MOLECULAR

i?itLmIcAL
P-LOGY
ELSEVIER

Molecular and Biochemical Parasitology 83 (1996) 141- 152

DNA markers define two major phylogenetic lineages of

Trypanosoma

CTUZi

Ricardo P. Souto, Octavia Fernandesb, AndrCa M. Macedo, David A. Campbelld,

Bianca Zingales**
Departamenro de Bioquimica. Institute de Quimica, Universidade de &io Paula, Caixa Postal 26077, CEP 05599-970.
Sio Paula, Brazil
bDepartamento de Bioquimica e Biologia Molecular, Institute Oswaldo Cru: (FIOCRUZ) and Departamento de Patologia. UERJ.
Rio de Janeiro, Brazil
Depariamento de Bioyuimica e Imunologia, Insritulo de Citkcias BioMgicas, UFMG, Belo Horkonte, Brazil
Departmenr of Microhio1og.v and Immunologic and Molecular BiologJj Institute, UCLA School of Mediche. Los Angeles, USA

Received 24 June 1996; revised 11 September 1996; accepted 17 September 1996

Abstract
Parasitic protozoa within the taxon T~ypano~orna cruzi are considered to be derived from multiple clonal lineages,
and show broad genetic diversity as a result of propagation
with little or no genetic exchange. We have analyzed a
wide sample of T. cruzi isolates from vertebrate and invertebrate
hosts by PCR amplification
of a ribosomal RNA
gene sequence, a mini-exon gene sequence and random amplified polymorphic
DNA (RAPD). Amplification
of the
distinct rDNA and mini-exon gene sequences indicated a dimorphism within both of the tandemly-repeated
genes: 125
or 110 bp products for rDNA and 300 or 350 bp products for the mini-exon. Within individual isolates, one of three
associations was observed: the 125 bp rDNA product with the 300 bp mini-exon product (defined as group l), the 110
bp rDNA product with the 350 bp mini-exon
product (defined as group 2) and the presence of both rDNA
amplification
products with the mini-exon group 1 product (group l/2). The RAPD analysis showed variability
between individual isolates, however, tree analysis clearly indicated the presence of two major branches. Interestingly,
the rDNA/mini-exon
group 2 isolates correlated precisely with one branch of the RAPD-derived
tree; group I and
group l/2 isolates correlated with the other branch. Our studies show a clear division of T. cruzi into two major
lineages presenting a high phylogenetic
divergence. Hypotheses are discussed to explain the origin of the two lineages
as well as isolates that are hybrid for group 1 and 2 rDNA markers. Copyright 0 1996 Elsevier Science B.V.
Keywords:

Mini-exon

genes; RAPD;

Ribosomal

RNA genes;

Abbreviations: ME. mini-exon; PCR, polymerase

ribosomal
RNA genes; rRNA, ribosomal
RNA.
*Corresponding

author.

Tel.:

0166-685 1!96/$15.00 Copyright

PfISO166-6851(96)02755-7

+ 55 11 8183812;

0 1996 Elsevier

fax:

Trypanosoma

chain reaction: RAPD,

+ 55 11 8155579;

Science B.V. All rights

e-mail:

reserved

cruzi

randomly

amplified polymorphic

bszodnas@quim.iq.usp.br

DNA;

rDNA,

142

R.P. Souto et al. / Molecular and Biochemical Parasitology 83 (1996) 141-152

1. Introduction

The protozoan
Trypanosoma cruzi is the
causative agent of Chagas disease, which affects
18 million people in Latin America, and has a
broad host range, infecting wild and domestic
animals. In humans, a striking feature of Chagas
disease is the variety of clinical presentations that
include the indeterminate form, cardiomyopathies
and digestive alterations. The severity and symptoms of the disease vary in different geographical
regions. It has been suggested that such variability
may be the result either of heterogeneity among
the T. cruzi isolates or of the host immune response.
Early studies on the population genetics of T.
cruzi revealed substantial
isozymic variability
among isolates defining three major groups or
zymodemes [I ,2]. Subsequently, these studies
were extended to include 15 gene loci coding for
enzymes from 121 T. cruzi isolates encompassing
both the domestic and sylvatic cycles with a broad
geographical distribution (United States to Southern Brazil and Chile) [3]. This analysis increased
the inferred isozyme genotype variability of this
organism to 43 natural isozyme strains or
clonets, that could not be grouped in few natural
clusters. Strong linkage disequilibrium observed
with these 15 markers suggested that sexual reproduction is rare or absent in T. cruzi. The main
conclusion of these studies was that the population structure of this parasite is clonal rather than
sexual. As a consequence, the present genetic and
biological variability of this parasite might be
explained by the isolated evolution of multiple
clonal lines [4].
Restriction-fragment-length
polymorphism
in
kinetoplast DNA of T. cruzi [5] and DNA fingerprinting [6] also showed genetic heterogeneity in
the parasite, however analysis of the ribosomal
RNA genes (rDNA) indicated dimorphic [7]
rather than polymorphic heterogeneity. Furthermore, a recent review [8] of isoenzyme and RAPD
markers suggested that T. cruzi might be subdivided into two major phylogenetic lineages, each
retaining considerable heterogeneity.
We have demonstrated that a sequence of approximately 100 bp located at the 3 end of the

24%~ ribosomal RNA (rRNA) gene of T. cruzi is

dimorphic and can be used as a target for sensitive parasite detection. Dimorphism in the 24Sc(
rRNA target region allowed the division of 16 T.
cruzi isolates into two groups [7]. This observation
was confirmed by others [9] in 18 isolates from
North America. In the present study we have
analyzed 88 isolates of T. cruzi derived from
infected humans, insects or animals from the sylvatic cycle originating in Brazil, Bolivia, Chile,
Venezuela and Argentina and including ten zymodemes of the most representative
groups
(clonets) identified by [4]. In addition, we herein
describe a dimorphism in the T. cruzi mini-exon
gene repeat [lo- 131 that correlates with the ribosomal DNA genotypes [7]. Analysis of these two
independent nuclear markers and confirmatory
RAPD data prove the division of T. cruzi into
two major phylogenetic lineages. Hypotheses concerning the origin of the observed genotypes are
discussed.

