castillo et al 2005

RAPID DIVERSIFICATION OF SOUTH AMERICAN TUCO-TUCOS(CTENOMYS; RODENTIA, CTENOMYIDAE): CONTRASTINGMITOCHONDRIAL AND NUCLEAR INTRON SEQUENCES

ANIBAL H. CASTILLO, MARIA NOEL CORTINAS, AND ENRIQUE P. LESSA*

Laboratorio de Evolucion, Facultad de Ciencias, Universidad de la Republica, Igua 4225, Montevideo 11400, UruguayPresent address of MNC: Forestry and Agricultural Biotechnology Institute, University of Pretoria, Pretoria 0002, South Africa

Subterranean tuco-tucos (genus Ctenomys) are a speciose group of South American hystricognath rodents, often

taken as an example of explosive speciation. The 4th intron of the rhodopsin gene (567 bp) and partial sequence

of the 2nd intron of the vimentin gene (403 bp) were used to assess phylogenetic relationships among 20 species

of Ctenomys and 3 octodontid species. Some of the main groups of Ctenomys species previously reported in the

literature (e.g., the ‘‘boliviensis’’ group) are confirmed, as is the lack of resolution of basal nodes. This star-like

pattern of diversification of tuco-tucos was recovered with both the new nuclear dataset and an expanded

mitochondrial dataset, providing further evidence that Ctenomys underwent a phase of rapid diversification early

in its history.

Key words: Ctenomys, introns, nuclear sequences, rates of diversification, tuco-tucos

The study of patterns and rates of diversification is of great

importance in evolutionary biology. With the outgrowth of

DNA sequencing techniques, molecular phylogenetics has

gained increasing relevance in the study of rates of diversifi-

cation. These rates are associated to natural processes such

as adaptive radiation, species selection, or extinction, among

others (Heard and Mooers 2002). Methods that use phylo-

genies to estimate diversification rates provide opportunities to

identify taxonomic groups that have had a history of successful

diversification and, more generally, identify areas of the

phylogeny in which diversification has been particularly rapid.

These observations, in turn, can help identify correlates and

possible causes of successful radiations (e.g., Agapow and

Purvis 2002; Chan and Moore 2002).

The species diversity of underground South American tuco-

tucos (genus Ctenomys; Rodentia, Caviomorpha, Ctenomyi-

dae) has attracted students of speciation and macroevolution.

This genus is distributed throughout the southern half of South

America, from the Andes to the Atlantic Ocean, including the

south of Brazil and Peru, Paraguay, Bolivia, Uruguay,

Argentina and Chile. By some estimates, they represent 45%

of all species of underground rodent lineages of the world, the

remaining species being distributed among 20 genera (Lacey

et al. 2000; see also Reig et al. 1990). This genus consists

mainly of solitary species, although some are social or

semisocial (Lacey et al. 1997). With a diversity of over 56

living species (Reig et al. 1990) and an antiquity in the fossil

record of approximately 2 million years (Mones and Castiglioni

1979), Ctenomys has been referred to as a case of explosive

speciation (Reig et al. 1990). More recently, Verzi (2002)

reported Ctenomys fossils in the Chapadmalalan, implying an

age greater than 3 million years (Schultz et al. 1998). In

addition, tuco-tucos exhibit great karyotypic diversity, ranging

from 2n ¼ 10 (C. steinbachi) to 2n ¼ 70 (C. pearsoni—Cook

et al. 1990; Ortells 1995; Reig et al. 1990). There have been

attempts to associate this chromosomal diversity with rapid

speciation (Ortells and Barrantes 1994; Reig 1970).

Previous work aimed at testing the hypothesis of an

increased rate of diversification in Ctenomys relative to their

sister taxon, the octodontids (Cook and Lessa 1998; Lessa and

Cook 1998; Mascheretti et al. 2000; Slamovits et al. 2001; but

see also Honeycutt et al. 2003) was done on the basis of

mitochondrial cytochrome b (Cytb) sequences. These studies

generally indicated a burst of diversification early in the history

of the genus. However, these results must be viewed cautiously

because they are all based on the mitochondrial Cytb gene, and

hence do not constitute independent evidence. The appeal of

mtDNA for phylogenetic studies lies in its rapid coalescence

time, predominantly maternal inheritance, high interspecific

levels of variation, and readily available primers and protocols

for amplification and sequencing. Despite these advantages,

mtDNA often shows high levels of homoplasy, as expected

* Correspondent: [email protected]

� 2005 American Society of Mammalogistswww.mammalogy.org

170

Journal of Mammalogy, 86(1):170–179, 2005

from its rapid rate of evolution. Moreover, Cytb appears to

have a complex mode of evolution (Slamovits et al. 2001)

and fails to fit a molecular clock in tuco-tucos and allied

octodontids (Cook and Lessa 1998).

To further examine the pattern of diversification of tuco-

tucos, we have obtained nucleotide sequences of introns from 2

nuclear genes. These introns are not linked to the mitochondrial

genome and therefore allow a phylogenetic assessment in-

dependent from those based on mitochondrial loci. The utility of

introns in phylogenetic and population genetic inference has

been investigated (DeBry and Seshadri 2001; Lessa 1992; Lessa

and Applebaum 1993; Swofford et al. 1996) and compared to

that of mitochondrial loci (Prychitko and Moore 2000; Rock-

man et al. 2001; Springer et al. 2001). In addition, we assembled

an expanded dataset that included all tuco-tuco species for

which mitochondrial Cytb sequences are available in GenBank.

Both datasets provide further support for the idea that the tuco-

tuco clade has undergone a rapid expansion early in its history.

MATERIALS AND METHODS

Specimens examined, DNA extraction, amplification, and se-quencing.—We used the same set of species as Cook and Lessa

(1998), with the exclusion of Ctenomys sociabilis. The following

species are represented by other specimens (collector and collector

number in parentheses) than those used in Cook and Lessa (1998):

C. torquatus (CA 654), C. pearsoni (CA 722), C. mendocinus(FC 5521), C. flamarioni (T 29—see D’Elıa et al. 1999, for details on

these specimens); and C. rionegrensis (EV 1064), a specimen

deposited in the collection of the Laboratorio de Evolucion, Facultad

de Ciencias, Universidad de la Republica, Montevideo, Uruguay, from

El Abrojal, Rıo Negro, Uruguay (see Wlasiuk et al. 2003 for details on

this locality). Following Lessa and Cook (1998), outgroup taxa were

Spalacopus cyanus, Octodon degus, and Tympanoctomys barrerae.

