes and geneproducts. These technologies are what might be calledthe modern-day molecular biologist’s toolkit, and includedatabase mining, genetic-marker identification, whole-genome expression analysis and genetic-manipulationtechniques. This book does an excellent job of describingmany of these in detail. The first six chapters consider thegeneration and in silico analysis of large, genome-scaledatasets. One of the must-read chapters for anyone who isinterested in using genome sequence data is chapter two

(Annotation of parasite genomes; Berriman, M. andHarris, M.), which gives a clear and concise guide to asubject that might seem confusing to many researchers.As stated by the authors, there are different standards ofannotation, from automated to manual (the ‘gold’ stan-dard), and an appreciation by researchers of the methodsused during the annotation process is essential for anassessment of the reliability of annotation. Realization ofthis concept by the scientific community would make lifemuch easier for many of us at genome-sequencing centers!

Subsequent chapters describe techniques for gener-ating genetic markers such as minisatellites and singlenucleotide polymorphisms, and whole-genome expressionsystems such as microarrays. Most of these chapters stopshort of discussion of data analysis after successfulcompletion of experiments, probably because this rep-resents a whole topic unto itself that would double the sizeof each chapter. Several low-throughput molecular biologytechniques such as rapid amplification of cDNA ends,chromosome separation by pulsed-field gel electrophore-sis, and chromosome digestion, fragmentation and map-ping are presented towards the end of the book in a veryvaluable series of chapters.

A cross-section of parasite species is used in thechapters, from the relatively well-studied Plasmodium,Trypanosoma and Toxoplasma species to less-studiedorganisms such as Trachipleistophora hominis – amember of the phylum Microspora. Just how transferableeach laboratory protocol is to other parasite species willdepend on the nature of the other species (e.g. intracellu-lar versus extracellular species, xenic versus axenicspecies). Parasites come in a variety of forms and sizesand, for this reason, it is not simple to compile a volumeof widely applicable methods. However, these protocols

Page 18

Update TRENDS in Parasitology Vol.21 No.7 July 2005 315

provide a starting point, at least, for the laboratoryresearcher, and it is hoped that they can be tweaked andmodified to suit other pathogen species.

One complaint is that some of the chapters could havebeen ordered better: for example, clustering together thechapters that describe the production of genetic markersfor genotyping (chapters eight, nine and 13). Also, con-sidering the field day that molecular evolutionary bio-logists are having with the mountain of new sequence data(which has earned them the tongue-in-cheek title of‘bottom feeders’), a chapter about basic evolutionaryanalysis of parasites, including inferring molecular

Corresponding author: Lashley, F.R. ([email protected]).

www.sciencedirect.com

phylogeny from the construction of simple phylogenetictrees, would have been useful. However, these are minorcomplaints from a reviewer with a bias towards evolution-ary analysis of genome sequence data. Generally, I havenothing but praise for Parasite Genomics Protocols –which must be among the first of its kind to link thedisparate disciplines of parasitology, molecular biology,genomics and bioinformatics – and it would make a usefuladdition to the modern-day parasitology laboratory.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.004

A return to tropical medicineOxford Handbook of Tropical Medicine (2nd edn) by Michael Eddleston, Robert Davidson, Robert Wilkinson and Stephen Pierini.

Oxford University Press, 2004. £24.95 (712 pages) ISBN 0198525095

Felissa R. Lashley

College of Nursing, Rutgers, The State University of New Jersey, 180 University Avenue, Suite 102, Newark, NJ 07102, USA

Two more editors have been added tothe second edition of the Oxford Hand-book of Tropical Medicine. In addition,nearly 30 experts with appointments invarious parts of the world have contri-buted their knowledge of broad areas.Miraculously, the editors have retainedthe compact style of this book, keeping itbelow 700 pages. The compact physical

dimensions and flexible but strong

water-repellant cover enables the practitioner to tuckthis book into a laboratory coat pocket.

The Oxford Handbook of Tropical Medicine (2nd edn) isorganized in a way that first sets the stage with the WorldHealth Organization (WHO; http://www.who.int/en/) andUnited Nations Children’s Fund (UNICEF; http://www.unicef.org/) approach to the integrated management ofchildhood illness in a concise manner that emphasizesteaching the mother. Next, the major serious diseasesfound in the tropics are covered, including malaria, HIVand other sexually transmitted infections, tuberculosis,diarrheal illness and acute respiratory infections. The restof the chapters cover systems such as cardiology andgastroenterology but, before these, there is a section aboutmultisystem diseases and infections that includes suchconditions as fever and rheumatoid arthritis. The finalsections are about nutrition, injuries – including burns –poisonings and snakebites (spider bites are not included),and immunizations. There are 32 color plates concen-trated into four pages that illustrate mainly microscopicpreparations of microorganisms, especially malarial para-sites. The white spaces in the book enable the reader toadd their own notations to the various topics covered.

The emphasis of this book is medicine in the field and,therefore, some topics that usually are discussed overmany pages, such as osteoarthritis in developed countries,are handled succinctly in this volume. In some cases, thesophistication of the content is uneven; for example, theHLA link in rheumatoid arthritis is mentioned in verygeneral, not up-to-date terms and without relating theinformation to its use in the tropical setting. In othersections such as the one about ophthalmology, excellentand succinct tables compare features for diagnosis andtreatment. The gastroenterology section also has particu-larly useful diagrams and tables, in addition to practicalpreventive information. The section about diarrhealdiseases is also particularly good and even includes asection dealing with situations that might be encounteredduring treatment. For example, the book discusses howone might respond if the mother of a child with diarrheabelieves that food should not be given to the child at thistime. In some sections, disease prevention is not a focusand is discussed only minimally. Many behavioral andcultural factors influence practices that can contribute tothe acquisition of disease, and these are difficult to cover ina book that must be useful in multiple continents and formultiple ways of living. However, in resource-poor areas,emphasizing prevention seems important. The nutritionsection is also strong, with many useful diagrams,management strategies and preventative aspects.

In summary, the Oxford Handbook of Tropical Medicine(2nd edn) will be useful for those practicing medicine andhealth care in the developing tropical world.

1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.pt.2005.05.001

Page 19

Trypanosoma equiperdum: master ofdisguise or historical mistake?Filip Claes1,2, Philippe Buscher2, Louis Touratier3 and Bruno Maria Goddeeris1

1Katholieke Universiteit Leuven Leuven, Faculty of Applied Bioscience and Engineering, Department of Animal Sciences,

Kasteelpark Arenberg 30, 3001 Leuven, Belgium2Institute of Tropical Medicine, Department of Parasitology, Nationalestraat 155, 2000 Antwerpen, Belgium3World Organization for Animal Health, 12 Rue de Prony, 75017 Paris, France

After 100 years of research, only a small number of

laboratory strains of Trypanosoma equiperdum exists,

and the history of most of the strains is unknown. No

definitive diagnosis of dourine can be made at the

serological or molecular level. Only clinical signs are

pathognomonic and international screening relies

on an outdated cross-reactive serological test (the

complement-fixation test) from 1915, resulting in

serious consequences at the practical level. Despite

many characterization attempts, no clear picture has

emerged of the position of T. equiperdum within the

Trypanozoon group. In this article, we highlight the

controversies that exist regarding T. equiperdum, and

the overlap that occurs with Trypanosoma evansi and

Trypanosoma brucei brucei. By revisiting the published

data, from the early decades of discovery to the recent

serological- and molecular-characterization studies, a

new hypothesis arises in which T. equiperdum no

longer exists as a separate species and in which current

strains can be divided into T. evansi (the historical

mistake) and Trypanosoma brucei equiperdum (the

master of disguise). Hence, dourine is a disease

caused by specific host immune responses to a

T. b. equiperdum or T. evansi infection.

What is dourine?

According to the World Organization for Animal Health(OIE; http://www.oie.int/eng/en_index.htm), ‘dourine is achronic or acute contagious disease of breeding solipedsthat is transmitted directly from animal to animal duringcoitus. The causal organism is Trypanosoma equiperdum.It is the only trypanosome that is not transmitted by aninvertebrate vector and it differs from other trypanosomesin that it is primarily a tissue parasite that rarely invadesthe blood. There is no known natural reservoir of theparasite other than infected equids. The clinical signs aremarked by periodic exacerbation and relapse, ending indeath or, possibly, recovery. Fever, local oedema of thegenitalia and mammary glands, cutaneous eruptions,incoordination, facial paralysis, ocular lesions, anaemia,and emaciation may all be observed. Oedematouscutaneous plaques, 5–8 cm in diameter and 1 cm thick,are pathognomonic.’ Because dourine is considered to be

Corresponding author: Claes, F. ([email protected]).Available online 26 May 2005

www.sciencedirect.com 1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved

incurable, seropositive horses should be slaughtered.The latest official reports from the OIE list the occur-rence of dourine in Botswana, Ethiopia, Germany,Kyrgyzstan, Namibia, Pakistan, Russia, South Africaand Uzbekistan.

The history of Trypanosoma equiperdum

A dourine-like disease was mentioned in early Arab textsbut the first recognized description of dourine in Europewas by Ammon and Dirkhausen who, in 1796, observedcases in a Prussian stud [1]. However, it was only in 1894that Rouget demonstrated the presence of T. equiperdumin the blood of an infected Algerian horse. However, thisparasite was lost before Rouget could reproduce thedisease in horses [2]. It was several years later thatBuffard and Schneider reproduced dourine in a horse afterthe subcutaneous inoculation of a parasite – isolated froma naturally infected Algerian horse – that was maintainedthrough several passages in experimentally infected dogs[3]. After confirmation of these results in 1900 [4], thistrypanosome was considered to be the causative organismof dourine; the name T. equiperdum was postulated byDoflein in 1901 [5]. Stabilates of this original strain are nolonger available. Because of the apparent difficultiesdetecting the parasite in some cases of dourine in Algeria,Chauvrat and Busy expressed their doubt aboutT. equiperdum being the causative organism [6]. Buffardand Schneider suggested in 1902 that the parasite mightcause surra or nagana, but not dourine. However, trypano-somes had been isolated from other cases of dourine inFrance, Hungary, Germany and Canada [5]. Other sourcesmention the possibility of symptomless carriers ofT. equiperdum in Canada and Russia [7,8].

Since the 19th century, dourine has occurred sporadic-ally in Europe. Around 1918, the disease was reportedonly in Russia, Turkey, Hungary and Spain. During WorldWar II, T. equiperdum was reintroduced into WesternEurope by Russian and Algerian horses, which were usedin the German army and in France, respectively [9]. Afterthe war, the disease was eradicated from Western Europeby systematic screening and control: clinical examination,confirmatory diagnosis by the complement-fixation test(CFT) and enforcement of sanitary measures, includingstamping out and treatment with high dosages of neo-arsphenamine in some cases.

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005

. doi:10.1016/j.pt.2005.05.010

Page 20

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005 317

Horses suffering from the clinical symptoms of dourinehave been reported recently in Ethiopia [10] and Mongolia[11] but, unfortunately, without successful parasite iso-lation. Several other attempts to isolate T. equiperdum inEastern Europe were also unsuccessful [12]. Thus,although this parasite has been present for more than100 years, most of its history has been lost; during the pastthree decades, nobody has isolated a new strain that couldbe used as reference material.

The availability of laboratory strains of Trypanosoma

equiperdum

Table 1 shows a list of T. equiperdum strains that aremaintained in laboratory collections and available forresearch. The Bordeaux Trypanosoma antigen type(BoTat) 1 clone is derived from the T. equiperdum strainInstitut Pasteur Paris. This strain arrived in Bordeaux in1961 and was maintained for ten years through serialpassages in mice; in 1971, it was cloned and kept in liquidnitrogen [13]. This strain was probably isolated from ahorse in Morocco in 1924. In 1976, the T. equiperdumOnderstepoort Veterinary Institute (OVI) strain wasisolated in South Africa from a horse showing clinicalsigns of dourine, in casu progressive emaciation andposterior paresis [14]. The Swiss Tropical Institute Baselstrain (STIB) 818 was isolated in China in 1979 from ahorse [15]; the strain was obtained after six months ofserial passages, in rabbits initially and then in mice.Unfortunately, no information was provided about theclinical signs of the original host. From the other strains,the exact history is unknown or doubtful. The Americanand Canadian strains were maintained by serial passagein rats at the National Veterinary Service Laboratories(http://www.aphis.usda.gov/vs/nvsl/) from 1923 to 1977and then kept in liquid nitrogen [16]. There are no recordsof their origin or whether they were isolated from horsessuffering from dourine. The names suggest that theyoriginated in the USA and Canada, respectively. Noreferences are available for the Alfort, Hamburg andStaatliches Veterinarmedizinisches Prufungsinstitut(SVP) strains. Nonetheless, they are thought to be strainsof T. equiperdum in that they are used as reference strainsfor dourine diagnosis in Germany (P.H. Clausen, personalcommunication). Antwerp Trypanosoma antigen type

Table 1. Available laboratory strains of Trypanosoma

equiperdum

Codea Country Host Year of

isolation

Refs

BoTat 1.1 Morocco Horse 1924 [13,40,42,48]

STIB 818 China Horse 1979 [15,36,40,42,48]

OVI South Africa Horse 1977 [14,35,42,48]

ATCC 30019 France Horse 1903? [17,35,37]

ATCC 30023 France Horse 1903? [17,35,37]

Am. stabilate America? Horse Unknown [16,48]

Can. stabilate Canada? Horse unknown [16,48]

AnTat 4.1 Unknown Unknown Unknown [35,48]

Alfort Unknown Horse 1949? [35,48]

Hamburg Unknown Unknown Unknown [35]

SVP Unknown Unknown Unknown [35,48]

TREU 2259 Unknown Unknown Unknown [49,50]aAbbreviations: Am., American; Can., Canadian; TREU, trypanosome research

Edinburgh University.

www.sciencedirect.com

(AnTat) 4 was cloned from a wild-type T. equiperdum atthe Institute of Tropical Medicine in Antwerp (Filip Claes,PhD thesis, Katholieke Universiteit Leuven, 2003).Within the American Type Culture Collection (ATCC;http://www.atcc.org), the dyskinetoplastic T. equiperdumstrain ATCC 30023 is derived from T. equiperdum strainATCC 30019 after a long in vitro laboratory cultivation[17]. As stated by Hoare and repeated by the OIE,the origin and identity of some laboratory strains ofT. equiperdum from dourine horses is so uncertain thatfresh isolates are desperately needed [18].

Clinical signs of dourine

Equines are considered to be the only natural host ofT. equiperdum. The disease in horses is chronic, persistsfor one or two years and is generally divided into threephases, although the clinical course can vary considerablyunder different conditions. The first period is character-ized by oedema, tumefaction and damage to the genitalia,and begins one to two weeks after infection. The secondstage of disease is pathognomonic for dourine. In thisperiod, typical cutaneous plaques or skin thicknesses canoccur, with sizes ranging from extremely small to handsized. Interestingly, these plaques have also been observedsporadically in animals infected with Trypanosoma evansi[19]. The third phase of dourine is characterized byprogressive anaemia, disorders of the nervous system –mainly paralysis of the hind legs and paraplegia – and,finally, death [20].

Experimental infections in horses through infusion intothe urogenital tract have been performed in South Africausing the OVI strain [14], in the USA using the Americanand Canadian strains [16] and in Kazakhstan* using awild-type strain. In the US study, none of the 20 infectedhorses developed typical clinical signs; they showed onlygeneral signs of trypanosomiasis. In South Africa andKazakhstan, however, the animals showed typical signs ofdourine, such as scrotal oedema, emaciation and posteriorparesis, but the presence of the pathognomonic dourineplaques were not reported by the authors. Apparently,differences in pathology are observed between animalsin these experimental infections but it remains unclearwhether the differences are related to the T. equiperdumstrain that is used or whether they are caused bydifferences in the host immune response.

Under laboratory conditions, dogs can develop dourine[21]. Different routes of infection (e.g. subcutaneous,intraperitoneal, intravenous, intraurethral and intra-vaginal transmission) were tested and all gave rise toobvious clinical signs of dourine [5]. Early experimentswith rabbits reported specific clinical signs of dourine[3,22]. By contrast, in recent experimental infections withT. equiperdum, rabbits developed clinical signs that couldnot be distinguished from those developed by rabbitsinfected with T. evansi [23].

Mice and rats can be infected with T. equiperdum but donot develop the ‘normal’ form of dourine, although allavailable laboratory strains of the parasite grow easily in

* G.D. Ilgekbayeva et al. Second Symposium of the New World on Trypanosomesand other Haemoparasites, Venezuela, October 1999.

Page 21

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005318

these animals (Theo Baltz, PhD thesis, University ofBordeaux 2, 1982). Ruminants seem to be refractory toinfection with T. equiperdum [5], but Wang producedclinical manifestations of dourine in sheep and goatsafter inoculation with a murine-adapted strain ofT. equiperdum [24]. Thus, although T. equiperdum occursnaturally only in equines, it seems that other animals canalso be susceptible, which suggests that these parasitesare strains of T. evansi or Trypanosoma brucei brucei andthat dourine pathology is strain related or host related.

Transmission and diagnosis

Generally, it is believed that natural transmission ofT. equiperdum parasites occurs only during copulation[20]. However, intravenous or intraperitoneal experimen-tal infections indicate that mechanical transmission bybloodfeeding flies cannot be excluded as a possible route ofinfection, even if the number of parasites in the blood isextremely low. Indeed, in recent experiments with cattleinfected with T. b. brucei, in the chronic phase – whenblood examinations and PCR of blood samples werenegative – tsetse flies could still be infected by feedingon these animals [25]. A similar phenomenon might alsooccur with T. equiperdum that is transmitted by blood-sucking flies.

Clinical signs of dourine can provide a strong indicationof the presence of the disease, as can its chronic evolution,but confirmatory diagnosis is needed. It is extremelydifficult to detect the parasite in the body fluids of infectedhorses; therefore, diagnosis of T. equiperdum infection isbased on serological evidence. Despite the development ofantibody and antigen enzyme-linked immunosorbentassays for T. equiperdum [26], CFT remains the onlyinternationally recommended test [27], although it doesnot distinguish clearly among T. equiperdum, T. evansiand T. b. brucei [28]. Indeed, because possible cross-reactions with T. evansi and T. b. brucei might occur, theseparasites cannot be distinguished from T. equiperdumunless the test samples originate from regions that arefree from T. evansi and T. b. brucei. Unfortunately, apartfrom South Africa and parts of Russia, countries in whichdourine is currently reported often lie within the distri-bution area of T. evansi. Therefore, it is essential todevelop tests that accurately differentiate dourine fromsurra infections in Equidae. To develop such tests, theavailable laboratory strains should first be character-ized properly using different serological and moleculartechniques.

Treatment

There are no officially approved drugs to treat horsessuffering from dourine, although some older publicationsmention experimental treatment of horses with naganoland neoarsphenamine [29], or quinapyramine sulfate [30].

International regulations currently impose the slaugh-tering of CFT-positive horses. Nevertheless, in vitrosensitivity of different T. equiperdum strains to suramin,diminazene, quinapyramine and melarsomine has beenreported [31]. Taking these results into account, it seemsthat it would be worthwhile to perform efficacy trials onthe treatment of T. equiperdum infections with the drugs

www.sciencedirect.com

currently used for T. evansi treatment in horses (suraminand melarsomine) or camels (cymelarsan). However,without reliable diagnostic tests, it will remain difficult,if not impossible, to assess cure rates in vivo in suchefficacy trials. Therefore, it might be useful to evaluatethe in vitro sensitivity of the available T. equiperdumstrains first.

Serological- and molecular-characterization studies

In the 1970s, research of the variable antigenic repertoireof one T. equiperdum strain (BoTat 1) defined somepreferentially expressed variable antigenic types (VATs)of T. equiperdum [13]. The occurrence of these BoTatVATs has not yet been examined in other strains ofT. equiperdum or trypanosome taxa (e.g. T. evansi).However, it has been shown that Rhode trypanosomeantigenic type (RoTat) 1.2 is a predominant VAT ofT. evansi against which antibodies can be detected in theserum of infected animals [32]. This has enabled thedevelopment of different serodiagnostic tests –based onthe RoTat 1.2 variable surface glycoprotein (VSG) – thathave been applied to different host species worldwide[33,34]. Experimental infection of rabbits indicates thatanti-RoTat-1.2 antibodies are also generated duringT. equiperdum infection [35]. Of 11 strains examined,only two (OVI and BoTat) did not express the RoTat 1.2VSG during a 35-day infection of rabbits. This is an earlyindication that there are at least two groups within theexisting T. equiperdum collection and that the largestgroup shares characteristics with T. evansi. Of the strainsexamined, only OVI had been shown to cause clinicalsymptoms of dourine in horses. Clinical and pathologicaldata are lacking for all other strains. Isoenzyme analysishas been usedtodemonstrate differencesamong 12T. evansistrains and one T. equiperdum strain (STIB 818). Of the16 enzymes tested, only malate dehydrogenase andalanine aminotransferase were different for some T. evansistrains. T. equiperdum, however, could not be distinguishedfrom some T. evansi strains [36].

Some authors suggest that the absence of maxicirclesfrom the kinetoplast DNA of T. evansi is a major differencebetween this parasite and T. equiperdum [37,38]. How-ever, because dyskinetoplastic strains exist in bothT. evansi and T. equiperdum, the validity of thischaracteristic is questionable. Moreover, Borst et al.stated in 1987 that ‘there is no theoretical reason whyT. evansi strains with a defective maxicircle could notexist’ [39]. Zhang et al. used Southern blot analysis todemonstrate differences among T. equiperdum (STIB 818,BoTat 1.1 and the OVI strain), T. b. brucei (one strain) andT. evansi stocks (15 strains). BoTat 1.1 and the SouthAfrican OVI strain clustered out from the T. evansi group.Moreover, this cluster shared more similarity withT. b. brucei than with the T. evansi cluster. AnotherT. equiperdum (STIB 818) was found within the T. evansigroup. The authors stated that this outlier STIB 818 couldreflect the limit of sensitivity of Southern blot onrestriction fragment length polymorphism or that itcould be due to the misclassification of this strain [40].Using the same DNA probes, a dissimilarity index of56% was observed between one T. equiperdum and one

Page 22

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005 319

T. evansi strain. However, this cluster analysis was basedon only one strain of each species and, unfortunately, theidentity of the strains used was not revealed in the article[41]. Microsatellite markers did not reveal generaldifferences between T. equiperdum and T. evansi. Of thethree T. equiperdum stocks tested (BoTat 1.1, STIB 818and the South African OVI strain), no species-specificalleles were found. The Chinese T. equiperdum (STIB818), however, shared common alleles with the ChineseT. evansi tested, and the BoTat 1.1 clone had identicalgenotypes with four loci of the KETRI 2480 T. evansistrain. This study highlights the heterogeneity within thestrains classified as T. equiperdum [42]. Internal tran-scribed spacer 1 analysis did not lead to species identifi-cation within the Trypanozoon group [43].

