Meeting Review

Royal Society Discussion Meeting: Utilisingthe genome sequence of parasitic protozoa21-22 March 2001 The Royal Society, 6 Carlton House Terrace, LondonSW1Y 5AG

Neil Hall*The Sanger Centre, Welcome Trust Genome Campus, Hinxton, Cambridge CB10 1SA, UK

*Correspondence to:The Sanger Centre, WelcomeTrust Genome Campus, Hinxton,Cambridge CB10 1SA, UK.E-mail: nh1@sanger.ac.uk

Abstract

Protozoan parasites cause some of the world’s most important diseases. Genome

sequencing information is rapidly being acquired and combined with new developments in

functional genome analysis to transform our understanding of parasites, and to enable new

approaches to combating the diseases they cause. Copyright # 2001 John Wiley & Sons,

Ltd.

Keywords: parasite; functional genomics; protozoa; apicomplexa; plasmodium; trypano-soma; leishmania

Introduction

To facilitate the dissemination of the exciting newdata, a public discussion meeting, hosted by TheRoyal Society, was organised by Professor JennieBlackwell, Professor Chris Newbold, Dr Mike Turnerand Professor Keith Vickerman FRS. The aim was toprovide an opportunity for researchers working onmany different parasites to discuss and co-ordinatetheir functional genomics approaches in the light ofthe wealth of sequence data emerging from thegenome centres. Since the Plasmodium falciparum(malaria) genome project was initiated in 1996, abroad range of other parasite genomes are now beingsequenced, resulting in a sea change in the way thatparasite biology can be investigated. The discussionmeeting was spilt into four half-day sessions; thisreport covers a cross-section of the papers presentedto an audience of over 270 delegates. A full list ofpresentations is provided in Table 1.

Presentations

To provide an example framework for howsequence data can be effectively utilised for func-tional genomics approaches to cell biology, SteveOliver (Manchester, UK) reported on the experience

of the yeast community, who obtained theircomplete genome sequence back in 1995. In hispaper ‘Functional genomics: lessons from yeast’ hesuggested one of the lessons for the parasitecommunity is that much can be achieved when thecommunity works together, whether this was in thearea of genome, transcriptome, proteome, ormetabolome. The yeast community saw globalexpression analysis and high-throughput geneknockouts as a pre-competitive resource. As theannotated yeast genome sequence preceded thedevelopment and application of DNA microarraytechnologies, the community had collaborated inhigh-throughput northern analysis on all of theyeast genes; genome-wide knockouts were also doneas part of a community-wide collaborative endea-vour. Laboratories acting independently couldnever have contemplated this work. It was encoura-ging that this model of ‘community research’ hadbeen adopted for the parasite sequencing projects,and was now being applied more and more toprotozoan functional genomics approaches. Thishas been considerably encouraged by the fundingagencies, whose policies are increasingly emphasis-ing the importance of multidisciplinary/centre col-laborations to answer biological questions.

Jennie Blackwell’s (Cambridge, UK) presentation‘From genomes to vaccines – Leishmania as a model’

Comparative and Functional GenomicsComp Funct Genom 2001; 2: 257–262.DOI: 10.1002 / cfg.88

Copyright # 2001 John Wiley & Sons, Ltd.

Tab

le1.Pre

senta

tions

Pro

f.Je

nnie

Bla

ckw

ell

Dep

t.of

Med

icin

e,U

nive

rsity

of

Cam

bridge

,U

KFr

om

geno

mes

tova

ccin

es:Le

ishm

ania

asa

model

Dr.

Dan

ielC

aruc

ciN

aval

Med

ical

Res

earc

hC

ente

r,U

SAT

echn

olo

gies

for

the

stud

yof

gene

and

pro

tein

expre

ssio

nin

Plas

mod

ium

Pro

f.St

eve

Oliv

erSc

hoolof

Bio

logi

calSc

ienc

es,U

nive

rsity

of

Man

ches

ter,

UK

Func

tiona

lge

nom

ics:

