Phylogenetic relationships among Robertsonian karyomorphs of Graomys griseoflavus (Rodentia, Muridae) by mitochondrial cytochrome b DNA sequencing

Hereditas 136: 130 – 136 (2002)

Phylogenetic relationships among Robertsonian karyomorphs of Graomys griseofla7us (Rodentia, Muridae) by mitochondrial cytochrome b DNA sequencing C. I. CATANESI1, L. VIDAL-RIOJA1, J. V. CRISCI2 and A. ZAMBELLI1 1

Laboratorio de Gene´tica Molecular, Instituto Multidisciplinario de Biologı´a Celular (IMBICE), La Plata, Argentina 2 Laboratorio de Sistema´tica y Biologı´a E6oluti6a, Museo de La Plata, Argentina Catanesi, C. I., Vidal-Rioja, L., Crisci, J. V. and Zambelli, A. 2002. Phylogenetic relationships among Robertsonian karyomorphs of Graomys griseofla6us (Rodentia, Muridae) by mitochondrial cytochrome b DNA sequencing. — Hereditas 136 : 130– 136. Lund, Sweden. ISSN 0018-0661. Received August 22, 2001. Accepted February 2, 2002 Graomys griseofla6us (Waterhouse 1837) is a phyllotine murid rodent with a Robertsonian autosomal polymorphism, having been described 2n = 42, 41, 38, 37, 36, 35 and 34 karyomorphs, and proposed a chromosomal divergence pathway accounted by four sequential Robertsonian fusions. Sequences of a fragment (422 bp long) of the cytochrome b (cyt b) mitochondrial gene and its 5% flanking region (tRNA Glu) were obtained for 19 Graomys griseofla6us from different karyomorphs to infer phylogenetic relationships by using maximum parsimony. Outgroups considered for this analysis were the phyllotine rodents Phyllotis xanthopygus and Eligmodontia typus cyt b sequences. Three trees were produced showing the 2n = 38 − 34 karyomorphs grouped in a single clade while the 2n = 42−41 animals formed a different one. This is in agreement with a hypothesis of a single origin for 2n= 38 −34 Robertsonian karyomorphs from the ancestral 2n =42. Andre´s Zambelli, IMBICE, C.C. 403, (1900) La Plata, Argentina. E-mail: [email protected]

Graomys griseofla6us (Waterhouse 1837) is a phyllotine murid rodent widely distributed in Argentina with Robertsonian autosomal polymorphisms, showing karyomorphs with diploid numbers equal to 42, 41, 38, 37, 36, 35 and 34 (ZAMBELLI et al. 1994). In Graomys griseofla6us, cytogenetic and molecular data supported the ancestrality of the 2n=42 karyomorph (GARDNER and PATTON 1976; ZAMBELLI et al. 1994; ZAMBELLI and VIDAL-RIOJA 1999) and it was proposed a chromosomal divergence pathway to explain the karyotype variability observed. Starting from 2n=42 karyomorph, two lines derived: one producing very low frequent 2n= 41 individuals and the other, 2n=38 specimens. The 2n= 38 karyomorphs were the consequence of two homozygous Robertsonian fusions (RF) that occurred in 2n=42 (RF15 –17 and RF16 –18) producing diploid number reduction. From 2n = 38, the 2n= 37, 36, 35 and 34 karyomorphs have appeared by a non-random downward sequence of Robertsonian fusions: RF1 – 6 and RF2 – 5 (Table 1; ZAMBELLI et al. 1994). In the area studied, the 2n= 42 individuals inhabit the ‘‘Espinal’’ and ‘‘Western Chaco’’ phytogeographic regions located in central Argentina, while the 2n =38, 37, 36, 35, 34 complex (2n= 38− 34) mainly occupies the ‘‘Monte’’ region in the westerncentral area of the country. There are no significant geographic barriers separating different populations of Graomys, in fact, narrow overlapping zones occur