2. Materials and methods

2.1. Parasites and nucleic acids isolation

Eighty-eight T. cruzi strains and clones (comprising recently collected or cryopreserved isolates
and cloned stocks) derived from naturally infected
hosts from Latin America were analyzed. Details
are presented in Table 1. Epimastigote forms were
grown in LIT (liver infusion tryptose) medium
[14] and DNA preparations were obtained by
phenol-chloroform extraction or simply by boiling
cells for 10 min in 10 mM Tris-HCl, pH 7.0; 1
mM EDTA and 150 mM NaCl. Southern blots of
Pst I-digested genomic DNA were hybridized with
a 24Sa-rDNA probe (GlOlp10[15]) and with a
mini-exon conserved probe (TcME, a 30-mer
oligonucleotide corresponding to positions 5-35
of the T. cruzi mini-exon gene [lo]).
2.2. DNA anlpliJication
Polymerase chain reaction (PCR) amplification
[16] of a divergent domain from 24Sa ribosomal
RNA gene was achieved with the primers 5-

143

R.P. So&o et al. / Molecular and Biochemical Parasitology 83 (1996) 141 -152
Table

Summary

of T. cru:i. isolates,

hosts.

Host

Isolate

Human

I IS (BR)
I82 (BR)
222 (BR)
226 (BR)
231 (BR)
1017 (BR)
9280 cll (BOL)
I3 379 cl7 (BOL)
Al38 (BR)
Ana (BR)
B147 (BR)
8167 (BR)
Basileu (BR)
Cl (BR)
CAI (ARC)
CAN111 cl1 (BR)
CBB cl3 (CH)
D143 (BR)
D207 (BR)
Esmeraldo cl3 (BR)
JM (BR)
M6241 cl6 (BR)
MAS cl1 (BR)
MC (BR)
MN cl2(CH)
MT01 (BR)
MT03 (BR)
MT05 (BR)
MT06 (BR)
MT07 (BR)
MT08 (BR)
MT09 (BR)
MTIO (BR)
MTII (BR)
MT12 (BR)
MT13 (BR)
MT14 (BR)
MT17 (BR)
MT18 (BR)
MT19 (BR)
MT20 (BR)
MT21 (BR)
MT24 (BR)
MT25 (BR)

n.a.. no amplification
Chile; VEN.
lsoenzyme

(country)

product;

nd.,

country

of origin

rDNA

ME

1,2

I
I
n.d.
I
nd.
2
I
nd.
I

2
I;2
2
2
I!?
2

and typing
Host

2
2

G (BR)
Gamba cll (BR)
Gamba3 (BR)
M563l cl5 (BR)
SC28 (BR)
SC33 (BR)
Cutia cll (BR)
Esquilo cl1 (BR)
M226 (BR)

Rodent

I
2

n.a.
nd.
I
I
2
2
I
I
I
I
I
1

I
I:2
2
2
I
I

Armadillo

P (BR)
1004 (BR)
1006 (BR)
1009 (BR)
1018 (BR)
1023 (BR)
Bug2148 cll (BR)
Bug2149 cl10 (BR)
CL (BR)
CL Brener (BR)
MT15 (BR)
MT16 (BR)
MT22 (BR)
MT23 (BR)
0PS22 (VEN)
SC43 cl1 (BOL)
SO3 cl5 (BOL)
SO34 cl4 (BOL)
SP104 cll (CH)
TU18 cll (BR)
Tulahuen (CH)
YuYu (BR)

Bug

I
I
I
I
I
I
I
I
I
I
I
not determined:

rDNA

(country)

Y (BR)
1001 (BR)
Cuica cll (BR)
Dm28 (VEN)
Dm30 (VEN)

Opossum

I
I
n.d.
I
I
I
n.d.
nd.
I

Isolate

MXCH88 (CH)
NR cl3 (CH)
OPS2l clll (VEN)
PI1 cl3 (BOL)
P209 cll (BOL)
RA (ARG)
SilvioXlO cll (BR)
Tatiana (BR)

1
1

I ;2

results

cl. indicates

cloned

organism;

ARC.

Argentina:

n.d.

I
17
I_
2
2
2
I
2
I
I
2
2
2
2
2
2
I
2
2
7
2
2
2
2
2
2
2
2
I,2
I2
I2
I
I
I
2
I
I
2
2
I:?
2
2
I
2
2

BOL, Bolivia; BR. Brazil; CH,

Venezuela.

classification

of stocks

[4]: clonet

I7 (SilvioXlO

cll),

I9 (I3 379 ~17. Cutia cll, Gamba

cll, P209 cll, SP104 cll, PI1 ~13,

OPS2l cll I). 20 (Cuica cll. Esquilo cll, SO34 ~14). 21 (OPS22). 27 (CAN111 cll), 30 (Esmeraldo
c13), 32 (CBB ~13. MASI cll.
MXCH88 ~12. TUl8 cl2), 35 (M6241 cl6), 36 (M5631 cl5), 39 (9280 cll, Bug2148 cll. Bug2149 ~110, MN cl?, NR ~13, SC43 cll, SO3
c15).