less

ons

from

yeas

tPro

fA

lan

Cow

man

Wal

ter

and

Eliz

aH

allIn

stitu

teof

Med

ical

Res

earc

h,

Mel

bour

ne,A

ustr

alia

Iden

tifica

tion

and

func

tiona

lan

alys

isof

pro

tein

sin

volv

edin

the

inva

sion

of

hum

aner

ythr

ocy

tes

by

the

mal

aria

par

asite

Plas

mod

ium

falc

ipar

um

Pro

f.D

avid

Roos

Uni

vers

ityof

Pen

nsyl

vani

a,U

SAM

inin

gth

ePl

asm

odiu

mge

nom

edat

abas

eto

defi

neorg

anel

lar

func

tions

Pro

f.St

eve

Bev

erle

yW

ashi

ngto

nU

nive

rsity

,St

Loui

s,U

SAA

naly

sis

of

Leishm

ania

gene

expre

ssio

nby

tran

sposo

ntr

appin

gan

d

expre

ssio

npro

filin

g

Pro

f.A

ndy

Wat

ers

Leid

enU

nive

rsity

Med

ical

Cen

tre,

The

Net

herlan

ds

Dem

ons

trat

ion

of

the

ort

holo

gous

natu

reof

the

geno

mes

of

Plas

mod

ium

berg

heian

dPl

asm

odiu

mfa

lcip

arum

:ex

plo

itatio

nto

under

stan

dpar

asite

inte

ract

ions

with

both

itsho

stan

dve

ctor

Pro

f.El

isab

etta

Ullu

Uni

vers

ityof

Yal

e,U

SAG

enet

icin

terf

eren

ceby

doub

le-s

tran

ded

RN

Ain

Tryp

anos

oma

bruc

ei

Pro

f.Ken

Stua

rtSe

attle

Bio

med

ical

Res

earc

hIn

stitu

tean

dU

nive

rsity

of

Was

hing

ton,

USA

The

editi

ngco

mple

xof

Tryp

anos

oma

bruc

ei

Dr

Dav

idSi

ble

yW

ashi

ngto

nU

nive

rsity

,U

SAG

enet

ican

alys

isof

viru

lenc

ein

toxo

pla

smosis

Pro

f.A

ndy

Tai

tW

ellc

om

eC

entr

eof

Mole

cula

rPar

asito

logy

,

Uni

vers

ityof

Gla

sgow

Gen

etic

anal

ysis

of

phe

noty

pe

inTr

ypan

osom

abr

ucei

:a

clas

sica

l

appro

ach

topote

ntia

llyco

mple

xtr

aits

Pro

f.A

lan

Fairla

mb

Uni

vers

ityof

Dun

dee

Met

abolic

anal

yses

intr

ypan

oso

mes

and

mal

aria

Pro

fR

icha

rdM

oxo

nIn

stitu

teof

Mole

cula

rM

edic

ine,

Uni

vers

ityof

Oxf

ord

Func

tiona

lge

nom

ics

of

pat

hoge

nic

bac

teria

258 Meeting Review

Copyright # 2001 John Wiley & Sons, Ltd. Comp Funct Genom 2001; 2: 257–262.

demonstrated the power and pitfalls of DNAmicroarray technology as a means of assessinggene function. The aim of the work is to identifygenes that are expressed in the metacyclic promas-tigote and amastigote forms of the parasite. It is themetacyclic stage promastigote that invades andsurvives as an amastigote form in the mammalianhost; hence genes expressed at these stages of thelife cycle provide possible targets for vaccines.Fortunately, the majority of the life-cycle stages ofLeishmania major can be mimicked in vitro, andRNA profiles can thus be obtained for a wide rangeof defined developmental points of the life cycle.Initial profiling studies (using EST-based datasets)have been conducted with arrays comprisingapproximately 1000 of the estimated 8000 Leishma-nia genes; as the proportion of the genomesequenced increases, this work will be extended toall predicted genes of Leishmania major. It was clearfrom the analyses performed so far that clusteringof expression profiles (using the EPCLUST soft-ware package developed by Jaak Vilo at theEuropean Bioinformatics Institute) could identifyseveral genes whose expression characteristics wereworthy of confirmation by Northern analysis andfurther study in vitro and in vivo.

One observation, also noted by a number ofresearchers including Steve Oliver, working onyeast, and Steve Beverley, who reported on hisgenome survey sequence (GSS) microarray studiesin Leishmania, was that small conceptual orpractical errors in performing microarray experi-ments frequently lead to misleading results. Allstressed the paramount importance of standardisa-tion of assay conditions, particularly if dataobtained from different laboratories/experimentswere ever to be compared.