Frozen (�808C) or ethanol preserved liver or muscle samples were

subjected to proteinase K digestion, NaCl precipitation of proteins, and

DNA precipitation with isopropanol, as outlined by Miller et al. (1988).

The 4th intron and flanking exon regions of the rhodopsin gene were

amplified by the polymerase chain reaction (PCR), and both strands of

the products were sequenced. Partial sequences of the second intron of

the vimentin gene were obtained using only the forward primer. PCR

amplifications were carried out in a reaction volume of 25 ll containing

1 U of Platinum Taq polymerase (Life Technologies, Carlsbad,

California), 12.5 ll of 1:100 total DNA dilution, 2.5–4.0 mM MgCl2,

0.2 lM of each primer and 0.8 mM of dNTPs (0.2 mM each).

Amplification conditions were the following: 33 cycles alternating

denaturation at 948C for 45 s, annealing at 538C for 45 s and extension

at 728C for 1 min, preceded by 2 min of polymerase activation at 948C

and followed by a final extension at 728C for 3 min. All PCR

experiments included negative controls. Primer Rho 4 (Table 1) was

designed based on a consensus rhodopsin sequence of several

vertebrates. Primers Rho 3c, IntRho 3, and Rho 4b were designed

based on the human rhodopsin sequence. Primers Vim F and Vim R are

those reported by Lyons et al. (1997) and Vim 2F is a variant of Vim F.

Aliquots (5 ll) of PCR products and negative controls were run in 5%

nondenaturing polyacrylamide gels and visualized by silver nitrate

staining according to Sanguinetti et al. (1994). Each remaining product

was purified using Sephadex G-50 columns (Sigma Aldritch, St. Louis,

Missouri), used as template in cycle sequencing reactions and run on 4%

denaturing polyacrylamide gels using an ABI 377 automated sequencer

(Applied Biosystems, Foster City, California). Sequences were aligned

using the programs Sequence Navigator (Applied Biosystems, Inc.,

Version 1.0.1) and ClustalX (Version 1.81—Thompson et al. 1994).

The sequences generated in this study were deposited in GenBank

(accession numbers AY622612–AY622657).

Data analysis.—Both nuclear loci were combined in a single matrix.

The most parsimonious trees were obtained using PAUP* (Swofford

2002), with equal weights for all sites and heuristic searches with 1,000

replicates, the steepest descent option, and TBR swapping algorithm. A

strict consensus of these trees was obtained. Fifty-six nested models of

molecular evolution were evaluated under the Maximum Likelihood

criterion, through likelihood ratio tests (Felsenstein 1981) with the

assistance of Modeltest (version 3.06—Posada and Crandall 1998).

The selected model was used to resolve a consensus of the most parsi-

monious trees by maximum likelihood (10 replicates), with the param-

eters previously obtained with Modeltest. Using the same conditions,

a Likelihood Ratio Test did not reject a molecular clock. Therefore,

branch lengths were adjusted according to this hypothesis (see

‘‘Results’’) and used for Lineages Through Time analysis with End-

Epi (Rambaut et al. 1997). In addition, the tests of Colless (1982) and

Chan and Moore (2002) were applied to the topologies obtained from

the combined nuclear data, as well as from the mitochondrial data using

SymmeTREE (version 1.1—Chan and Moore 2002). The same set of

procedures was then used separately for each nuclear locus.

Statistical support was assessed with MrBayes (Huelsenbeck and

Ronquist 2001), using the selected model conditions as priors, one

million generations, and four chains. Only the last 100,000 trees were

retained.

Mitochondrial cytochrome b gene sequences.—We re-examined the

pattern obtained by Lessa and Cook (1998) using 1 representative of

each species for which complete mitochondrial Cytb gene sequences

are now available in GenBank (Appendix I). We repeated the

analytical steps used for the nuclear data on this Cytb dataset.

A Likelihood Ratio Test allowed us to reject the molecular clock

hypothesis (see ‘‘Results’’), so we estimated relative branching times

with the nonparametric method of Sanderson (1997) using the

program TreeEdit, version 1.0 (http://evolve.zoo.ox.ac.uk).

RESULTS

Variation, compositional bias and models of molecularevolution.—Rhodopsin 4th intron and partial flanking exon

sequences were 567 bp in length for Ctenomys and 547 bp for

Octodon, Spalacopus and Tympanoctomys. Vimentin partial

second intron and 59 exon partial sequences were 403 bp in

length for Ctenomys and 414 bp for Octodon, Spalacopus, and

Tympanoctomys. The total combined alignment length was

983 bp. The maximum divergence within Ctenomys was 0.025

TABLE 1.—Primers used in this study to amplify and sequence parts

of the rhodopsin (Rho) and vimentin (Vim) genes.

Name Orientation Sequence (59 to 39) Reference

Rho 3c Forward ccc atc ttc atg acc atc cca gc This paper

IntRho 3 Forward atc atg atg aac aag cag gtg cct This paper

Rho 4 Reverse cag agg gtg gtg atc atg cag tt This paper

Rho 4b Reverse tag gcc ggg gcc acc tgg ctc gt This paper

Vim F Forward att gag att gcc acc tac agg Lyons et al. 1997

Vim 2F Forward att gag att gcc acc tac ag This paper

Vim R Reverse tga gtg ggt gtc aac cag ag Lyons et al. 1997

February 2005 171CASTILLO ET AL.—RAPID DIVERSIFICATION OF TUCO-TUCOS

(C. rionegrensis–C. sp. ‘‘Ita’’) for rhodopsin, 0.027 (C.torquatus–C. lewisi) for vimentin, and 0.164 (C. conoveri–C. leucodon) for Cytb (Table 2). The maximum divergences

between Ctenomys and octodontids were 0.157 (S. cyanus–

C. frater), 0.163 (T. barrerae–C. conoveri) and 0.307

(O. gliroides–C. lewisi) for rhodopsin, vimentin, and Cytb,

respectively. Average nucleotide composition was rhodopsin

(0.22, 0.28, 0.25, 0.25), vimentin (0.31, 0.22, 0.16, 0.31), and

Cytb (0.31, 0.26, 0.12, 0.31) for A, C, G and T, respectively.

