Experimental Parasitology 104 (2003) 26–32

www.elsevier.com/locate/yexpr

Trypanosoma brucei: Trypanosome strain typing using PCRanalysis of mobile genetic elements (MGE-PCR)

A. Tilley,a,* S.C. Welburn,a E.M. F�eevre,a E.J. Feil,b and G. Hidec

a Centre for Tropical Veterinary Medicine, Royal (Dick) School of Veterinary Science, University of Edinburgh, Easter Bush,

Roslin, Midlothian, Scotland EH25 9RG, UKb Department of Biology and Biochemistry, University of Bath, BA2 7AY, UK

c Centre for Molecular Epidemiology and Ecology, School of Environmental and Life Sciences, University of Salford, UK

Received 2 December 2002; received in revised form 22 May 2003; accepted 2 June 2003

Abstract

We describe the development of a single-primer amplification system, which uses the trypanosomal mobile genetic element

RIME as a molecular marker for the differentiation of Trypanosoma brucei stocks. Using a well-characterised set of T. brucei

stocks from southeast Uganda, Kenya and Zambia, we have evaluated the application of this technique, termed MGE-PCR

(mobile genetic element PCR) for the typing of trypanosome strains. The technique revealed considerable variation between

stocks and was sufficiently specific to amplify trypanosomal DNA in the presence of host DNA. The results showed a clear

distinction between human-infective and non-human-infective stocks. Comparative studies on these stocks using markers for the

human serum resistance associated (SRA) gene, which identifies human-infective stocks, demonstrated complete agreement be-

tween MGE-PCR derived groups and human-infectivity status. Furthermore, MGE-PCR detects high levels of variability within

the T. b. brucei and T. b. rhodesiense groups and is therefore a powerful discriminatory tool for tracking individual T. brucei

genotypes and strains.

� 2003 Elsevier Science (USA). All rights reserved.

Keywords: Trypanosoma brucei brucei; Trypanosoma brucei rhodesiense; Molecular epidemiology; Mobile genetic element; Sleeping sickness; PCR;

Human-infectivity; Serum resistance associated (SRA) gene

1. Introduction

The epidemiology of sleeping sickness in East Africa

is complicated by the fact that the morphologically

identical human-infective Trypanosoma brucei rhodes-

iense and non-human-infective Trypanosoma brucei

brucei sub-species circulate together in domestic live-

stock, principally cattle reservoirs (Onyango et al., 1966,Hide et al., 1996). The recent development of a genetic

marker based on the detection of a serum resistance

associated (SRA) gene in Trypanosoma brucei rhodes-

iense (Welburn et al., 2001, Gibson et al., 2002) has

enabled the division of the two T. brucei sub-species and

therefore identification of animal reservoirs harboring

* Corresponding author. Fax: +44-131-445-5099.

E-mail address: Aimee.Tilley@ed.ac.uk (A. Tilley).

0014-4894/$ - see front matter � 2003 Elsevier Science (USA). All rights res

doi:10.1016/S0014-4894(03)00114-0

the human-infective sub-species T. b. rhodesiense. Ad-

ditional techniques, however, which have the ability to

discriminate further to strain level, are required to track

the origins and movement of trypanosome strains in-

volved in sleeping sickness epidemics (Fevre et al., 2001;

Welburn and Odiit, 2002).

Previously developed molecular technologies for

tracking trypanosome strains have focused on the use ofisoenzyme characterisation (Gibson et al., 1980; Tait,

1980) and RFLP analysis (Hide et al., 1991). Whilst

successful, these techniques are complex, time consum-

ing and require a considerable amount of parasite

material. The recent development of a number of PCR-

based techniques using markers based on minisatellites

(MacLeod et al., 2000, 2001) RAPD (Mathieu-Daude

et al., 1995), AFLP (Masiga et al., 2000), and MGE-PCR (Hide and Tilley, 2001) have attempted to over-

come many of these obstacles. Minisatellite markers and

erved.

A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32 27

MGE-PCR can be used in the presence of host or vectorDNA and as such are particularly good tools for

tracking parasite strains.

MGE-PCR comprises a single primer PCR technique

that generates variable banding patterns using markers

based on the trypanosomal mobile genetic RIME.