Blackwell’s group are also piloting the use ofDNA vaccines in mice as a possible high-throughput tool for identifying new vaccine targets.Initially, this was performed using cDNAs that arethought to be either predominantly, or exclusively,expressed in the amastigote stage of the parasite.Initial experiments were conducted using a poolingstrategy, but when single DNA samples frompotentially ‘protective’ pools were used, differentresults were obtained; DNA vaccines are now beingscreened individually.

Dan Carucci (Naval Medical Research Center,USA) spoke on ‘Technologies for the study of geneand protein expression in Plasmodium’. He intro-duced some of the new techniques that are available

to analyse the genomes, transcriptomes and pro-teomes of parasites. These technologies are beingutilised by The Naval Medical Research Center todevelop potential vaccines for the malaria parasitePlasmodium falciparum.

The Carucci lab has now made microarray DNAchips for chromosomes 2, 3, 12 and 14 ofP. falciparum and they plan to extend this to theentire malaria genome over the next two years.These microarrays can be used to study effects ofdrugs on parasite growth, mechanisms of drugresistance, mechanisms of antigenic variation andgenes involved in cell invasion. However, Caruccistressed that for these experiments to be mean-ingful, numerous replicate experiments must becarried out and that changes in expression levelsneed to be shown to be statistically significant.

While microarrays can give a good indication ofgene expression profiles within the cell, this does notnecessarily provide a measure of protein dynamics.The Carucci lab is using high-throughput proteo-mics to study the protein content of parasite cells.The traditional workhorse of proteomics, the 2Dgel, is cumbersome, can be difficult to reproducereliably, and only displays soluble proteins; henceCarucci has adopted a new technique, capillaryliquid chromatography coupled with tandem massspectrometry, as an alternative. This combinedapproach facilitates high-throughput analysis ofthe protein content of a cell or tissue, without thepossibly limiting step of electrophoresis.

Carucci also made the point that any laboratorywishing to get heavily involved in proteomics ormicroarray analysis will need specialised bioinfor-matics tools to analyse the data efficiently. Hislaboratory uses an in-house relational database tolink information from the malaria genome, micro-array experiments and proteomics experiments.

David Roos, (Pennsylvania, USA) in his paper‘Mining the Plasmodium genome database to defineorganellar function’ discussed how Toxoplasmagondii can be used as a useful model for molecularstudies of other Apicomplexan parasites. Studies inToxoplasma have elucidated the function of theapicoplast, which up until recently had remained amystery.

The apicoplast is an organelle, unique to apicom-plexan parasites, which is associated with the apicalcomplex of the cell. It has such significant similarityto a plant chloroplast that it is believed to havearisen from a secondary endosymbiotic event of anancient ancestor of modern plants. The apicoplast

Meeting Review 259

Copyright # 2001 John Wiley & Sons, Ltd. Comp Funct Genom 2001; 2: 257–262.

has a 35 kb genome that encodes mostly house-keeping genes, yet it has since been validated as aprophylactic drug target. By data mining theP. falciparum genome sequence, Roos and hiscolleagues were able to identify a number of genesthat were predicted to be localised in the apicoplast;this was achieved by identifying genes exhibitingsimilarity to chloroplast-encoded genes in plants, ordisplayed other ‘‘plant-like’’ characterisics. TheRoos group were then able to test if the proteinproducts from these, and other genes targeted to theapicoplast are by using a range of GFP fusionconstructs in Toxoplasma. To date 150 nuclear-encoded plastid genes have been identified, givingan almost complete picture of plastid metabolism.

This work has demonstrated how well designedbioinformatic studies, validated by ‘‘wet lab’’ experi-mentation, can be a powerful and efficient approachto functional studies. To facilitate this, the Rooslab has established a web-accessible Plasmodiumgenome database [6], PlasmoDB (www.plasmodb.org).

The phenomenon of RNA interference (RNAi)has provided a very useful tool for functionalstudies of diploid organisms, such as Drosophila [5]and C. elegans [4]. In her presentation ‘Geneticinterference by double-stranded RNA in Trypano-soma brucei’, Elisabetta Ullu (Yale, USA) demon-strated how it has also been put to use in theT. brucei parasite. RNAi uses double-strandedRNA molecules to down-regulate levels ofmRNA in cells. This technique has a number ofadvantages over traditional knockouts, as it avoidsthe need to transfect organisms twice to knock outboth copies of a gene. Also, mutant phenotypescan be rescued using regulated promoters tomediate the degree of interference.