For the nuclear markers, Kimura’s 2P (Kimura 1980) model

with a gamma distribution of rates at variable sites represented

an appropriate compromise between increasing the complexity

of the model and maximizing the accuracy of parameter esti-

mation, as indicated by Likelihood Ratio Tests. The transition/

transversion ratio was estimated at 2.298 with an alpha shape

parameter for the gamma distribution of 0.387. With the same

criterion, the choice for Cytb was TrN model (Tamura and Nei

1993), a submodel of GTR (Rodrıguez et al. 1990), with

a proportion of invariable sites estimated at 0.508, gamma

distribution of rates at variable sites with alpha ¼ 1.383, and

nucleotide substitution rates of [A-G] ¼ 14.84, [C-T] ¼ 18.54,

[A-C] ¼ [A-T] ¼ [C-G] ¼ [G-T] ¼ 1.00.

Phylogeny.—Several arrangements suggested by the nuclear

markers (Fig. 1) are congruent with previous results based on

mitochondrial Cytb sequences. These include 1) a clade con-

taining C. lewisi–C. frater, C. sp. ‘‘Llathu,’’ and C. conoveri;2) a boliviensis group including C. boliviensis ‘‘Robore’’–C. steinbachi, C. boliviensis, and C. goodfellowi; 3) a clade

of two Patagonian species, C. coyhaiquensis and C. haigi;4) a C. sp. ‘‘Ita’’–C. sp. ‘‘Monte’’–C. sp. ‘‘Minut’’ clade;

5) a clade formed with C. mendocinus–C. rionegrensis and

C. flamarioni, the so-called ‘‘mendocinus’’ group; and 6) a clade

formed by C. pearsoni and C. torquatus. As in earlier Cytbanalyses (Cook and Lessa 1998; D’Elıa et al. 1999; Lessa and

Cook 1998; Mascheretti et al. 2000; Slamovits et al. 2001), the

relationships between these clades, and more generally those

near the base of the tuco-tuco clade, are poorly supported.

Patterns of diversification.— In the case of the nuclear

markers, the Likelihood Ratio Test did not reject the hypothesis

of a molecular clock. This allowed the analysis of the rates of

diversification using a plot of lineages through time (Rambaut

et al. 1997), which uses branch lengths as estimates of times of

divergence. The corresponding semilogarithmic plot suggests

clades with a significant increase (P , 0.05 and P , 0.01) in

the rate of diversification; these branches are concentrated near

the base of the tuco-tuco clade, suggesting that a rapid

expansion occurred early in the history of the genus (Fig. 2).

The same analysis performed on individual nuclear matrices

yielded similar results (not shown).

In the case of the expanded Cytb data, the Likelihood Ratio

Test rejected the molecular clock hypothesis, so the estimation

of the relative branching times was based on the nonparametric

method of Sanderson (1997). Again, the corresponding semi-

logarithmic plot suggests a significant increase in the rate of

diversification near the base of the tuco-tuco clade (Fig. 3). For

a description of the clades suggested by this locus, see

Slamovits et al. (2001).

The topological indices of tree imbalance proposed by

Colless (1982) and by Chan and Moore (2002) were not signif-

icant for trees derived from all of our data sets (mitochondrial,

combined nuclear, and individual nuclear loci; results not

shown).

DISCUSSION

The main goal of this paper was to address with independent

evidence the pattern and rate of diversification of lineages in

the genus Ctenomys using Cytb. This entails 2 related issues,

namely 1) whether the tree is significantly imbalanced in favor

of the tuco-tuco relative to their sister taxon, the octodontids,

and 2) whether the tuco-tuco clade shows a burst of diversi-

fication early in its history.

Many of the phylogenetic groups suggested by our new

nuclear sequence data agree with those found in earlier anal-

yses. A clade containing C. lewisi–C. frater, C. sp. ‘‘Llathu,’’C. conoveri was described by D’Elıa et al. (1999), Lessa and

Cook (1998), Mascheretti et al. (2000), and Slamovits et al.

(2001) based on Cytb variation. The ‘‘boliviensis’’ group was

also indicated by these authors, and previously by Gardner

(1991), in a coevolution study of 6 Ctenomys species and

parasite nematodes of the genus Paraspidodera. An allozyme

analysis by Cook and Yates (1994) suggested C. boliviensis to

be paraphyletic relative to C. steinbachi. C. pearsoni and

C. torquatus were found to be sister species by Cook and Lessa

(1998) and D’Elıa et al. (1999). The Patagonian species

C. coyhaiquensis and C. haigi had already been presented as

sister species by Cook and Lessa (1998), D’Elıa et al. (1999),

and Slamovits et al. (2001). The ‘‘mendocinus’’ group was

reported by Ortells (1995) on the basis of chromosomal data, and

analyzed in some detail by D’Elıa et al. (1999) using sequences.

The clade including C. sp. ‘‘Ita’’–C. sp. ‘‘Monte’’–C. sp.‘‘Minut’’ was presented by Cook and Lessa (1998), although the

position of C. sp. ‘‘Minut’’ inside this clade is uncertain (Lessa

and Cook 1998; Slamovits et al. 2001). Relationships within the

outgroup taxa, with S. cyanus and O. degus being more closely

related to each other than any of them to T. barrerae, are the

same as those suggested by previous analyses based on Cytbsequences (e.g., Lessa and Cook 1998), and fit well with more

comprehensive analyses of octodontid relationships (e.g.,

Honeycutt et al. 2003). Analyses performed with individual

TABLE 2.—Descriptive statistics for the 2 nuclear introns and Cytb.Divergences were computed using Kimura 2PþG for the nuclear loci

and a TrNþIþG for Cytb.