in some regions (THEILER and BLANCO 1996a; TIRANTI 1998). In the laboratory, matings between 2n=42/41, 38/37, 38/36 and 37/37 karyomorphs resulted in F1 and F2 fertile progenies while matings between 2n= 42/38, 42/37 and 42/36 individuals failed to breed or gave sterile hybrids heterozygous for RF15 –17 and RF16 –18 (ZAMBELLI et al. 1994; THEILER and BLANCO 1996a,b). Among the total wild animals sampled during 15 years (about 150) heterozygous specimens for these RFs were never reported. This finding was attributed to a possible relationship between the chromosomal rearrangements and the mechanism of reproductive isolation. We suggested that gametic cell precursors bearing the RF15 –17 and 16–18 in heterozygous state fail to segregate during meiosis, affecting the fertility of the heterozygous individuals (ZAMBELLI et al. 1994). Based on allozyme and reproductive behaviour analyses, THEILER and BLANCO (1996b) and THEILER et al. (1999) revised the taxonomic status of Graomys and reassigned them to two sibling species: Graomys centralis for the 2n=42 specimens, and Graomys griseofla6us for the 2n =38−36 complex. In this revision these authors did not include the 2n= 41, 35 and 34 specimens. The homozygous RF15 –17 and RF16 –18 are the common chromosomal feature in the 2n=38−34 complex, and their presence may be correlated with the Nucleolar Organizer Regions (NOR) pattern and

mtDNA phylogenetics of Graomys griseofla6us

Hereditas 136 (2002)

satellite DNA organization. Analysis of NOR locations both by silver staining (Ag-NOR) and in situ hybridization revealed that the 2n=42 exhibit highly variable Ag-NOR patterns both in number and chromosome location, while 2n=38 − 34 karyomorphic group showed a single Ag-NOR pattern. The latter animals underwent two NOR deletions in reference to the 2n=42 karyomorphs, one of which would be the consequence of a Robertsonian fusion and the other would be produced by unequal crossing-over mechanism (ZAMBELLI and VIDAL-RIOJA 1996). On the other hand, the Graomys chromosomal divergence has been correlated with molecular organization of two satellite DNA families (EG250 and Hpa3.2). When all karyomorphs were compared, a clear differentiation between 2n=42 −41 and 2n= 38 − 34 was found at the level of methylation pattern of the EG250 satellite, and the molecular organization of Hpa3.2 satellite, which is much more abundant in the 2n=38 − 34 group than in the 2n = 42−41 (ZAMBELLI and VIDAL-RIOJA 1999). Despite the accumulated evidence about the 2n= 42 and 38−36 karyomorphs represent different species, TIRANTI (1998) argued that the nomenclatorial situation of Graomys griseofla6us was not addressed, particularly the assessment of the available names centralis, griseofla6us, edithae and medius (REDFORD and EISENBERG 1992; BRAUN 1993; MUSER and CARLETON 1993; THEILER and BLANCO 1996b). Mitochondrial DNA is a rapidly evolving molecule, maternally inherited and not linked to nuclear markers. Therefore, areas of congruence of phylogenetic trees based on nuclear and mitochondrial DNA markers may reflect past relationships. Mitochondrial cytochrome b (cyt b) DNA sequence variation has provided phylogenetic resolution for several orders of mammalian taxa (IRWIN et al. 1991; SMITH and PATTON 1991, 1993; NACHMAN et al. 1994; STEPPAN 1998). Table 1. Robertsonian fusions found in each karyomorph of G. griseofla6us 2n

RF1–6

RF2–5

RF15–17

RF16–18

42 41 38 37 36 35 34

– Ht – Ht Hm Hm Hm

– – – – – Ht Hm

– – Hm Hm Hm Hm Hm

– – Hm Hm Hm Hm Hm

Fig. 1. Map of Argentina showing the localities where karyomorphs of G. griseofla6us (indicated between parentheses) were collected.

In this work, variability of cyt b sequences from all Graomys karyomorphs was correlated with the Robertsonian chromosomal divergence, and the phylogenetic relationships among them were drawn.