144

R.P. Souto et al. 1 Molecular and Biochemical Parasitology 83 (1996) 141-1.52

AAGGTGCGTCGACAGTGTGG
(D71) and 5TTTTCAGAATGGCCGAACAGT
(D72), as described in [7]. After heat denaturation (4 min at
94C), samples were submitted to 30 cycles of
three temperatures (94C/l min, 6OC/l min and
72C/l min). The reaction product was run on
7.5% polyacrylamide gels and alkaline blotted to
Zeta Probe membranes. Southern blots were hybridized to probes a 1 and x2 which are complementary, respectively, to the 125 and 110 bp
amplified sequences [7]. Densitometry of the autoradiograms was obtained in a Shimadzu CS900 Scanner.
Amplification of part of the intergenic region of
T. cruzi mini-exon genes [lo- 131 was achieved
using a pool of three oligonucleotides (see Fig.
2(A)): 5-GTGTCCGCCACCTCCTTCGGGCC
(TC 1, group 1-specific),S-CCTGCAGGCACACGTGTGTGTG
(TC2, group 2-specific), and 5CCCCCCTCCCAGGCCACACTG
(TC, common
to groups 1 and 2) and the following thermal
profile: 94C/l min; 27 cycles of 94C/30 s, 55C/30
s, 72C/30 s; 72C/lO min. Amplification products
were analyzed in 1.5% agarose gels.
RAPD amplification was performed with three
primers arbitrarily selected from laboratory stocks:
M 13 Forward-40 (M 13F) (5-GTTTTCCCAGTCACGAC),
/igt 11F (5-GACTCCTGGAGCCCG) and L15996 (5- CTCCACCATTAGCACCAAAGC), the last one devised to amplify D loop
from human mitochondrial DNA [17]. Parasite
DNA (1 ng) was amplified in 10 ~1 of reaction
mixture with Tuq DNA polymerase (a gift from
Cenbiot, Brazil) according to published conditions
[18]. Following amplification, 3 ~1 of the reaction
was electrophoresed in 5% polyacrylamide gels and
silver stained.
2.3. DNA cloning sequencing
The ribosomal rDNA amplification products
were cloned into the HincII site of M13mp19 and
the mini-exon amplification products were cloned
into pCRI1 (InVitrogen). DNA sequencing was
performed using Klenow or the Sequenase Kit
(Amersham).

2.4. RAPD data analysis

Gel photographs were enlarged to an 18 x 24
cm size, scored by eye and the band patterns
entered by hand in a computer program DNAPOP described by Pena and Nunes (Fingerprint
News 2:7-8, 1990). The parameter D = 1 -S
(where S= 2nxy/(nx + ny); nx and ny being the
number of bands in the isolate x and y and nxy,
the number of shared bands between isolates I
and y) was used as genetic distance measure between all pairs of isolates. The phenetic tree
based on D values was constructed by using UPGMA algorithm from the MEGA computer program [19].

3. Results
3.1. Sequence dimorphism in the 24Sa rDNA in
T. cruzi isolates

DNA from T. cruzi was amplified by PCR

employing oligonucleotides D71 and D72 and
analyzed by gel electrophoresis followed by ethidium bromide staining. All eighty-eight isolates
gave rise to PCR amplification products of 125 bp
(group l), 110 bp (group 2) or both (group l/2)
(Table 1). A typical amplification experiment is
shown in Fig. l(A). Hybridization of the products
with group l- or 2-specific probes [7] confirmed
the respective identities of the products (Fig. l(B)
and (C)). In every instance, the 125 and 110 bp
products were found to hybridize with the group
1 and 2 probes respectively, indicating that size
alone is a sufficient marker for group affiliation.
This result was consistent with previous observations that the 24Sc( rDNA target region allows the
division of T. cruzi isolates into two groups [7]. In
a portion of the samples from human (6/53) and
bug (4/21) sources both the 125 and 110 bp
products were amplified (cf. Fig. 1, lanes 1, 3 and
8). These isolates were designated as group l/2
(see Table 1). It should be noted that six of these
samples were from cloned cell isolates, ruling out
the possibility of mixed populations of parasites.
We have previously reported the sequences of
the 24Sa rDNA target region of five isolates of

145

R.P. Souto et (11.i Molecular and Biochemical Parasitology 8.1 (1996) 141~ 152

group 1 and four isolates of group 2, concluding

that within each group high sequence homogeneity was found [7]. The alignment of representative
sequences of group 1 (GenBankTM accession number M28885, positions
1891-2013)
and group 2
( GenBankTM accession number L14468) indicated
a considerable
sequence divergence. In the present
study the 110 bp products
of three additional
isolates of group 2 and of three isolates of group
1,12were cloned into M 13mp 19 and several clones
from each amplification
reaction were sequenced
(data not shown). Alignment
of the sequences of
these isolates as well as those of isolates already
analyzed [7] indicate that within each of the three
groups the sequence shows 98- 100% identity with
minor variation in homopolymeric
regions, confirming
previous
observations
[7]. Comparative
alignment
of the 125 bp (group 1) and 110 bp
(group 2) sequences shows 75585% identity.
3.2. Mini-exon

gem

typing of T. cruzi isolutrs

Our second approach to the characterization

T. cruzi isolates is based on PCR amplification

of
of

- 141
- 105
- 78

- 141
- 105
- 78

- 141
- 10.5
- 78

of rDNA target sequence from T.