Rhodopsin Vimentin Cytb

Length (bp) 568 415 1140

Variable sites 91 69 474

Parsimony-informative sites 53 39 381

Maximum divergence within Ctenomys 0.025 0.027 0.164

Total maximum divergence 0.157 0.163 0.307

Homoplasy index of MP trees 0.101 0.077 0.603

Transition/transversion ratio 2.204075 2.168872 n/aa

a Under the TrN model (Tamura and Nei, 1993) used for Cytb transition rates are

[A-G] ¼ 14.84; [C-T] ¼ 18.54, all transversions are assigned a rate of 1.00.

172 JOURNAL OF MAMMALOGY Vol. 86, No. 1

nuclear loci yielded similar results with respect to the clades

obtained, although the relationships between these clades varied

(results not shown). These variable branches coincide with those

bearing low statistical support (Fig. 1). The principal differences

between our results and those previously reported in the

literature concern the relationships among these clades (see

above), as such relationships continue to be poorly supported by

both nuclear and mitochondrial data. According to the nuclear

loci, C. leucodon does not appear to be sister to the remaining

tuco-tucos, as is the case when Cytb is analyzed. This taxon had

been proposed by Osgood (1946) to constitute the subgenus

Haptomys, based on its distinct skull morphology and pro-

cumbent incisors. In part, this distinction was supported by

Lessa and Cook (1998) and D’Elıa et al. (1999) based on

mitochondrial data, although they noted that C. leucodon was 1

of many lineages stemming from a poorly resolved base of the

tuco-tuco clade. The nuclear data provide no further evidence to

support the recognition of Haptomys as a valid subgenus of

Ctenomys. In the case of C. conoveri, assigned to the monotypic

subgenus Chacomys by Thomas (1916), our analyses agree with

earlier ones in that this species belongs to a clade that includes C.

lewisi, C. frater, and C. sp. ‘‘Llathu.’’ From a phylogenetic

standpoint, therefore, there are no grounds to retain Chacomysas a valid, monotypic subgenus.

Early burst of diversification in Ctenomys.—As in earlier

analyses (Lessa and Cook 1998), we found no significant

imbalance in the topological test developed by Colless (1982).

The most recent statistics developed by Chan and Moore (2002)

were not significant either. Thus, the observed imbalance of

standing diversity in favor of tuco-tucos relative to octodontids

does not necessarily require different underlying rates of

diversification. These negative results hold for both the nuclear

and the expanded mitochondrial datasets. However, these

statistics were developed for completely sampled phylogenies.

As we cannot fulfill this requirement (see below), the latter

conclusions about imbalance might be questionable (Chan and

Moore 2002). More generally, tests based on imbalance make

use of topological data, but do not use branch lengths or other

estimates of time. They therefore have limited power.

Other approaches exploit additional phylogenetic information

and can therefore identify specific regions of the phylogeny that

show rapid successions of branching events. A basal polytomy

FIG. 1.—Analysis of phylogenetic relationships within species of Ctenomys based on combined intron sequences under maximum likelihood.

Numbers above branches indicate Bayesian posterior probabilities, assessed with MrBayes (Huelsenbeck and Ronquist 2001—See ‘‘Materials and

Methods’’ for details).


in the radiation of Ctenomys genus had been reported previously

on the basis of mitochondrial Cytb nucleotide sequences (D’Elıa

et al. 1999; Lessa and Cook 1998; Mascheretti et al. 2000;

Slamovits et al. 2001). This phylogenetic pattern was once again

revealed by the nuclear loci used in this study (see Figs. 1 and 2).

Regardless of the phylogenetic method of inference, the base of

the tuco-tuco clade is poorly resolved and essentially collapses

to a basal polytomy (Fig. 1). Moreover, the increase in the

relative rate of branching revealed by the lineages through time

plot is consistent with a burst of diversification early in the

history of the genus (Fig. 2). The expanded Cytb dataset shows

a similar pattern (Fig. 3), although departures from a molecular

clock required us to use non-parametric rate smoothing to

estimate branch lengths (Sanderson 1997).

This star-like phylogenetic pattern has been difficult to

interpret for several reasons. First, although the observed

limited resolution near the base of the tuco-tuco clade might

reflect a true star phylogeny, it might simply reflect an uneven

distribution of phylogenetic signal across the phylogeny,

attributable to the mode of evolution of the gene (Slamovits

et al. 2001). Saturation of codon changes at third positions, for

example, might produce the observed pattern. Lessa and Cook

(1998) debated the issue and noted that in a related taxon, the

spiny rats of the family Echimyidae, Lara et al. (1996) had

observed another apparent burst of diversification using Cytbsequence data. Lessa and Cook (1998) also noted that the

echimyid burst appeared to be much older than the tuco-tuco

case, in line with inferences from the fossil record (Verzi 2002,

FIG. 2.—a) Maximum likelihood tree for the combined intron sequences with branch lengths estimated using a molecular clock. Lines

corresponding to lineages that show significantly high relative rates of cladogenesis are highlighted as follows: gray (P , 0.01) and dotted (P ,

0.05). b) Corresponding lineages through time plot. Time units are arbitrary but proportional to time according to the molecular clock hypothesis.


and references therein; see also Huchon and Douzery 2001).

However, this observation does not rule out saturation or other

gene-specific, rather than taxon-specific, explanations of the

phylogenetic pattern, as those authors recognized.

Our newly available nuclear sequences provide a way out of

this conundrum by independently revealing a pattern entirely

consistent with that obtained with Cytb. The simplest explana-

tion is that both nuclear and mitochondrial data reflect a burst of

FIG. 3.—a) Maximum likelihood tree for the Cytb sequences with branch lengths estimated using the non-parametric method of Sanderson

(1997). Lines corresponding to lineages that show significantly high relative rates of cladogenesis are highlighted as follows: gray (P , 0.01) and

dotted (P , 0.05). b) Corresponding lineages through time plot. Time units are arbitrary but proportional to time according to the non-parametric

rate smoothing method.


diversification early in the history of the tuco-tucos. Given the

low levels of homoplasy of our nuclear intron sequences (see

Table 2), it is not reasonable to invoke an effect of saturation.