MATERIALS AND METHODS Nineteen Graomys griseofla6us specimens representing all karyomorphs were collected by field trapping in Table 2. Geographic procedence and amount of collected specimens of G. griseofla6us karyomorphs analysed Locality

2n

No. of specimens

Santiago Temple (31°23% S, 63°25% W) Chamical (30°21% S, 66°19% W) General Belgrano (31°59% S, 64°34% W)

42 41 42 38 37 37 36 36 35 34

4 1 5 2 1 1 1 2 1 1

Salicas (28°22% S, 67°03% W) Mendoza (32°53% S, 68°51% W)

Ht: heterozygous Hm: homozygous –: absence of the RF

131

132

C. I. Catanesi et al.

the following localities of Central Argentina (Fig. 1; Table 2): Santiago Temple and General Belgrano in Co´ rdoba Province; Chamical and Salicas (approximately 600 km northwest from Santiago Temple area) in La Rioja Province; and Mendoza in Mendoza Province (approximately 450 km south from Salicas area and 600 km west from Santiago Temple). All individuals were karyotyped as previously described (ZAMBELLI et al. 1994). Total DNA was obtained from fixed liver as described elsewhere (ZAMBELLI and VIDAL-RIOJA 1995). A 422 bp mitochondrial DNA fragment, consisting of 379 bp of the 5% cyt b segment and its flanking region of glutamic acid-tRNA gene (43 bp), was PCR-amplified using L14724 and H15149 primers (KOCHER et al. 1989) and the PCR SuperMIX High Fidelity kit (Gibco BRL, Life Technologies). DNA was amplified in 1 cycle of 94°C for 2 min and 35 cycles of 94°C for 45 sec, 56°C for 50 sec and 72°C for 60 sec, with a final extension of 72°C for 5 min. The 422 bp-PCR fragments were ligated directly to the T-Easy vector (Promega) using T4 DNA ligase (Promega). The ligation mix was used to transform XL1Blue (Stratagene) E. coli strain and recombinant clones were selected in LB-Amp-Xgal plaques. The recombinant clones were isolated and both strands sequenced by the dideoxy chain termination method using 35S-dATP, universal sequencing primers and T7 DNA polymerase kit (Pharmacia). Sequences were run on 6% denaturing polyacrylamide gels. Cyt b fragment sequences obtained for each of the 19 Graomys griseofla6us individuals analysed in this work were deposited in the GenBank under the following accession numbers: AF281202 to AF281219. Phylogenetic analysis of the aligned cyt b sequences was performed using the parsimony algorithms and tree analyses (exact search) options of the PAUP software program, version 4.0 (SWOFFORD 1999). The polarity of characters was determined by the outgroup comparison method (MADDISON et al. 1984). The phyllotine species Phyllotis xanthopygus (accession number U86830; STEPPAN 1998) and Eligmodontia typus (accession number AF108692; SMITH and PATTON 1999) were alternatively included in the outgroup or in the ingroup. Variable nucleotide positions were equally weighted and treated as unordered characters. Branch and bound search strategy was used to find the shortest tree(s). Bootstrap analysis with 100 replications (FELSENSTEIN 1988) was performed to provide a sense of the support of each clade. The consistency and retention indexes were calculated. Sequence similarity between two individual sequences was estimated with BLASTN Program, version 2.0.11 (ALTSCHUL et al. 1990). Nucleotide

Hereditas 136 (2002)

Table 3. Silent and replacement nucleotide-changes for comparisons at 379 bp cyt b fragment among all G. griseofla6us karyomorphs. TS: transition; TV: trans6ersion Codon base position

1st

2nd

3rd

Total

TS

5 3 0 1 8.0

0 1 0 2 0.5

30 1 5 0 6.2

35 5 5 3 5.0

Silent Replacement TV Silent Replacement TS/TV ratio

diversity values were estimated using the Arlequin program (SCHNEIDER et al. 1997). The fixation index (Fst) was calculated based on nucleotide diversity within and between populations according to HUDSON et al. (1992).