Fig. I. PCR amplification
uuzi isolates. Lanes IllO.
Bug2149 ~110, CL, SO3 ~15, Y,
Esmeraldo ~13, SilvioXlO cll, Tulahuen. NR ~13, Dm28. negative control. (A) 7.5% polyacrylamide
gel. ethidium bromide
stained. (B) and (C) DNA blot of the gel shown in panel (A)
hybridized
to probes rl and x2. respectively.
Molecular
size
markers in bp derived from pUCl9 digested with Sau3AI are
assigned on the right. The rDNA group of each isolate is
indicated.

Table 2
Percent similarity between the mini-exon
spacer regions of T.cruzi isolates.

gene non-transcribed

Isolate

CL

Yb

SilvioX lo

Tulahuenj

CL
Y
SilvioXlO
Tulahuen

100

97.9
100

57.9
58.3
100

58.6
57.3
90.3
100

GenBankTM accession numbers:

L X62674 [12], X00632 [IO].

U578984

[ 1 I], b K0263 1 [ 131,

a portion
of the tandemly-repeated
mini-exon
gene. Previous experiments,
which showed nonstoichiometric
hybridization
of an intergenic
region probe
from SilvioXlO
to the amplified
products
from the CL and Dm28 strains [12],
suggested
heterogeneity
of this region between
different isolates. In light of this result. and because the intergenic
region has been useful in
distinguishing
among other very closely related
trypanosomatids
such as Lrishrnanic~ spp. [20], we
further analyzed DNA sequences of the T. crux
mini-exon
gene repeats
[lo- 131. Comparative
alignment
of these sequences suggested the presence of two discrete groups that shared identical
39 bp exons and similar 73 bp introns ( z 98%
identity), but which showed only 57759% identity
between the intergenic regions (range 4844494 bp)
(Table 2).
We exploited the differences in the intergenic
regions by designing oligonucleotides
that could
be used in PCR to specifically amplify a portion
of the mini-exon
gene from each of the two
groups. Using a common
oligonucleotide
corresponding to sequences conserved between all three
isolates and two different oligonucleotides
corresponding to unique sequences in each group (see
Fig. 2(A)), we were able to amplify the predicted
300 bp product from isolate CL and the 350 bp
product from stocks SilvioXlO and Dm28 (data
not shown). To determine if other T. cruzi isolates
could be clustered in the same two groups, we
applied the assay to additional
stocks (Table 1).
We found that the assay could be generally applied to T. crud isolates in that all clones but one
(M6241 ~16) yielded either the 300 bp or 350 bp
product. At this point we noticed a correlation

146

R.P. Souto et al. I Molecular

and Biochemical Parasito1og.v 83 (1996) 141-152

between the mini-exon gene products and the

rDNA groupings [7]. To be consistent, we designated the 300 bp product as group 1 and the 350
bp product as group 2. Subsequently. 67 DNA
samples were tested by both the ribosomal RNA
gene assay (Fig. 1 and Table 1) and the mini-exon
gene assay (Fig. 2(B) and Table 1).
To summarize the cumulative results, we found
a strict correlation between the two nuclear markers in 57/67 isolates: 38 isolates were rDNA and
mini-exon group 1 and 19 stocks were rDNA and
mini-exon group 2. Nine isolates that were rDNA
group l/2 were typed as mini-exon group 1. The
remaining isolate (SC43 cl1 ) was rDNA group 2

A
TC

12345678

:g

- 298
- 174

Fig. 2. (A) Schematic representation

of the T. cruzi mini-exon
gene PCR typing assay. The black box represents the 39 bp
mini-exon, the shaded box represents the intron sequence and
the line represents the intergenic region (mean length + 484
bp; determined
by the PILEUP
routine
in the UWGCG
package).
The positions
of type l-specific (106- 128). type
-specific (51-72) and common (inverse complement
of 381402) oligonucleotides
used for PCR are indicated by arrows.
(B) Ethidium bromide staining of mini-exon gene PCR products amplified from 7. cru;i type 1 (300 bp) and type 2 (350
bp) isolates after resolution
by agarose gel electrophoresis.
Genomic
DNA samples were from the following
T. cruzi
isolates: (lanes t-8) G, Dm28, Esmeraldo ~13, Silvio X10 cll.
CAl, Basileu, Bl67 and Al38. The mini-exon group of each
isolate is indicated. Molecular size markers in bp derived from
pUC19 digested with Hurl11 are assigned on the right. Higher
molecular weight products
represent amplification
of tandem