The variability of transferrin receptor genes(ESAG6/7) in T. evansi, T. equiperdum and T. b. bruceiwas studied by comparing several T. b. brucei or T. evansistrains with one T. equiperdum strain (STIB 818) [44,45].Both studies attest that the transferrin gene sequencesobtained from STIB 818 are a subcluster within thesequences of T. b. brucei or T. evansi. These studies merelyconfirm that multiple transferrin receptor genes arepresent in all genomes and that these genes are relatively

80 82 84 86 88 90 92 94 96 98

100

100

100

100

100

100

100

100

95

97

77

77

53

8575

76

61

71

75

63

97

Figure 1. Unweighted pair groupmethodwith arithmeticmean homology tree. The tree (b

RAPD ILO 525 and MEGA [48]. Numbers at nodes are the percentages of 1000 bootstrap

www.sciencedirect.com

conserved in the three Trypanosoma species. Both articlesconclude that the variability of the transferrin receptor ismore limited in T. equiperdum and that this might berelated to the limited host specificity of this parasite.However, this conclusion has been contested recently bySalmon et al., who proved that transferrin receptors arenot host-species specific [46]. The examples that have beenmentioned illustrate the problems related to the differen-tiation of the three species and the limited statisticalpower of some published results because only a few strainswere compared. Because of these scattered data and theuse of only a few, mostly different strains of T. equiperdumin the analyses, it is difficult (if not impossible) to drawmeaningful conclusions about T. equiperdum from theseobservations.

Recently, a putative PCR that is specific to T. evansiwas developed based on the RoTat 1.2 VSG. Nine of the11 so-called T. equiperdum strains tested positive in thisPCR [47]. Moreover, the two T. equiperdum strains thatlacked the RoTat 1.2 VSG gene (BoTat and OVI) alsoseemed to be different from the other tested populationsin random amplified polymorphic DNA (RAPD) and inmultiple endonuclease genotyping approach (MEGA)(Figure 1). Indeed, they resembled T. b. brucei more

TRENDS in Parasitology

100 Strain

CAN 86 K

AnTat 3.1

SVP

Can. stabilate

RoTat 1.2

Stock Colombia

Zagora

ATCC 30019

ATCC 30023

STIB 818

Am. stabilate

Stock Vietnam

Merzouga

KETRI 2480

Alfort

AnTat 4.1

AnTat 5.2

AnTat 17.1

KETRI 2994

OVI

BoTat 1.1

AnTat 2.2

Species

T. evansi

T. evansi

T. equiperdum

T. equiperdum

T. evansi

T. evansi

T. evansi

T. equiperdum

T. equiperdum

T. equiperdum

T. equiperdum

T. evansi

T. evansi

T. evansi

T. equiperdum

T. equiperdum

T. b. brucei

T. b. brucei

T. b. brucei

T. equiperdum

T. equiperdum

T. b. brucei

Refs

[23,32,48]

[23,32,48]

[34,48]

[16,48]

[23,32,48]

[23,32,48]

[23,32,48]

[17,35,37]

[17,35,37]

[15,36,40,42,48]

[16,48]

[23,32,34,48]

[23,32,48]

[23,32,48]

[35,48]

[35,48]

[42,48]

[48]

[48]

[14,35,40,42,48]

[13,40,42,48]

[48]

ased on percentage similarity) was obtained by combining the data fromRAPD 606,

replicates in which these nodes appeared.

Page 23

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005320

closely, whereas all other T. equiperdum strains clusteredin a homogenous group of T. evansi [48]. This finding is inaccordance with those of Zhang et al. [40], who alsoobserved a different cluster for BoTat 1.1 and the OVIstrain compared with the STIB 818 strain, whichclustered within the T. evansi group. Moreover, also infunctional genome analysis (microarrays), the expressionprofiles for T. equiperdum BoTat and OVI resemble thoseof T. b. brucei rather than those of T. evansi (F. Claes,unpublished).

From these new insights, it seems that dourine can becaused by particular strains of T. equiperdum that areclosely related to T. b. brucei, and that most so-calledT. equiperdum strains cluster serologically and molecu-larly within T. evansi. Nonetheless, one must keep in mindthat, irrespective of the strain used, the clinical outcome ofan infection might depend entirely on the immunologicalresponse of the host.

Concluding remarks

Published data that have been gathered using differentserological and molecular methods do not enable consist-ent discrimination between T. evansi and T. equiperdum,and several questions remain. Recent data indicate thatthe T. equiperdum collection is not homogenous and thatmore attention should be paid to the differences betweenthe so-called T. equiperdum strains.

Furthermore, we hypothesize that some T. equiperdumstrains are actually T. b. brucei or members of a subspeciesof T. brucei (Trypanosoma brucei equiperdum) and thatall other T. equiperdum strains are misidentified andare, in fact, T. evansi. Consequently, we propose analternative definition of dourine: a chronic contagiousdisease of solipeds caused by T. b. brucei or T. evansi thatis transmitted directly from animal to animal duringcoitus or by insect vector.

This hypothesis can be investigated by performingexperimental infections of horses with T. equiperdum andby comparing clinical signs with the pathology of con-firmed T. evansi and T. b. brucei strains. Controlledexperimental infections of horses, preferably with newisolates, might lead to a better insight into the pathologyof dourine. Therefore, obtaining new isolates is anessential step for a better understanding of the parasitethat causes this disease. The close relationship betweenT. evansi and T. b. brucei makes it necessary to continuethe characterization of both species, in particular usingmolecular or serological markers for the different groupsof T. equiperdum and T. b. brucei. Nevertheless, for suchcharacterization to be successful, studies should beperformed on a large collection of well-documentedT. equiperdum strains. When a reliable diagnostic testbecomes available, it will be possible to screen trypano-cidal drugs for efficacy against this parasite in experi-mental infections.

Acknowledgements

F.C. is funded by the Institute for the Promotion of Innovation by Scienceand Technology in Flanders, and his study received financial support fromthe International Livestock Research Institute (Nairobi). We thank RetoBrun, Peter-Henning Clausen and David Kinker for providing

www.sciencedirect.com

Trypanosoma equiperdum strains. We are particularly indebted to thelate Joyce Hagebock, who provided us with materials and information.

References

1 Zwick, W. and Knuth, P. (1928) Beschaelseuche oder dourine.Handbuch der Pathog. Mikro-org. (Kolle, Kraus u Uhlenhuth) 7,1434–1476

2 Rouget, J. (1896) Contribution a l’etude du trypanosome desmammiferes. Ann. Inst. Pasteur 10, 716–728

3 Buffard, M. and Schneider, G. (1900) Le trypanosome de la dourine.Archives Parasitologie 3, 124–133

4 Nocard, E. (1900) Sur les rapports qui existent entre la dourine et lesurra ou le nagana. Bull. Acad. Med. 64, 154–163

5 Laveran, A. and Mesnil, F. (1912) Trypanosomes et Trypanosomiases,Masson et Co., Paris

6 Buffard, M. and Schneider, G. (1902) Note sur l’existence en Algeried’un trypanosome autre que la dourine. Rec. med. veter 9, 721–727

7 Watson, E.A., ed. (1920) Dourine in Canada, Dept of Agric., Health ofAnimals Branch, Ottawa

8 Kasansky, I.I. (1958) Methods of combating trypanosomes of animals(Su-auru and dourine) in the USSR. Bull. OIE 49, 149–170

9 Saurat, P. (1946) La dourine: son existence actuelle en France. Rec.Med. Vet. 122, 289–309

10 Alemu, T. et al. (1997) The use of enzyme linked immunosorbentassays to investigate the prevalence of Trypanosoma equiperdum inEthiopian horses. Vet. Parasitol. 71, 239–250

11 Clausen, P-H. et al. (2003) A field study to estimate the prevalenceof Trypanosoma equiperdum in Mongolian horses. Vet. Parasitol. 115,9–18

12 Zablotskij, V.T. et al. (2003) The current challenges of dourine:difficulties in differentiating Trypanosoma equiperdum within thesubgenus Trypanozoon. Rev. Sci. Tech. 22, 1087–1096

13 Capbern, A. et al. (1977) Trypanosoma equiperdum: etude desvariations antigeniques au cours de la trypanosomose experimentaledu lapin. Exp. Parasitol. 42, 6–13

14 Barrowman, P.R. (1976) Experimental intraspinal Trypanosomaequiperdum infection in a horse. Onderstepoort J. Vet. Res. 43,201–202

15 Lun, Z.R. et al. (1992) Kinetoplast DNA and molecular karyotypes ofTrypanosoma evansi and Trypanosoma equiperdum from China. Mol.Biochem. Parasitol. 50, 189–196

16 Hagebock, J.M. et al. (1993) Evaluation of agar gel immunodiffusionand indirect fluorescent antibody assays as supplemental tests fordourine in equids. Am. J. Vet. Res. 54, 1201–1208

17 Hajduk, S.L. (1976) Demonstration of kinetoplast DNA in dyskineto-plastic strains of Trypanosoma equiperdum. Science 191, 858–859

18 Hoare, C.A., ed. (1972) The Trypanosomes of Mammals, BlackwellScience

19 Brun, R. et al. (1998) Trypanosoma evansi and T. equiperdum:distribution, biology, treatment and phylogenetic relationship. Vet.Parasitol. 79, 95–107

20 Stephen, L.E. (1986) Trypanosomiasis. A veterinary perspective,Pergamon Press

21 Theis, J.H. and Bolton, V. (1980) Trypanosoma equiperdum: move-ment from the dermis. Exp. Parasitol. 50, 317–330

22 Nocard, E. (1901) Sur les rapports qui existent entre la dourine et lesurra ou le nagana. C.R. Soc. Biol., Paris 64, 464–466

23 Claes, F. et al. (2002) The expression of RoTat 1.2 variable antigen typein Trypanosoma evansi and T. equiperdum. Ann. N. Y. Acad. Sci. 969,174–179

24 Wang, Z-L. (1988) The similarities and differences of the character-istics between T. equiperdum and T. evansi. Bull. Vet. Col. 8, 300–303

25 Van Den Bossche et al. (2004) Recirculation of Trypanosoma bruceibrucei in cattle after T. congolense challenge by tsetse flies. Vet.Parasitol. 121, 79–85

26 Katz, J. et al. (2000) Procedurally similar competitive immunoassaysystems for the serodiagnosis of Babesia equi, Babesia caballi,Trypanosoma equiperdum, and Burkholderia mallei infection inhorses. J. Vet. Diagn. Invest. 12, 46–50

27 Watson, E.A. (1915) Dourine and the complement fixation test.Parasitology 8, 156–183

Page 24

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005 321

28 Robinson, E.M. (1926) Serological investigations into some diseases ofdomesticated animals in South Africa caused by trypanosomes. Rep.Direct. Vet. Educ. Res., Dept. Agr. S. Afr. 9, 11–12

29 Ciuca, A. (1933) La dourine. Bull. OIE. 7, 168–17230 Vaysse, J. and Zottner, G. (1950) Contribution a l’etude de la

chimiotherapie et de la chimioprevention de la dourine parl’antracyde. Bull. OIE. 34, 172–179

31 Zhang, Z.Q. et al. (1992) In vivo and in vitro sensitivity ofTrypanosoma evansi and Trypanosoma equiperdum to diminazene,suramin, MelCy, quinapyramine and isometamidium. Acta Trop. 50,101–110

32 Verloo, D. et al. (2001) General expression of RoTat 1.2 variableantigen type in Trypanosoma evansi isolates of different origin. Vet.Parasitol. 97, 183–189

33 Dia, M.L. et al. (1997) Some factors affecting the prevalence ofTrypanosoma evansi in camels in Mauritania. Vet. Parasitol. 72,111–120

34 Verloo, D. et al. (2000) Comparison of serological tests for Trypano-

soma evansi natural infections in water buffaloes from NorthVietnam. Vet. Parasitol. 92, 87–96

35 Claes, F. et al. (2003) The expression of RoTat 1.2 variable surfaceglycoprotein (VSG) in Trypanosoma evansi and T. equiperdum. Vet.

Parasitol. 116, 209–21636 Lun, Z.R. et al. (1992) The isoenzyme characteristics of Trypanosoma

evansi and Trypanosoma equiperdum isolated from domestic stocks inChina. Ann. Trop. Med. Parasitol. 86, 333–340

37 Frash, A.C. et al. (1980) The kinetoplast DNA of Trypanosoma

equiperdum. Biochim. Biophys. Acta 607, 397–41038 Riou, G.F. and Saucier, J.M. (1979) Characterization of the molecular

components in kinetoplast–mitochondrial DNA of Trypanosomaequiperdum. Comparative study of the dyskinetoplastic and wildstrains. J. Cell Biol. 82, 248–263

39 Borst, P. et al. (1987) Kinetoplast DNA of Trypanosoma evansi. Mol.

Biochem. Parasitol. 23, 31–38

ScienceDirect collection reache

Elsevier recently announced that six million articles are now ava

milestone in electronic scientific, technical and medical publishing

an unsurpassed volume of information fr

ScienceDirect’s extensive and unique full-text collection covers

Tetrahedron and the full suite of Trends and Current Opinion journ

unsurpassed searching and linking function

The rapid growth of the ScienceDirect collection is due to the inte

addition to the Backfiles – heritage collections in a number of disci

Elsevier’s journals back to volume one, issue one, is the addition o

Also available online for the first time are six Cell titles’ long-await

important historic development

The six-millionth article loaded onto ScienceDirect entitled "Gene

was authored by Benjamin M. Shykind and colleagues from the

Hughes Medical Institute, College of Physicians and Surgeons at C

Elsevier’s leading journal Cell, Volu

www.science

www.sciencedirect.com

40 Zhang, Z.Q. and Baltz, T. (1994) Identification of Trypanosoma evansi,Trypanosoma equiperdum and Trypanosoma brucei brucei usingrepetitive DNA probes. Vet. Parasitol. 53, 197–208

41 Hide, G. et al. (1990) The identification of Trypanosoma bruceisubspecies using repetitive DNA sequences. Mol. Biochem. Parasitol.39, 213–226

42 Biteau, N. et al. (2000) Characterization of Trypanozoon isolates usinga repeated coding sequence and microsatelite markers. Mol. Biochem.Parasitol. 105, 185–201

43 Desquesnes, M. et al. (2001) Detection and identification of Trypano-soma of African livestock through a single PCR based on internaltranscribed spacer 1 of rDNA. Int. J. Parasitol. 31, 610–614

44 Isobe, T. et al. (2003) The transferrin receptor genes of Trypanosomaequiperdum are less diverse in their transferrin binding site thanthose of broad-host range Trypanosoma brucei. J. Mol. Evol. 56,377–386

45 Witola, W.H. et al. (2005) Genetic variability in ESAG6 genes amongTrypanosoma evansi isolates and in comparison to other Trypanozoonmembers. Acta Trop. 93, 63–73

46 Salmon, D. et al. (2005) Trypanosoma brucei: growth differences indifferent mammalian sera are not due to the species-specificity oftransferrin. Exp. Parasitol. 109, 188–194

47 Claes, F. et al. (2004) Variable surface glycoprotein RoTat 1.2 PCR as aspecific diagnostic tool for the detection of Trypanosoma evansiinfections. Kinetoplastid Biol. Dis. 3, 3

48 Claes, F. et al. (2003) How does Trypanosoma equiperdum fit into theTrypanozoon group? A cluster analysis by random amplified poly-morphic DNA (RAPD) and multiple endonuclease genotypingapproach (MEGA). Parasitology 126, 425–431

49 Wassal, D.A. et al. (1991) Comparative evaluation of enzyme-linkedimmunosorbent assay (ELISA) for the serodiagnosis of dourine. Vet.Parasitol. 39, 233–239

50 Bishop, P. et al. (1995) Trypanosoma equiperdum: detection oftrypanosomal antibodies and antigen by enzyme-linked immuno-sorbent assay. Br. Vet. J. 151, 715–720

s six million full-text articles

ilable on its premier electronic platform, ScienceDirect. This

means that researchers around the globe will be able to access

om the convenience of their desktop.

over 1900 journals, including titles such as The Lancet, Cell,

als. With ScienceDirect, the research process is enhanced with

ality, all on a single, intuitive interface.

gration of several prestigious publications as well as ongoing

plines. The latest step in this ambitious project to digitize all of

f the highly cited Cell Press journal collection on ScienceDirect.

ed Backfiles, containing more than 12,000 articles highlighting

s in the field of life sciences.

Switching and the Stability of Odorant Receptor Gene Choice"

Dept. of Biochemistry and Molecular Biophysics and Howard

olumbia University. The article appears in the 11 June issue of

me 117, Issue 6, pages 801–815.

direct.com

Page 25

SAGE and the quantitative analysis ofgene expression in parasitesDavid P. Knox and Philip J. Skuce

Moredun Research Institute, Pentlands Science Park, Bush Loan, Penicuik, Midlothian, UK, EH26 0PZ

The nature of an organism is defined by the genes that it

expresses. Genome- and expressed-sequence-tag (EST)

sequencing projects are underway for many of themajor

parasites of humans and animals. These provide essen-

tial datasets that delineate the genes present in an

organism and, in the case of ESTs, some quantitative

information on gene expression. The temporal and

quantitative analysis of gene expression is essential to

fully exploit these datasets and define the biology of the

parasite at the molecular level. Here, we discuss the

application of serial analysis of gene expression (SAGE)

for this purpose. SAGE is a technique that allows the

rapid, quantitative analysis of thousands of transcripts.

It complements microarray analysis with the advantage

that it is affordable for standard laboratories. It provides

a platform to define complete metabolic pathways and

has been applied to study responses to drug treatment

and the molecular events that are associated with

arrested larval development.

Analysis of gene expression

Genome- and expressed-sequence-tag (EST) sequencingprojects have been initiated, and some completed, forseveral protozoan and metazoan parasites of humansand of animals; a comprehensive list of completed andongoing sequencing projects can be viewed at http://www.genomesonline.org [1]. Now that we are entering the post-genome era, there is a need to develop and apply methodsfor the detailed analysis of gene expression in parasites todefine their biology at the molecular level. Such analysesmight help to identify gene products that are crucial forparasite survival and might be novel targets for drugs andvaccines. The comprehensive analysis of gene expressionin different parasite stages, different tissue sites andduring exposure to developing host immunity shouldprovide insight into the host–parasite relationship.

To date, comparative studies have employed methodo-logies such as differential colony/plaque hybridization,suppressive subtractive hybridization, mRNA differentialdisplay and real-time PCR [2]. Each method has advan-tages and disadvantages but the major drawback of eachis that they examine a relatively small number of genes atone time. DNA-microarray technology can extend suchstudies to a genome-wide scale but this potential isrestricted by the limited genome coverage of many ofEST datasets. For example, there are w17 500 ESTs

Corresponding author: Knox, D.P. ([email protected]).Available online 26 May 2005

www.sciencedirect.com 1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved

available for the ovine abomasal nematode Haemonchuscontortus in NEMBASE2 (http://www.nematodes.org) [3],which is equivalent to w30% predicted gene coverage;that 70% gene coverage is lacking potentially compro-mises the output from array analysis of gene expression.EST analysis remains an efficient tool for gene discoveryand it has been applied to most parasites of medical andveterinary importance. EST datasets provide insights intothe molecular aspects of tissue organization, signalling,host interactions and immune evasion [4,5], and the regu-lation of gene expression during development [4–6].

Serial analysis of gene expression

Serial analysis of gene expression (SAGE) is a sequence-based, gene-expression-profiling technique [7] that can beused to obtain complete transcriptional profiles ofexpressed genes, even when these are unknown. Unlikemicroarray approaches, it does not require knowledge ofthe sequences to be analyzed, which is an advantage formany aspects of parasite biology. Microarrays areregarded as closed profiling platforms because the dataobtained are limited to either a predetermined or knownset of genes on the array. However, SAGE is not restrictedin this manner and is regarded as an open platform thatcan identify previously unknown genes. The number ofSAGE tags (defined below) that are sequenced is animportant consideration that has implications for the cost-effectiveness and scale of comparative analyses. Althoughone might assume that the more tags that are sequenced,the more representative the gene-expression profile, thisis not necessarily the case. For example, in a recent studyof gene expression in HeLa cells, a total of 80 000 tagswere sequenced and the transcript profile compared withprofiles generated from smaller subsets that represent2000, 4000, 10 000, 20 000 and 40 000 tags [8]. Comparingsubsets revealed virtually no difference in tag proportionsof highly abundant genes, with discrepancies only evidentfor tags with an abundance of !3 in the smaller subsets[8]. The PCR-amplification step of SAGE (see later andFigure 1) reduces the amount of mRNA that is requiredcompared with microarrays [9], with SAGE librariesconstructed from considerably less material (50–500-ngRNA) [9–11] and even from a single cell [12]. However, aPCR step is likely to introduce amplification bias thataffects the acquisition of quantitative data. In this regard,it is noteworthy that SAGE and microarray data correlate[13], but SAGE has the advantage that its digital outputallows statistical comparisons between SAGE libraries

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005

. doi:10.1016/j.pt.2005.05.011

Page 26

AAAAA

AAAAA

AAAAA

AAAAA

AAAAA

AAAAA

AAAAA

AAAAA

Target organism,tissue or cell

(i)

(ii)

(iii)

TRENDS in Parasitology

Sequence-linked tags

Figure 1. The SAGE process. (i) Double-stranded cDNA is synthesized from mRNA

using a biotinylated (light green) oligo(dT) primer, and the cDNA cleaved using a

restriction enzyme with a 4-bp-recognition sequence, termed an anchoring enzyme

(AE), which cleaves every 256 bp on average. NlaIII is the AE that is used most

commonly. The 3 0 portion of the cleaved cDNA is recovered by binding to

streptavidin-coated beads (brown). (ii) The reactionmixture is then divided into two

portions and two independent linkers ligated to each portion using NlaIII-cohesive

termini. The linkers contain type IIS recognition sites (BsmFI is used most

commonly) near the NlaIII site. Because BsmFI cuts w10 bp 3 0 to the recognition

site of the AE to generate the unique oligonucleotide tag, it is termed the tagging

enzyme (TE). The released tags are recovered, blunt-ended and the two portions

mixed and ligated tail-to-tail via their mRNA-derived termini (ditags). The products

are PCR-amplified using forward- and reverse-primer-binding sites that are

included in the original adaptor sequences. After cleavage with the AE (NlaIII)

and gel purification, ditags are ligated to obtain concatemers that are cloned into a

plasmid vector for sequencing. The more tags per clone, the more efficient the

sequencing effort. (iii) Bioinformatics are used to isolate individual tag sequences

for identifying and quantifying tags. Gene-expression profiles are then generated.