RESULTS The 5% cyt b segment and its flanking region from all Graomys karyomorphs, E. typus and P. xanthopygus, were aligned. In the analysed cyt b gene region (379 bp long), 48 variable positions were found within Graomys karyomorphs, as detailed in Table 3. Of the total variable positions found, 36 (75.0%; 35 silent and 1 replacement) were variable third positions of codons, 9 (18.7%; 5 silent and 4 replacement) were variable first positions, and 3 (6.3%; 3 replacement) involved second position changes. When more distant taxa are compared, the number of transversions increases, while the transition/ transversion (TS/TV) ratio decreases (SMITH and PATTON 1991). For instance, comparison of cyt b fragments of 401 bp from 12 different species of Akodon showed a TS/TV ratio of 7.7 (SMITH and PATTON 1991). When all Graomys karyomorphs were compared to each other, it was found a total of 40 transitions and 8 transversions, with a TS/TV ratio of 5.0 (Table 3). The parsimony equally weighted analysis showed 54 uninformative variable positions and 60 informative, and yielded 3 most parsimonious trees with 148 steps excluding uninformative characters. The monophyly of the Graomys karyomorphs analysed had a good support (bootstrap value of 98%) with respect to E. typus and P. xanthopygus, both of which remained out of the Graomys cluster in all parsimonious trees obtained. All of the trees showed two well-defined clades (bootstrap values of 100%): one including the 2n= 42−41 karyomorps and the other including the 2n=38− 34 (Fig. 2). These two clades were also found in the strict consensus tree (Fig. 3). Next shortest tree analysis yielded 122 trees with 149

Hereditas 136 (2002)

mtDNA phylogenetics of Graomys griseofla6us

133

Fig. 2. Phylograms yielded by the maximum parsimony analysis of cyt b sequences of G. griseofla6us; consistency index: 0.858, retention index: 0.938. Numbers and letters indicate diploid number and procedence, respectively. Subscript numbers indicate different individuals. ST: Santiago Temple; GB: General Belgrano; Ch: Chamical; Sa: Salicas; Mz: Mendoza. Phy: Phyllotis xanthopygus; Eli: Eligmodontia typus.

steps. Strict consensus tree of the 125 trees (122 trees with 149 steps and 3 trees with 148 steps) still showed the same two major clades grouping the 2n= 42− 41 and the 2n= 38− 34 karyomorphs. The trees obtained using UPGMA, Neighbor-Joining, and maximum likelihood showed as well two clades, with the same members as the ones included in the maximum parsimony analysis (data not shown).

Within the 2n= 42− 41 animals studied, cyt b sequence similarities ranged from 95 to 100% (average value 97.9%) and within 2n= 38− 34 from 95 to 99% (average value 97.3%). Comparison between 2n= 42− 41 and 2n= 38− 34 karyomorphic groups showed sequence similarities which ranged from 86 to 91% with an average value of 88.6%. When Graomys cyt b sequences of both 2n= 42− 41 and 2n= 38− 34 karyomorphs were compared to those of different

134

C. I. Catanesi et al.

murid species, the average values of sequence similarity were: 82.1% to Phyllotis xanthopygus, 81.4% to Eligmodontia typus and 78.6% to Akodon oli6aceous (accession number M35693; SMITH and PATTON 1991). Total nucleotide diversity was 0.04939 0.0038. T test comparison of nucleotide diversity within 2n= 42− 41 (0.00469 0.0010) and within 2n= 38 − 34 (0.008090.0017) showed no significant difference between them (p= 0.126; a= 0.05). The Fst statistics was 6.55%, slightly lower than the mean Fst value calculated by THEILER et al. (1999) based on 11 polymorphic allozyme loci (mean Fst = 7.24%). DISCUSSION The three trees obtained by comparison of mitochondrial cyt b sequence fragments from all Graomys karyomorphs showed two well-defined clades: one including 2n=42 −41 animals (bootstrap 100%) and the other with 2n=38 −34 individuals (bootstrap 100%). Moreover, comparison to Eligmodontia and Phyllotis supported the monophyly of all Graomys karyomorphs (bootstrap 98%). Although nuclear and mitochondrial genomes have quite different evolutionary dynamics, the cyt b differentiation between 2n =42−41 and 38− 34 karyomorphic groups was concordant with the described interkaryomorphic dif-

Fig. 3. Strict consensus tree of cyt b sequences of G. griseofla6us.