RIME is a trypanozoon-specific mobile genetic element,

present in the T. brucei genome at up to 400 copies

(Hasan et al., 1984; Kimmel et al., 1987; Murphy et al.,1987). A high copy number, combined with polymor-

phisms in length and genomic distribution of the ele-

ment make RIME a useful marker for investigating

strain diversity within T. brucei. In this paper, we de-

scribe the development of MGE-PCR for analysis of

positional variation of RIME, via amplification of the

flanking regions of the RIME element in T. brucei, and

discuss the application of MGE-PCR to the analysis ofrelationships between trypanosome strains isolated from

human sleeping sickness foci.

2. Materials and methods

2.1. Collection of parasites

Trypanosome DNA was extracted from stocks of

T. brucei collected in 1988–1992 from the Busoga focus

of southeast Uganda and the Luangwa Valley focus,

Zambia as previously described (Hide et al., 1994, 1998,

Welburn et al., 2001). KINUHBW and BUGH1 were

isolated from humans and KINUC21 from a cow, in

the Tororo district, and BWP1 was isolated from a pig

in the Kamuli district of Uganda in 2000. DNA fromthese stocks was isolated from bloodstream forms using

the QIAamp DNA blood minikit (Qiagen, Crawley,

UK), following parasite amplification in mice and iso-

lation of trypanosomes from mouse blood by anion

exchange on DEAE cellulose (Lanham and Godfrey,

1970).

2.2. PCR analysis of length variation of RIME between

stocks

Analysis of length variation of the RIME element

was carried out using primers designed to sequences

towards either end of the RIME element. The primers

were labeled RIME A (50 AGC GTC TTG CTG

GGA CAC CGT 30) and RIME B (50 AGC CAT

CAC CGT AGA GCC CTG 30). PCR amplificationwas performed using a Robocycler (Stratagene). Re-

action volumes of 50 ll contained 0.5 lM of each

primer, 1 lM of dNTP mix containing 0.25 lM of

each dNTP and 2U (5U/ll) of Supertaq DNA poly-

merase (HT Biotechnology, UK) in the manufacturers

buffer containing 1.5mM MgCl2. PCR cycles were as

follows; a single cycle at 94 �C for 5min, followed by

30 cycles at 94 �C for 45 s, 55 �C for 1min and 72 �Cfor 2min.

2.3. MGE-PCR analysis of flanking regions of RIME

To investigate positional variation of RIME, via

analysis of the flanking regions of the element, MGE-

PCR was performed via a single primer technique using

primer REV B (50CAG GGC TCT ACG GTG ATGGCT 30), which anneals to the 30 end of the trypanoso-

mal mobile genetic element RIME. Reaction volumes of

50 ll were used, as described above, with the exception

that the volume of the primer was doubled. The PCR

amplification program was as follows: 1 cycle of 94 �Cfor 10min, 10 cycles of 94 �C for 45 s, 60 �C for 1min

and 72 �C for 2min, followed by 25 cycles of 94 �C for

45 s, 56 �C for 1min and 72 �C for 2min.

2.4. Phylogenetic analysis of banding patterns

PCR products were run on a 1.5% agarose gel, stained

with ethidium bromide and visualized using an Alpha

1220 gel imaging system (Flowgen). Measurement of

bands was performed using the Flowgen Alpha imaging

system. Multiple exposures of gels were taken to ensuremaximum detection of low intensity bands and clear

resolution of high intensity bands. PCR amplification

and electrophoresis of samples were repeated to ensure

consistency and reproducibility of banding patterns. A

table of consensus banding patterns was created from the

multiple exposures and each band was documented. A

binary bar code was allocated to each sample indicating

presence or absence of each band. The clustering calcu-lator (www2.biology.ualberta.ca/jbrzusto/cluster.php)

was used to group strains based on Jaccard similarity

analysis of the presence and absence of each band in each

sample, whilst the un-weighted arithmetic average clus-

tering method was used to determine the average dis-

tance between samples in distinct clusters. Results of the

cluster analysis were presented as a dendogram using the

TREEVIEW program (Page, 1996). Each stock wascross-referenced with its human-infectivity status, as

previously defined using human-serum-sensitivity tests,

RFLP methods (Hide et al., 1994) and presence of the

SRA gene (Welburn et al., 2001).