and mini-exon group 1. We analyzed six different

subclones derived from SC43 cl1 and in all cases
the same discrepancy between both markers was
observed.
3.3. Ribosonlal RNA and mini-exon genes
organizution
Genomic arrangements of ribosomal RNA and
mini-exon genes were analyzed in some isolates
from rDNA groups 1, 2 and l/2 (Fig. 3). Nuclear
DNA was digested with PstI, since this enzyme
cleaves the 24Sc( subunit of the rDNA cistron of
T. cmsi [15] and differentially cleaves mini-exon
genes from different strains of the parasite [l I].
The Southern blot was hybridized to a probe
containing the 24%~ rRNA gene from the Y strain
[15] and to a 30-mer oligonucleotide (TcME) from
T. cruzi mini-exon gene.
Data in Fig. 3(A) show that the rDNA probe
hybridizes to a single band of 3.8 kb in two
isolates of group 1 (lanes 1 and 2) and to a band
of 3.6 kb in two isolates of group 2 (lanes 6 and
7) indicating that structural differences in other
variable regions of the 24%~ rRNA gene in each
group exist. The hybridization pattern of three
clones of rDNA group l/2 (lanes 335) shows two
bands of sizes corresponding to fragments of
group 1 and 2, respectively. Densitometry of the
hybridization signals in isolates of group l/2 indicates eight to ten-fold greater copy number of the
group 2-rDNA genes relative to group l-rDNA
genes. Fig. 3(B) shows the hybridization of the
same blot to probe TcME. In isolates of miniexon and rDNA group 1 (lanes 1 and 2) and
isolates of rDNA group l/2 (lanes 3-5) hybridization is observed to a high molecular mass DNA
that contains the entire cluster of mini-exon genes.
In isolates of group 2 (lanes 6 and 7) hybridization to 0.6 and 0.9 kb bands is observed. The 0.6
kb fragment represents the repeating unit of the
mini-exon genes, while the 0.9 kb band may correspond to flanking or orphon genes [21].
Taken as a whole the Southern blot analysis
clearly indicates that the overall structure of both
the rDNA cistron and the mini-exon repeating
units differ between isolates of group 1 and isolates of group 2, supporting group division.

R.P. Souto et al.

/ Molecular and Biochemiccd Parasitology

1234567

147

83 (1996) 141-15-1

1234567

- 1.58 - 1.38 -

- 0.95 - 0.83 -

rDNA

l/2

l/2

l/2

ME

Fig. 3. Southern blot of Pstl-digested

DNA from T. crrc~i isolates (lanes l-7): Y. Basileu, Bug 2149 ~110, NR ~1.3. SO3 ~15, Dm28
and G. (A) Hybridization
to GlOlplO
probe (24s~ rDNA gene from T. cr~i).
(B) Hybridization
to TcME probe (mini-exon
sequence conserved in T. cruzi). Molecular size markers in kb derived from Act857 DNA digested with &oRI and Hind111 are
assigned. The rDNA and mini-exon (ME) groups of each isolate are indicated.

3.4. RAPD

unalysis

of T. cm5

isolates

RAPD has been described as a powerful technique for obtaining

hypervariable
DNA markers
and for establishing
genetic relationships
for T.
cruzi isolates [l&22]. In the present work we
examined
the genomic structure
of 25 T. cruzi
isolates
by RAPD
analysis
in a double-blind
study. The major goal of this analysis was to
investigate
whether there was a broader genetic
correlation
between
parasite
isolates
and the
rDNA
and mini-exon-based
division
into two
groups. Fig. 4 shows a typical RAPD
profile
obtained with the Rgtl 1F primer. The same analy-

sis was done with the other two primers (data not
shown). We detected on average
19 + 4 bands
varying from 11 to 26, depending
on the primer
and isolate analyzed (Fig. 4). In general the patterns observed were distinct for any two stocks,
although
a lower degree of variability
was detected among
isolates
belonging
to the same
group defined by rDNA/mini-exon
genes. Overall
51% of bands were shared among all isolates
studied; 66% of bands were shared among isolates
from mini-exon
group 1 and 72% were shared
among isolates from mini-exon
group 2. On the
other hand, less than 35% of bands were shared
between isolates of the two groups, indicating that

R.P. Souto et al. / Molecular and Biochemical Parasitology 83 (1996) 141-152

148

2,036
1,636

1,018

298

rDNA

1 112112112 1

DA;

1111111111111112222222222

2 112112112112 2

Fig. 4. RAPD profiles obtained with i.gtl 1F primer. Isolate designations are indicated above the respective lane. On the left the
migration of DNA fragments (in bp) of 1 kb ladder (BRL) is shown. The rDNA and mini-exon (ME) groups are indicated.

parasites within the same group are more closely

related to each other than to isolates of the other
group.
The phenogram resulting from RAPD data
analysis using the three primers showed two main
branches (Fig. 5). In the upper branch are clustered isolates from mini-exon and rDNA group 1
as well as stocks from rDNA group l/2 (lineage
l), while isolates from mini-exon and rDNA
group 2 are in the lower branch (lineage 2).
Isolate SC43cll (mini-exon group 1 and rDNA
group 2) was clustered in the upper branch, consistent with the mini-exon and not with the rDNA
grouping.

4. Discussion

In this study, a large selection of T. cruzi isolates from domestic and sylvatic cycles and different geographic
areas
were analyzed
for

polymorphism
in independent
DNA markers
(rDNA, mini-exon genes and anonymous DNA
sequences) indicating a division of the isolates
into two discrete groups. The recognition of a
dimorphic grouping within these isolates represents a relatively stable marker in the evolution of
the parasite and provides a different perspective
on the findings of other researchers, who describe
a great genotypic and phenotypic variability
within T. cruzi.
The hypervariability
observed
during the
molecular characterization of T. cruzi, by multilocus enzyme electrophoresis
[4,22], schizodeme
analysis [S], molecular karyotyping [23] and DNA
fingerprinting [6] has been described as being consistent with no natural clustering and subdivision
into multiple groupings [4]. The above approaches, which reflect differences in organization
of nuclear and kinetoplast genomes, record undefined genetic changes that may be theoretically
as simple as a single nucleotide change at one

R.P.

Souto

et al.