Hereditas 136 (2002)

ferences on NOR pattern and repetitive DNA sequences organization. The house mouse Mus domesticus constitutes one of the most conspicuous and largely studied mammal model for Robertsonian (Rb) chromosomal differentiation. Hypotheses to explain the evolutionary origin of Mus Rb races (which range in diploid number from 2n= 22 to 2n= 38) have taken two main forms. The classical view states that many or most Rb populations arose independently of each other, each presumably from a close standard karyotype (2n= 40) population, suggesting that the mechanism producing the Rb fusions is present throughout the species (BRITTON-DAVIDIAN et al. 1989; NACHMAN et al. 1994; RIGINOS and NACHMAN 1999). The second hypothesis proposes that Rb populations arose in restricted areas from which they spread to the present locations (WINKING et al. 1988; BAUCHAU 1990; NACHMAN et al. 1994; RIGINOS and NACHMAN 1999). This proposal suggests that all Rb populations may be more closely related to each other than to a standard karyotype population. According to this hypothesis, the mechanism producing the Rb rearrangements should be present in a single M. domesticus lineage. Strict consensus tree obtained for Mus by comparison of mitochondrial DNA sequences (the control region and the ND3 gene region) showed that Rb races were not grouped in the same clade, but forming 6 major different clades that also included standard karyotype individuals. Same topology of this tree was obtained using different tree building methods, and using different weighting schemes for transitions, transversions, or insertion/deletions. Based on this mitochondrial DNA polyphyletic relationships, it was proposed the independent origin of M. domesticus Rb populations (NACHMAN et al. 1994). The population analysis of microsatellite markers linked to Rb5.15 was concordant with mtDNA phylogenetic relationships, rejecting also the single origin of Mus Rb races, and suggesting that Rb5.15 may have arisen at least twice (RIGINOS and NACHMAN 1999). Contrarily to M. domesticus, in Graomys, trees from cyt b sequences showed all G. griseofla6us Rb karyomorphs grouped in a single clade, while the ancestral 2n= 42 animals and the 2n= 41 karyomorph, formed a different one. This is consistent with the hypothesis of a single origin for Rb karyomorphs (2n=38− 34 complex). THEILER et al. (1999) compared the mean number of alleles per locus (A) between 2n= 42 karyomorph (G. centralis) and 2n= 38− 36 complex (G. griseofla6us) and found very similar A values (1.57 and 1.53, respectively). Since A value is particularly affected by population bottlenecks (NEI et al. 1975), these authors proposed

Hereditas 136 (2002)

that fixation of RF15– 17, 16– 18 and 1– 6 in 2n= 38 −36 occurred without such effect. In opposition to Theiler’s proposal, the concordance between our nuclear and mitochondrial data support the single origin of the Graomys Rb races, not ruling out the occurrence of a founder effect during the chromosomal differentiation. The clades showed by cyt b trees did not exhibit concordance with the geographical distribution of the analysed individuals, probably because among the areas occupied by Graomys karyomorphs there are no geographic barriers (WALKER and AVISE 1998). So far, the 2n=38 −34 karyomorps have extended their geographic distribution further than the ancestral 2n= 42 individuals. Therefore, THEILER et al. (1999) suggested that some selective advantage (probably acquired by generation of new coadapted gene complexes in the fused chromosomes) would have allowed fast dispersal of the new species (2n=38 −36) during one of the periods of colonization in the ‘‘Monte’’ region, an unexploited zone by the 2n=42 karyomorph. From the present results one question emerges: why 2n =42 individuals having the current 2n= 38 −34 cyt b consensus haplotype were not found. AVISE (1986) proposed a theoretical model of stochastic extintion of matriarcal lineages encompassing speciation events. Accordingly, the probability is high that sibling species will be polyphyletic in matriarchal ancestry for about 2– 4 k generations after the speciation (where k is the carrying capacity of each sibling species). Only later, as lineage sorting through random extinction continues, the probability greatly increases that the sibling species will appear monophyletic with respect to one another (AVISE 1986). In agreement to this model, the cyt b tree obtained showed that 2n=42 − 41 group is monophyletic respect to 2n=38 −34 (bootstrap 100% for both clades). An approximated and accepted time scale considers that the molecular evolutionary rate of rodent mtDNA equals 7–8% substitution per 1 million years (CATZEFLIS et al. 1992). Comparison between 2n=42 −41 and 2n= 38 − 34 karyomorphic groups showed around 11% average base substitution, indicating that the divergence event among them would had occured approximately 1.5 million years ago. Probably, the 2n= 38 − 34 chromosome evolution occurred during this time involved the stochastic extintion of some matriarcal lineages and the establishment of the current consensus mitochondrial haplotype (shared by all Robertsonian karyomorphs), which clearly differs from the 2n=42 −41 consensus haplotype. The analysis of cyt b sequence of Graomys griseofla6us contributed to bring light onto the