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005 323

from the same source. This is augmented by the availabilityof commercial SAGE kits that include several quality-control steps. Comparing microarray experiments is moredifficult because of the possibility of random errors andsystemic errors associated with different laboratories [14].

www.sciencedirect.com

How SAGE works

SAGE methodology is based on three key principles [15].First, a short sequence tag (9–14 bp), which is defined by aspecific restriction endonuclease site (termed the anchor-ing enzyme) (Figure 1) at a fixed distance from the poly(A)tail, contains sufficient information to identify the mRNAtranscript from which it originates. Theoretically, a 10-bptag can give 410 different sequence combinations [16].Second, concatenation of the tags allows efficient analysisof transcripts in a serial manner because SAGE uses serialprocessing such that 25–50 tags are analyzed from eachlane of a DNA sequencer. Third, the number of times agiven tag is observed equates to the expression level of itsoriginal transcript.

The 10-bp tag is not sufficient to identify all the genes inthe unannotated human genome [17]. This problem hasbeen overcome by developing a ‘LongSAGE’ protocol thatemploys a different, type IIS restriction enzyme, MmeI,which cuts 17 bp 3 0 from the anchoring site. The increasein tag length increases tag uniqueness to R99% andreduces the impact of tag-sequencing errors. Correctassignation of the tag is an essential step in SAGE andis restricted by the quality of the databases that aresearched, with full-length transcript sequences preferred.Many SAGE tags are not found in available cDNA data-bases and, because tags can represent two separatecDNAs, it is possible that they might be mis-assigned.The deficiencies in tag assignment and ways to overcomethem have been addressed recently [18]. The assignationof tags is improved dramatically by increasing the lengthof the tag sequence. A new technique, SuperSage [19],generates 26-bp tags. This allows the simultaneous quan-tification of gene expression in the host and pathogen,which has clear potential application in parasitology.Continuing efforts to increase the length of the tagsequences, and improvements in gDNA- and full-lengthcDNA sequence databases will improve the accuracy andpower of SAGE.

Of primary importance to many potential applicationsin parasitology are technical modifications that improvethe efficiency of library production and require lessmRNA. SAGE-Lite [19] includes an additional PCRamplification step of the initial cDNA that might intro-duce amplification bias, whereas in MicroSAGE [20] theenzymatic steps are performed on mRNA bound to a solidphase. Another variant, MiniSAGE [21], has been used toprofile gene expression in human fibroblasts from 1 mgtotal RNA without extra PCR amplification. This pro-cedure uses Phase Lock Gele to purify DNA, employs alower linker concentration, which increases the efficiencyof ditag production and uses an mRNA Capture Kite tocarry out the initial steps of the SAGE protocol frommRNA isolation through to tag release (Figure 1).Processing takes place in one tube, thereby reducing thesample losses that might occur between steps. There is anote of caution, however; SAGE is technically demandingand reproducibility is an issue, especially because, typic-ally, SAGE libraries are sampled only once. In addition,laboratory contamination with PCR-amplified ditags is arisk. Recently, Invitrogen have marketed a convenient kit,in both standard and long format, that contains all the

Page 27

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005324

necessary quality control reagents and is sufficient toprepare five SAGE libraries. This kit helps address most ofthe technical problems and reduces the variation in thequality of the library, which improves the outcome ofSAGE experiments.

Gene discovery

The sequence files generated during SAGE analysis areanalyzed using SAGE2000 software (http://www.sagenet.org), which extracts the ditag sequences, checks forduplicates (a potential artefact of ditag amplification)and quantifies the individual tags. The identity of the tagscan then be sought by conventional BLAST searches. Therelative abundance of tags in different SAGE libraries canbe compared directly and subjected to rigorous statisticalanalysis. Examples of a particular SAGE tag representingmore than one gene have been observed and, in addition,more than one tag can encode a given gene if there arealternative 3 0-splice sites. Correct assignation of tags isinfluenced by factors such as sequencing errors and singlenucleotide polymorphisms, whereas PCR and cloninginefficiencies influence the abundance of individual tags,which is problematical when quantifying low-abundancetags. Therefore, it is important to verify the identity ofgenes-of-interest with independent techniques such asreal-time PCR [22], northern blots and in situ hybridiz-ation. Real-time PCR procedures [23] can also be used toextend the sequences of unknown tags, usually in the5 0 direction, and so allow their analysis by conventionalmolecular biology techniques. This is vital for parasiteprojects in which the background sequence informationavailable for homology searches is either limited ornonexistent. Thus, SAGE has potential as a powerfulgene-discovery tool to analyze host–parasite interactions.

SAGE applications in parasite biology

SAGE has been used widely in studies of human cancer,immunology, physiology and developmental biology(for review, see Ref. [24]) (http://www.sagenet.org) andit is a powerful tool for studying aspects of parasiteimmunology.

For example, the genes that are expressed in a rat mastcell line before and after stimulation with immunoglobulinE have been surveyed using SAGE [25]. This study hasidentified several novel, constitutively expressed genesand transcripts that are expressed differentially inresponse to antigen-induced clustering of the IgE receptor.Among the former are a cytokine macrophage inhibitoryfactor, neurohormone receptors and components of theexocytotic machinery [25].

Intraepithelial lymphocytes (IELs) of the gut are aprimary immune barrier against infection by orallyacquired infections, and adoptive transfer of gut IELsprotects mice against infection with Toxoplasma gondii[26]. IELs are evolutionarily conserved T cells that areimplicated in immunity to intestinal parasitism, butwhat they do and how they do it is largely unknown.SAGE analysis of IELs from murine gut has identified15 574 unique transcripts that reflect an activated butresting, Th1-skewed, cytolytic and immunoregulatoryphenotype [27].

www.sciencedirect.com

SAGE has been applied recently to several organismsincluding Plasmodium falciparum [28,29] and the free-living nematode, Caenorhabditis elegans [29]. ApplyingSAGE to the erythrocytic stages of P. falciparum,3D7-strain parasites [28,29], shows that the techniquecan be applied to AT-rich genomes without changing theenzymes that are needed to generate representativeSAGE libraries. The major metabolic pathways inPlasmodium in normal culture conditions are definedpartially by the analysis of 8335 tags that represent 4866genes. Several high-abundance tags are derived frompreviously uncharacterized open-reading frames, whichdemonstrates the potential of SAGE in genome annota-tion. Of the highly abundant tags, 17% are derived fromantisense transcripts, which has implications for thetranscriptional regulation of the Plasmodium genome[29] and, probably, the genomes of other Apicomplexa.SAGE has been used also to analyse the response of anasexual-stage population exposed to chloroquine and hasdetected the differential regulation of O100 transcripts[30]. Although some responsive loci had been implicatedalready in the mechanism of chloroquine action and/orresistance, the SAGE study [30] identified several unex-pected responses to drug exposure. For example, there is asignificant decrease in the transcripts that are involved inmitochondrial metabolism, and differential regulation ofcytoskeletal components and cell-surface markers. Inaddition, the study detected a relatively large number ofunknown ORFs.

SAGE has been used to quantify and compare geneexpression in a dauer population and non-dauer (mixed-stage) population of C. elegans, withw150 000 SAGE tagsanalyzed from each population [31]. These tags identified11 130 C. elegans genes (19 100 are predicted). Genesthat have been implicated previously in longevity areexpressed abundantly in the dauer library. However, 2016genes were detected only in the dauer library, whichindicates a complex gene-expression profile. This studyprovides evidence of alternative chromatin packaging indauer populations and identifies a novel, highly expressedgene that encodes a protein that might interact witheither telomeres or telomere-associated proteins. In addi-tion, the abundance of antisense mitochondrial tran-scripts (2% of all tags), indicates that there is anantisense-mediated regulatory mechanism in the mito-chondria of C. elegans.

SAGE has been evaluated as a tool for quantifying geneexpression in the ovine abomasal nematode H. contortusfor which there is relatively limited EST information(w30% gene coverage) [32]. The EST dataset is contiged,with the number of ESTs in clusters (which representindividual genes) available at NEMBASE2. Beforeembarking on large-scale comparisons, the study wasinitiated to establish: (i) how SAGE data and EST datafrom the same life-cycle stage compare; (ii) how easilySAGE tags can be assigned to genes when the completegenome sequence is not available; and (iii) whether it ispossible to extend the sequences of unknown SAGE tagsto facilitate their identification. In total, 2825 tagsequences were analyzed from adults harvested 28-dayspost-infection and the identity of the encoding gene

Page 28

TRENDS in Parasitology

No hit

C. elegansKey:

RibosomalHypothetical

MetabolicAntigen Others

Reproductive

Cytoskeleton

Digestive enzyme

Figure 2. The relative abundance of ESTs and SAGE tags derived fromHaemonchus

contortus, grouped according putative function. The ESTs are from library 8613 of

NEMBASE2 (http://www.nematodes.org/).

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005 325

ascribed to 63% of these tags. Furthermore, the relativeabundance of these genes, categorized arbitrarily onthe basis of function, was comparable to that in anequivalent adult EST dataset comprising 2317 ESTs(Figure 2). In addition, tag sequences could be extendedreadily and, thus, identified using a tag-based primer andreal-time PCR.

In an extensive analysis of the transcriptome ofSchistosoma mansoni [33], SAGE has been used tovalidate the conclusion from EST analysis that w50% ofall genes in S. mansoni are expressed in the adult stage.

Concluding remarks

SAGE is a powerful tool for gene discovery and defininggene expression in parasites and host immune cells. SAGEdoes not require a fully annotated genome, but this doesenhance the accuracy of the output, it is quantitative andcan be applied to small amounts of starting material.Practically, it is within the scope of laboratories withaccess to automated sequencing. Combined with a tech-nique such as laser-capture microdissection [34], SAGEmight be used to analyse comprehensively the dynamics ofparasite- and host-gene expression during the tissueinvasive stages of the parasite life-cycle. However, SAGEshould not be viewed as a stand-alone technology but asone that complements and enhances microarray experi-ments. As the use of these techniques becomes moreextensive, we need to develop methods to interrogate theoutputs of both with a common query. This will enhancethe definition of molecular pathways and interactions and,combined with proteomic approaches, will enhance under-standing of the host–parasite interaction as a system.

Acknowledgements

We thank Peter O’Shaughnessy from the University of Glasgow (UK) forhis helpful advice before commencing the work described in Ref. [32].

www.sciencedirect.com

References

1 Parkinson, J. et al. (2003) 400,000 nematode ESTs on the Net. Trends

Parasitol. 19, 283–2862 Knox, D.P. (2004) Technological advances and genomics in metazoan

parasites. Int. J. Parasitol. 34, 139–1523 Parkinson, J. et al. (2004) NEMBASE: a resource for parasitic

nematode ESTs. Nucleic Acids Res. 32, D427–D4304 Verjovski-Almeida, S. et al. (2003) Transcriptome analysis of the

acoelomate human parasite Schistosoma mansoni. Nat. Genet. 35,148–157

5 Cleary, M.D. et al. (2002) Toxoplasma gondii asexual development:identification of developmentally regulated genes and distinct pat-terns of gene expression. Eukaryot. Cell 1, 329–340

6 Li, L. et al. (2003) Gene discovery in the Apicomplexa as revealed byEST sequencing and assembly of a comparative gene database.Genome Res. 13, 443–454

7 Velculescu, V.E. et al. (2000) Analysing uncharted transcriptomes withSAGE. Trends Genet. 16, 423–425

8 Yamamoto, M. et al. (2001) Use of serial analysis of gene expression(SAGE) technology. J. Immunol. Methods 250, 45–66

9 Virlon, B. et al. (1999) Serial microanalysis of renal transcriptomes.Proc. Natl. Acad. Sci. U. S. A. 96, 15286–15291

10 Neilson, L. et al. (2000) Molecular phenotype of the human oocyte by

PCR-SAGE. Genomics 63, 13–2411 Datson, N.A. et al. (1999) MicroSAGE: a modified procedure for serial

analysis of gene expression in limited amounts of tissue.Nucleic Acids

Res. 27, 1300–130712 Schober, M.S. et al. (2001) Serial analysis of gene expression in a

single cell. Biotechnology (N. Y.) 31, 1240–124213 Ishii, M. et al. (2000) Direct comparison of GeneChip and SAGE on the

quantitative accuracy in transcript profiling analysis. Genomics 68,136–143

14 Nadon, R. and Shoemaker, J. (2002) Statistical issues with micro-arrays: processing and analysis. Trends Genet. 18, 265–271

15 Velculescu, V.E. et al. (1995) Seria1 analysis of gene expression.Science 270, 484–487

16 Patino, W.D. et al. (2002) Serial analysis of gene expression: technicalconsiderations and applications to cardiovascular biology. Circ. Res.91, 565–569

17 Saha, S. et al. (2002) Using the transcriptome to annotate the genome.Nat. Biotechnol. 20, 508–512

18 Pleasance, E. et al. (2003) Assessment of SAGE in transcriptidentification. Genome Res. 13, 1203–1215

19 Peters, D.G. et al. (1999) Comprehensive transcript analysis in small

quantities of mRNA by SAGE-lite. Nucleic Acids Res. 27, e3920 Matsumura, H. et al. (2005) SuperSAGE. Cell. Microbiol. 7, 11–1821 Datson, N.A. et al. (1999) MicroSAGE: a modified procedure for serial

analysis of gene expression in limited amounts of tissue.Nucleic Acids

Res. 27, 1300–130722 Ye, S.Q. et al. (2000) miniSAGE: gene expression profiling using serial

analysis of gene expression from 1 microgram total RNA. Anal.

Biochem. 287, 144–15223 van den Berg, A. et al. (1999) Serial analysis of gene expression: rapid

RT-PCR analysis of unknown SAGE tags. Nucleic Acids Res. 27, e1724 Tuteja, R. and Tuteja, N. (2004) Serial Analysis of Gene Expression:

Applications in Human Studies. J. Biomed. Biotechnol. 2004, 113–12025 Chen, H. et al. (1998) Characterization of gene expression in resting

and activated mast cells. J. Exp. Med. 188, 1657–166826 Buzoni-Gatel, D. et al. (1997) Adoptive transfer of gut intraepithelial

lymphocytes protects against murine infection with Toxoplasma

gondii. J. Immunol. 158, 5883–588927 Shires, J. et al. (2001) Biological insights into TCRgammadeltaC and

TCRalphabetaC intraepithelial lymphocytes provided by serialanalysis of gene expression (SAGE). Immunity 15, 419–434

28 Munasinghe, A. et al. (2001) Serial analysis of gene expression (SAGE)

in Plasmodium falciparum: application of the technique to AT richgenomes. Mol. Biochem. Parasitol. 113, 23–34

29 Patankar, S. et al. (2001) Serial analysis of gene expression inPlasmodium falciparum reveals the global expression profile oferythrocytic stages and the presence of anti-sense transcripts in themalarial parasite. Mol. Biol. Cell 12, 3114–3125

Page 29

Opinion TRENDS in Parasitology Vol.21 No.7 July 2005326

30 Gunasekera, A.M. et al. (2003) Drug-induced alterations in geneexpression of the asexual blood forms of Plasmodium falciparum.Mol.Microbiol. 50, 1229–1239

31 Jones, S.J. et al. (2001) Changes in gene expression associated withdevelopmental arrest and longevity in Caenorhabditis elegans.Genome Res. 11, 1346–1352

32 Skuce, P.J. et al. (2005) An evaluation of serial analysis of gene

Endea

the quarterly magaziand philosophy

You can access EndeScienceDirect, whecollection of beaut

articles on the historreviews and edito

featuri

Selling the silver: country house libraries and the hisCarl Schmidt – a chemical tourist in V

The rise, fall and resurrection of gMary Anning: the fossilist as

Caroline Herschel: ‘the unqScience in the 19th-century

The melancholy of an

and comin

Etienne Geoffroy St-Hillaire, Napoleon’s Egyptian camLosing it in New Guinea: The voyage o

The accidental conservatPowering the porter brFemale scientists in fi

and much, muc

Locate Endeavour on ScienceDirect

www.sciencedirect.com

expression (SAGE) in the parasitic nematode, Haemonchus contortus.Parasitology 130, 1–7

33 McManus, D.P. et al. (2004) Schistosome transcriptome analysis at thecutting edge. Trends Parasitol. 20, 301–304

34 Semblat, J.P. et al. (2002) Laser capture microdissection of Plasmo-dium falciparum liver stages for mRNA analysis. Mol. Biochem.Parasitol. 121, 179–183

vour

ne for the historyof science

avour online viare you’ll find aifully illustratedy of science, bookrial comment.

ng

tory of science by Roger Gaskell and Patricia Faraictorian Britain by R. Stefan Rossroup selection by M.E. Borelloexegete by T.W. Goodhueuiet heart’ by M. Hoskinzoo by Oliver Hochadelatomy by P. Fara

g soon

paign and a theory of everything by P. Humphriesf HMS Rattlesnake by J. Goodmanionist by M.A. Andreiewery byJ. Sumnerlms by B.A. Jones

h more . . .

(http://www.sciencedirect.com)

Page 30

Methods for assessing the burden ofparasitic zoonoses: echinococcosisand cysticercosisHelene Carabin1, Christine M. Budke2, Linda D. Cowan1, A. Lee Willingham III3,4 and

Paul R. Torgerson2

1College of Public Health, University of Oklahoma Health Sciences Center, 801 Northeast 13th Street, Room 303, Oklahoma City,

OK 73104, USA2Institute of Parasitology, University of Zurich, Winterthurerstrasse 266a, CH-8057 Zurich, Switzerland3WHO/FAO Collaborating Center for Parasitic Zoonoses, Danish Center for Experimental Parasitology, Royal Veterinary and

Agricultural University, Dyrelaegevej 100, DK-1870 Frederiksberg C, Denmark4People, Livestock and the Environment Thematic Programme, International Livestock Research Institute, PO Box 30709,

00100 Nairobi, Kenya

Glossary

Cysticercosis and echinococcosis cause illness and

productivity losses in human and agricultural animal

populations. Recent studies suggest that these diseases

have large societal impacts on endemic areas. Estimates

of burden provide essential, evidence-based data for

conducting cost–benefit and cost–utility analyses that

will secure political will, and financial and technical

resources. To evaluate the burden, the monetary and

non-monetary impacts of these zoonoses on human

health, agriculture and society must be considered

comprehensively. In this article, we review the frame-

work used to assess the burden of cysticercosis and

echinococcosis, and the data needed to estimate the

extent of the problem for societies.

Alveolar echinococcosis: an infection or disease of humans or animals caused

by the larvae of Echinococcus multilocularis.

Cystic echinococcosis: an infection or disease of humans or animals caused by

the larvae of Echinococcus granulosus.

Cysticercosis: an infection or disease of humans or animals caused by the

larvae of Taenia spp. In this article, the term refers to infection of humans or

pigs with Taenia solium cysticercosis.

Decision-tree analysis: a method of organizing epidemiological data into

infections and the frequency of their consequences.

Disability-adjusted life year: in simplest terms, this can be considered a lost

healthy year of life and is a non-monetary measure of disease burden. It takes

into account the severity of the syndrome and its duration, thus levelling the

playing field when comparing acute and chronic conditions. A DALY also has

the same value in poor and rich countries.

Disability weight: a score between 0 and 1 that is assigned to a condition

depending on the degree of debilitation.

Direct costs: costs such as carcass condemnation or medical costs arising

directly from the treatment of infection.

Health-adjusted life year: an umbrella term for a family of measures of

population health that includes, for example, DALYs and QALYs.

Indirect costs: costs such as production deficits or wage losses arising

indirectly from infection.

Metacestodosis: infection with the larval stage of a cestode.

Monte Carlo sampling technique: a method that can be employed in cost

analysis when exact estimates are unknown. Repeated samples are taken over a

probability distribution based on the known information.

Neurocysticercosis: a neurological disease caused by invasion of the CNS by

larvae of T. solium.

Quality-adjusted life year: a population measure of health. A year of full health

is equivalent to 1 QALY, whereas death corresponds to 0 QALYs. Disease

conditions are graded on a continuous scale between these two extremes.

The impact of cysticercosis and echinococcosis

Human cysticercosis (see Glossary) and echinococcosisresult in mortality, morbidity and economic losses inhuman and animal populations. Because the total societalimpact is often unknown, several initiatives have startedto assess the burden of these infections in both monetaryand non-monetary terms.