Further phylogenetic analysis of 26 representative

strains was undertaken using data generated from

analysis of similarities and differences in banding pat-

terns. The data were converted into a distance matrixof percentage identity (p value), and these distances

were used to reconstruct a tree using the Minimum

Evolution algorithm as implemented in MEGA version

2.1 (Kumar et al., 2001). The interior branch test with

2000 replicates (Nei et al., 1985) was used to examine

the reliability of the tree in terms of the clustering of

strains.

28 A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32

3. Results

3.1. PCR analysis of length variation of RIME between

T. brucei stocks

To determine whether variation in length classes of

the RIME element occurred between stocks of T. brucei,

primers RIME A and RIME B were designed to either

end of the element. PCR amplification using theseprimers produced a range of band sizes, but minimal

band variation between stocks was observed, with major

bands occurring at 312, 515, 848, and 1425 bp in all

stocks (Fig. 1). This lack of intra-specific variation

suggested that analysis of length variation of the RIME

element was an unsuitable method for characterizing

T. brucei stocks.

3.2. MGE-PCR analysis of flanking regions of RIME

Investigations into the positional variation of RIME

were undertaken using analysis of the flanking regions

of the element. Amplification of these regions using a

standard PCR method would require knowledge of

those flanking sequences for primer design. To enable

the measurement of positional variation without priorknowledge of the DNA sequence flanking the element

we developed a single primer approach termed MGE-

PCR. The technique utilises a combination of specific

Fig. 1. Amplification of the RIME element in stocks of T. brucei from

the Busoga and Zambian foci. Representative gel showing banding

patterns from amplification of the RIME element in stocks of T. brucei

from the Busoga and Zambian foci. Lane 1 contains a marker for

which band sizes (base pairs) are indicated to the left of the gel. Lane

12 contains a negative water control. Lane 5 contains sample Z208; a

human-infective sample from the Zambian focus. Remaining lanes

contain the following range of T. brucei stocks from the Busoga focus

and are designated (+) human-infective or ()) non-human-infective.

Lane 2, UGL (+); lane 3, UGA (+); lane 4, UGK (+); lane 6, Bumanda

25 ()); lane 7, Buteba 135 ()); lane 8, Bumanda 146 ()); lane 9,

Magola 18 ()); lane 10, Iyolwa 116 ()); and lane 11, Fly 97 ()).

priming within the element, priming with adjacent ele-ments and mispriming in flanking regions, resulting in

the generation of variable banding patterns between

trypanosome stocks (Hide and Tilley, 2001).

MGE-PCR amplification using primer REV B re-

vealed considerable variation in banding patterns be-

tween stocks (Fig. 2). Band sizes were measured from

multiple gel exposures and a table of composite banding

profiles for each stock was produced (not shown).Replicate amplifications were checked against the com-

posite profiles to ensure consistency and reproducibility

of the technique. Banding patterns from replicate am-

plifications were identical.

Specificity of the technique was demonstrated by the

lack of amplification of host (tsetse, cow, human, and

mouse) DNA (Fig. 3a). Furthermore, amplification of

trypanosome DNA was not inhibited by the presence ofhost DNA (Fig. 3a), as demonstrated by identical

banding patterns resulting from amplification of try-

panosome DNA with and without the presence of host

DNA (Fig. 3a).

The reproducibility of the technique was further

confirmed by comparing banding patterns from dupli-

cate DNA samples made from the same stocks but ex-

tracted at different times and under different conditions.Samples isolated in 1988–1992 had been passaged twice

through mice then fed to tsetse flies, from which pro-

cyclic cultures were obtained. Half of the sample had

Fig. 2. MGE-PCR. Representative gel showing banding patterns from

MGE-PCR amplification of a range of T. brucei stocks from the Bu-

soga focus using primer REV B. Lane 1 contains a marker for which

band sizes (base pairs) are indicated to the left of the gel. Lane 15

contains a negative water control. Lanes 2–3 contain human-infective

samples KINU C21 and BUG H1, respectively, both of which were

collected from the Busoga focus in 2000. Remaining lanes contain the

following samples collected from the Busoga focus in 1989–1992. (+)

Denotes human-infective (T. b. rhodesiense) stocks, ()) denotes non-

human-infective (T. b. brucei) stocks. Lane 4, UGK (+); lane 5, Mela 2

(+); lane 6, Mela 71 (+); lane 7, Mela Pig 1 ()); lane 8, Mela Pig 2 ());lane 9, UGB 88 (+); lane 10, UGC 88 (+); lane 11, UGL (+); lane 12,

UG 89/9 (+); lane 13, UG 89/5 (+); and lane 14, UG 89/3 (+).