mtDNA phylogenetics of Graomys griseofla6us

135

origin of the Robertsonian karyomorphs; however, one of the aspects not still clarified is the phylogeny among the 2n =38−34 derived karyomorphs, specially the relationships among the 2n= 38, 37 and 36 karyomorphs and the RF2–5-carrying 2n=35 and 34 karyomorphs. ACKNOWLEDGEMENTS We thank to Dr. James Wilgenbusch for his criticism on the phylogenetic analysis. This work was supported by grants from CONICET and CIC, Argentina. C. I. Catanesi is a scholar from CONICET.

REFERENCES Altschul SF, Gish W, Miller W, Myers EW and Lipman DJ, (1990). Basic local alignment search tool. J. Mol. Biol. 215: 403 –410. Avise JC, (1986). Mitochondrial DNA and the evolutionary genetics of higher animals. Philos. Trans. R. S. Lond. Ser. B 312: 325 – 342. Bauchau V, (1990). Phylogenetic analysis of the distribution of chromosomal races of Mus musculus domesticus Rutty in Europe. Biol. J. Linn. Soc. 41: 171 –192. Braun JK, (1993). Systematic relationships of the tribe Phyllotini (Muridae: Sigmodontinae) of South America. Special Publication, Oklahoma Museum of Natural History, Norman, Oklahoma, p. 1 – 50. Britton-Davidian J, Nadeau JH, Croset H and Thaler L, (1989). Genic differentiation and origin of Robertsonian populations of the house mouse (Mus musculus domesticus Rutty). Genet. Res. 53: 29 –44. Catzeflis FM, Aguilar J-P and Jaeger J-J, (1992). Muroid rodents: phylogeny and evolution. Tree 7: 122 –126. Felsenstein J, (1988). Phylogenies from molecular sequences: Inference and reliability. Annu. Rev. Genet. 22: 521 –565. Gardner AL and Patton JL, (1976). Karyotypic variation in oryzomine rodents (Cricetidae) with comments on chromosomal evolution in the neotropical cricetine complex. Occas. Papers Mus. Zool., Lousiana State Univ. 49: 1 –48. Hudson RR, Slatkin M and Maddison WP, (1992). Estimation of levels of gene flow from DNA sequence data. Genetics 132: 583 –589. Irwin DMT, Kocher TD and Wilson AC, (1991). Evolution of the cytochrome b in mammals. J. Mol. Evol. 32: 128 –144. Kocher TD, Thomas WK, Meyer A, Edwards SV, Pa¨ a¨ bo S, Villablanca FX and Wilson AC, (1989). Dynamics of mitochondrial DNA evolution in animals: Amplification and sequencing with conserved primers. Proc. Nat. Acad. Sci. USA 86: 6196 – 6200. Maddison WP, Donoghue J and Maddison DR, (1984). Outgroup analysis and parsimony. Syst. Zool. 33: 83 – 103. Muser GG and Carleton MD, (1993). Family Muridae. In: Mammal Species of the World. A taxonomic and geographic reference (eds DE Wilson and DAM Reeder). Smithsonian Institution Press, Washington, p. 501 –755. Nachman WM, Boyer SN, Searle JB and Aquadro CF, (1994). Mitochondrial DNA variation and evolution of Robertsonian chromosomal races of house mouse, Mus domesticus. Genetics 136: 1105 – 1120.