Molecular

and Biochemicul

Purasitologv

83 (1996)

141-15-7

149

the high level of DNA sequence conservation

within each group (approx. 98% for both genes)
reflects diversity among genes (and gene families)
and is consistent with criteria that have been used
to define phylogenetic grouping.
A strict correlation was observed for 57,67
samples that were analyzed simultaneously by
rDNA and mini-exon genes. This suggests that
group 1 and group 2 represent two major clades
in T. crxi. This conclusion is strengthened by the
RAPD analysis that shows approximately 35% of
bands shared between isolates from the two
groups, contrasting with 66 and 72% of bands
shared within isolates of group 1 or 2, respectively. Consequently the mini-exon and rDNA
typing, associated with the RAPD profiles, clearly
indicate two major phylogenetic lineages.
Studies of several T. cruzi stocks by multilocus
enzyme electrophoresis indicated a highly polymorphic population structure with a large number
HEIDWli of zymodemes [4]. However, initial studies by
Miles et al. [1,2] classified T. cruzi populations
into three classical zymodemes Zl, 22 and 23.
Size fractionation
of chromosomal
bands by
pulse-field gel electrophoresis and hybridization to
different DNA probes indicated a certain correlation between karyotype pattern and the classical
zymodeme (Zl and 22) division [23]. In the
present work several isolates classified as Zl and
22 were studied and we conclude that 22 stocks
belong to lineage 1, while stocks classified as Zl
fall into lineage 2. The position of 23 in our
grouping cannot be established since only one
isolate (CAN111 cll) belonging to this zymodeme
was analyzed. Concerning the major isoenzyme
strains 19, 20 and 39, which are widespread over
large ecogeographical areas [4], our typing assays
indicate that clonet 39 belongs to lineage 1, while
clonets 19 and 20 belong to lineage 2.
Although our data show two groups, they do
not rule out the possibility of additional groupings. For example, we were unable to amplify
with the three primers approach (see Section 2)
part of the non-transcribed spacer of the miniFig. 5. UPGMA
tree based on the proportion
of bands not
exon gene from isolate M6241 cl6 (see Table 1).
shared among the taxa. For each isolate the corresponding
due to additional DNA sequence variation.
rDNA and mini-exon (ME) groups are indicated (see Table I).
Cloning of the full-length gene shows that alThe scale refers to the D values obtained
with the three
arbitrary
primers.
though the common TC oligonucleotide-binding

locus (in the case of isoenzyme differences) or as

complex as major changes at multiple loci (in the
case of differences revealed by DNA fingerprinting and karyotyping). Our approach to the analysis of genetic diversity in T. crmi isolates has been
to study very specific details of two gene families
that have proven to be useful as markers of (1)
evolution in most cell types, the ribosomal RNA
genes (for kinetoplastida see [24,25]) and (2) genomic variation in trypanosomatids,
the miniexon gene [20,26]. Our measure of diversity
among T. cruzi isolates consists of 1lo- 125 bp of
the 24% rRNA gene [7], which gives rise to a
functional product and approximately 500 bp of
the mini-exon gene non-transcribed spacer region
[12] that are well defined and can be quantitated
readily. Thus, the constant levels of DNA sequence variation between groups 1 and 2 of T.
crux (20/0for rDNA and 40% for mini-exon) and

150

R.P. Souto et al. I Molecular

and Biochemical

site is conserved, the TC 1 oligonucleotide-binding

site is absent and the sequence of the TC2-binding
site is not conserved (N. Sturm and D.A.C., unpublished results).
In addition, lo/88 isolates were anomalous
since amplification of both 125 and 110 bp rDNA
sequences was obtained (group l/2; see Table 1).
These isolates are mini-exon group 1 (Table 1)
and by RAPD have a genetic structure of group 1
(Fig. 5). Analysis by Southern blot confirms the
presence of both types of rDNA genes in group
l/2 stocks (Fig. 3(A)). Further, it indicates major
structural divergences in the rDNA cistron of
stocks of groups 1 and 2. supporting the conclusion of genotype division. Ribosomal DNA variation in T. cruzi has also been reported by
riboprinting analysis [9]. Another feature of group
l/2 is a close relationship to isozyme strain 39 [4].
Seven stocks belonging to clonet 39 were analyzed
and only SC43 cl1 does not belong to group l/2
(Table 1). The other four isolates classified as
group l/2 (stocks 115, B147, 226 and 1023) have
not been submitted to isoenzyme analysis. Clonet
39 is interesting because it remains unchanged
over a broad geographical range (Chile, Brazil,
Bolivia) and maintains heterozygosity at several
loci [4] suggesting that natural selection has favored this stable genotypic combination.
Various hypotheses could explain the origin of
the two major T. cvuzi lineages. Nevertheless we
favor two of them: (1) groups 1 and 2 represent
two ancestors and group l/2 has arisen from a
genetic transfer event from group 2 to 1 and (2) a
group 2 ancestor has originated the groups 1 and
l/2 by long term clonal evolution.
The first hypothesis is based on divergence of
two lineages (Fig. 6(A)). According to this hypothesis, the independent evolution of genotypes
1 and 2 would justify the straight correlation of
rDNA, mini-exon and RAPD markers. The origin
of rDNA group l/2 isolates could be explained by
an eventual transference of rDNA from individuals of group 2 to 1 (Fig. 6(A), event A) either by
mating or by horizontal gene transfer events. Supporting the hypothesis of genetic exchange in T.
cruzi, two recent reports suggest the formation of
hybrid organisms both in sylvatic parasite populations [27] and in sympatric clinical isolates [28].

Parasito1og.y 83 (1996) 141-152

Fig. 6. Schemes corresponding to two hypotheses concerning

the evolution of T. cruzi. groups. (A) Gene transfer hypothesis.
(B) Long term clonal evolution hypothesis.