The persistence of this polytomy across independent loci

provides further evidence of a rapid clade expansion early in

the history of Ctenomys. Interestingly, a recent reexamination

of the echimyid case, based on additional taxa and longer

mitochondrial sequences, also substantiates a true star phylog-

eny for spiny rats (Leite and Patton 2002). Both of these highly

diverse clades occur within the superfamily Octodontoidea, the

most diverse group of caviomorph rodents (Honeycutt et al.

2003; Rowe and Honeycutt 2002). Reig (1970) and Vrba and

Gould (1986) have argued that the high species diversity of tuco-

tucos relative to the monotypic coruro (Spalacopus cyanus)

might result from species-level differences (e.g., population

structure, levels of gene flow), rather than different organismic

adaptations. Although our results suggest a rapid process of

diversification of tuco-tucos, they cannot tell us whether species

selection or other processes, including accidents of history

leading to different opportunities for diversification or classical,

organismic adaptations, have caused the observed levels of

diversity. In fact, it is not possible at this point to say whether

coruros and tuco-tucos differ in population structure and levels

of gene flow. The necessary data are only available for selected

tuco-tucos (e.g., C. rionegrensis—Wlasiuk et al. 2003) and

lacking altogether for coruros.

Using a molecular clock, we can attempt to date the ap-

pearance of the Ctenomys clade. According to the fossil record,

the split between ctenomyids and octodontids dates to 9.8

million years ago (mya—Marshall and Sempere 1993; Quintana

1994; Verzi 2002, and references therein). Based on this

calibration point, the age of the Ctenomys clade is estimated at

1.3 million years (my; individual loci yielded similar results: 1.1

my for Rho intron, 1.7 my for Vim intron). These are sur-

prisingly low estimates, given recent records of fossil Ctenomysmore than 3 my in age (Verzi 2002). On the other hand, the

expanded Cytb data set results in a corresponding estimate of 3.7

my, based on the same calibration point and Sanderson’s (1997)

nonparametric rate smoothing. This estimate is not in great

disagreement with paleontological data. However, several

sources of error render all of these estimates uncertain. These

are discussed in Hillis et al. (1996) and Benton and Ayala (2003),

and include, among others, the stochastic nature of the molecular

clock. Additional limitations in our estimation include the use of

a single calibration point, incomplete sampling of the current

diversity of the genus, and the inability to include fossil species.

In sum, the ages presented in this paper can be regarded as rough

estimations of the date of the appearance of the genus.

Introns in phylogenetics.—Mitochondrial genomes evolve

faster than nuclear genomes. However, uncertainty remains

about the amount of phylogenetic information that can be

revealed by each type of locus. The core of this uncertainty

concerns the levels of variation, homoplasy, and resolution of

each type of locus. There are many examples of taxa where

nuclear loci provide more support to certain nodes than

mitochondrial loci due to lower levels of homoplasy (Adkins et

al. 2001; Honeycutt et al. 2003; Prychitko and Moore 2000;

Sheldon et al. 2000; Sota and Vogler 2001; Springer et al.

2001; but see also Drovetsky 2002; Pereira et al. 2002;

Rockman et al. 2001). Intron sequences used in this paper

display low levels of variation but also concomitantly low

levels of homoplasy in comparison to mitochondrial Cytb data

(Table 2). In addition, our intron data do not display strong

compositional bias, which can also complicate phylogenetic

inference. The pattern of variation found in intron sequences of

Ctenomys could be fit to a relatively simple model of molecular

evolution. The appropriateness of a simple model, such as

K2PþG for nuclear introns contrasts strongly with mitochon-

drial Cytb, which requires a much more complex model to

account for its evolutionary features. Introns are noncoding

nuclear sequences and therefore are less constrained by

purifying selection and compositional bias. Moreover, a Likeli-

hood Ratio Test failed to reject the molecular clock hypothesis,

suggesting rate homogeneity across lineages for these nuclear

loci. In turn, the lineages through time analysis was facilitated

by these convenient features.

In sum, nuclear introns are valuable for phylogenetic

reconstructions, as it has already been proposed theoretically

and empirically (see Adkins et al. 2001; DeBry and Seshadri

2001; Lessa 1992; Prychitko and Moore 2000; Rowe and

Honeycutt 2002; Ruvolo 1997; Sheldon et al. 2000; Springer et

al. 2001). Introns appear to be promising molecular regions to

reconstruct the phylogenetic history of groups of relatively

recent origin, such as Ctenomys.

There has been great debate about the issue of the

combinability of different datasets (Kluge 1989; Miyamoto

and Fitch 1995; see Huelsenbeck et al. 1996, for a review).

Many authors argue for the use of this type of approach (Adkins

et al. 2001; Flores-Villela et al. 2000; Pereira et al. 2002;

Prychitko and Moore 2000; Rowe and Honeycutt 2002; Ruvolo

1997; Sota and Vogler, 2001). Some authors favor combined

data sets because phylogenetic analyses are expected to be more

robust as more data are included (Pereira et al. 2002). Others,

however, claim that incongruence between individual loci

reconstructions argues against combined analysis (Drovetsky

2002). In the present analysis an attempt was made to combine

both nuclear and mitochondrial data partitions. Such analyses

(not shown) simply recovered the strongly supported clades, as

well as their collapse at the base of the tuco-tuco radiation.

With the expansion of molecular techniques, many issues are

being addressed in a molecular phylogenetic framework. New

statistical approaches to the study of diversification rates have

taken advantage of the growing proliferation of molecular data

for diverse taxa (Agapow and Purvis 2002; Chan and Moore

2002; Heard and Mooers 2002; Purvis and Agapow 2002; see

also McConway and Sims 2004). The results and conclusions

presented in this paper regarding the early burst of diversifica-

tion documented in these highly speciose subterranean rodents

await further data and statistical testing.

RESUMEN

Los tuco-tucos (genero Ctenomys) son un grupo de numerosas

especies de roedores histricognatos sudamericanos. Con


frecuencia se los menciona como un ejemplo de especiacion

explosiva. Se utilizo el cuarto intron del gen de la rodopsina

(567 pb) y secuencias parciales del segundo intron del gen de

la vimentina (403 pb) para examinar las relaciones filogeneticas

entre 20 especies de Ctenomys y tres octodondidos. Algunos de

los grupos principales de especies de Ctenomys previamente

reportadas en la literatura (e.g., el grupo ‘‘boliviensis’’) se

confirman, ası como la falta de resolucion de los nodos basales.