136

C. I. Catanesi et al.

Nei M, Maruyama T and Chakraborty R, (1975). The bottleneck effect and genetic variability in natural populations. Evolution 29: 1 –10. Redford KH and Eisenberg JF, (1992). Mammals of the Neotropics. V. 2. The southern cone: Chile, Argentina, Uruguay, Paraguay. Univ. Press, Chicago. Riginos C and Nachman MW, (1999). The origin of a Robertsonian chromosomal translocation in house mice inferred from linked microsatellite markers. Mol. Biol. Evol. 16: 1763 –1773. Schneider S, Kueffer JM, Roessli D and Excoffier L, (1997). Arlequin ver. 1.1: A software for population genetic data analysis. Genetics and Biometry Laboratory, Univ. of Geneva, Switzerland. Smith MF and Patton JL, (1991). Variation in the mitochondrial cytochrome b sequence in natural populations of South American akodontine rodents (Muridae: Sigmodontinae). Mol. Biol. Evol. 8: 85 –103. Smith MF and Patton JL, (1993). Diversification of South American murid rodents: Evidence from mitochondrial DNA sequence data for the akodontine tribe. Biol. J. Linn. Soc. 50: 149 –177. Smith MF and Patton JL, (1999). Phylogenetic realtionships and the radiation of sigmodontine rodents in South America: evidence from cytochrome b. J. Mammal. Evol. 6: 89–128. Steppan SJ, (1998). Phylogenetic relationships and species limits within Phyllotis (Rodentia: Sigmodontinae): concordance between mtDNA sequence and morphology. J. Mammal. 79: 573 –593. Swofford DL, (1999). PAUP Phylogenetic analysis using parsimony. Version 4.0 beta 2. Sinauer Associates, Sunderland, Massachusetts. Theiler GR and Blanco A, (1996a). Patterns of evolution in Graomys griseoflavus (Rodentia, Muridae). III. Olfactory

Hereditas 136 (2002)

discrimination as a premating isolation mechanism between cytotypes. J. Exp. Zool. 274: 346 – 350. Theiler GR and Blanco A, (1996b). Patterns of evolution in Graomys griseoflavus (Rodentia, Muridae). II. Reproductive isolation between cytotypes. J. Mammal. 77: 776– 784. Theiler GR, Gardenal CN and Blanco A, (1999). Patterns of evolution in Graomys griseoflavus (Rodentia, Muridae). IV. A case of rapid speciation. J. Evol. Biol. 12: 970– 979. Tiranti SI, (1998). Cytogenetics of Graomys griseoflavus (Rodentia: Sigmodontinae) in central Argentina. Z. Sa¨ ugetierkunde 63: 32 –36. Walker DE and Avise JC, (1998). Principles of phylogeography as ilustrated by fresh water terrestrial turtles in the southeastern United States. Annu. Rev. Ecol. Syst. 29: 23– 58. Winking H, Dulic B and Bulfield G, (1988). Robertsonian karyotype variation in the European house mouse, Mus musculus; survey of present knowledge and new observations. Z. Sa¨ ugetierkunde 53: 148 – 161. Zambelli A and Vidal-Rioja L, (1995). Molecular analysis of chromosomal polymorphism in the South American cricetid, Graomys griseoflavus. Chrom. Res. 3: 361 – 367. Zambelli A and Vidal-Rioja L, (1996). Loss of nucleolar organizer regions during chromosomal evolution in the South-American cricetid Graomys griseoflavus. Genetica 98: 53 – 57. Zambelli A and Vidal-Rioja L, (1999). Molecular events during chromosomal divergence of the South American rodent Graomys griseoflavus. Acta Theriol. 44: 345 – 352. Zambelli A, Vidal-Rioja L and Wainberg R, (1994). Cytogenetic analysis of autosomal polymorphism in Graomys griseoflavus (Rodentia, Cricetidae). Z. Sa¨ ugetierkunde 59: 14– 20.