Fig. 3. Specificity and reproducibility of MGE-PCR. (a) Representative gel showing specificity of MGE-PCR amplification of trypanosome DNA in

the presence of host DNA. Lane 1 contains a marker, for which band sizes (bp) are shown to the left of the gel. Lanes 2–4 contain tsetse, mouse and

cow DNA, respectively. Amplification of these samples produced no bands. Lanes 5 and 6 contain trypanosome DNA mixed with cow DNA and

mouse DNA, respectively. Lane 7 contains pure trypanosome DNA, without the presence of host DNA and demonstrates that amplification of host

DNA mixed with trypanosome DNA produced banding patterns that were identical to those of pure trypanosome DNA. Lane 8 contains a negative

water control. (b) Representative gel showing reproducibility of MGE-PCR on samples from which DNA was extracted on 2 separate occasions

using different extraction techniques. Lane 1 contains a marker for which band sizes are shown to the left of the gel. Lanes 2 and 5 represent banding

patterns, from two different trypanosome stocks, from MGE-PCR amplification of procyclic DNA prepared in 1988–1992 using standard phenol

chloroform extraction techniques. Lanes 3 and 6 represent banding patterns, respectively, from the same two stocks shown in lanes 2 and 5, from

MGE-PCR amplification of DNA extracted using a Qiagen kit following 10 years of cryopreservation in procyclic form. Lane 4 contains a negative

water control.

A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32 29

been cryopreserved and stored in liquid nitrogen for 10

years, after which time the DNA was extracted using a

QIAamp DNA blood minikit (Qiagen, Crawley, UK).

DNA from the remainder of the sample was extracted

prior to cryopreservation, using standard phenol–chlo-

roform techniques, resuspended in sterile distilled water

and refrigerated. Banding patterns were identical in all

samples analysed (Fig. 3b).

3.3. Phylogenetic analysis of banding patterns

The variation in banding patterns was further

analysed by the generation of a dendogram, which

represents relationships between the stocks based on

the similarities and differences in banding patterns. The

resulting dendogram showed a division of strains intothree distinct groups (Fig. 4). Groups 2 and 3 represent

human-infective and non-human-infective strains from

southeast Uganda, respectively, and group 1 comprises

human-infective strains isolated from an unrelated fo-

cus in Zambia. More specifically, group 2 contains

isolates sampled from the Busoga focus in 1988–1992

together with a new set of samples (labelled b) collected

from humans and cattle in villages within the samearea in 2000. Identical banding patterns were observed

in both cases, and all possessed SRA, thereby dem-

onstrating the long-term stability of human-infective

genotypes in this region. This was further confirmed by

the reconstruction of a minimum evolution tree of 26

representative strains; the branch leading to the

human-infective strains is well supported by the inte-

rior branch test implying that this cluster is indeed

monophyletic (see Fig. 5).

4. Discussion

In this paper, we have discussed the development andapplication of MGE-PCR as a tool for trypanosome

strain typing. Using analysis of banding patterns gen-

erated from MGE-PCR amplification of flanking re-

gions of the trypanosomal mobile genetic element

RIME, we have demonstrated the division of a range of

T. brucei stocks into 3 distinct groups reflecting human-

infectivity status and geographical origin:

Groups 2 and 3 comprise T. brucei stocks fromUganda and represent non-human-infective and human-

infective stocks, respectively, as previously determined

by human-serum-sensitivity, RFLP and SRA. These

well-defined groups clearly demonstrate the ability of

MGE-PCR to differentiate T. b. brucei and T. b. rho-

desiense. Furthermore, evidence of differentiation within

sub-species is demonstrated by the separation of human-

infective Zambian strains into group 1. The distinctionof human-infective isolates from Uganda and Zambia

supports previous evidence that the strains circulating in

these geographically distinct foci differ considerably and

is consistent with the hypothesis that these isolates have

arisen independently (Hide et al., 1991, Macleod et al.,

2001).