Analysis of other genetic markers would indicate

if other sets of genes besides rDNA sequences
have also been transferred in group l/2 and might
help to discriminate between sexual and non-sexual gene transfer in T. cruzi.
According to the second hypothesis (Fig. 6(B))
the common ancestor would have rDNA and
mini-exon genes similar to the modern group 2
(Fig. 6(B), left branch). In the other branch, ancestral type 2 mini-exon genes mutated to group 1
pattern (Fig. 6(B), event A). Subsequently, the
type 1 rDNA gene pattern could have been acquired (Fig. 6(B), event B) originating the modern
stocks having both rDNA sequences (group l/2)
and being mini-exon 1. The present rDNA 1 and
mini-exon 1 stocks would have appeared by loss
of type 2 rDNA cistrons (Fig. 6(B), event C).
In both hypotheses the genotype of isolate
SC43 cl1 (rDNA type 2 and mini-exon type 1)
could have derived from rDNA group l/2 following the complete loss of rDNA type 1 genes (Fig.
6(A), event B and Fig. 6(B), event D). In fact,
SC43 cl1 belongs to the same isoenzyme classification of group l/2 (clonet 39) and is located in the
upper branch of RAPD dendrogram with group
l/2 isolates (Fig. 5). A preferential loss of type 1
rDNA may also have happened in group l/2
isolates, as suggested by Southern blot experiments (Fig. 3(A)) indicating that l/2 stocks have
approximately
lo-fold more copies of type 2
rDNA genes than type 1 genes.
Whichever hypothesis actually describes the
origin of the two lineages of T. cvu~i, both the
conservation of multigene families in one given
lineage and the divergence among the two lineages

R.P. Smtro et al. i Moledar

and Biochemical

are products of evolution of multigene families as

cohesive units [29]. Isolates from group l/2 could
be particularly
useful in the study of specific
mechanisms
of copy homogenization
in multigene
families.
The molecular typing based on PCR-amplification of rDNA and mini-exon sequences represents
specific and simple methodological
approaches for
defining the two major lineages. The phylogenetic
distance
between the two lineages of T. cruzi
revealed in this study may be sufficient to define
two taxa. However
at this point subspeciation
would not be an advantage
for the study of
Chagas disease because it confers no predictive
value for epidemiological
studies or determination
of the disease potential
of the isolates. For the
purpose of clarity and communication
among researchers who are studying the distinct biological
properties of the parasite we hereby designate the
major branches as lineage 1 and lineage 2.

Acknowledgements
We are deeply indebted to Dr Michel Tibayrenc
(ORSTOM)
for critical reading of the manuscript
and insightful discussion
of the data. We would
like to thank Dr Michel Tibayrenc,
Dr Christian
Barnab
and Dr Simone Breni&e from ORSTOM
and Professor JosC Rodrigues Coura and his colleagues from FIOCRUZ
for providing
us with
several
parasite
isolates;
Dr Wim
Degrave
(FIOCRUZ)
and Dr Antonio Teixeira (UnB) for
use of their laboratory
facilities; Professor SCrgio
Pena
(UFMG),
Professor
Carlos
Morel
(FIOCRUZ),
Professor
Erney Camargo
(USP)
and Dr Nancy Sturm (UCLA) for critical suggestions and encouragements.
This work was funded
by FundaCgo de Amparo ri Pesquisa do Estado de
Sdo Paulo (FAPESP)
and Conselho National
de
Pesquisa (CNPq). D.C. is a Burroughs Wellcome
New Investigator
in Molecular Parasitology.

References
[I] Miles. M.A.. Souza, A., Povoa. M., Shaw, J.J.. Lainson.
R. and Toye,