Este patron de diversificacion en forma de filogenia estrella fue

hallado tanto con los datos nucleares como con una base de datos

mitocondriales ampliada, proveyendo evidencia adicional de

que Ctenomys paso por una fase de diversificacion rapida en su

historia temprana.

ACKNOWLEDGMENTS

We thank J. A. Cook (University of New Mexico), T. L. Yates

(Museum of Southwestern Biology), J. L. Patton (Museum of

Vertebrate Zoology), R. L. Honeycutt (Texas A&M University), and

F. Bozinovic (Universidad de Chile) for providing DNA and tissue

samples. We also thank G. D’Elıa, J. A. Cook, J. L. Patton, and 2

anonymous reviewers for comments on earlier versions of the

manuscript, and D. H. Verzi for clarifying the Ctenomys fossil record.

Financial support was provided by the Comision Sectorial de

Investigacion Cientıfica, Universidad de la Republica, Uruguay.

LITERATURE CITED

ADKINS, R. M., E. L. GELKE, D. ROWE, AND R. L. HONEYCUTT. 2001

Molecular phylogeny and divergence time estimates for major

rodent groups: evidence from multiple genes. Molecular Biology

and Evolution 18:777–791.

AGAPOW, P. M., AND A. PURVIS. 2002. Power of eight tree shape

statistics to detect non-random diversification: A comparison by

simulation of two models of cladogenesis. Systematic Biology

51:866–872.

BENTON, M. J., AND F. J. AYALA. 2003. Dating the tree of life. Science

300:1698–1700.

CHAN, K. M., AND B. R. MOORE. 2002. Whole-tree methods for

detecting differential diversification rates. Systematic Biology

51:855–865.

COLLESS, D. H. 1982. Phylogenetics: the theory and practice of

phylogenetic systematics II (book review). Systematic Zoology

31:100–104.

COOK, J. A., AND T. L. YATES. 1994. Systematic relationships of the

Bolivian tuco-tucos, genus Ctenomys (Rodentia: Ctenomyidae).

Journal of Mammalogy 75:583–599.

COOK, J. A., AND E. P. LESSA. 1998. Macroevolutionary patterns of

diversification in subterranean rodents. The case of the South

American tuco-tucos (genus Ctenomys, Rodentia: Octodontidae).

Evolution 52:1521–1527.

COOK, J. A., S. ANDERSON, AND T. L. YATES. 1990. Notes on Bolivian

mammals. 6: The genus Ctenomys (Rodentia: Octodontidae) in the

highlands. American Museum Novitates 2980:1–27.

D’ELIA, G., E. P. LESSA, AND J. A. COOK. 1999. Molecular phylogeny

of tuco-tucos, genus Ctenomys (Rodentia, Octodontidae) evaluation

of the mendocinus species group and the evolution of the

asymmetric sperm. Journal of Mammalian Evolution 6:19–38.

DEBRY, R. W., AND S. SESHADRI. 2001. Nuclear intron sequences for

phylogenetics of closely related mammals: an example using the

phylogeny of Mus. Journal of Mammalogy 82:280–288.

DROVETSKI, S. V. 2002. Molecular phylogeny of grouse: individual

and combined performance of w-linked, autosomal, and mitochon-

drial loci. Systematic Biology 51:844–854.

FELSENSTEIN, J. 1981. Evolutionary trees from DNA sequences:

a maximum likelihood approach. Journal of Molecular Evolution

17:368–376.

FLORES-VILLELA, O., K. M. KJER, M. BENABIB, AND J. W. SITES, JR.

2000. Multiple data sets, congruence and hypothesis testing for

the phylogeny of basal groups of the lizard genus Sceloporus(Squamata, Phrynosomatidae). Systematic Biology 49:713–739.

GARDNER, S. L. 1991. Phyletic coevolution between subterranean

rodents of the genus Ctenomys (Rodentia: Histricognathi) and

nematodes of the genus Paraspidodera (Heterakoidea: Aspidoder-

idae) in the Neotropics: temporal and evolutionary implications.

Zoological Journal of the Linnean Society 102:169–201.

HEARD, S. B. AND A. O. MOOERS. 2002. Signatures of random and

selective mass extinctions in phylogenetic tree balance. Systematic

Biology 51:889–897.

HILLIS, M., B. MABLE, AND C. MORITZ. 1996. Application of molecular

systematics: The State of the field and a look to the future. Pp. 407–

514, in Molecular systematics, 2nd edition (M. Hillis, C. Moritz,

and B. Mable. eds.). Sinauer Associates, Inc., Publishers, Sunder-

land, Massachusetts.

HONEYCUTT, R. L., D. L. ROWE, AND M. H. GALLARDO. 2003.

Molecular systematics of the South American caviomorph rodents:

relationships among species and genera in the family Octodontidae.

Molecular Phylogenetics and Evolution 26:476–489.

HUCHON, D. AND E. J. DOUZERY. 2001. From the Old World to the New

World: a molecular chronicle of the phylogeny and biogeography of

hystricognath rodents. Molecular Phylogenetics and Evolution

20:238–251.

HUELSENBECK, J. P., J. J. BULL, AND C. W. CUNNINGHAM. 1996.

Combining data in phylogenetic analysis. Trends in Ecology and

Evolution 11:152–157.

HUELSENBECK, J. P., AND F. RONQUIST. 2001. MrBayes: Bayesian

inference of phylogeny. Bioinformatics 17:754–755.

KIMURA, M. 1980. A simple method for estimating evolutionary rate of

base substitutions through comparative studies of nucleotide

sequences. Journal of Molecular Evolution 16:111–120.

KLUGE, A. G. 1989. A concern for evidence and a phylogenetic

hypothesis of relationships among Epicrates (Boidae, Serpentes).

Systematic Zoology 38:7–25.

LACEY, E. A., S. H. BRAUDE, AND J. R. WIECZOREK. 1997. Burrow

sharing by colonial tuco-tucos (Ctenomys sociabilis). Journal of

Mammalogy 78:556–562.