z Z208 z Z210 n Gup1043 n Gup2078 n Gup1042 n Gup1052 a Mawero31 a Fly48 a UGL b KINUHBW* b KINUC21* b BUGH1* a Masaba. P. a Aaja. M. a Taka. H. a Namwiza. F. a UGK a UGC88 a UGB88 a UGA88 a UGA a UG89/9 a UG89/5 a UG89/3 a UG89/2 a Mela71 a UG89/10 a UG89/8 a Apollo. M. a Fly73 a Papol278 a UG89/1 a UGM a Oketch. E. a UGC a UGG88 a UGI a Mela2 a UGJ a UGH a Katerema116 a Mawero42 a Mela27 a UGE a Bumanda25 a Magola18 a Iyolwa116 a Mawero80 a Mawero85 a Buteba135 a Fly97 a Tira27 a Bumanda146 a MelaPig1 a MelaPig2 a Iyolwa147 a Katerema311 a Iyolwa125 a Mawero32 a Mawero65 a Mela3 a Tira17 b BWP1

% similarity

Fig. 4. Dendogram representing relationships between T. brucei stocks. Dendogram was constructed using cluster analysis of banding patterns of

T. brucei stocks, generated from MGE-PCR analysis, using single primer REV B. Stocks labeled (a) were collected from the Busoga focus in 1989–

1991, stocks labeled (b) were collected from the same focus in 2000. Stocks labeled (n) were collected in the Nyanza (Kenya) region of the Busoga

focus in the 1960s and stocks labeled (z) were collected from the Zambian focus. Group 1 represents human-infective samples from the Zambian

focus, group 2 represents SRA positive isolates (or previously designated T. b. rhodesiense by resistant to lysis by human serum), group 3 represents

SRA negative (human-serum-sensitive, T. b. brucei) isolates.

30 A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32

Differentiation within sub-species to strain level is

further demonstrated by the observation of contrasting

population structures for T. b. brucei and T. b. rhodes-

iense stocks, with the latter exhibiting a far more re-

stricted number of genotypes, resulting in a

predominantly clonal population (Tibayrenc and Ayala,2002, Tibayrenc et al., 1991). The monophyletic clus-

tering of these human-infective strains was further in-

vestigated by the reconstruction of a minimum evolution

tree, which confirmed that the branch leading to the

cluster of identical human-infective strains was well

supported, thereby providing additional evidence for a

monophyletic cluster. This was further supported by the

observation that T. b. rhodesiense strains isolated from

humans and cattle in the Busoga focus in 2000, wereidentical to those isolated from the humans and cattle in

the same focus in 1989, thereby illustrating the remark-

able stability of T. b. rhodesiense strains within sleeping

Fig. 5. Phylogenetic tree representing relationships between 26 representative T. brucei stocks. Phylogenetic tree of 26 representative strains

reconstructed on the basis of presence or absence of each amplified band. The data were converted into a distance matrix of percentage identity

(p value), and these distances were used to reconstruct a tree using the Minimum Evolution algorithm as implemented in MEGA version 2.1 (Kumar

et al., 2001). The reliability of this tree and support for the branch leading to the cluster of human-infective strains was demonstrated using the

interior branch test with 2000 replicates.

A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32 31

sickness foci and reinforcing the important reservoir role

of cattle in maintaining human-infective strains.

The sub-specific and strain differentiation generated by

MGE-PCR amplification of T. brucei stocks is consistent

with previously reported results using RFLP techniques

(Hide et al., 1994). However, MGE-PCR has advantages

over RFLP techniques in terms of rapidity and lowerparasite material requirements and as the RIME element

has been found on several T. brucei chromosomes

(Blackwell and Melville, 1999; El-Sayed et al., 2000), a

relatively large proportion of the trypanosome genome

can be assessed for variation using MGE-PCR.

An important feature of MGE-PCR is its specificity

in the presence of host DNA, conferring significant

advantages over typing techniques such as RAPD,which generate similar levels of variation, but are re-

stricted to use on purified parasite preparations. The

ability to amplify trypanosome DNA in the presence of

host DNA favours future adaptation of MGE-PCR to

direct analysis of field samples containing large amounts

of host DNA. The stability of the technique when ap-

plied to a variety of sample preparations is further

supported by the observation that identical banding

patterns were generated from DNA extracted from

procyclic isolates collected from Tororo in 1989 and

bloodstream isolates collected in 2000 from the same

focus. This reproducibility between different forms of

the parasite and different DNA extraction methods is

particularly important in terms of comparing strainsfrom new and established foci (archived DNA), in order

to determine the origins of outbreaks, spread of disease

and establishment of new foci.