Human cystic echinococcosis (CE) (Figure 1) occurswhen a person is infected with the larval stage ofEchinococcus granulosus following ingestion of eggs andcan result in a substantial lesion, most commonly in theliver or lungs [1]. Humans are aberrant intermediatehosts for Echinococcus multilocularis, which causesalveolar echinococcosis (AE): a primary infiltrative lesionin the liver, withmetastases in advanced cases [1]. HumanCE and, particularly, AE are difficult and expensive totreat. CE in domestic animals can result in major losses ofedible offal and productivity (for review, see Ref. [2]).

Humans acquire taeniosis by eating undercooked porkthat is contaminated with cysticerci, the larval form ofTaenia solium. In low-income countries, pigs often roam

Corresponding author: Torgerson, P.R. ([email protected]).Available online 26 May 2005

www.sciencedirect.com 1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved

where people live. They eat human faeces containing eggsthat form larval cysts after migration. In humans, themetacestode can also establish itself following ingestionof eggs, and a principal site of migration is the centralnervous system (CNS), resulting in neurocysticercosis(NCC). NCC has been reported to be the most commonhelminth disease of the CNS and the most frequentpreventable cause of epilepsy in low-income countries [3].NCC can also manifest with severe headaches, blindness,hydrocephalus, chronic meningitis, symptoms due tospace-occupying CNS lesions, and dementia [4,5].

Review TRENDS in Parasitology Vol.21 No.7 July 2005

Taeniosis: an infection of humans caused by the adult stages of Taenia spp.

(in this article, T. solium specifically).

. doi:10.1016/j.pt.2005.05.009

Page 31

Figure 1. An African child with severe abdominal echinococcosis. Photograph by

Peter Schantz.

Review TRENDS in Parasitology Vol.21 No.7 July 2005328

Cysticercosis–taeniosis is extremely important in small-holder farming communities [6,7] because it exacts a tripleprice in terms of: (i) disease of humans; (ii) loss of meatproducts; and (iii) substantial reduction in the householdincome of farmers.

* H. Carabin et al., abstract 179, 53rd Annual Meeting of the American Society ofTropical Medicine and Hygiene, Miami Beach, November 2004.

Measuring morbidity and mortality rates

The first step in measuring the impact of infectious agentsis to assess their frequency of occurrence and any causalassociation with specific symptoms and/or death [8,9].Therefore, an initial distinction must be made betweenthe occurrence of infection and the occurrence of disease(or symptoms) or death among infected individuals.

The prevalence of human and pig infection has beenestimated using serology [10]. It is important to dis-tinguish tests based on antibody detection – whichmeasure current and past infections, or even detectmaternally derived antibody – from serological tests

www.sciencedirect.com

based on antigen detection, which usually represents agood measure of current infection. In most cases, however,the exact time of infection is not known. Serum antibodytests for both CE and AE are well established forindividual patient diagnosis and mass screening [11].For animal cysticercosis and echinococcosis infection,postmortem examinations can be used for diagnosis. Forporcine cysticercosis, macrolesions on the tongue can alsobe used but the sensitivity of this method is no morethan 50–55% [12].

Human metacestodosis can be diagnosed by sympto-matology or imaging techniques. However, these tech-niques count only cases presenting for treatment and,therefore, underestimate the true incidence of disease.Imaging methods such as ultrasound and computedtomography (CT) scan have also been used to estimatethe prevalence of echinococcosis andNCC, respectively [13].

To obtain better estimates of the risk (or cumulativeincidence) of infection over time or the incidence ofinfection per person–year at risk, a cohort study is needed.In this design, a group of individuals initially free frominfection is followed for a set period and new infections areidentified. Alternatively, estimates could be obtained froma nested case-control study design [14]. These studies arecostly and have not, to our knowledge, been conducted inhuman populations to estimate the incidence of infectionwith echinococcosis or cysticercosis. However, if trans-mission is thought to be stable over a period of time, theage-specific incidence of human symptomatic cases can beused as an estimate of the cumulative incidence of disease.

All measures that rely on diagnostic tests will present adegree of error that must be accounted for to estimate thereal frequency of infection [15]. Bayesian techniques canbe used to adjust for multiple sources of error and havebeen used to estimate the adjusted prevalence of porcinecysticercosis [10]. With echinococcosis in livestock, theprevalence of infection increases with age at a rate that isdependent on the prevailing infection pressure [16]. Thus,the prevalence in livestock must be stratified by age toanalyse potential production losses.

Decision-tree analyses

Information about population frequencies of infectiontypes and frequencies of associated diseases can beorganized with decision-tree analysis [17] (Figure 2). Asmore information becomes available about a disease, thedecision tree becomes more complicated. The decisiontree can be complemented further by information aboutthe average costs per case of each event represented by thebranches. The average cost per case of each event ismultiplied by the end probability of each branch. The sumof all these values corresponds to the average cost per caseof the disease of interest (Box 1). For example, the averagecost per case of measles in Canada was US$254 in 2000[18] and the cost of cysticercosis in Eastern Cape Province(South Africa) was estimated to be US$14.9 million for apopulation of 7.1 million in 2004*.

Page 32

2.3%

0.015% Death?

97.8% Disability

Surgical CE?

99.985%0.99985

Humans

Yes

Yes

Permanent

Temporary

0.000003375

0.000135628

1.09969E-05

No

No

TRENDS in Parasitology

92.5%

7.5%

0

Figure 2. Decision-analysis tree for estimating the monetary burden of CE in humans in Tunisia. Circles correspond to chance nodes (defined by the probability or incidence

rate of the event occurring) and triangles represent end nodes. The number at the top of each branch shows the proportion of each event occurring at that point in the tree.

The total proportion of cases in each group is given at the right of each branch. The values at the end of each branch represent the prevalence of this particular end-point. In

this example, the annual incidence rate of surgical CE is 0.015 per 100 person–years. Of these cases, 2.3%die. Thus, the overall probability of death in human surgical CE cases

in Tunisia corresponds to 0.015% multiplied by 2.3%, or 0.000345%.

Review TRENDS in Parasitology Vol.21 No.7 July 2005 329

Animal health costs

The direct cost of CE is the loss of edible offal (mainly liver)from infected animals. The diseased parts of the organ aretrimmed or the whole organ might be condemned. In thecase of cysticercosis, the whole carcass will typicallybe condemned at slaughterhouses [19]. However, mostpigs in poor countries are home slaughtered [20] andsome parts of the infected animals are sold, usually at alower price [19].

For CE, the largest costs might be the indirect costs,including reductions in live weight-gain, milk yield,fertility rates and value of wool or other products. Becausesome of these deficits are estimated to be R10%, theyrepresent the most serious CE-attributable losses toagriculture. These losses might be difficult to estimatebecause of limited numbers of controlled studies. Never-theless, available data (for review, see Ref. [21]) suggestthat these losses are important. In Jordan, indirect lossesrepresent up to 70% of total livestock losses that areattributable to CE [22]. AE seems to have little economiceffect on livestock, although there are reports of infectionsin farm animals [23].

Box 1. Monetary valuation of disease in both humans and

animals

The overall monetary burden of zoonoses can be calculated using

the following expression [26] (Equation I):

XSsZ1

XAaZ1

Na;sba;sXXxZ1

px ;a;sCx ;a;s

!" #(I)

This corresponds to the additive societal costs for all affected species

(S) across all age groups (A). For the age–species-specific population

of size (Na,s), with the age–species-specific annual incidence (ba,s),

there is an age–species proportion (px,a,s) of infected individuals with

symptoms x. The treatment and consequences of each of these

symptoms have a monetary burden of Cx,a,s. Ideally, the whole

spectrum of symptoms in humans and animals is included. In reality,

data about all of these elements are rarely available. A societal

approach has been used to try to value each symptom of echino-

coccosis and cysticercosis. This means that the monetary burden

will include both direct and indirect costs in humans and animals.

www.sciencedirect.com

Few studies to assess the monetary burden of swinecysticercosis have been conducted. In Mexico, economiclosses due to porcine cysticercosis represent more than50% of the national investment in swine production and,in Latin America, these losses have been estimated tobe US$164 million [24]. Another estimate from Mexico,which assumes a prevalence of 1.55% of cysticercosis inpigs, led to an estimated annual loss due to condemnationof carcasses (total or partial) of US$43 310 524 [25]. Thereare no readily available data about the effect of T. soliumon swine productivity in terms of growth, fertility and/orfecundity, and the evaluation of these would requirecontrolled experimental studies.

To calculate the animal health losses, a decision tree(Figure 2) can document different types of productiondeficit, and Monte Carlo sampling techniques representthe uncertainty associated with these data. This approachhas shown overall animal health costs per year to be US$4million to 11 million in Tunisia [26] and US$2.3 million to6.3 million in Jordan [22].

Human health costs

Direct costs associated with the diagnosis and treatmentof patients can be calculated from a representative sampleof case records and by costing the range of tests andtreatments those patients receive. In health care systemsin which the costs of diagnosis and treatment are fundedprivately, there will be charges for each intervention. In apublicly funded health service, the costs can be calculatedfrom internal auditing procedures [27].

In endemic areas, 2–4% of the population might beaffected by NCC [28]. In many poor countries, 10% of acuteneurological cases are patients with NCC [29]. In Mexico,5.4% of hospitalizations were due to NCC, and 10.3% ofautopsies undertaken in the National NeurosurgicalInstitute reported pathological evidence of NCC [30].Furthermore, NCC was the final diagnosis for 25% ofindividuals presenting for brain tumours [30]. Accordingto data collected in Peru, there would be an estimated23 512 to 39 186 inhabitants with symptomatic NCC

Page 33

Review TRENDS in Parasitology Vol.21 No.7 July 2005330

among who 18 809 to 31 349 would be associated withseizures [13]. In 1982, it was estimated that the cost perNCC patient spent by public health institutions in Mexicowas US$2173 [30]. In the USA, Roberts et al. [31]suggested that the total annual costs for hospitalization(US$6539 per case) and income loss (US$1416 per case)caused by NCC in the estimated 1100 cases per year wouldbe US$8 750 490 per year. In Mexico, estimates of wagelosses associated with NCC in 1982 were US$255 millionannually [30]. Treatment costs in Brazil have beenestimated to be approximately US$85million [31]. Severalstudies in Asia and Africa have reported the costs ofepilepsy but none has been published about the specificcosts associated with NCC [32–34]. Most estimates do notinclude the indirect costs of lost productivity, which wouldrequire case-control or, preferably, cohort studies. Inhumans, it has been reported that 69–96% of symptomaticNCC cases have one or more seizures, and some willdevelop epilepsy. Epilepsy is a syndrome with consider-able social, psychological, economical and physical impacton a community, and patients with epilepsy suffer from adecreasing quality of life according to the frequency oftheir seizures [35]. The mean annual cost for all medicalservices associated with one epilepsy case, estimated froma study conducted in northeastern USA, was US$9617 [36].

Mean direct costs of treating a case of human echino-coccosis have been calculated to be US$524 in Jordan andUS$10 215 in the UK [22,27]. The differences reflect thedifferent wealth of the two nations and, hence, the costs ofgoods and services. Indirect costs might be considerable.These include the long-term ill health of individuals whohave been treated for echinococcosis or who are affectedby undiagnosed disease. The mortality rate for CE isgenerally reported to be 1–2% in cases that undergosurgical treatment but it is much higher for AE [1]. Thecase fatality rate of NCC has been recorded as 5.8% inCalifornia (USA) [37] and 6.7% in Oregon (USA) [38].Fatal cases should be costed, and the value of human lifehas been calculated in several ways. The capital approachis the easiest to calculate and equates the value of lifewith the present value of future lost output (as proxiedby earnings and other labour costs). Alternatively, the‘willingness to pay’ approach [39] looks at the maximumamount of money that an individual is willing to pay forreduction of mortality risk.

Uncertainty and sensitivity analyses

When estimating the national societal burden of infection,one must find estimates of the average parameters ofinterest. If the estimates of these parameters from fieldstudies are biased, adding a variance around the pointestimate will not correct the bias and a correction (withuncertainty) based on judgment might need to be done.For example, cross-sectional studies tend to be conductedin endemic areas in which the prevalence estimates willrepresent the upper limit of the national prevalence.There is also a large amount of uncertainty related to theproportion of infected cases that will develop differentsymptoms or experience productivity losses. This is truefor animal and human cases. Finally, the monetary valuesattributed to the diagnosis and treatment of each symptom

www.sciencedirect.com

and, for example, the daily salary of a farmer or homemakerin a low-income country that would be lost with a reductionin productivity are virtually always uncertain.

Monte Carlo or Latin Hypercube methods can be usedto model uncertain parameters. The distribution of theseparameters must be selected with care. When there isno knowledge about a parameter, a uniform distributioncan be selected. More-informative distributions can beselected when more data become available. For example,when the prevalence is estimated, a beta distribution canbe used that takes into account the sample size and thenumbers affected to model the true prevalence in thepopulation. The overall monetary impact of CE in Tunisiawas estimated to be US$14.7 million per annum [95%credible interval (CI) US$10.4 million to 19.0 million]using uniform probability distributions, and US$10.7million (95% CI US$3.4 million to 18.8 million) whendefined distributions were used [26]. In Jordan, theoverall annual monetary impact of CE was estimated tobe US$3.8 million (95% CI US$2.6 million to 6.5 million)using this analytical technique [22]. Even when there is ahigh degree of uncertainty, such an approach can be usefulbecause the lower limits of the stochastic output of the costestimates provide an upper limit to the allowable costs fora proposed control program.

Non-financial approaches for estimating the burden

of disease

Non-financial methods have been developed to estimatehuman disease burden and are often referred to under theumbrella term ‘HALYs’ (health-adjusted life years) [40].HALYs are summary measures of population health thatenable the combined impact of morbidity and death to beconsidered simultaneously. These methods assess themorbidity of disease states so that comparisons betweendifferent diseases can be made. They avoid costing theeffects of disease financially but can be used to prioritizeresources to control human disease. However, the use ofHALYs has several disadvantages (Box 2) and might notalways be appropriate when estimating the societalburden of zoonoses.

Disability-adjusted life years (DALYs) are the preferreddisease-burdenmeasure of theWorld Health Organization(WHO; http://www.who.int/en/) and were first constructedfor the Global Burden of Disease (GBD) study [41,42]. Toquantify DALYs, disability weights (Table 1) are assignedto each morbidity adjusted for their duration. Forexample, a healthy individual has a disability weight of0 and no loss of DALYs, whereas a fatal condition has aweight of 1. The number of DALYs lost is calculated forthe remaining life expectancy at the age of onset of thecondition. The effectiveness of intervention strategies iscalculated as being the number of DALYs estimated tooccur because of a given condition in the absence ofintervention minus the number of DALYs expected ifcontrol measures were implemented. However, thisdisregards additional benefits to agriculture of diseasecontrol (e.g. anthelmintic treatment of dogs reducesechinococcosis incidence in both humans and sheep).Nevertheless, if a total societal financial analysis isundertaken, the true cost effectiveness of control, in

Page 34

Box 2. A critical review of burden-assessment methods

HALYs† Non-financial approaches ignore the agricultural impact of zoonotic

infections. Although cost sharing can apportion burdens between the

health and agricultural sectors, it needs a total financial analysis to

enable allotment of HALYs and, hence, estimate cost sharing for cost-

effectiveness analysis [43].

† Disability classes (used in DALYs) do not provide adequate

information for evaluating the impact or true burden of disease [49].

DALYs can undervalue disability in low-income countries. For

example, the differences in quality of life for paraplegics in impover-

ished African countries compared with those in an industrialized

nation are disregarded [50].

† Different types of HALY give different results. For example, there

were differences in estimates of disease burden for five common

medical conditions made using both DALYs and QALYs, and in rank

order of the illnesses [51].

† HALYs do not give priority to those who are worse off, they

discriminate against people with limited treatment potential and they

do not account for qualitative differences in outcomes (e.g. life saving

versus health improving) [40].

† A proportion of non-infectious or chronic conditions due to

infections might not be recognized, thus underestimating the disease

burden of infectious diseases [52].

† The practice of discounting future life – implying that a life saved

today will be worth multiple lives saved in the future – might be

interpreted as bias and/or morally inappropriate [49].

† Weighting disabilities for infections with delayed symptoms might

not be straightforward if based on cross-sectional data rather than

cohort data.

† Other limitations of DALYs can be found in Ref. [49].

Monetary burden† The cost of treating each end-component of the decision tree (see

Figure 2 in main text) can be extremely variable, difficult to obtain and

confounded by the widespread practice of bartering in poor countries.

† The value of labour and health costs vary between different

economies, whereas DALYs are a uniform measure and, therefore,

should have the same value in poor and rich countries [40].

† Productivity losses for children, homemakers and the unemployed

can be difficult to estimate. However, studies of lymphatic filariasis in

India have estimated working-time losses and costed them as either

wage losses or opportunity losses for non-economically active

workers [53,54]. From these, total costs were estimated.

† Conditions such as epilepsy involve stigmatization and social iso-

lation [55]. Assigning a monetary value to this is not straightforward

and might be captured more easily with HALY measures.

Review TRENDS in Parasitology Vol.21 No.7 July 2005 331

terms of DALYs saved, can be estimated by implemen-tation of cost sharing between sectors proportional to theoverall benefit of each sector [43]. Neither humancysticercosis nor echinococcosis was evaluated in theGBD study. Research to address this deficit is nowunderway for human echinococcosis and, in the future,the burden of disease due to cysticercosis could becalculated using similar methodology.

Quality-adjusted life years (QALYs) and DALYs aresimilar conceptually but use different scales. For example,QALYs rank 1 as perfect health and 0 as death and,thus, have a reverse scale compared with DALYs. There-fore, interventions would aim to minimize DALYs butmaximize QALYs.

Echinococcosis and DALYs

The potential impact of disease on afflicted individualsmust be taken into consideration when constructing aDALY. Two studies suggested that patients surgicallytreated for CE had a significant decrease in quality of life[22,27]. Furthermore, in Kyrgyzstan, patients presentingfor treatment of CE had twice the unemployment rate ofthe general population [44]. The SF-12v2e health survey,

Table 1. Examples of disability weights used to calculate DALYsa

Degree of morbidity

Healthy

Limited ability to perform at least one activity in one of the following areas

occupation

Limited ability to perform most activities in one of the following areas: rec

occupation

Limited ability to perform most activities in two or more of the following are

occupation

Limited ability to perform most activities in all of the following areas: recre

occupation

Requires assistance with instrumental activities of daily living such as mea

Requires assistance with activities of daily living such as eating, personal h

DeadaThe societal burden of disease, in terms of DALYs, takes into account the prevalence

distribution and gender distribution of disease, and the life expectancy at the time of di

www.sciencedirect.com

which is a generic measure of general health andwellbeing, has recently been used in China to evaluatethe extent of morbidity due to previously undiagnosedechinococcosis [45]. The results demonstrate significantdeficits in all health categories of subjects diagnosed withechinococcosis compared with healthy gender- and age-matched controls. Further studies have demonstratedthat subjects with echinococcosis are also more likely to bein a lower income bracket than cross-matched healthycontrols [46]. These results justify the use of DALYs tocalculate human disease burden.

Disability weights were constructed for AE and CEbased on reported disability weights for liver cancer:a disease of similar symptomatology. DALYs were thencalculated using these weights, with the likely outcomesbased on several different reports. The outcomes of thedisease were assigned probabilities in a multinomialdistribution according to the observed outcomes in thelimited numbers of trials with chemotherapy (because freealbendazole treatment was the only therapy option avail-able to this population). Disability weights were assignedusing a weight of 0.200 (Dutch weight for clinicallydisease-free cancer) for an improved outcome [47], a

Disability weight

0

: recreation, education, procreation or 0.096

reation, education, procreation or 0.220

as: recreation, education, procreation or 0.400

ation, education, procreation or 0.600

l preparation, shopping or housework 0.810

ygiene or toilet use 0.920

1

of disease, the duration (at the disability weight defined for that condition), age

agnosis. Data from Ref. [41].

Page 35

Review TRENDS in Parasitology Vol.21 No.7 July 2005332

disability weight of 0.239 (the GBD weight for pre-terminal liver cancer) for those with stable disease, anda disability weight of 0.809 (the GBD weight for terminalliver cancer) for disease that worsened. Fatal cases wereassigned a disability weight of 1.

In the calculation of DALYs lost, stochasticity wasintroduced because of the uncertainty of assigning dis-ability weights and the uncertainty in the point preva-lence estimates. Monte Carlo sampling techniques wereused (see earlier). Using these methods, the total numbersof DALYs lost because of echinococcosis were estimated tobe 50 933 (95% CIs 41 995 to 61 026) in Shiqu county(China). This consisted of w32 978 (95% CIs 25 019 to42 422) DALYs lost because of AE and w17 955 (14 268 to22 1238) DALYs lost because of CE, and suggests anaverage of w0.81 DALYs lost per person in Shiqu County[45]. By comparison, the GBD study suggested that theaverage number of DALYs lost per person because ofinfectious and non-infectious diseases wasw0.18 in Chinaas a whole. This demonstrates that, in localized highlyendemic areas, echinococcosis can be among the leadingcauses of illness. In addition, using a cost-sharingapproach for controlling echinococcosis, the cost perDALYaverted in China – attributable to the public healthsector – is approximately US$10–12 [46], which is withinthe most cost-effective band of the WHO of less thanUS$25 per DALYaverted [48]. In addition to these DALYsaverted, the costs of intervention to the livestock sectorare, at most, only 25% of the potential economic benefits toanimal health [46].

Concluding remarks

Cysticercosis and echinococcosis contribute to high levelsof human morbidity and some mortality, and livestockproduction losses in many parts of the world. Control ofthese zoonoses should be prioritized because they arepreventable diseases. There are examples of effectivecontrol of these diseases; porcine cysticercosis has beeneliminated from parts of the industrialized world, andechinococcosis has been eliminated from Iceland, Tasmaniaand New Zealand. In addition, recent studies suggest thatthe control of echinococcosis could be extremely costeffective in highly endemic areas such as China. Inmany endemic areas, a lack of accurate estimates ofdisease burden hampers the setting of priorities forcontrol of infectious diseases because of finite resources.Progress in assessing the burden of echinococcosis andcysticercosis has been reviewed and some difficulties inburden measurement have been highlighted. In thisarticle, we have also indicated where more research isneeded to improve the estimates of disease burden ofparasitic zoonoses in both humans and animals.