LACEY, E. A., J. L. PATTON, AND G. N. CAMERON (eds.). 2000. Life

underground: the biology of subterranean rodents. The University

of Chicago Press, Chicago, Illinois.

LARA, M. C., J. L. PATTON, AND M. N. DA SILVA. 1996. The

simultaneous diversification of South American echimyid rodents

(Hystricognathi) based on complete cytochrome b sequences.


LEITE, Y. L., AND J. L. PATTON. 2002. Evolution of South American

spiny rats (Rodentia, Echimyidae): the star-phylogeny hypothesis

revisited. Molecular Phylogenetics and Evolution 25:455–464.

LESSA, E. P. 1992. Rapid surveying of DNA sequence variation in

natural populations. Molecular Biology and Evolution 9:323–330.

LESSA, E. P., AND G. APPLEBAUM. 1993. Screening techniques for

detecting allelic variation in DNA sequences. Molecular Ecology

2:119–129.

LESSA, E. P., AND J. A. COOK. 1998. The molecular phylogenetics of

tuco-tucos (genus Ctenomys, Rodentia: Octodontidae) suggests an


early burst of speciation. Molecular Phylogenetics and Evolution

9:88–99.

LYONS, L. A., T. F. LAUGHLIN, N. G. COPELAND, N. A. JENKINS, J. E.

WOMACK, AND S. J. O’BRIEN. 1997. Comparative anchor tagged

sequences (CATS) for integrative mapping of mammalian genomes.

Nature Genetics 15:47–56.

MARSHALL, L. G. AND T. SEMPERE. 1993. Evolution of the Neotropical

Cenozoic land mammal fauna in its geochronologic, stratigraphic,

and tectonic context. Pp. 329–392 in Biological relationships

between Africa and South America (P. Goldblatt, ed.). Yale

University Press, New Haven, Connecticut.

MASCHERETTI, S., P. MIROL, M. GIMENEZ, C. BIDAU, J. CONTRERAS, AND

J. SEARLE. 2000. Phylogenetics of the speciose and chromosomally

variable genus Ctenomys (Ctenomyidae, Octodontoidea), based on

mitochondrial cytochrome b sequences. Biological Journal of the

Linnean Society 70:361–376.

MCCONWAY, K. J., AND H. J. SIMS. 2004. A likelihood-based method

for testing for nonstochastic variation of diversification rates in

phylogenies. Evolution 58:12–23.

MILLER, S. A., D. D. DYKES, AND H. F. POLESKY. 1988. A simple

salting procedure for extracting DNA from human nucleated cells.

Nucleic Acid Research 16:1215.

MIYAMOTO, M. M., AND W. M. FITCH. 1995. Testing species

phylogenies and phylogenetic methods with congruence. System-

atic Biology 44:64–76.

MONES, A., AND L. R. CASTIGLIONI. 1979. Additions to the knowledge

on fossil rodents of Uruguay (Mammalia: Rodentia). Paleontologi-

sche Zeitschrift 53:77–87.

ORTELLS, M. O. 1995. Phylogenetic analysis of G-banded karyotypes

among the South American subterranean rodents of the genus

Ctenomys (Caviomorpha: Octodontidae), with special reference to

chromosomal evolution and speciation. Biological Journal of the


ORTELLS, M. O., AND G. E. BARRANTES. 1994. A study of genetic

distances and variability in several species of the genus Ctenomys(Rodentia: Octodontidae) with special reference to a probable

causal role of chromosomes in speciation. Biological Journal of the


OSGOOD, W. H. 1946. A new octodont rodent from the Paraguayan

Chaco. Fieldiana: Zoology 31:47–49.

PEREIRA, S. L., A. J. BAKER, AND A. WAJNTAL. 2002. Combined

nuclear and mitochondrial DNA sequences resolve generic relation-

ships within the Cracidae (Galliformes, Aves). Systematic Biology

51:946–958.

POSADA, D., AND K. A. CRANDALL. 1998. Modeltest: testing the model

of DNA substitution. Bioinformatics 14:817–818.

PRYCHITKO, T. M., AND W. S. MOORE. 2000. Comparative evolution of

the mitochondrial cytochrome b gene and nuclear b-fibrinogenintron 7 in woodpeckers. Molecular Biology and Evolution

17:1101–1111.

PURVIS, A., AND P. M. AGAPOW. 2002. Phylogeny imbalance:

taxonomic level matters. Systematic Biology 51:844–854.

QUINTANA, C. A. 1994. Ctenominos primitivos (Rodentia, Ocotodon-

tidae) del Mioceno de la Provincia de Buenos Aires, Argentina.

Boletın de la Real Academia Espanola de Historia Natural (Seccion

Geologıa) 89:19–23.

RAMBAUT, A., P. H. HARVEY, AND S. NEE. 1997. End-Epi: an

application for inferring phylogenetic and population dynamical

processes from molecular sequences. Computer Applications in the

Biosciences 13:303–306.

REIG, O. A. 1970. Ecological notes on the fossorial octodontid rodent

Spalacopus cyanus (Molina). Journal of Mammalogy 51:592–601.

REIG, O. A., C. BUSCH, M. O. ORTELLS, AND J. R. CONTRERAS. 1990.

An overview of evolution, systematics, population biology,

cytogenetics, molecular biology and speciation in Ctenomys. Pp.

71–96, in Evolution of subterranean mammals at the organismal

and molecular levels (E. Nevo and O. A. Reig, eds.). Alan R. Liss,

Inc., New York.

ROCKMAN, M. V., D. M. ROWELL, AND N. N. TAIT. 2001. Phylogenetics

of Planipapillus, lawn-headed onychophorans of the Australian

Alps, based on nuclear and mitochondrial gene sequences.


RODRIGUEZ, F., J. L. OLIVER, A. MARIN, AND J. R. MEDINA. 1990. The

general stochastic model of nucleotide substitution. Journal of

Theoretical Biology 142:485–501.

ROWE, D. L., AND R. L. HONEYCUTT. 2002. Phylogenetic relationships,

ecological correlates, and molecular evolution within the cavioidea

(Mammalia, Rodentia). Molecular Biology and Evolution 19:

263–277.