A current limitation of the technique is its inability to

provide data on samples that contain mixed genotypes.

These mixtures can be detected using minisatellites

markers (MacLeod et al., 2000, 2001). Further work is

required to address the issue of detection of mixed ge-notypes using MGE-PCR. However, MGE-PCR gen-

erates a level of intra-specific variation sufficient to

perform large-scale analyses of strain variation within

T. brucei populations, using a single step PCR reaction.

MGE-PCR represents a rapid and specific molecular

method for the differentiation of T. brucei stocks based

on comparative genomics at the intra specific level.

32 A. Tilley et al. / Experimental Parasitology 104 (2003) 26–32

Analogous studies with Toxoplasma spp. (Terry et al.,2001) suggest that this technique can be adapted for use

as a strain typing method for a wide variety of organ-

isms. The application of MGE-PCR to the analysis of

samples from a recent and geographically distinct out-

break of sleeping sickness and their comparison to

samples from established foci is currently in progress

and will further determine the efficacy of MGE-PCR as

a tool for large-scale investigations into the origins andspread of sleeping sickness.

Acknowledgments

We thank Professor Ian Maudlin, Dr. Rebecca Terry

and Dr. Kim Picozzi for helpful discussions and thesleeping sickness patients for provision of samples. We

would also like to thank the University of Salford, The

Royal Society, The Wellcome Trust, The Medical Re-

search Council and The Department for International

Development (DFID) for funding. The views expressed

in this work do not necessarily reflect those of the

funding bodies.

References

Blackwell, J., Melville, S., 1999. Status of protozoan genome analysis:

trypanosomatids. Parasitology 118 (Suppl.), S11–S14.

El-Sayed, N., Hegde, P., Quackenbush, J., Melville, S., Donelson, J.,

2000. The African trypanosome genome. International Journal of

Parasitology 30, 329–345.

Fevre, E.M., Coleman, P.G., Odiit, M., Magona, J.W., Welburn, S.C.,

Woolhouse, M., 2001. The origins of a new Trypanosoma brucei

rhodesiense sleeping sickness outbreak in eastern Uganda. Lancet

358, 603–604.

Gibson, W.C., Marshall, T.F., Godfrey, D.G., 1980. Numerical

analysis of enzyme polymorphism. A new approach to the

epidemiology and taxonomy of trypanosomes of the genus

Trypanozoon. Advances in Parasitology 18, 175–245.

Gibson, W.C., Backhouse, T., Griffiths, A., 2002. The human serum

resistance associated gene is ubiquitous and conserved in Trypan-

osoma brucei rhodesiense throughout East Africa. Infection,

Genetics and Evolution 1, 207–214.

Hasan, G., Turner, M.J., Cordingley, J.S., 1984. Complete nucleotide

sequence of an unusual mobile element from Trypanosoma brucei.

Cell 37, 333–341.

Hide, G., Tilley, A., 2001. Use of mobile genetic elements as tools for

molecular epidemiology. International Journal for Parasitology 31,

599–602.

Hide, G., Angus, S., Holmes, P.H., Maudlin, I., Welburn, S.C., 1998.

Comparison of Trypanosoma brucei strains circulating in an

endemic and epidemic sleeping sickness focus. Experimental

Parasitology 89, 21–29.

Hide, G., Tait, A., Maudlin, I., Welburn, S.C., 1996. The origins,

dynamics and generation of Trypanosoma brucei rhodesiense

epidemics in East Africa. Parasitology Today 12, 50–55.

Hide, G., Welburn, S.C., Tait, A., Maudlin, I., 1994. Epidemiological

relationships of Trypanosoma brucei stocks from South East

Uganda: evidence for different population structures in human

infective and non-human infective isolates. Parasitology 109,

95–111.

Hide, G., Buchanan, N., Welburn, S.C., Maudlin, I., Barry, J., Tait,

A., 1991. Trypanosoma brucei rhodesiense: characterisation of

stocks from Zambia, Kenya and Uganda using repetitive DNA

probes. Experimental Parasitology 72 (4), 430–439.