AcknowledgementsWe thank INTAS (INTAS 01 500, 01 505 and 03 51 5661), the NIH, the USNational Science Foundation (1R01TW01565–01) and the WHO for theirfinancial support. We also thank Dirk Engels, Meghan Majorowski andTheodore Nash for helpful discussions.

References

1 Pawlowski, Z.S. et al. (2001) Echinococcosis in humans: clinicalaspects, diagnosis and treatment. In WHO/OIE Manual on

www.sciencedirect.com

Echinococcosis in Humans and Animals: a Public Health Problem ofGlobal Concern (Eckert, J. et al., eds), pp. 20–71, World Organisationfor Animal Health

2 Torgerson, P.R. (2003) Economic effects of echinococcosis. Acta Trop.85, 113–118

3 Roman, G. et al. (2000) A proposal to declare neurocysticercosisan international reportable disease. Bull. World Health Organ. 78,399–406

4 Prabhakar, S. et al., eds (2002). Taenia solium Cysticercosis. FromBasic to Clinical Science, CABI Publishing

5 Dixon, H.B.F. and Lipscomb, F.M. (1961) Cysticercosis: an analysisand follow-up of 450 cases. In Medical Research Council SpecialReport Series 299, pp. 1–58, HM Stationary Office

6 Mafojane, N.A. et al. (2003) The current status of neurocysticercosisin Eastern and Southern Africa. Acta Trop. 87, 25–34

7 Phiri, S. et al. (2003) The emergence of Taenia solium cysticercosisin Eastern and Southern Africa as a serious agricultural problem andpublic health risk. Acta Trop. 87, 13–24

8 Merson, M.H. et al., eds (2001) International Public Health. Diseases,

Programs, Systems, and Policies, Aspen Publishers9 Thomas, J.C. and Weber, D.J., eds (2001) Epidemiologic Methods for

the Study of Infectious Diseases, Oxford University Press10 Dorny, P. et al. (2003) Immunodiagnostic tools for human and porcine

cysticercosis. Acta Trop. 87, 79–8611 Eckert, J. and Deplazes, P. (2004) Biological, epidemiological, and

clinical aspects of echinococcosis, a zoonosis of increasing concern.Clin. Microbiol. Rev. 17, 107–135

12 Pouedet, M.S. et al. (2002) Epidemiological survey of swine cysticer-cosis in two rural communities of West-Cameroon. Vet. Parasitol. 106,45–54

13 Bern, C. et al. (1999) Magnitude of the disease burden fromneurocysticercosis in a developing country. Clin. Infect. Dis. 29,1203–1209

14 Neutral, R.R. and Drolette, M.E. (1978) Estimating exposure-specificdisease rates from case-control studies using Bayes’ theorem. Am.J. Epidemiol. 108, 214–222

15 Basanez, M-G. et al. (2004) Bayesian statistics for parasitologists.Trends Parasitol. 20, 85–91

16 Torgerson, P.R. and Health, D.D. (2003) Transmission dynamicsand control options of Echinococcus granulosus. Parasitology 127,S143–S158

17 Haddix, A.C. et al., eds (2003) Prevention Effectiveness. A Guide toDecision Analysis and Economic Evaluation, Oxford University Press

18 Carabin, H. et al. (2002) The average cost of measles cases and adverseevents following vaccination in industrialised countries. BMC PublicHealth. DOI: 10.1186/1471-2458-2-22 (www.biomedcentral.com)

19 Zoli, A. et al. (2003) Regional status, epidemiology and impact ofTaenia solium cysticercosis in Western and Central Africa. Acta Trop.87, 35–42

20 Joshi, D.D. et al. (2003) Improving meat inspection and controlin resource-poor communities: the Nepal example. Acta Trop. 87,119–127

21 Torgerson, P.R. et al. (2000) Estimating the economic effects of cysticechinococcosis: Uruguay, a developing country with upper-middleincome. Ann. Trop. Med. Parasitol. 94, 703–713

22 Torgerson, P.R. et al. (2001) Estimating the economic effects of cysticechinococcosis. Part 3: Jordan, a developing country with lower-middle income. Ann. Trop. Med. Parasitol. 95, 595–603

23 Deplazes, P. and Eckert, J. (2001) Veterinary aspects of alveolarechinococcosis – a zoonosis of public health significance. Vet. Parasitol.98, 65–87

24 Schantz, P.M. et al. (1993) Potential eradicability of taeniasis andcysticercosis. Bull. Pan Am. Health Organ. 27, 397–403

25 Acevedo-Hernandez, A. (1982) Economic impact of porcine cysticer-cosis. In Cysticercosis: Present State of Knowledge and Perspectives(Flisser, A. et al., eds), pp. 63–68, Academic Press

26 Majorwoski, M.M. et al. (2005) Echinococcosis in Tunisia: a costanalysis. Trans. R. Soc. Trop. Med. Hyg. 99, 268–278

27 Torgerson, P.R. and Dowling, P.M. (2001) Estimating the economiceffects of cystic echinococcosis. Part 2: an endemic region in the UnitedKingdom, a wealthy industrialized economy. Ann. Trop. Med. Para-sitol. 95, 177–185

Page 36

Review TRENDS in Parasitology Vol.21 No.7 July 2005 333

28 Goodman, K.A. et al. (1999) Case-control study of seropositivity forcysticercosis in Cuenca, Ecuador. Am. J. Trop. Med. Hyg. 60, 70–74

29 Garcia, H.H. et al. (1995) Factors associated with Taenia soliumcysticercosis: analysis of nine hundred forty-six Peruvian neurologicpatients. Am. J. Trop. Med. Hyg. 52, 145–148

30 Velasco-Suarez, M. et al. (1982) Human cysticercosis: medical–socialimplications and economic impact. In Cysticercosis: Present State ofKnowledge and Perspectives (Flisser, A. et al., eds), pp. 47–51,Academic Press

31 Roberts, T. et al. (1994) Economic losses caused by food-borne parasiticdiseases. Parasitol. Today 10, 419–423

32 Nsengyumva, G. et al. (2003) Cysticercosis as a major risk factor forepilepsy in Burundi, East Africa. Epilepsia 44, 950–955

33 Thomas, S.V. et al. (2001) Economic burden of epilepsy in India.Epilepsia 42, 1052–1060

34 Sawhney, I.M. et al. (1996) Evaluation of epilepsy management in adeveloping country: a prospective study of 407 patients. Acta Neurol.

Scand. 94, 19–2335 van Hout, B. et al. (1997) Relationship between seizure frequency and

costs and quality of life of outpatients with partial epilepsy in France,Germany, and the United Kingdom. Epilepsia 38, 1221–1226

36 Griffiths, R.I. et al. (1999) Payer costs of patients diagnosed withepilepsy. Epilepsia 40, 351–358

37 Sorvillo, F.J. et al. (1992) Cysticercosis surveillance: locallyacquired and travel related infections and detection of intestinaltapeworm carriers in Los Angeles County. Am. J. Trop. Med. Hyg. 47,365–371

38 Townes, J.M. et al. (2004) Neurocysticercosis in Oregon, 1995–2000.Emerg. Infect. Dis. 10, 508–510

39 Gyrd-Hansen, D. (2003) Willingness to pay for a QALY. Health Econ.12, 1049–1060

40 Gold, M.R. et al. (2002) HALYs and QALYs and DALYS, oh my:similarities and differences in summary measures of populationhealth. Annu. Rev. Pub. Health 23, 115–134

41 Murray, C.J.L. (1994) Quantifying the burden of disease: the technicalbasis for disability-adjusted life years. Bull. World Health Organ. 72,429–445

Five things you might no

1.

Elsevier is a founder member of the WHO’s HINARI and AGORA ini

access to scientific literature. More than 1000 journals, including the T

at significantly re

2.

The online archive of Elsevier’s premier Cell Press journal collection w

recent archive, including Cell,Neuron, Immunity and Current Biology

sites 12 months after artic

3.

Have you contributed to anElsevier journal, book or series?Did you kn

stand-alone CDs when ordered directly from us?

+1 800 782 4927 (US) or +1 800 460 3110

or +44 1865 474 010 (

4.

Elsevier has a long tradition of liberal copyright policies and formany y

and the posting of final papers on internal servers. Now, Elsevier has

the final text version of their papers on both their person

5.

The Elsevier Foundation is a knowledge-centered foundationmaking

culturally rich global organization, the Foundation has funded, for ex

Philadelphia, provided storybooks to children in Cape Town, sponsor

Brigham and Women’s Hospital and given funding to the 3rd Intern

www.sciencedirect.com

42 Murray, C.J.L. and Lopez, A.D., eds (1996) The Global Burden ofDisease: A Comprehensive Assessment of Mortality and DisabilityfromDisease, Injuries, and Risk Factors in 1990 and Projected to 2020,Harvard University Press

43 Roth, F. et al. (2003) Human health benefits from livestock vaccinationfor brucellosis: case study. Bull. World Health Organ. 81, 867–876

44 Torgerson, P.R. et al. (2003) Human cystic echinococcosis in Kyrgy-stan: an epidemiological study. Acta Trop. 85, 51–61

45 Budke, C.M. et al. (2004) Utilization of DALYs in the estimation ofthe disease burden of echinococcosis for a high endemic region of theTibetan plateau. Am. J. Trop. Med. Hyg. 71, 56–64

46 Budke, C.M. et al. Economic effects of echinococcosis on a highlyendemic region of the Tibetan plateau. Am. J. Trop. Med. Hyg.(in press)

47 Stouthard, M.E.A. et al. (2000) Disability weights for diseases:a modified protocol and results for a Western European region. Eur.J. Pub. Health 10, 24–30

48 World Health Organization (1996) Report of the Ad Hoc Committeeon Health Research Relating to Future Intervention Options(TDR/Gen/96.1), World Health Organization

49 Anand, S. and Hanson, K. (1997) Disability-adjusted life years:a critical review. J. Health Econ. 16, 685–702

50 Allotey, P. et al. (2003) The DALY, context and the determinants ofthe severity of disease: an exploratory comparison of paraplegia inAustralia and Cameroon. Soc. Sci. Med. 57, 949–958

51 Gold, M.R. and Muennig, P. (2002) Measure-dependent variation inburden of disease estimates: implications for policy. Med. Care 40,260–266

52 Zou, S. (2001) Applying DALYs to the burden of infectious diseases.Bull. World Health Organ. 79, 267–268

53 Babu, B.V. and Nayak, A.N. (2004) Treatment costs and work time lossdue to episodic adenolymphangitis in lymphatic filariasis patientsin rural communities of Orissa, India. Trop. Med. Int. Health 8,1102–1109

54 Ramaiah, K.D. et al. (2000) The economic burden of lymphaticfilariasis in India. Parasitol. Today 16, 251–253

55 Baker, G.A. (2002) The psychosocial burden of epilepsy. Epilepsia 43(Suppl. 6), 26–30

t know about Elsevier

tiatives, which enable the world’s poorest countries to gain free

rends and Current Opinion collections, will be available for free or

duced prices.

ill become freely available from January 2005. Free access to the

, will be available on both ScienceDirect and the Cell Press journal

les are first published.

ow that all our authors are entitled to a 30%discount on books and

For more information, call our sales offices:

(Canada, South & Central America)

rest of the world)

ears has permitted both the posting of preprints on public servers

extended its author posting policy to allow authors to freely post

al websites and institutional repositories or websites.

grants and contributions throughout the world. A reflection of our

ample, the setting up of a video library to educate for children in

ed the creation of the Stanley L. Robbins Visiting Professorship at

ational Conference on Children’s Health and the Environment.

Page 37

Comparative folate metabolism inhumans and malaria parasites (part II):activities as yet untargeted or specificto PlasmodiumAlexis Nzila1,2,3, Steve A. Ward3, Kevin Marsh4,5, Paul F.G. Sims6 and John E. Hyde6

1Kenya Medical Research Institute and Wellcome Trust Collaborative Research Program, Wellcome Trust Research Laboratories,

PO Box 43640, Nairobi GPO 00100, Kenya2Department of Pharmacology and Therapeutics, University of Liverpool, Liverpool, UK, L69 3BX3Liverpool School of Tropical Medicine, Pembroke Place, Liverpool, UK, L53 QA4Kenya Medical Research Institute and Wellcome Trust Collaborative Research Program, Centre for Geographical Medicine

Research, PO Box 230, Kilifi, Kenya5Nuffield Department of Medicine, John Radcliffe Hospital, Oxford, UK, OX3 9DU6Faculty of Life Sciences, University of Manchester, PO Box 88, Manchester, UK, M60 1QD

The folate pathway represents a powerful target for

combating rapidly dividing systems such as cancer cells,

bacteria and malaria parasites. Whereas folate meta-

bolism in mammalian cells and bacteria has been

studied extensively, it is understood less well in malaria

parasites. In two articles, we attempt to reconstitute the

malaria folate pathway based on available information

from mammalian and microbial systems, in addition to

Plasmodium-genome-sequencing projects. In part I, we

focused on folate enzymes that are already used

clinically as anticancer drug targets or that are under

development in drug-discovery programs. In this article,

we discussmammalian folate enzymes that have not yet

been exploited as potential drug targets, and enzymes

that function in the de novo folate-synthesis pathway of

the parasite – a particularly attractive area of attack

because of its absence from the mammalian host.

Folate enzymes

Folate derivatives (FDs) are important cellular cofactorsinvolved in supplying one-carbon (C1) units for threemajor metabolic pathways: the biosynthesis of (i) meth-ionine, (ii) purines and (iii) pyrimidines; pathways (ii) and(iii) are essential for DNA generation. A C1 unit is alsorequired for the initiation of protein synthesis in mito-chondria through formylation of methionine. Rapidlydividing cells such as tumors, bacteria and malariaparasites rely heavily on the availability of FDs fortheir growth. Thus, the inhibition of enzymes involved inthese processes greatly affects cell division, throughinhibition of DNA and protein synthesis. This featurehas been exploited for the development of antifolatedrugs against cancer cells and microbial infections,including malaria.

Corresponding author: Nzila, A. ([email protected]).

www.sciencedirect.com 1471-4922/$ - see front matter Q 2005 Elsevier Ltd. All rights reserved

Folate is a generic term that comprises nine FDs –folic acid (FA), dihydrofolate (DHF), tetrahydrofolate(THF), 5,10-methenyltetrahydrofolate (5,10-CHC-THF),5,10-methylenetetrahydrofolate (5,10-CH2-THF),5-methyltetrahydrofolate (5-CH3-THF), 5-formyltetra-hydrofolate (5-CHO-THF), 10-formyltetrahydrofolate(10-CHO-THF) and 5-formiminotetrahydrofolate(5-NHaCH-THF) – that are found in mammalian andmicrobial cells [1]. In this article, we discuss themammalian folate enzymes that have not yet beenexploited as targets for drug discovery and their statusin malaria parasites, in addition to enzymes of the de novofolate-synthesis pathway, which are not found in mam-malian cells. As in part I [1], we have exploited malarial,bacterial and yeast genome information to identifyputative malaria candidate enzymes that have not yetbeen described. The aim of collating this informationis to provide a realistic and useful model of the likelycomposition of the malaria folate pathway and a firmerbasis for future evaluation of potential drug targets.

Folate enzymes not yet targeted in cancer studies and

their status in Plasmodium

The folate enzymes described in this section have not yetbeen tested as potential targets in tumor cells. Althoughmost of these enzymes have regulatory roles in folatemetabolism, some are involved in the synthesis andmetabolism of amino acids (methionine, glutamate andhistidine), making them good potential targets for drugdevelopment. We consider their role in mammalianmetabolism and the available evidence of their existencein Plasmodium.

Methylenetetrahydrofolate reductase and methionine

synthase

Methylenetetrahydrofolate reductase (MTHFR)(EC 1.5.1.20) mediates one of the most important reactions

Review TRENDS in Parasitology Vol.21 No.7 July 2005

. doi:10.1016/j.pt.2005.05.008

Page 38

Review TRENDS in Parasitology Vol.21 No.7 July 2005 335

in folate metabolism: synthesizing 5-CH3-THF from5,10-CH2-THF (reaction 13 in Figure 2 of Ref. [1]).Thereafter, the methyl group of 5-CH3-THF is transferredby methionine synthase (MS) (EC 1.16.1.8) to homo-cysteine, to generate methionine (reaction 14 in Figure 2of Ref. [1]). These reactions are the sole source ofmethionine, which, besides its other roles, functions asthe precursor of S-adenosylmethionine, the methyl-groupdonor in w100 reactions [2]. MS from humans (amongother organisms) requires a form of vitamin B12 (cyano-cobalamin) as a cofactor to which the methyl group is firsttransferred from 5-CH3-THF to form methylcyanoco-balamin, before transfer to homocysteine. Deficiency ofMTHFR activity, as a result of point mutations in its gene,is associated with a decrease in the remethylation ofhomocysteine, leading to hyperhomocystemia: a defectwith serious and diverse clinical consequences [3,4]. To thebest of our knowledge, no attempt has been made to usethis enzyme as an antifolate target.

5-CH3-THF is the most prevalent FD in Plasmodiumfalciparum [5] and in human serum. Although radiolabel-ing studies show that salvage of 5-CH3-THF by theparasite occurs [6], there is also biochemical evidence ofthe presence of both MTHFR and MS [6,7]. Thus,significant levels of MTHFR activity were detected inthree Plasmodium species [6], and MS was partiallypurified and characterized from extracts of P. falciparum[7]. Because P. falciparum can be cultured with normalgrowth rates in methionine-depleted medium [6], it canclearly derive this amino acid from hemoglobin degra-dation and/or by de novo synthesis. However, unlike thehost, the parasite is not necessarily dependent uponMTHFR for supply of the 5-CH3-THF that is required formethionine synthesis because adequate levels of thiscofactor are normally present in host plasma. A relativelack of importance of this activity to the parasite would beconsistent with the failure of basic local alignment searchtool (BLAST) searches, using a wide range of prokaryoticand eukaryotic probes, to detect an MTHFR ortholog inPlasmodium, but this apparent absence conflicts with thebiochemical data. Moreover, the parasite requires MS, andthe failure of similar searches for the gene encoding thisactivity, using both cobalamin-dependent and -indepen-dent enzyme sequences as probes, is perhaps more likelyto indicate that the plasmodial enzymes are highlydivergent rather than absent. If so, they might representvaluable targets after they are identified.

10-Formyltetrahydrofolate dehydrogenase

10-Formyltetrahydrofolate dehydrogenase (FTHFD)(EC 1.5.1.6) in mammals consists of two functionaldomains: a hydrolase that removes the formyl groupfrom 10-CHO-THF, and an NADP(C)-dependent dehydro-genase that oxidizes this group to CO2 [8,9] (reaction 24bin Figure 2 of Ref. [1]). FTHFD is important for theregulation of 10-CHO-THF pools in purine synthesisand for the removal of formate. The hydrolase activity isalso found in bacteria (EC 3.5.1.10; also known as10-formyltetrahydrofolate deformylase), in which it pro-vides the major source of formate for the synthesis of5 0-phosphoribosyl-N-formylglycinamide in the purine

www.sciencedirect.com

pathway. This enzyme type has not been reported inmalaria parasites, and BLAST searches with bacterialdeformylase sequences and the mammalian dehydro-genases provide no evidence of its presence, which,again, is consistent with the dependence of Plasmodiumon purine salvage.

Glutamate formiminotransferase and

formiminotetrahydrofolate cyclodeaminase

Glutamate formiminotransferase (GFT) (EC 2.1.2.5)and formiminotetrahydrofolate cyclodeaminase (FCD)(EC 4.3.1.4), expressed as a single polypeptide in allknown organisms, mediate consecutively two importantreactions in the metabolism of histidine and glutamate.GFT catalyzes the synthesis of 5-NHaCH-THF andglutamate from THF and formiminoglutamate (FiGlu)(reaction 16 in Figure 2 of Ref. [1]). FiGlu is a product ofhistidine metabolism; thus, these reactions control bothhistidine and glutamate levels, in addition to supplying aC1 unit to the folate pathway. The 5-NHaCH-THFproduced by GFT is further converted to 5,10-CHC-THFin the presence of FCD (reaction 17 in Figure 2 of Ref. [1]),which is then converted to either 10-CHO-THF or5,10-CH2-THF, both of which are C1 donors. Thus, therole of this part of the pathway is to provide an additionalsource of such groups. The 3D structure of this enzymecomplex has been resolved [10,11]; however, so far, nostudies have been devoted to the screening of GFT or FCDinhibitors. It could be that the malaria parasite does notrequire the extra capacity to provide C1 groups that thiscomplex provides because no gene from any of thePlasmodium databases is identified in BLAST searchesusing either bacterial or vertebrate GFT–FCD probes.

Methenyltetrahydrofolate synthetase

Methenyltetrahydrofolate synthetase (MTHFS), alsoknown as 5-formyltetrahydrofolate cycloligase(EC 6.3.3.2), catalyzes the irreversible ATP-dependentconversion of 5-CHO-THF to 5,10-CHC-THF (reaction 18in Figure 2 of Ref. [1]). The reverse reaction, leading tothe synthesis of 5-CHO-THF, is carried out by serinehydroxylmethyltransferase (SHMT) (reaction 19 inFigure 2 of Ref. [1]), the same enzyme involved in reaction9 of this Figure, in which 5,10-CH2-THF is generated fromserine and THF. As mentioned previously, 5-CHO-THFdoes not seem to have a major biological role in the cell[12], so MTHFS is not considered a good target for drugdevelopment. However, 5-CHO-THF, also known as folinicacid or leucovorin, is used as an adjuvant with antifolatesto increase their therapeutic index in the treatment ofcancer. MTHFS is the sole enzyme that enables theincorporation of leucovorin into the folate pathway; thisreaction is, therefore, crucial in cancer therapy for thereduction of toxicity to normal cells of the antifolateinhibitor [13].