P.J. (1978) lsozymic

heterogeneity

of TV?,-

Parasitology
parlosorna

8.1 (19Y6) 141-152

cruzi

in the first autochtonous

151
patients

with

Chagas disease in Amazonian

Brazil. Nature 272. 819821.
[2] Miles, M.A.. Lanham. SM., Souza, A.A. and P6voa, M.
(1980) Further enzymic characters
of Trypanosoma wrzi
and their evaluation
for strain identification.
Trans. R.
Sot. Trop. Med. Hyg. 74. 221-337.
[3] Tibayrenc.
M.. Ward. P.. Moya. A. and Ayala. F.J.
( 1986) Natural populations
of Tr~punosornu wk.
the
agent of Chagas disease. have a complex
multiclonal
structure.
Proc. Nat]. Acad. Sci. USA 83, 293~ 296.
[4] Tibayrenc.
M. and Aydla, F. (1988) Isoenzyme variability
in Tr~pamwoma cruzi. the agent of Chagas disease: genetical. taxonomical
and epidemiological
significance. Evolution 42 777-292.
[S] Morel. C.. Chiari.
E.. Camargo.
E.P.. Mattei,
D.M.,
Romanha.
A.J. and Simpson. L. (1980) Strains and clones
of Tr~panosoma cruzi can be characterized
by restriction
endonucleases
fingerprint
of kinetoplast
DNA molecules.
Proc. Natl. Acad. Sci. USA 77. 6810~6814.
[6] Macedo.
A.M., Martins,
M.S.. Chiari.
E. and Pena.
S.D.J. (1992) DNA fingerprinting
of Tryparlosoma crux: a
new tool for characterization
of strains and clones. Mel,
Biochem. Parasitol. 55. 147- 154.
[7] Souto. R.P. and Zingales, B. (1993) Sensitive detection
and strain classification
of Tr~~panowmtr cr~c:i by amphfication of a ribosomal
RNA sequence.
Mol. Biochem.
Parasitol. 62, 45- 51.
[8] Tibayrenc,
M. (1995) Populations
genetics of parasitic
protozoa
and other microorganisms.
Ad\. Parasitol.
36.
47-115.
[91 Clark. C.G. and Pung. O.J. (1994) Host specificity of
ribosomal
DNA variants
in sylvatic Tr>pawsma
crrrzi
from North America. Mol. Biochem. Parasitol. 66. l75179.
[lOI De Lange. T.. Berkvens. T.M.. Veerman. H.J.G.. Frasch.
A.C.C.. Barry, J.D. and Borst. P. (1984) Comparison
of
the genes coding for the common 5 terminal sequence of
messenger
RNAs in three trypanosome
species. Nucl.
Acids Res. 13, 4431~-4443.
C.. Taylor, Z.A. and Buck. G.A. (1989)
[Ill McCarthy-Burke,
characterization
of the spliced leader genes and transcripts in Tr~~pano.vom~ cruzi. Gene 82, I 77.m 189.
[I:1 Murthy, V.K., Dibbern. K.M. and Campbell. D.A. (1992)
PCR amplification
of mini-exon genes differentiates
T,-j,par7osor~7cr crxi
from Tr~~l~amwmtr rcu7q4. Mol. Cell.
Probes 6. 237-243.
1131 Milhausen. M.. Nelson, R.G.. Sather, S., Selkirk, M. and
Agabian.
N. (1984) Identification
of a small RNA containing the trypanosome
spliced leader: a donor of shared
5 sequences of trypanosomatid
mRNAs? Cell 38. 721
779.
u41 Castellani, 0.. Ribeiro. L.V. and Ferreird. J.F. (1967)
Differentiation
of Tr~~pamwm~a cwzi in culture. J. Protozool. 14, 447-451.
[I51 Arruda. M.V., Reinach. F.C.. Colli. W. and Zingales. B.
(1990) Sequence of the 24s alpha ribosomal
RNA gene

R.P. Souto et al. 1 A4oleculur and Biochemical Parasitology 83 (1996) 141-152

152
and characterization

of a corresponding pseudogene from

Trypanosoma cruzi. Mol. Biochem. Parasitol. 40, 35-42.

[16] Saiki, R.K.. Gelfand, D.H., Stoffel, S., Scharf, S.J.,

Higuchi, R., Horn, G.T., Mullis, K.B. and Erlich, H.A.
(1988) Primer-directed enzymatic amplification of DNA
with a thermostable DNA polymerase. Science 239, 4877
491.
[17] Vigilant, L., Pennington, R., Harpending, H., Kocher.
T.D. and Wilson, AC. (1989) Mitochondrial DNA sequences in single hairs from a southern African population. Proc. Natl. Acad. Sci. USA 86, 9350-9354.
[18] Steindel, M.. Dias Neto, E., Menezes, C.L.P., Romanha,
A.J. and Simpson, A.J.G. (1993) Random amplified polymorphic DNA analysis of Trypanosoma cruzi strains.
Mol. Biochem. Parasitol. 60. 71-80.
[19] Kumar, S., Tamurd, K. and Nei, M. (1993) Mega: Molecular Evolutionary Genetics Analysis. Version 1.0, The
Pennsylvania State University, University Park, PA16802
[20] Fernandes. O., Murthy, V.K., Kurath. U., Degrave,
W.M. and Campbell, D.A. (1994) Mini-exon gene variation in human pathogenic Leishmania species. Mol.
Biochem. Parasitol. 66, 261-271.
[21] Parsons, M., Nelson, R.G. and Agabian, N. (1986) The
trypanosome spliced leader small RNA gene family:
stage-specific modification of one of several similar dispersed genes. Nucleic Acids Res. 14. 1703~1718.
[22] Tibayrenc, M., Neubauer, K., Barnabe, C., Guerrini F..
Skarecky, D. and Ayala. F.J. (1993) Genetic characteriza-

tion of six parasitic protozoa: Parity between randomenzyme

DNA
typing
multilocus
primer
and
electrophoresis. Proc. Natl. Acad. Sci. USA 90, 13351339.
[23] Henriksson. J., Petterson, U. and Solari, A. (1993) Trypanosoma cruzi: correlation between karyotype variability
and isoenzyme classification. Exp. Parasitol. 77, 334-348.
[24] Fernandes, A.P., Nelson. K. and Beverley, S.M. (1993)
Evolution of nuclear ribosomal RNAs in kinetoplastid
protozoa: perspectives on the age and origins of parasitism. Proc. Natl. Acad. Sci. USA 90, 11608-t 1612.
[25] Maslov, D.A., Avila, H.A., Lake, J.A. and Simpson, L.
(1994) Evolution of RNA editing in kinetoplastid protozoa. Nature 368, 3455348.
[26] Sturm. N.R., Fernandes, 0. and Campbell, D.A. (1995)
The mini-exon gene of three phytomonas isolates that
differ in plant tissue tropism. FEMS Microbial. Lett. 130,
177-182.
[27] Carrasco. H.J., Frame, LA., Valente, S.A. and Miles,
M.A. (1996) Genetic exchange as a possible source of
genomic diversity in sylvatic populations of Trypanosoma
cruzi. Am. J. Trop. Med. Hyg. 54, 4188424.
[28] Bogliolo, A.R.. Lauria-Pires, L. and Gibson, W.C. (1996)
Polymorphism in Trypunosoma cruzi: evidence of genetic
recombination. Acta Trop. 61, 31-40.
[29] Dover, G.A. (1986) Molecular drive in multigene families:
how biological novelties arise, spread and are assimilated.
Trends Genet. 2. 1599165.