RUVOLO, M. 1997. Molecular phylogeny of the hominoids: inferences

from multiple independent DNA sequence data sets. Molecular

Biology and Evolution 14:248–265.

SANDERSON, M. J. 1997. A nonparametric approach to estimating

divergence times in the absence of rate constancy. Molecular

Biology and Evolution 14:1218–1232.

SANGUINETTI, C. J., E. DIAS NETO, AND A. J. SIMPSON. 1994. Rapid

silver staining and recovery of PCR products separated on

polyacrylamide gels. Biotechniques 17:914–921.

SCHULTZ, P. H., M. ZARATE, W. HAMES, C. CAMILION, AND J. KING.

1998. A 3.3-Ma impact in Argentina and possible consequences.

Science 282:2061–2063.

SHELDON, F. H., C. E. JONES, AND K. G. MCCRACKEN. 2000. Relative

patterns and rates of evolution in heron nuclear and mitochondrial

DNA. Molecular Biology and Evolution 17:437–450.

SLAMOVITS, C. H., J. A. COOK, E. P. LESSA, AND M. S. ROSSI. 2001.

Recurrent amplifications and deletions of satellite DNA accompa-

nied chromosomal diversification in South American tuco-tucos

(genus Ctenomys, Rodentia: Octodontidae): a phylogenetic

approach. Molecular Biology and Evolution 18:1708–1719.

SOTA, T., AND A. P. VOGLER. 2001. Incongruence of mitochondrial and

nuclear gene trees in the Carabid beetles Ohomopterus. Systematic

Biology 50:39–59.

SPRINGER, M. S., ET AL. 2001. Mitochondrial versus nuclear gene

sequences in deep-level mammalian phylogeny reconstruction.

Molecular Biology and Evolution 18:132–143.

SWOFFORD, D. L. 2002. PAUP*. Phylogenetic analysis using parsi-

mony (* and other methods). Version 4.0b10. Sinauer Associates,

Inc., Publishers, Sunderland, Massachusetts.

SWOFFORD, D. L., G. J. OLSEN, G. J. WADELL, AND D. M. HILLIS. 1996.

Phylogenetic Inference. Pp. 407–514, in Molecular systematics. 2nd

edition (M. Hillis, C. Moritz, and B. Mable eds.). Sinauer

Associates, Inc., Publishers, Sunderland, Massachusetts.

TAMURA, K., AND M. NEI. 1993. Estimation of the number of

nucleotide substitutions in the control region of mitochondrial DNA

in humans and chimpanzees. Molecular Biology and Evolution

10:512–526.

THOMAS, O. 1916. Two new Argentine rodents, with a new sub-

genus of Ctenomys. Annals and Magazine of Natural History 18:

303–306.

THOMPSON, J. D., D. G. HIGGINS, AND T. J. GIBSON. 1994. CLUSTAL

W: improving the sensitivity of progressive multiple sequence

alignment through sequence weighting, position-specific gap

penalties and weight matrix choice. Nucleic Acid Research

22:4673–4680.


VERZI, D. H. 2002. Patrones de evolucion morfologica en Ctenomyinae

(Rodentia: Octodontidae). Mastozoolgıa Neotropical 9:309–328.

VRBA, E. S., AND S. J. GOULD. 1986. The hierarchical expansion of

sorting and selection: sorting and selection cannot be equated.

Paleobiology 12:217–228.

WLASIUK, G., J. C. GARZA, AND E. P. LESSA. 2003. Genetic and

geographic differentiation in the Rio Negro tuco-tuco (Ctenomysrionegrensis): inferring the roles of migration and drift from

multiple genetic markers. Evolution 57:913–926.

Submitted 22 March 2004. Accepted 21 June 2004.

Associate Editor was Eric A. Rickart.

APPENDIX IComplete mitochondrial DNA cytochrome b sequences used in this

study. Genbank accession numbers and publications reporting each

sequence are provided.

Ctenomys argentinus; AF370680; Slamovits et al. 2001. C.australis; AF370697; Slamovits et al. 2001. C. boliviensis ‘‘Robore’’;AF007039; Lessa and Cook 1998. C. boliviensis; AF007038; Lessa

and Cook 1998. C. conoveri; AF007054; Slamovits et al. 2001.

C. coyhaiquensis; AF119112; D’Elıa et al. 1999. C. flamarioni;AF119107; D’Elıa et al. 1999. C. frater; AF007046; Lessa and Cook

1998. C. fulvus; AF370688; Slamovits et al. 2001. C. goodfellowi;AF007050; Lessa and Cook 1998. C. haigi; AF007063; Lessa and

Cook 1998. C. latro; AF370704; Slamovits et al. 2001. C. leucodon;

AF007056; Lessa and Cook 1998. C. lewisi; AF007049; Lessa and

Cook 1998. C. magellanicus; AF370690; Slamovits et al. 2001.

C. maulinus; AF370702; Slamovits et al. 2001. C. mendocinus;

AF007062; Lessa and Cook 1998. C. opimus; AF007042; Lessa and

Cook 1998. C. pearsoni; AF119108; D’Elıa et al. 1999. C. porteousi;AF370682; Slamovits et al. 2001.

C. rionegrensis; AF119114; D’Elıa et al. 1999. C. sp. ‘‘Ita’’;AF007047; Lessa and Cook 1998. C. sp. ‘‘Llathu’’; AF007048; Lessa

and Cook 1998. C. sp. ‘‘Minut’’; AF007053; Lessa and Cook 1998.

C. sp. ‘‘Monte’’; AF007057; Lessa and Cook 1998. C. steinbachi;AF007043; Lessa and Cook 1998. C. talarum; AF370689; Slamovits

et al. 2001. C. torquatus; AF119109; D’Elıa et al. 1999. C. tuconax;

AF370684; Slamovits et al. 2001. C. tucumanus; AF370691;

Slamovits et al. 2001.

Octodon degus; AF007059; Lessa and Cook 1998. Octodontomysgliroides; AF370706; Slamovits et al. 2001. Spalacopus cyanus;

AF007061; Lessa and Cook 1998. Tympanoctomys barrerae;

AF007060; Lessa and Cook 1998.


castillo et al 2005

Documents