Kimmel, B.E., Ole-MoiYoi, O.K., Young, J.R., 1987. Ingi, a 5.2 kb

dispersed sequence element from Trypanosome brucei that carries

half of a smaller mobile element at either end and has homology

with mammalian lines. Molecular and Cellular Biology 7,

1465–1475.

Kumar, S., Tamura, K., Jakobsen, I.B., Nei, M., 2001. in: MEGA2:

Molecular Evolutionary Genetics Analysis, 2.1 ed. Arizona State

University, Tempe, Arizona.

Lanham, S.M., Godfrey, D.G., 1970. Isolation of salivarian trypan-

osomes from man and other mammals using DEAE-cellulose.

Experimental Parasitology 28, 521–534.

Macleod, A., Welburn, S.C., Maudlin, I., Turner, C.M.R., Tait, A.,

2001. Evidence for multiple origins of human infectivity in

Trypanosoma brucei revealed by minisatellite variant repeat map-

ping. Journal of Molecular Evolution 52, 290–301.

MacLeod, A., Tweedie, A., Welburn, S.C., Maudlin, I., Turner,

C.M.R., Tait, A., 2000. Minisatellite marker analysis of Trypan-

osoma brucei: reconciliation of clonal, panmictic and epidemic

population genetic structures. Proceedings of the National Acad-

emy of Sciences 97, 13442–13447.

Masiga, D.K., Tait, A., Turner, C.M.R., 2000. Amplified restriction

fragment length polymorphism in parasite genetics. Parasitology

Today 16, 350–353.

Mathieu-Daude, F., Stevens, J., Welsh, J., Tibayrenc, M., McLelland,

M., 1995. Genetic diversity and population structure of Trypano-

soma brucei: clonality versus sexuality. Molecular and Biochemical

Parasitology 72, 89–101.

Murphy, N.B., Pays, A., Tebabi, P., Coquelet, H., Guyaux, M.,

Steinert, M., Pays, E., 1987. Trypanosoma brucei repeated element

with unusual structural and transcriptional properties. Journal of

Molecular Biology 195, 855–871.

Nei, M., Stephens, J.C., Saitou, N., 1985. Methods for computing the

standard errors of branching points in an evolutionary tree and

their application to molecular data from humans and apes. Mol.

Biol. Evol. 2, 66–85.

Onyango, R., Van Hoeve, K., De Raadt, P., 1966. The epidemiology

of Trypanosoma rhodesiense sleeping sickness in Alego location,

Central Nyanza, Kenya. I. Evidence that cattle may act as reservoir

hosts of trypanosomes infective to man. Transactions of the Royal

Society of Tropical Medicine and Hygiene 60, 175–182.

Page, R.D.M., 1996. TREEVIEW: an application to display phylog-

enetics trees on personal computers. Computer Applications in the

Biosciences 12, 357–358.

Tait, A., 1980. Evidence for diploidy and mating in trypanosomes.

Nature 287, 536–538.

Terry, R.S., Smith, J.E., Duncanson, P., Hide, G., 2001. MGE-PCR: a

novel approach to the analysis of Toxoplasma gondii strain

differentiation using mobile genetic elements. International Journal

for Parasitology 31, 155–161.

Tibayrenc, M., Ayala, F.J., 2002. The clonal theory of parasitic

protozoa; 12 years on. Trends in Parasitology 18, 405–410.

Tibayrenc, M., Kjellberg, F., Arnaud, J., Oury, B., Brenvere, S.F.,

Darde, M.L., Ayala, F.J., 1991. Are eukaryotic microorganisms

clonal or sexual? A population genetics vantage. Proceedings of the

National Academy of Sciences USA 88, 5129–5133.

Welburn, S.C., Odiit, M., 2002. Recent developments in human

African trypanosomiasis. Tropical and Travel Associated Diseases,

477–484.

Welburn, S.C., Picozzi, K., Fevre, E.M., Coleman, P.G., Odiit, M.,

Carrington, M., Maudlin, I., 2001. Identification of human-

infective trypanosomes in animal reservoirs of sleeping sickness

in Uganda by means of serum-resistance-associated (SRA) gene.

Lancet 358, 2017–2019.