Although MTHFS has not yet been reported in malariaparasites, and BLAST searches with bacterial and planthomologs fail to identify any candidate genes, radiolabeledfolinic acid is taken up and processed efficiently by theparasite, providing a much better source of exogenousfolate in culture than the usual supplementation with folic

Page 39

Review TRENDS in Parasitology Vol.21 No.7 July 2005336

acid [14,15]. This is consistent with earlier observationsthat the in vitro activity of some antimalarial antifolatedrugs tested in the presence of 5-CHO-THF is decreased[16,17]. In the apparent absence of an MTHFS-encodinggene, it remains to be determined how 5-CHO-THF is usedso efficiently by the parasite. Conceivably, host erythro-cytes could convert 5-CHO-THF to other FD forms such asDHF or THF that could be salvaged by the parasite.

Glycine-cleavage system

The glycine-cleavage system (GCV), or the glycinedecarboxylase complex (GDC), is a tetrafunctional enzymecomplex found in mitochondria that consists of P-protein(glycine decarboxylase; EC 1.4.4.2), L-protein (lipoamidedehydrogenase; EC 1.8.1.4), H-protein (carrier of thelipoamide chain) and T-protein (THF aminomethyltrans-ferase; EC 2.1.2.10). These mediate four consecutivereactions that oxidatively cleave glycine and, with THF,lead to the production of CO2, NH3 and 5,10-CH2-THFwith electron transfer to produce NADH (reaction 10 inFigure 2 of Ref. [1]). Thus, this system affects the synthesisof a C1-donor group, as does the SHMT-mediated conver-sion of serine to glycine (see later). However, several linesof evidence indicate that the primary role of the GCV is tocontrol the metabolism of glycine [18]. On this basis,inhibition of the enzyme system would perturb glycinemetabolism, which could lead to the blocking of cell growth.

In P. falciparum, genes encoding significantly similarproducts for all but one of the proteins of the GCV can beidentified, and transcripts from them are detected inmature forms from in vitro cultures [19]. Genes encoding aP. falciparum T-protein (PF13_0345) and an H-protein(PF11_0339) were identified by protein similarities of48%–56% to relevant eukaryotic orthologs (e.g. yeast,Arabidopsis thaliana, Anopheles gambiae and human).However, a gene encoding a product with significantsimilarity to known P-proteins has yet to be found. Despitethis, the existence of a GCV complex in P. falciparum isfurther indicated by the discovery of two different genesthat encode products similar to the L-protein, one with ahigh signal-peptide score that indicates a mitochondrialisoform (PFL1550w), and the other with a strong signa-ture peptide for the apicoplast (PF08_0066): a situationthat mirrors the mitochondrial and chloroplast forms seenin plants. The relative importance of the SHMT and GCVenzymes has not been addressed directly in either bac-terial or eukaryotic systems. However, the shmt gene ispresent in genomes from which gcv genes are absent and itis part of the minimal gene set of genomes with the lowestcoding capacity sequenced to date.

Dimethylglycine dehydrogenase and sarcosine

dehydrogenase

Dimethylglycine dehydrogenase (DGDH) (EC.1.5.99.2)and sarcosine dehydrogenase (SDH) (EC 1.5.99.1) (reac-tions 11 and 12 in Figure 2 of Ref. [1]) catalyze the transferof one C1 unit, for the synthesis of 5,10-CH2-THF, fromdimethylglycine and methylglycine (sarcosine), respect-ively. Both enzymes use THF as a cofactor, and the productof the first reaction, sarcosine, is the substrate of thesecond. These enzymes are expressed exclusively in

www.sciencedirect.com

mitochondria, in which they have important roles incholine metabolism and as key components in the glycine–sarcosine cycle that regulates the ratio S-adenosylhomo-cysteine:S-adenosylmethionine. This ratio is believed tobe important for modulating the plethora of transmethyl-ation reactions that involve S-adenosylmethionine as themethyl group donor [2].

In P. falciparum, BLAST searches yield an ambiguouspicture because bacterial and eukaryotic probes for bothDGDH and SDH give relatively strong (and, often, sole)hits (with probability values up to 2e-13) to the same gene,PF13_0345, the product of which is annotated as having35% identity to 92% of the mitochondrial precursor of theGCV T-protein described previously. However, BLASTsearches identified only the C-terminal half of thePF13_0345 product as having sequence similarity to theDGDH and SDH probes, indicating that assignment ofPF13_0345 as being a gene that encodes a GCV T-proteinis likely to be a more reliable prediction. Considering thatone of the other key genes of the GCV seems to be missing,however (see earlier), it cannot be excluded that theparasite might lack a functioning GCV and that theproduct of PF13_0345 functions in the demethylationreactions of modified glycine rather than in C1 transferfrom the methylamine group of the H-protein conjugate.These alternatives require biochemical testing.

Enzymes of de novo folate synthesis

Apart from that involved in reaction 1 in Figure 2 of Ref.[1], all enzymes of de novo folate synthesis (reactions 2–6of Figure 2 of Ref. [1]) are specific to malaria parasitesbecause the host cannot synthesize folate de novo. There-fore, this part of the pathway could offer excellent targetsfor antimalarial drug development and, indeed, thedihydropteroate synthase (DHPS) (EC 2.5.1.15) activity(reaction 5 of Figure 2 of Ref. [1]) has long been exploitedas the target of the sulfa drugs (sulfonamides and sulfones).

GTP cyclohydrolase I

The first enzyme of folate synthesis, GTP cyclohydrolase I(GTPC) (EC 3.5.4.16), catalyzes the conversion of GTP todihydroneopterin triphosphate (DHNP) (reaction 1 inFigure 2 of Ref. [1]) and has a crucial role in humanphysiology. DHNP is not used for folate production inmammalian cells but it is the precursor for the synthesisof neopterin and biopterin derivatives. The reduced formof biopterin, tetrahydrobiopterin (THB), is a cofactor forthe synthesis of nitric oxide, which is a messengerinvolved in the regulation of many reactions and thepathology of several diseases. THB is also a cofactor in thehydroxylation of the aromatic amino acids (phenylalanine,tyrosine and tryptophan) [20,21].

GTPC activity has been measured in Plasmodiumspecies [22], and the gene characterized from P. falciparum[23] and other malarial species (GenBank accessionnumbers AF043557, AF486639, AF486640, AY582138,AY604168 and AY458431). In P. falciparum, transcriptionpeaks during the early trophozoite stage, which isconsistent with its role in the synthesis of folate molecules.In other microorganisms, this enzyme controls de novosynthesis of folate and regulates the cell cycle [24] and,

Page 40

Review TRENDS in Parasitology Vol.21 No.7 July 2005 337

therefore, might be a good target for drug development.However, no systematic search for GTPC inhibitors hasbeen made so far. Attempts to knock out the gene encodingthis protein in P. falciparum are in progress, using atransfection system that has been developed specificallyto explore the folate pathway [25] and that has beenemployed successfully to disable DHPS activity [14]. Thesuccess of these experiments would provide validation ofGTPC as a potentially useful drug target.

Dihydroneopterin aldolase

DHNP, the triphosphate product of the GTPC reaction, ishydrolyzed to the free hydroxyl form; this reaction hasbeen postulated to involve a non-enzymatic loss of pyro-phosphate followed by nonspecific phosphatase activitythat removes the third phosphate group [26] (reaction 2of Figure 2 of Ref. [1]). The resulting substrate is con-verted to 2-amino-4-hydroxy-6-hydroxymethyldihydrop-terin in the presence of dihydroneopterin aldolase(DHNA) (EC 4.1.2.25) (reaction 3 of Figure 2 of Ref. [1]).Because radiolabeled GTP or guanosine precursors areultimately converted to folate in the parasite [22], theexistence of DHNA as the mediator of a key step along thispathway would seem to be mandatory, yet its identifi-cation, at both gene and protein level, remains elusive.Despite considerable efforts both experimentally andin silico, we have, as yet, been unable to identify aplausible candidate. At present, the possibilities are(i) that this protein, which is poorly conserved in otherorganisms, is so divergent in Plasmodium that it isunrecognizable, even using bioinformatics tools that areconsiderably more sophisticated than BLAST; (ii) that itsgene is fragmented into too many small exons for openreading frame (ORF) prediction programs to handle; or(iii) that it is absent from the parasite, which is less likely.

Hydroxymethyldihydropterin pyrophosphokinase and

dihydropteroate synthase

Hydroxymethyldihydropterin pyrophosphokinase (PPPK,or HPPK) (EC 2.7.6.3) catalyzes the diphosphorylation of2-amino-4-hydroxy-6-hydroxymethyldihydropterin (reac-tion 4 in Figure 2 of Ref. [1]). The resulting compound thencondenses with p-aminobenzoate (PABA) to generatedihydropteroate, mediated by DHPS (reaction 5 inFigure 2 of Ref. [1]). PABA can be obtained either byde novo synthesis through the shikimate pathway or bysalvage from the host plasma. PPPK and DHPS occur as abifunctional protein in Plasmodium [27,28], other proto-zoa and plants. Together with dihydrofolate reductase(DHFR), DHPS is a longstanding target of choice forantimalarial antifolates. The sulfa drugs are structuralanalogs of PABA and their inhibition of DHPS functions inpotent synergy with anti-DHFR inhibitors, justifyingtheir use in the antimalarial antifolate combinationstypified by sulfadoxine–pyrimethamine (SP) and Lapdapw[1]. Similar to the situation for the dhfr gene [1], a smallnumber of mutations in dhps (principally in codons 437,540 and 581) contributes to the observed clinical resist-ance of parasites to SP [29,30]. Although DHPS is a goodtarget for further drug development, there have been fewstudies on the identification of new antimalarial DHPS

www.sciencedirect.com

inhibitors. Several compounds are being analyzed aspotent anti-DHFR agents but the development of thesecompounds will require their use in combination withother inhibitors to increase potency further and to helpretard the onset of clinical resistance. Novel anti-DHPSinhibitors could fulfill this role.

Dihydrofolate synthase

Dihydrofolate synthase (DHFS) (EC 6.3.2.12) catalyzesthe conversion of dihydropteroate to dihydrofolate byaddition of a single L-glutamate moiety; this reaction(reaction 6 of Figure 2 of Ref. [1]) is the final step inde novo folate synthesis. The gene encoding P. falciparumDHFS has been characterized [23,31] and is expressed asa bifunctional protein that also exhibits folylpolygluta-mate synthase (FPGS) activity, which adds furtherglutamate residues to the molecule [1]. So far, no attempthas been made to target DHFS activity in the malariaparasite but this protein could provide a powerful targetfor new drugs. Unlike the bifunctional molecules DHFR–TS(thymidylate synthase) and PPPK–DHPS, in which dis-crete domains carry the individual activities, both theDHFS and FPGS activities are likely to be mediatedby residues that are distributed throughout the entiremolecule [31]. Thus, successful inhibitors are likely toblock both activities, and investigations of the inhibition ofDHFS activity are currently underway.

Concluding remarks

In the two parts of this review (this article and Ref. [1]), wehave systematically examined the enzymes that catalyzereactions of the folate pathway to the extent that thispathway is understood in other organisms, in particularthe human host of the malaria parasite P. falciparum.With regard to both anticancer and antimalarial agents,we have categorized actual and potential drug targets interms of their current clinical or experimental status.From this comprehensive analysis, we argue that, of the23 enzyme activities described from mammalian systems,only six [GTPC, DHFR, TS, SHMT, FPGS and methionyl-tRNA formyltransferase (MTFT)] can be ascribed to themalaria parasite with confidence. Another three (PPPK,DHPS and DHFS) are found only in the parasite and,based on biochemical evidence, three more are thought tobe present (DHNA, MTHFR and MS) but have not yetbeen identified at the gene or protein level. A further twoenzyme activities (the T- and L-proteins of the GCV) seemto be part of an incomplete pathway (Table 1).

Thus, we can conclude that the most promising noveltargets at present, based purely on their firm assignment,would be TS, DHFS–FPGS, SHMT, GTPC, PPPK and,possibly, MTFT. The sequence relationship (expressed asamino acid identities) of those with human orthologsdecreases in the order TS (50%)OSHMT (43%)OGTPC(17%)OFPGS (16%)OMTFT (11%) [23]. By comparison,DHFR has an identity of 25% with its host ortholog andprovides a paradigm for the successful development ofinhibitors with marked differential binding to the host andparasite molecules. These values are, admittedly, onlycrude indicators of potential use because, ideally, such

Page 41

Table 1. Folate-associated enzymes in Plasmodiuma

Enzyme EC number Gene locus in

Plasmodium

falciparumb

GenBank

accession

numberc

Reaction in

Figure 2 of

Ref. [1]

GTPC 3.5.4.16 PFL1155w AF043557 1

PPPK–DHPS 2.7.6.3–

2.5.1.15

PF08_0095 Z31584 4 and 5

DHFS–FPGS 6.3.2.12–

6.3.2.17

PF13_0140 AF161264 6 and 7

DHFR–TS 1.5.1.3–

2.1.1.45

PFD0830w J03028 8 and 15

SHMT 2.1.2.1 PFL1720w AF195023 9

MTFT 2.1.2.9 MAL13P1.67 CAD52276 21

MTHFR 1.5.1.20 ?d None 13

MS 1.16.1.8 ?d None 14

GCV T-protein 2.1.2.10 PF13_0345 CAD52774 10

GCV H-protein None PF11_0339 AAN35923 10

GCV L-protein 1 1.8.1.4 PFL1550w AAN36396 10

GCV L-protein 2 1.8.1.4 PF08_0066 CAD51214 10aTable lists enzymes for which there is definite or reasonable evidence from

experimental and/or sequence analysis.b3D7 clone (see http://plasmodb.org).cMultiple accession numbers for different strains of P. falciparum and different

Plasmodium species are not shown.dActivities reported (see main text) but genes not yet identified.

Review TRENDS in Parasitology Vol.21 No.7 July 2005338

comparisons should be between the regions of themolecule to which the inhibitors bind.

Of these parasite activities, TS, DHFR, SHMT, FPGSand MTFT have been considered in detail in the contextof their mammalian counterparts [1]. However, all of thereactions, except the first, involved in de novo folatesynthesis are absent from host cells. In bacterial systems,in which such synthesis is the norm, mutations ordeletions of genes in this pathway generally result innonviability. Moreover, crystal structures exist for all ofthe enzymes involved. GTPC is regarded as being apotentially good target for rational drug design, althoughthere are concerns that the well-conserved active sitewould be too similar to that of human GTPC for effectivediscrimination [32]. Inhibitors of Escherichia coli DHNAhave been investigated [33] but, as described previously,this enzyme is still unidentified in Plasmodium. PPPK isalso seen as being a particularly attractive target [32] andinhibitors based on binding to its two substrate pocketshave been described [34]. In addition to the clinicallyvalidated sulfa drugs, high-potency pterin analogs havebeen developed against DHPS [35], and inhibitors ofDHFS have been investigated in E. coli [36] and Neisseria[37]. Again, with the latter enzyme, careful product designwould be required to achieve sufficient selectivity withrespect to human FPGS because of the likely similaritiesin active-site geometries. Overall, however, this area ofthe folate pathway should represent potentially fertileground for antimalarial drug discovery. A counterargu-ment would be that, unlike many bacteria, the parasitecan salvage folate from external sources. This is known toantagonize DHPS inhibitors strongly when tested in vitro[17,38] but the extent to which salvage from the host canmeet the needs of the parasite in vivo is still uncertain.Reinforcing the view that this area of folate metabolismshould be investigated more thoroughly, recent transfec-tion studies indicate that blockage of biosynthesis cannotbe compensated for completely by salvage of exogenousfolate and, thus, an element of de novo synthesis, possibly

www.sciencedirect.com

located in a separate cellular compartment, seems to beessential for parasite growth [14]. Moreover, by reducingthe uptake of exogenous folate by including a transportinhibitor such as probenecid [39,40], the effectiveness ofthe traditional antifolates can be much enhanced and thisrepresents a promising area for further exploration. Wealso note that, after GTPC, all subsequent enzymes inthe de novo pathway use pterin derivatives as substratesbut efforts to test pterin analogs as potential inhibitorsof the different steps in folate synthesis are relativelyrare [41], possibly owing to the generally poor solubilityof these compounds.

Although determining the genome sequences ofP. falciparum and the other Plasmodium species hasbeen useful – for example, indicating the likely presence ofan MTFT activity that might be exploited [1] – furtherfunctional studies are required before the picture of thefolate-pathway components can be considered complete.However, it seems clear that the principal role of thispathway in these parasites is geared to the production ofthymidine because, of the additional roles seen inmammals and other organisms, purine synthesis is absentand amino acids are derived mostly from hemoglobindegradation and salvage from the host. The enzymes thatare involved in thymidine biosynthesis are well definedand it is here that short-term to medium-term effortsshould be applied to validate these targets and assesstheir potential use.

AcknowledgementsWe thank Enrique Salcedo for communicating unpublished research ofthe glycine-cleavage system. This work was supported by the WellcomeTrust (grants 056769, 056845, 062372, 067201 and 073896), BBSRC(grant 36/JE616379) and the NIH (Fogarty International grant TW01186). A.N. and K.M. thank the Wellcome Trust for personal support.S.A.W. thanks the Wellcome Trust for institutional support. We apologizeto those authors whose work was not cited because of space limitations.

References

1 Nzila, A. et al. Comparative folate metabolism in humans and malariaparasites (part I): pointers for malaria treatment from cancerchemotherapy. Trends Parasitol. DOI:10.1016/j.pt.2005.04.002 (www.sciencedirect.com)

2 Wagner, C. (1995) Biochemical role of folate in cellular metabolism. InFolate in health and disease (Bailey, L.B., ed.), pp. 23–42, MarcelDekker

3 Schwahn, B. and Rozen, R. (2001) Polymorphisms in the methylene-tetrahydrofolate reductase gene: clinical consequences. Am.J. Pharmacogenomics 1, 189–201

4 Bailey, L.B. and Gregory, J.F., III. (1999) Polymorphisms of methyl-enetetrahydrofolate reductase and other enzymes: metabolic signifi-cance, risks and impact on folate requirement. J. Nutr. 129, 919–922

5 Krungkrai, J. et al. (1989) De novo and salvage biosynthesis ofpteroylpentaglutamates in the human malaria parasite, Plasmodiumfalciparum. Mol. Biochem. Parasitol. 32, 25–37

6 Asawamahasakda, W. and Yuthavong, Y. (1993) The methioninesynthesis cycle and salvage of methyltetrahydrofolate from host redcells in the malaria parasite (Plasmodium falciparum). Parasitology107, 1–10

7 Krungkrai, J. et al. (1989) Characterization of cobalamin-dependentmethionine synthase purified from the human malarial parasite,Plasmodium falciparum. Parasitol. Res. 75, 512–517

8 Chumanevich, A.A. et al. (2004) The crystal structure of the hydrolasedomain of 10-formyltetrahydrofolate dehydrogenase: mechanism ofhydrolysis and its interplay with the dehydrogenase domain. J. Biol.Chem. 279, 14355–14364

Page 42

Review TRENDS in Parasitology Vol.21 No.7 July 2005 339

9 Krupenko, S.A. and Wagner, C. (1999) Aspartate 142 is involved inboth hydrolase and dehydrogenase catalytic centers of 10-formyl-tetrahydrofolate dehydrogenase. J. Biol. Chem. 274, 35777–35784

10 Kohls, D. et al. (1999) Crystallization and preliminary X-ray analysisof the formiminotransferase domain from the bifunctional enzymeformiminotransferase-cyclodeaminase. Acta Crystallogr. D Biol. Crys-tallogr. 55, 1206–1208

11 Kohls, D. et al. (2000) The crystal structure of the formiminotrans-ferase domain of formiminotransferase-cyclodeaminase: implicationsfor substrate channeling in a bifunctional enzyme. Structure Fold.Des. 8, 35–46

12 Anguera, M.C. et al. (2003) Methenyltetrahydrofolate synthetaseregulates folate turnover and accumulation. J. Biol. Chem. 278,29856–29862

13 Barredo, J. et al. (1999) Folates as chemotherapeutic modulators. InAntifolate Drugs in Cancer Therapy (Jackman, A., ed.), pp. 323–337,Humana Press

14 Wang, P. et al. (2004) Transfection studies to explore essential folatemetabolism and antifolate drug synergy in the human malariaparasite Plasmodium falciparum. Mol. Microbiol. 51, 1425–1438

15 Wang, P. et al. (2004) Genetic and metabolic analysis of folate salvagein the human malaria parasite Plasmodium falciparum. Mol.Biochem. Parasitol. 135, 77–87

16 Kinyanjui, S.M. et al. (1999) The antimalarial triazine WR99210 andthe prodrug PS-15: folate reversal of in vitro activity againstPlasmodium falciparum and a non-antifolate mode of action of theprodrug. Am. J. Trop. Med. Hyg. 60, 943–947

17 Wang, P. et al. (1999) Utilization of exogenous folate in the humanmalaria parasite Plasmodium falciparum and its critical role inantifolate drug synergy. Mol. Microbiol. 32, 1254–1262

18 Kisliuk, R.L. (1999) Folate biochemistry in relation to antifolateselectivity. In Antifolate Drugs in Cancer Therapy (Jackman, A., ed.),pp. 13–36, Human Press

19 Le Roch, K.G. et al. (2004) Global analysis of transcript and proteinlevels across the Plasmodium falciparum life cycle. Genome Res. 14,2308–2318

20 Hobbs, A.J. et al. (1999) Inhibition of nitric oxide synthase as apotential therapeutic target. Annu. Rev. Pharmacol. Toxicol. 39,191–220

21 Kaufman, S. (1993) New tetrahydrobiopterin-dependent systems.Annu. Rev. Nutr. 13, 261–286

22 Krungkrai, J. et al. (1985) Guanosine triphosphate cyclohydrolase inPlasmodium falciparum and other Plasmodium species. Mol. Bio-chem. Parasitol. 17, 265–276

23 Lee, C.S. et al. (2001) Characterization of three genes encodingenzymes of the folate biosynthetic pathway in Plasmodium falci-parum. Parasitology 122, 1–13