The mechanisms behind gene regulation byRNAi are not well understood; however Ullu andher colleagues have come a long way towardselucidating them. They have identified siRNA(small interfering RNA) species in Trypanosomesthat are believed to interact with polyribosomes,thus preventing translation. A high proportion ofthese siRNAs are derived from the retrotransposon-like repeat sequences INGI and SLACS, suggestingthat RNAi may play a role in the regulation ofthese elements.

The RNAi data for T. brucei are encouraging,but questions remain. Why, for example, is RNAiproving so hard to demonstrate in the relatedkinetoplastid Leishmania? Time will no doubt tell,

but irrespective of this, the observation of RNAi inparasites underlines that they are not only ofinterest on account of their medical importance,but can frequently provide useful experimentalmodel systems for investigating fundamental biolo-gical phenomena. This point was also exemplifiedby Ken Stuart in his presentation entitled ‘Theediting complex of Trypanosoma brucei’.

In his presentation ‘Genetic analysis of phenotypein T. brucei: a classical approach to potentiallycomplex traits’, Andy Tait (Glasgow, UK), spokeabout the relevance of genetic studies in the post-genomic era. While genetic analysis is typically nota high throughput technique for functional ana-lyses, it has its advantages in that the work isbiologically driven with a defined phenotype thatusually can be chosen for its scientific relevance,such as drug resistance or reduced infectivity;complex phenotypes are notoriously difficult tostudy by gene knockouts or RNAi.

Tait and his colleagues now have a genetic mapof Trypanosoma brucei mini-satellite and micro-satellite markers at a resolution of 10 cM onchromosomes I and II, and a partial map ofchromosome IV. They also have 140 AFLPmarkers, giving a total map covering 1/3 of thegenome.

The Tait group intends to complete this map overthe next few years. Stocks of T. brucei have beenidentified that differ in their drug sensitivity, humaninfectivity and virulence, and these parental stocksare being crossed to build up a series of progenypanels. Linkage analysis is to be used to determinethe genetic bases underlying these important phe-notypes.

In his presentation ‘Demonstration of the ortholo-gous nature of the genomes of Plasmodium bergheiand Plasmodium falciparum: exploitation to under-stand parasite interactions with both its host andvector’ Andrew Waters (Leiden, The Netherlands)spoke about how comparative genomics is startingto aid our understanding of the structure andcontent of malaria genomes. The genomes ofPlasmodium spp have been shown to be highlysyntenic [2,1]. Synteny is a measure of geneticconservation, which may refer to entire chromo-somes or simply neighbouring segments of DNA. Itdoes not, as is commonly thought, simply measuregene order.

Prior to genome sequencing, synteny was mea-sured by hybridising DNA probes to pulsed field gelelectrophoresis (PFGE)- separated chromosomes. It

260 Meeting Review

Copyright # 2001 John Wiley & Sons, Ltd. Comp Funct Genom 2001; 2: 257–262.

was observed that rodent malarias were highlysyntenic, with almost all probes hybridising toequivalent chromosomes in all species analysed.There is also significant synteny between rodentmalarias and the human parasite, P. falciparum.

Waters demonstrated how synteny on a smallerscale (i.e. conservation over small genetic distances)can aid the annotation and analysis of genomesequence data.

Waters and his collaborators have fullysequenced a YAC clone containing DNA from therodent malaria species Plasmodium berghei. Thesequence obtained from this clone has been care-fully analysed and the primary transcipt RNAsplicing characteristics of the 6 predicted genesannotated. The P. berghei YAC is highly syntenicwith a contig sequenced and assembled by The

Institute of Genomic Research (TIGR, Washington)

from chromosome 10 of P. falciparum. When thetwo sequences are compared, it is clear that thecoding regions are considerably more conservedthan non-coding regions, hence it is possible todirectly compare the predicted genes and splicingpatterns between the two species.

Importantly, this work demonstrates that given awell-studied sequence from one Plasmodium species,conserved DNA synteny will enable us to identifystructural components and coding regions in otherspecies. Variations from this ‘rule’ will serve as flagsfor areas containing potentially interesting biologi-cal features (e.g. host restriction). Also, even twosyntenic, but otherwise uncharacterised sequencescould give clues to the location and structure ofgenes within both of them, due to the inherentconservation between the coding regions.

In his presentation ‘Metabolic analysis in trypano-

somes and malaria’, Alan Fairlamb (Dundee, UK)concentrated on the analysis of metabolic pathwaysin parasitic organisms. Such analysis is critical forthe identification of potential drug targets. Intheory, the availability of a completely annotatedgenome sequence should provide investigators witha complete metabolome that should in turn providea wide array of possible drug targets. However,Fairlamb went on to explain that the analysis ofsuch a complex data set requires well-designed andwell-curated metabolism databases.

As an example, Fairlamb discussed the poly-amine biosynthesis pathway in the kinetoplastidsLeishmania and Trypanosoma; in these protozoanparasites, the enzyme trypanothione reductase is

present, ‘‘replacing’’ the glutathione reductasefound in the mammalian pathway [3].

This difference results in a unique metabolite,trypanothione (N1, N8-bis(glutathionyl)-spermidine),which has taken on many of the protective and anti-oxidant functions normally ascribed to glutathionein mammalian cells. Inhibitor studies have validatedthe parasite pathway as a possible drug target.However in some public metabolic pathway data-bases, this novel pathway is not represented, despitethe breadth of knowledge available.

This is an unsatisfactory state of affairs, ascomponents of pathways that are unique to, ordifferent in, medically relevant organisms are morethan likely to be candidate drug targets. Intensifieddialogue between the research communities, andthose curating the public databases, is a priority forall scientists involved in the field of parasitefunctional genomics.

Chris Newbold (Oxford, UK) closed by thankingeveryone for their contribution to the discussions,and the Royal Society for hosting the meeting. Inkeeping with the number of different organisms thatwere the subject of the talks over the two days, hesuggested that perhaps one of the most importantmessages from the meeting was that a great deal ofinformation would come from comparative geno-mics. With so many genomes being sequenced nowby centres around the world, parasitic protozoaoffer an exciting opportunity to understand genomeevolution, possibly more so than in any other areaof eukaryotic biology. It was also very importantthat the communities take on board the concept ofpre-competitive research, particularly if the hugequantity of sequencing, expression and proteomicsdata, to cite but three examples, were to be used totheir full potential.

References

1. Carlton JM, Galinski MR, Barnwell JW, Dame JB. Karyo-

type and synteny among the chromosomes of all four species

of human malaria parasite. Mol Biochem Parasitol 1999; 101:

23–32.

2. Carlton JM, Vinkenoog R, Waters AP, Walliker D. Gene

synteny in species of Plasmodium. Mol Biochem Parasitol

1998; 93: 285–294.

3. Fairlamb AH, Cerami A. Metabolism and functions of

trypanothione in the Kinetoplastida. Annu Rev Microbiol

1992; 46: 695–729.

4. Fire A, Xu S, Montgomery MK, Kostas SA, Driver

SE, Mello CC.Potent and specific genetic interference by

Meeting Review 261

Copyright # 2001 John Wiley & Sons, Ltd. Comp Funct Genom 2001; 2: 257–262.

double-stranded RNA in Caenorhabditis elegans.: Nature 1998;

391: 806–811.

5. Hammond SM, Bernstein E, Beach D, Hannon GJ. An RNA-

directed nuclease mediates post-transcriptional gene silencing

in Drosophila cells. Nature 2000; 404: 293–296.

6. The Plasmodium Genome Database Collaborative. Plas-

moDB: An integrative database of the Plasmodium falciparum

genome. Tools for accessing and analyzing finished and

unfinished sequence data. Nucleic Acids Res 2001; 29: 66–69.

The Meeting Reviews of Comparative and Functional Genomics aim to present a commentary on the topicalissues in genomics studies presented at a conference. These reviews are invited and each represents apersonal critical analysis of the current reports and aim at providing implications for future genomicsstudies.

262 Meeting Review

Copyright # 2001 John Wiley & Sons, Ltd. Comp Funct Genom 2001; 2: 257–262.