24 Witter, K. et al. (1996) Cloning, sequencing and functional studies ofthe gene encoding human GTP cyclohydrolase I. Gene 171, 285–290

25 Wang, P. et al. (2002) Rapid positive selection of stable integrantsfollowing transfection of Plasmodium falciparum. Mol. Biochem.Parasitol. 123, 1–10

www.sciencedirect.com

26 De Saizieu, A. et al. (1995) Enzymic characterization of Bacillussubtilis GTP cyclohydrolase I. Evidence for a chemical dephosphory-lation of dihydroneopterin triphosphate. Biochem. J. 306, 371–377

27 Triglia, T. and Cowman, A.F. (1994) Primary structure and expressionof the dihydropteroate synthetase gene of Plasmodium falciparum.Proc. Natl. Acad. Sci. U. S. A. 91, 7149–7153

28 Brooks, D.R. et al. (1994) Sequence variation of the hydroxymethyldi-hydropterin pyrophosphokinase: dihydropteroate synthase gene inlines of the human malaria parasite, Plasmodium falciparum, withdiffering resistance to sulfadoxine. Eur. J. Biochem. 224, 397–405

29 Wang, P. et al. (1997) Resistance to antifolates in Plasmodiumfalciparum monitored by sequence analysis of dihydropteroatesynthetase and dihydrofolate reductase alleles in a large number offield samples of diverse origins. Mol. Biochem. Parasitol. 89, 161–177

30 Triglia, T. et al. (1998) Allelic exchange at the endogenous genomiclocus in Plasmodium falciparum proves the role of dihydropteroatesynthase in sulfadoxine-resistant malaria. EMBO J. 17, 3807–3815

31 Salcedo, E. et al. (2001) A bifunctional dihydrofolate synthetase–folylpolyglutamate synthetase in Plasmodium falciparum identifiedby functional complementation in yeast and bacteria. Mol. Biochem.Parasitol. 112, 239–252

32 Bermingham, A. and Derrick, J.P. (2002) The folic acid biosynthesispathway in bacteria: evaluation of potential for antibacterial drugdiscovery. Bioessays 24, 637–648

33 Zimmerman, M. et al. (1977) Inhibitors of folate biosynthesis. 1.Inhibition of dihydroneopterin aldolase by pteridine derivatives.J. Med. Chem. 20, 1213–1215

34 Shi, G. et al. (2001) Bisubstrate analogue inhibitors of 6-hydroxy-methyl-7,8-dihydropterin pyrophosphokinase: synthesis and bio-chemical and crystallographic studies. J. Med. Chem. 44, 1364–1371

35 Lever, O.W., Jr. et al. (1986) Inhibitors of dihydropteroate synthase:substituent effects in the side-chain aromatic ring of 6-{[3-(aryloxy)-propyl]amino}-5-nitrosoisocytosines and synthesis and inhibitorypotency of bridged 5-nitrosoisocytosine-p-aminobenzoic acid ana-logues. J. Med. Chem. 29, 665–670

36 Ho, R.I. et al. (1976) A simple radioassay for dihydrofolate synthetaseactivity in Escherichia coli and its application to an inhibition study ofnew pteroate analogs. Anal. Biochem. 73, 493–500

37 Pongsamart, S. et al. (1984) Characterization and inhibition ofdihydrofolate synthetase from Neisseria gonorrhoeae. Mol. Cell.Biochem. 59, 165–171

38 Watkins, W.M. et al. (1985) Antagonism of sulfadoxine and pyri-methamine antimalarial activity in vitro by p-aminobenzoic acid,p-aminobenzoylglutamic acid and folic acid. Mol. Biochem. Parasitol.14, 55–61

39 Nzila, A. et al. (2004) Therapeutic potential of folate uptake inhibitionin Plasmodium falciparum. Trends Parasitol. 20, 109–112

40 Nzila, A. et al. (2003) Chemosensitization of Plasmodium falciparumby probenecid in vitro. Antimicrob. Agents Chemother. 47, 2108–2112

41 Guiney, D. et al. (2003) Syntheses of highly functionalised6-substituted pteridines. Org. Biomol. Chem. 1, 664–675

Page 43

Emerging technologies for thedetection and genetic characterizationof protozoan parasitesPaul T. Monis1, Steven Giglio1, Alexandra R. Keegan1 and R.C. Andrew Thompson2

1Australian Water Quality Centre, South AustralianWater Corporation, Private Mail Bag 3, Salisbury, South Australia 5108, Australia2WHO Collaborating Centre for the Molecular Epidemiology of Parasitic Infections, Veterinary and Biomedical Sciences,

Murdoch University, Murdoch, Western Australia 6150, Australia

The development and adaptation of new technologies

for the genetic characterization and identification of

parasites continue to accelerate, providing an increasing

number of research and analytical tools. We review

emerging technologies that have applications in this

area, including real-time PCR and microarrays, and

discuss the fundamental principles of some of these

technologies and how they are applied to characterize

parasites. We give special consideration to the appli-

cation of genetic data to biological questions, where

selection of the most appropriate technique depends on

the biological question posed by the investigator.

Assessing technologies

Technologies are being developed rapidly to detect avariety of analytes in clinical, environmental and bio-security settings. The application of some of thesetechnologies to parasites and microbial pathogens hasbeen reported [1,2], but information on emerging tech-nologies is difficult to access because development of thetechnology is on-going and, in many cases, commerciallysensitive. In this review, we provide an overview ofemerging technologies that have applications for thedetection and genetic characterization of parasites, andoutline the principles by which some of the technologiesoperate. We focus on more mature technologies that areavailable commercially, rather than those that are indevelopment and not available to the wider researchcommunity. Two broad areas are covered in detail,amplification-based techniques and hybridization-basedtechniques. We give a brief overview of new technologiesthat will have potential applications if they become readilyaccessible. In addition, we consider how the resulting datacan be applied appropriately to answer biological questions.

Amplification techniques and applications

PCR has enabled advances in many areas of research andit is a fundamental platform technology for geneticidentification and characterization. This technique hasbeen used for almost 20 years and warrants no furtherdescription because it is covered in detail in basic mol-ecular biology textbooks. Several alternative strategies to

Corresponding author: Monis, P.T. ([email protected]).

www.sciencedirect.com 1471-4922/$ - see front matter. Crown Copyright Q 2005 Published b

amplify nucleic acids [e.g. nucleic-acid-sequence-basedamplification (NASBA), ligase chain reaction (LCR), stranddisplacement amplification, Q-b replicase-mediated ampli-fication and linear linked amplification] have been deve-loped [3], but none is used as widely as PCR. The mostlikely reasons for this are cost and ease of use, with mostalternative systems lacking the simplicity and cost-effectiveness of PCR. Of the alternatives, NASBA andLCR have clear advantages over PCR for particular appli-cations. Possibly the greatest revolution in amplificationtechnology since the advent of PCR has been the develop-ment of systems that monitor amplification in real-time.

Real-time amplification

Developed in the early 1990s [4], real-time PCR providesresearchers and diagnostic laboratories with additionaltools for disease diagnosis, identification of species,quantifying gene expression and monitoring infectionloads during therapy. The availability and flexibility ofthe technology was limited originally, but within the lastsix years instrumentation has become more reliable,flexible and affordable, resulting in widespread uptake ofthis technology. All current real-time-detection systemsdetect the amplification of either DNA or RNA usingfluorescent chemistries, most of which are describedbelow. The specific requirements of the different chem-istries must be considered during the design and develop-ment of the assay, and detailed information on currentinstrumentation, detection chemistries and assay designis included a recently published textbook [5].

The primary advantages of real-time PCR over conven-tional PCR are that it provides high-throughput analysisin a closed-tube format (no post-PCR handling isrequired); that it can be used for quantitation over abroad dynamic range; and that it can be used to dif-ferentiate DNA fragments by analyzing the melting curveof DNA. Quantitation exploits the proportional relation-ship between the threshold cycle at which exponentialamplification is detected (Ct) and the starting number ofcopies of the target nucleic-acid fragment. The amplifica-tion of appropriate DNA standards enables the construc-tion of a standard curve and estimation of the gene-copynumber from the Ct of an unknown sample [6]. Forsome detection chemistries, amplified fragments can be

Review TRENDS in Parasitology Vol.21 No.7 July 2005

y Elsevier Ltd. All rights reserved. doi:10.1016/j.pt.2005.05.012

Page 44

TRENDS in Parasitology

Key:

R

R

R

R

RR

R

R

R

R

R

(a) (b)

(c) (d)

(e) (f)

(g) (h)

(i) (j)

(k) (l)

R

Fluorescingreportermolecule

Fluorescingintercalatingdye

Donormolecule

Minor-groovebinder

Intercalatingdye

Quenchingmolecule

Reportermolecule

Taq DNApolymerase

Figure 1. Commonly used, real-time PCR chemistries. Intact Taqman probes do not

fluoresce because of the proximity of the reporter and quencher molecules (a) but

they produce a fluorescent signal following hydrolysis by Taq polymerase, which

releases the reporter molecule (b). The secondary structure of MGB Eclipse probes

holds the quencher and reporter molecules in close proximity so that they do not

fluoresce (c) but, once bound to target DNA, the secondary structure is lost, and the

quencher and reporter are separated sufficiently to enable fluorescence (d). The

stem-loop structure of molecular beacons hold the quencher and reporter in close

proximity, which prevents fluorescence when not bound to the target DNA (e) but

enables fluorescence once binding occurs (f). FRET probes anneal to target

sequences and energy transfer from the donor molecule to the reporter molecule

(g) increases fluorescence of the reporter molecule (h). Intercalating dyes used in

iFRET do not fluoresce unless bound to double-stranded DNA and do not function

as energy donors in the unbound state (i). Energy transfer occurs following

intercalation of the dye and binding of the iFRET probe, resulting in fluorescence (j).

Unbound dsDNA-specific intercalating dyes are poorly fluorescent (k) but produce

a large increase in fluorescence after binding to double-stranded DNA (l).

Review TRENDS in Parasitology Vol.21 No.7 July 2005 341

characterized further by analysis of the DNA meltingcurve, which measures the dissociation kinetics of eitherthe entire amplified fragment (in the case of intercalatingdyes) or the probe–target hybrid. Plotting the firstderivative of the melting curve versus temperatureenables the melting temperature of the product to bedetermined, which is affected by the GC content of thefragment and the absolute order of the bases in thesequence. This detects genetic variation in products inwhich the number of base differences relative to the size ofthe fragment is sufficient to affect the melting temperature.Generally, larger fragments requiremorebasedifferences toproduce a detectable difference in melting temperature,which is why probe-based chemistries are better fordetection of single nucleotide polymorphisms (SNPs).

Real-time detection chemistries

Taqman probes (hydrolysis probes and 5 0 nuclease assay)[7] are one of the earliest real-time chemistries to bedeveloped, producing fluorescence following hydrolysis byTaq DNA polymerase. These probes have a 5 0 fluorescentreporter and a 3 0 quencher molecule. The proximity ofthe reporter and quencher prevents fluorescence by thereporter when it is exposed to the appropriate wavelengthof light. The probe binds to target DNA during PCRannealing (Figure 1a) and it is hydrolyzed by theexonuclease activity of the DNA polymerase during DNAsynthesis at 60 8C (Figure 1b). This separates the reporterand quencher, and enables detection of the reporter. Theincrease in signal after each cycle is cumulative, thus,unlike other real-time detection chemistries, Taqmanprobes enable signal amplification in addition to DNAamplification provided by PCR. More recently, Taqmanprobes have been enhanced by incorporating a minor-groove-binding molecule at the 3 0 end of the probe. Thisimproves binding of the primer (by increasing its meltingtemperature) and enables the use of shorter probes, whichincreases the efficiency [8]. Taqman assays have been usedto detect numerous parasites including Cryptosporidium[9,10],Giardia [11], Leishmania infantum [12],Myxoboluscerebralis [13], Plasmodium falciparum [14] and Toxo-plasma gondii [15]. Taqman probes are one of the mostwidely used chemistries because assay design is relativelysimple and assays are generally robust. However, Taqmanprobes do not confirm that the correct fragment has beenamplified (other than by running a gel). As a result, it isimportant to validate the specificity of primer–probecombinations. Taqman assays can be multiplexed readilyby using probes with different colored fluorophores; themaximum number of assays that can be multiplexeddepends on the instrument used (an upper limit of six isavailable on some instruments).

In contrast to Taqman probes, minor-groove binder(MGB) Eclipse probes [16] have a quencher and minor-groove-binding molecule at the 5 0 end and a reportermolecule at the 3 0 end. In solution, MGB Eclipse probes donot fluoresce (Figure 1c), but when a probe anneals totarget DNA it unfolds, creating sufficient distancebetween the reporter and quencher to enable fluorescence(Figure 1d). Unlike Taqman probes, MGB Eclipse probesare not hydrolyzed because of the MGB at the 5 0 end.

www.sciencedirect.com

Molecular beacons [17] have a similar mode of action toMGB Eclipse probes, but a stem-loop structure is used tohold the reporter and quencher in close proximity whenthe probe is not bound to its target (Figure 1e,f ). It isclaimed that MGB Eclipse probes are more stable andhave a better signal-to-noise ratio than molecular beacons(http://www.epochbio.com/products/mgbe_how_it_works.htm). Both MGB Eclipse probes and molecular beaconscan be used for DNA-melting-curve analysis. Althoughreports of the use of MGB Eclipse probes are limited, they

Page 45

TRENDS in Parasitology

Primer withT7 promoter 5′

AMV reversetranscriptase,RNase H

RNA

cDNA withT7 promoter

Antisense RNA

AMV reverse transcriptase,RNase H

T7 RNA polymerase

Figure 2. In NASBA, either mRNA or rRNA (red wavy line) is converted to cDNA

(blue) that contains a T7 promoter (yellow). This cDNA functions as a template for

the production of many copies of antisense RNA (blue wavy lines). Each transcript

is a template for the production of additional cDNA templates.

Review TRENDS in Parasitology Vol.21 No.7 July 2005342

are useful for SNP typing [18]. Molecular beacons havebeen used for several pathogens such as Bordetella [19]and Hepatitis C [20], particularly in combination withNASBA [21], but have only been applied to Plasmodium[22] in terms of parasites.

Fluorescence- (or Forster) resonance-energy-transfer(FRET)-based assays [23] rely on the energy transferbetween a donor fluorophore and a reporter fluorophore,which are located 3 0 and 5 0 respectively, on separateprobes (Figure 1g,h). Fluorescence is detected only whenthe probes hybridize adjacent to each other on the targetDNA. FRET probes can be used for both quantitative PCRand DNA-melting-curve analysis and are particularlyuseful for species discrimination and SNP analysis.iFRET [24] is a variation of FRET that uses an inter-calating dye specific for double-stranded DNA as the donorfluorophore, and a single probe with a reporter molecule(Figure 1i,j). iFRET seems to have advantages over FRETin terms of cost and signal strength [24] but has not beenadopted widely. FRET assays have been used to detectToxoplasma [25], to detect and discriminate betweenspecies and genotypes of Cryptosporidium [26], and todetect, differentiate and quantitate Leishmania [27].Although a powerful tool, FRET assays are more cumber-some to optimize than Taqman assays and are compro-mised by too many starting copies of target DNA, whichaffects annealing of the probes and prevents accuratequantitation and DNA-melting-curve analysis.

Intercalating dyes that are specific for double-strandedDNA are possibly the most commonly used real-time PCRchemistry. SYBR Green I [28] is the current industrystandard, but the use of dyes such as BEBO [29] and LCGreen [30] has been described. Intercalating dyes are themost cost-efficient chemistry and, thus, are the mostpopular option among researchers. Although relativelycheap, SYBRGreen is difficult to use because it can inhibitPCR [31] and is difficult to optimize. In addition, SYBRGreen seems to have limited use for multiplex PCRbecause of preferential binding to specific amplicons [32].Intercalating dyes detect any double-stranded DNA. Thisis an advantage because they can be used for any assay,but a disadvantage because all nonspecific amplicons andprimer-dimers are detected. However, amplification of thecorrect fragment can be confirmed by DNA-melting-curveanalysis, and nonspecific products excluded from analysisby raising the acquisition temperature above the meltingtemperature of the nonspecific products. Numerous appli-cations of SYBR Green I have been reported, includinghigh-throughput screening for antimalarial drugs [33],genotyping Cryptosporidium parvum [34] and detectingLeishmania [35] and Trypanosoma brucei [36].

NASBA

NASBA is an isothermal amplification method that ismodeled on viral replication and makes use of reversetranscriptase, RNA polymerase and target-specific pri-mers with an added T7 promoter sequence [37]. Thiscombination of enzymes and primers targets a specificRNA transcript and produces a DNA fragment with a T7promoter that forms a template for the production of moreRNA transcripts (Figure 2). These transcripts are then

www.sciencedirect.com

used to produce additional DNA fragments with T7promoters. The reaction is self-sustaining and the tech-nique has been called self-sustained sequence replication.Potentially, this amplification method is more useful thanreverse-transcription PCR because it is not affected by thepresence of DNA, which means that RNA quantitation canbe conducted from crude extracts. Although it has beenavailable for almost as long as PCR, NASBA has not beenadopted as widely by researchers, probably because it isdifficult to prepare reliable, in-house master mixes andcommercial kits are relatively expensive. More recently,NASBA has received increased attention because it can bemonitored in real-time in combination with molecularbeacons [38] and it has been used to quantify Plasmodium[22] and viruses [39].

LCR

The LCR was also developed shortly after PCR [40]. In itscurrent format, LCR uses a thermostable DNA ligase andfour primers; two, adjacent forward primers and theircomplements. The primers are ligated when they hybrid-ize next to each other and, once ligated, function astemplates for further ligation. DNA ligases are highlyspecific and do not tolerate base mismatches (unlike DNApolymerases). This makes LCR superior to PCR for thedetection of SNPs, which has been its chief application[41]. Products of the LCR are detected in real-time byusing either FRET probes as LCR primers [23] or primersthat are designed to form molecular beacons once ligated.Considering that this technique can detect in real-time, itis unclear why it is not adopted more widely.

Hybridization techniques and applications

Hybridization techniques have been used to identifyparasites for many years. Fluorescent in situ hybridiz-ation (FISH) is in widespread use, and microarray-basedassays are being developed to detect multiple pathogens ina single sample. Nucleic-acid hybridization has also been

Page 46

TRENDS in Parasitology

Solid support

Capture probe

Sandwich hybridization

Detergent lysis of liposome,deposition of dye

Nucleic acid targetLiposome containing dyeand capture probe

Figure 3. Schematic of liposome-based lateral flow biosensor. Sandwich hybrid-

ization is used to capture target nucleic acid (RNA amplified by NASBA in the case

of the CryptoDetect system) and the reporter liposome, which contains a dye.

Following capture and washing, the liposomes are lyzed using detergent, which

produces a signal.

Review TRENDS in Parasitology Vol.21 No.7 July 2005 343

used in some of themoremature biosensor platforms, suchas the BIACORE and AMBRI systems.

FISH

The FISH technique (for review, see Ref. [42]) usesfluorescently-labeled probes [either DNA or peptidenucleic acid (PNA)] that hybridize to complementarynucleic acid targets in whole cells to enable the directdetection of organisms in complex communities. PNAprobes are pseudopeptides that hybridize to complemen-tary nucleic acid targets (DNA and RNA) with betterspecificity and stability than DNA probes [43] and canbind to DNA–DNA hybrids to form triplexes [44]. Themain disadvantage of PNA probes is that they areexpensive to synthesize. Applying FISH to whole cellsgives information about microbial identity, cell mor-phology, abundance and spatial distribution of individualtarget species (for review, see Ref. [45]). In protozoanparasites, FISH has been applied to study the partitioningand chromosome composition of nuclei inGiardia [46] andto detect arthropozoonotic species and genotypes ofCryptosporidium and Giardia [47]. PNA probes havebeen used for the direct FISH detection of Africantrypanosomes [48]. One disadvantage of FISH, particu-larly with environmental samples, is nonspecific bindingof the probe to debris, which results in excessive back-ground fluorescence. This can be overcome using eithermolecular beacons [49] or FRET probes [50] because theseonly fluoresce when bound to their specific targetsequence. Traditionally, FISH has been used to targetabundant transcripts, such as rRNA, to produce a signalthat is detected by fluorescence microscopy. The develop-ment of systems such as tyramide signal amplification(TSA) [51] enables detection of low-copy transcripts andsingle-copy genes. TSA is based on the horseradishperoxidase-catalyzed deposition of labeled tyramine mol-ecules at sites of probe binding. TSA detection mightincrease the signal by up to 100-fold compared withconventional fluorescent probes and improves spatialresolution [51].

Microarrays

DNA microarrays provide a powerful tool for the parallelanalysis of multiple genes and gene transcripts. Micro-arrays are arrays of either cDNAs or oligonucleotidesthat are either spotted onto a glass microscope slide orsynthesized on a silicone chip. DNA or mRNA extractedfrom cells or tissues is labeled with specific fluorescentmolecules and hybridized to the microarray DNA. Toimprove sensitivity, samples are often pre-amplified byPCR, which can also be used sample labeling by using afluorescently labelled primer in the PCR. The resultingimage of fluorescent spots is visualized in a confocalscanner and digitized for quantitative analysis [52].Microarrays have been used to detect and discriminatebetween a range of parasites, including different speciesand genotypes of Entamoeba, Giardia and Cryptospori-dium in a single assay [53]. The microarray assay is moresensitive than commonly used PCR technology.

The main application of microarrays has been to studygene expression in hosts (e.g. mosquitoes [54]) and

www.sciencedirect.com

parasites (e.g. Plasmodium [55]). Microarrays have beenused to study differential gene expression in differentlifecycle stages of T. brucei [56], gender-associated geneexpression in Schistosoma japonicum [57] and geneexpression during asexual development of T. gondii [58].

Universal arrays (ZipCode)

A novel approach to using microarrays has come with thedevelopment of ‘universal microarrays’ that exploit theproperties of LCR and make use of unique sequence tagscalled ZipCodes [59]. Each unique ZipCode oligonucleotideis spotted to a known address on a microarray. The LCRassay is designed so that the 5 0 primer is labeledfluorescently at its 5 0 end, and the 3 0 primer, which isligated to the 5 0 primer, has the complementary ZipCodesequence at its 3 0 end. Ligation products, which aregenerated only in the presence of the specific template,contain both the fluorescent label and the complementaryZipCode sequence, which enables detection on the uni-versal array. This technique holds great promise as ascreening assay for multiple pathogens and has beenused to discriminate between bacteria [60] and iden-tify viruses [61].

Emerging technologies

Numerous technologies are being developed to identifyand characterize pathogens. Some, such as laboratory-on-a-chip devices and biosensors, make use of amplification

Page 47

Review TRENDS in Parasitology Vol.21 No.7 July 2005344

and either hybridization or ligand–receptor interactions.Another approach uses matrix-assisted laser desorption–ionization time-of-flight (MALDI TOF) mass spectrometryto identify pathogens based on the detection of diagnosticproteins [62], but a database that contains the massinformation of the diagnostic biomarkers is needed to useMALDI TOF as an identification tool [63]. Few of thesetechnologies are readily available commercially for use inparasite diagnosis and those that are (e.g. MALDI TOFand the Doodlebug biochip that is described later) requireexpensive, specialized equipment and/or the establish-ment of appropriate sample databases. One commerciallyavailable biosensor platform is the Biacore system, whichuses surface plasmon resonance (http://www.biacore.com/lifesciences/technology/following_interaction/index.html).This technology detects changes in mass on the surface ofthe biosensor chip and can, therefore, be used to studyreceptor–ligand binding, including DNA hybridization.The Biacore system has been applied largely to the studyof pathogen–host interactions and the binding character-istics of drug targets, but there is a report of its use todetect Listeria [64].

Another biosensor platform that is rapidly approachingcommercial reality is the AMBRI system, which usesion-channel switches [65]. Like the Biacore system, theAMBRI biosensor can detect pathogens by either DNAhybridization or specific antibodies (http://www.ambri.com/Content/display.asp?screenZ206 and http://www.ambri.com/Content/display.asp?screenZ223). The Crypto-Detect system is a biosensor that detects Cryptosporidiumby combining liposome technology developed at CornellUniversity with NASBA (Figure 3), but its commercialavailability is not clear (http://www.ibi.cc/cryptospori-dium_parvum.htm). A promising technology for parasiteidentification is the Doodlebug biochip, which makes useof surface-enhanced Raman spectroscopy and has been

Table 1. Characterizing genetic diversity in parasitesa,b

Function Purpose

Discrimination above species level Taxonomy and phylogen

Discrimination between species Taxonomy, diagnosis an

Discrimination between intraspecific

variants, strains and genotypes

Population genetics, bre

(e.g. cross- versus self-fe

specificity, molecular ep

conservation (e.g. predic

pathogens) and biosecur

emerging pathogens)

Discrimination between individual isolates,

clonal lineages and subgenotypes, and

ecological interactions within host

‘Fingerprinting’ and mol

(tracking transmission o

sources of infection, risk

interactions and course o

Genetic markers, and linking phenotype and

genotype

Identifying phenotypic tr

epidemiological significa

virulence, infectivity and

aDepending on the parasites studied and the level of variation that is detected by a partic

the function (modified from Ref. [70]).bAbbreviations: AFLP, amplified-fragment-length polymorphism; GDH, glutamate dehy

PCR-coupled restriction-fragment-length polymorphism; PFGE, pulsed-field gel elect

difference analysis; SSCP, single-strand conformation polymorphism; TPI, triose phosp

www.sciencedirect.com

used to characterize waterborne pathogens includingCryptosporidium [66]. This system analyzes individualcells to generate a spectral fingerprint that differentiatesbetween species of Cryptosporidium and organismsexposed to different environmental conditions. However,further work is required to demonstrate robustness andperformance on real samples.

Possibly the most promising area of emerging technol-ogy is nanotechnology, which is driving the development oflaboratory-on-a-chip systems (for review, see Refs [67,68]).Systems are being developed that miniaturize conven-tional and real-time-amplification systems to analyze sub-microliter volumes extremely rapidly (e.g. nanoliterTaqmanmultiplex PCR [69]). Such systems should furtherreduce costs of real-time technology and enable point-of-care analysis in clinical applications.

Applying new technologies to biological questions

Molecular techniques enable interspecific and intraspeci-fic detection and characterization of parasites, so formingthe basis for most molecular epidemiological investi-gations. However, characterizing such genetic diversityrequires the selection of appropriate molecular tools todistinguish genetic variants at the required hierarchicallevel [70]. Increasing emphasis is given to definingappropriate regions of DNA that can be used for this, inaddition to the most appropriate techniques (Table 1). Thechoice of DNA region and locus, and technique, should bebased on the research question that is asked. Each regionwill have a range over which it can usefully detect geneticvariation, and the choice is often easier if there is alreadysome information on genetic variation at the level belowthe one of interest. For example, a molecular tool todiscriminate between species will be easier to establishbased on data on intraspecific variation. This will ensurethat the region of DNA provides consistent results

Tools

y Highly conserved coding regions such as

small subunit rDNA and somemitochondrial

genes

d epidemiology Moderately conserved regions such as

coding mitochondrial genes, ITS rDNA and

other loci (e.g. house-keeping genes that

encode GDH, TPI, HSP and actin)

eding systems

rtilization), host

idemiology,

ting susceptibility to

ity (exotic and

Variable regions such as allozymes, RAPD,

AFLP, PFGE and PCR–RFLP

ecular epidemiology

f subgenotypes,

factors, competitive

f infection)

Fingerprinting techniques such as

minisatellites and microsatellites, and SSCP

aits of clinical and

nce including

drug sensitivity

Genotype linked to phenotype by genetic

maps, RDA, and sequencing and/or real-time

PCR of genes that are thought to be linked to

phenotypic traits

ular approach, there might be overlap between the tools (regions of DNA) used and

drogenase; HSP, heat shock protein; ITS, internal transcribed spacer; PCR–RFLP,

rophoresis; RAPD, random amplified polymorphic DNA; RDA, representational

hate isomerase.

Page 48

Review TRENDS in Parasitology Vol.21 No.7 July 2005 345

throughout the species and across related species, andprovides a pattern that is distinct from other species [71].

Highly conserved regions can reveal information abouttaxonomic relationships between species whereas mode-rately conserved regions can differentiate strains andclosely related species; moderately variable regions canreveal population genetic structure whereas highly vari-able regions enables tracking of particular isolates in apopulation; and ‘mapping’ genetic markers can be usedto find markers that correlate genotype to phenotype(Table 1). Sequencing detects the highest level of variation(down to changes in individual base pairs) whereastechniques based on restriction-fragment-length poly-morphisms only detect changes that affect restrictionsites, but can be used to sample across the entire genome.Characterizing genetic variation also depends on appro-priate, rigorousanalysis, the reliability ofwhich isenhancedwhen several unlinked genetic loci are used [71].

There are many examples that describe the develop-ment and appropriate use of molecular tools for parasites.For example, recent research has identified several gene-tic loci that reproducibly differentiate between speciesof Cryptosporidium and Giardia (for review, see Refs[72,73]). Such research provides a database of genes thatcan be used to characterize and compare parasite isolates.Problems arise when isolates, particularly those withinteresting phenotypic features, are characterized usingdifferent molecular tools that do not enable meaningfulcomparison with established databases [74]. Using mol-ecular tools to characterize genetic diversity does notmean that more traditional methods of discrimination areno longer valuable. Indeed, molecular characterization isoften not possible in some field situations. However, it canbe used to demonstrate whether phenotypic characters,particularly morphology, have a genetic basis, thus,enabling future studies to rely on less sophisticated butmore practical discriminatory procedures [75].

Concluding remarks

Several technologies are being developed that can be usedto identify and characterize parasites. Of these, real-timeamplification and array-based technologies have foundwidespread use in general parasitological research andclinical diagnostics. Although biosensors and laboratory-on-a-chip devices show promise as future tools foridentifying and characterizing parasites, they are not yetavailable to the general research community and so theiruse is limited. When using new a technology, it is import-ant to consider how it can be applied to answer a specificbiological question, and whether it provides data thatenables meaningful analysis and interpretation.

References

1 Schaad, N.W. et al. (2003) Advances in molecular-based diagnosticsin meeting crop biosecurity and phytosanitary issues. Annu. Rev.Phytopathol. 41, 305–324

2 Cai, H.Y. et al. (2003) Molecular genetic methods in the veterinaryclinical bacteriology laboratory: current usage and future appli-cations. Anim. Health Res. Rev. 4, 73–93

3 Monis, P.T. et al. (2002) Molecular biology techniques in parasiteecology. Int. J. Parasitol. 32, 551–562

www.sciencedirect.com

4 Higuchi, R. et al. (1992) Simultaneous amplification and detection ofspecific DNA sequences. Biotechnology (N. Y.) 10, 413–417

5 Edwards, K. et al., eds (2004) Real-Time PCR, An Essential Guide,Horizon Bioscience, Wymondham

6 Saunders, N.A. (2004) In Edwards, K., Logan, J. and Saunders, N.(eds.), Real-Time PCR, An Essential Guide. Horizon Bioscience,Wymondham

7 Heid, C.A. et al. (1996) Real time quantitative PCR. Genome Res. 6,986–994

8 Kutyavin, I.V. et al. (2000) 3 0-minor groove binder-DNA probesincrease sequence specificity at PCR extension temperatures. NucleicAcids Res. 28, 655–661

9 Higgins, J.A. et al. (2001) Real-time PCR for the detection ofCryptosporidium parvum. J. Microbiol. Methods 47, 323–337

10 Keegan, A.R. et al. (2003) Cell culture-Taqman PCR assay forevaluation of Cryptosporidium parvum disinfection. Appl. Environ.Microbiol. 69, 2505–2511

11 Bertrand, I. et al. (2004) Improved specificity for Giardia lamblia cystquantification in wastewater by development of a real-time PCRmethod. J. Microbiol. Methods 57, 41–53

12 Rolao, N. et al. (2004) Quantification of Leishmania infantumparasites in tissue biopsies by real-time polymerase chain reactionand polymerase chain reaction-enzyme-linked immunosorbent assay.J. Parasitol. 90, 1150–1154

13 Kelley, G.O. et al. (2004) Evaluation of five diagnostic methods for thedetection and quantification of Myxobolus cerebralis. J. Vet. Diagn.Invest. 16, 202–211

14 Blair, P.L. et al. (2002) Transcripts of developmentally regulatedPlasmodium falciparum genes quantified by real-time RT-PCR.Nucleic Acids Res. 30, 2224–2231

15 Jauregui, L.H. et al. (2001) Development of a real-time PCR assay fordetection of Toxoplasma gondii in pig and mouse tissues. J. Clin.Microbiol. 39, 2065–2071

16 Afonina, I.A. et al. (2002) Minor groove binder-conjugated DNA probesfor quantitative DNA detection by hybridization-triggered fluor-escence. Biotechniques 32, 940–949

17 Piatek, A.S. et al. (1998) Molecular beacon sequence analysis fordetecting drug resistance in Mycobacterium tuberculosis. Nat.Biotechnol. 16, 359–363

18 Belousov, Y.S. et al. (2004) Single nucleotide polymorphism geno-typing by two colour melting curve analysis using the MGB EclipseProbe System in challenging sequence environment. Hum. Genomics1, 209–217

19 Poddar, S.K. and Le, C.T. (2001) Bordetella pertussis detection byspectrofluorometry using polymerase chain reaction (PCR) and amolecular beacon probe. Mol. Cell. Probes 15, 161–167

20 Yang, J.H. et al. (2002) Real-time RT-PCR for quantitation of hepatitisC virus RNA. J. Virol. Methods 102, 119–128

21 Greijer, A.E. et al. (2002) Multiplex real-time NASBA for monitoringexpression dynamics of human cytomegalovirus encoded IE1 and pp67RNA. J. Clin. Virol. 24, 57–66

22 Schneider, P. et al. (2005) Real-Time Nucleic Acid Sequence-BasedAmplification is more convenient than Real-Time PCR for quantifi-cation of Plasmodium falciparum. J. Clin. Microbiol. 43, 402–405

23 Chen, X. and Kwok, P.Y. (1999) Homogeneous genotyping assays forsingle nucleotide polymorphisms with fluorescence resonance energytransfer detection. Genet. Anal. 14, 157–163

24 Howell, W.M. et al. (2002) iFRET: an improved fluorescence system forDNA-melting analysis. Genome Res. 12, 1401–1407

25 Simon, A. et al. (2004) Use of fluorescence resonance energy transferhybridization probes to evaluate quantitative real-time PCR fordiagnosis of ocular toxoplasmosis. J. Clin. Microbiol. 42, 3681–3685

26 Limor, J.R. et al. (2002) Detection and differentiation of Crypto-sporidium parasites that are pathogenic for humans by real-timePCR. J. Clin. Microbiol. 40, 2335–2338

27 Schulz, A. et al. (2003) Detection, differentiation, and quantitation ofpathogenic Leishmania organisms by a fluorescence resonance energytransfer-based real-time PCR assay. J. Clin. Microbiol. 41, 1529–1535

28 Becker, A. et al. (1996) A quantitative method of determining initialamounts of DNA by polymerase chain reaction cycle titration usingdigital imaging and a novel DNA stain. Anal. Biochem. 237, 204–207

29 Bengtsson, M. et al. (2003) A new minor groove binding asymmetriccyanine reporter dye for real-time PCR. Nucleic Acids Res. 31, e45

Page 49

Review TRENDS in Parasitology Vol.21 No.7 July 2005346

30 Wittwer, C.T. et al. (2003) High-resolution genotyping by ampliconmelting analysis using LCGreen. Clin. Chem. 49, 853–860

31 Nath, K. et al. (2000) Effects of ethidium bromide and SYBR Green Ion different polymerase chain reaction systems. J. Biochem. Biophys.Methods 42, 15–29

32 Giglio, S. et al. (2003) Demonstration of preferential binding ofSYBR Green I to specific DNA fragments in real-time multiplex PCR.Nucleic Acids Res. 31, e136

33 Smilkstein, M. et al. (2004) Simple and inexpensive fluorescence-based technique for high-throughput antimalarial drug screening.Antimicrob. Agents Chemother. 48, 1803–1806

34 Widmer, G. et al. (2004) Genotyping of Cryptosporidium parvum withmicrosatellite markers. Methods Mol. Biol. 268, 177–187

35 Nicolas, L. et al. (2002) Real-time PCR for detection and quantitationof Leishmania in mouse tissues. J. Clin. Microbiol. 40, 1666–1669

36 Becker, S. et al. (2004) Real-time PCR for detection of Trypanosomabrucei in human blood samples. Diagn. Microbiol. Infect. Dis. 50,193–199

37 Guatelli, J.C. et al. (1990) Isothermal, in vitro amplification of nucleicacids by a multienzyme reaction modelled after retroviral replication.Proc. Natl. Acad. Sci. U. S. A. 87, 1874–1878

38 Leone, G. et al. (1998) Molecular beacon probes combined withamplification by NASBA enable homogeneous, real-time detection ofRNA. Nucleic Acids Res. 26, 2150–2155

39 Ayele, W. et al. (2004) Development of a nucleic acid sequence-basedamplification assay that uses gag-based molecular beacons todistinguish between human immunodeficiency virus type 1 subtypeC and C 0 infections in Ethiopia. J. Clin. Microbiol. 42, 1534–1541

40 Wu, D.Y. and Wallace, R.B. (1989) The ligation amplification reaction(LAR)–amplification of specific DNA sequences using sequentialrounds of template-dependent ligation. Genomics 4, 560–569

41 McNamara, D.T. et al. (2004) Development of a multiplex PCR-ligasedetection reaction assay for diagnosis of infection by the four parasitespecies causing malaria in humans. J. Clin. Microbiol. 42, 2403–2410

42 Amann, R. and Ludwig, W. (2000) Ribosomal RNA-targeted nucleicacid probes for studies in microbial ecology. FEMS Microbiol. Rev. 24,555–565

43 Chen, C. et al. (1999) Single base discrimination of CENP-B repeats onmouse and human Chromosomes with PNA-FISH. Mamm. Genome10, 13–18

44 Kuhn, H. et al. (1998) Kinetic sequence discrimination of cationicbis-PNAs upon targeting of double-stranded DNA. Nucleic Acids Res.26, 582–587

45 Brehm-Stecher, B.F. and Johnson, E.A. (2004) Single-cell micro-biology: tools, technologies, and applications. Microbiol. Mol. Biol.Rev. 68, 538–559

46 Yu, L.Z. et al. (2002) The two nuclei of Giardia each have completecopies of the genome and are partitioned equationally at cytokinesis.Eukaryot. Cell 1, 191–199

47 Graczyk, T.K. et al. (2003) Detection of Cryptosporidium parvum andGiardia lamblia carried by synanthropic flies by combined fluorescentin situ hybridization and a monoclonal antibody. Am. J. Trop. Med.Hyg. 68, 228–232

48 Radwanska, M. et al. (2002) Direct detection and identification ofAfrican trypanosomes by fluorescence in situ hybridization withpeptide nucleic acid probes. J. Clin. Microbiol. 40, 4295–4297

49 Xi, C. et al. (2003) Use of DNA and peptide nucleic acid molecularbeacons for detection and quantification of rRNA in solution and inwhole cells. Appl. Environ. Microbiol. 69, 5673–5678

50 Stender, H. et al. (1999) Fluorescence in situ hybridization assay usingpeptide nucleic acid probes for differentiation between tuberculousand nontuberculous mycobacterium species in smears of mycobac-terium cultures. J. Clin. Microbiol. 37, 2760–2765

51 van Gijlswijk, R.P. et al. (1997) Fluorochrome-labeled tyramides: usein immunocytochemistry and fluorescence in situ hybridization.J. Histochem. Cytochem. 45, 375–382

www.sciencedirect.com

52 Dennis, P. et al. (2003) Monitoring gene expression in mixed microbialcommunities by using DNAmicroarrays. Appl. Environ. Microbiol. 69,769–778

53 Wang, Z. et al. (2004) Detection and genotyping of Entamoebahistolytica, Entamoeba dispar,Giardia lamblia, and Cryptosporidiumparvum by oligonucleotide microarray. J. Clin. Microbiol. 42,3262–3271

54 Chen, H. et al. (2004) Microarray analysis for identification ofPlasmodium-refractoriness candidate genes in mosquitoes. Genome47, 1061–1070

55 Hall, N. et al. (2005) A comprehensive survey of the Plasmodium lifecycle by genomic, transcriptomic, and proteomic analyses. Science307, 82–86

56 El-Sayed, N.M. et al. (2000) The African trypanosome genome. Int.J. Parasitol. 30, 329–345

57 Fitzpatrick, J.M. et al. (2004) Gender-associated gene expression intwo related strains of Schistosoma japonicum. Mol. Biochem. Para-sitol. 136, 191–209

58 Cleary, M.D. et al. (2002) Toxoplasma gondii asexual development:identification of developmentally regulated genes and distinct pat-terns of gene expression. Eukaryot. Cell 1, 329–340

59 Gerry, N.P. et al. (1999) Universal DNA microarray method formultiplex detection of low abundance point mutations. J. Mol. Biol.292, 251–262

60 Busti, E. et al. (2002) Bacterial discrimination by means of a universalarray approach mediated by LDR (ligase detection reaction). BMCMicrobiol. 2, 27

61 Long, W.H. et al. (2004) A universal microarray for detection of SARScoronavirus. J. Virol. Methods 121, 57–63

62 Lay, J.O. (2001) MALDI-TOF mass spectrometry of bacteria. MassSpectrom. Rev. 20, 172–194

63 Demirev, P.A. et al. (2004) Bioinformatics-based strategies for rapidmicroorganism identification by mass spectrometry. Johns HopkinsApl. Technical Digest 25, 27–37

64 Leonard, P. et al. (2004) A generic approach for the detection of wholeListeria monocytogenes cells in contaminated samples using surfaceplasmon resonance. Biosens. Bioelectron. 19, 1331–1335

65 Cornell, B.A. et al. (1997) A biosensor that uses ion-channel switches.Nature 387, 580–583

66 Grow, A.E. et al. (2003). Evaluation of the Doodlebug biochip fordetecting waterborne pathogens. Report Number 00-HH-8-UR. WaterEnvironment Research Foundation, Alexandria, VA. pp. 88

67 de Mello, A.J. (2001) DNA amplification: does ‘small’ really mean‘efficient’? Lab on a Chip 1, 24N–29N

68 Gardeniers, J.G. and van den Berg, A. (2004) Lab-on-a-chip systemsfor biomedical and environmental monitoring. Anal. Bioanal. Chem.378, 1700–1703

69 Matsubara, Y. et al. (2004) On-chip nanoliter-volume multiplexTaqMan polymerase chain reaction from a single copy based oncounting fluorescence released microchambers. Anal. Chem. 76,6434–6439

70 Thompson, R.C.A. (2004) Molecular systematics and diagnosis. Vet.Parasitol. 125, 69–75

71 Constantine, C.C. (2003) Importance and pitfalls of molecular analysisto parasite epidemiology. Trends Parasitol. 19, 346–348

72 Monis, P.T. and Thompson, R.C. (2003) Cryptosporidium and Giardia-zoonoses: fact or fiction? Infect. Genet. Evol. 3, 233–244

73 Xiao, L. and Ryan, U.M. (2004) Cryptosporidiosis: an update inmolecular epidemiology. Curr. Opin. Infect. Dis. 17, 483–490

74 McDonnell, P.A. et al. (2003) Giardia duodenalis trophozoites isolatedfrom a parrot (Cacatua galerita) colonize the small intestinal tracts ofdomestic kittens and lambs. Vet. Parasitol. 111, 31–46

75 Harandi, M.F. et al. (2002) Molecular and morphological characteriz-ation of Echinococcus granulosus of human and animal origin in Iran.Parasitology 125, 367–373