Cytosystematics of the South African Aethomys (Rodentia: )

D.S. Visser and T.J. Robinson Research Institute, University of Pretoria, Pretoria

The G- and C-band chromosome patterns and the location of the genus Aethomys are morphologically similar of the nucleolus organizer regions (NORs) are presented for and yet form a karyotypically diverse group. As is presently A. namaquensis (2n = 24), A. granti (2n = 32) and A. understood, the genus comprises three South African species chrysophilus (2n = 44; 2n = 50). The presence of two of which two, A. namaquensis and A _ granti are placed in distinct cytotypes in what is conventionally recognized as A. chrysophilus is indicative of the presence of two discrete the subgenus Michaelamys and the third, A. chrysophilus, in species which, karyology apart, appear to be indistinguish­ the subgenus Aethomys. Because of the morphological able using existing identification keys. The chromosomal similarity of the species, the lack of intensive studies employing relationships of the South African species and the modem techniques and the relative rarity of some species (A. taxonomic implications of these data are discussed. grantl), there has been a degree of uncertainty regarding the s. Afr. J. Zool. 1986,21: 264 - 268 taxonomic relationships in this genus (Davis 1975). The present Die G- en C-bandchromosoompatrone en ligging van die investigation examines, through the use of chromosome­ nukleolus organiserende areas van A. namaquensis (2n = banding data, the various karyotypic changes that have 24), A. granti (2n 32) en A. chrysophilus (2n = 44; 2n accompanied the evolutionary divergence of the South African = = .

) 50) word beskryf. Die teenwoordigheid van twee Aethomys species. These data provide an unequivocal means 0

1 afsonderlike sitotipes in wat konvensioneel beskou word as for the delimitation of the taxa and contribute to the develop­ 0 A. chrysophilus, is aanduidend van die bestaan van twee 2

ment of a reliable identification scheme for these species.

d verskillende spesies wat, buiten kariologie, ononderskeibaar e t met behulp van bestaande identifikasiesleutels blyk te a wees. Die chromosomale verwantskappe van die Suid­ Material and Methods d ( Afrikaanse spesies en die taksonomiese implikasies van die Fibroblast cultures were established from skin or tail biopsies, r e data word bespreek. or disaggregated kidney tissue, using standard procedures h s S.·Afr. Tydskr. Dierk. 1986, 21: 264 - 268 i

l (Paul 1975). Trypsin G-banding and barium hydroxide C­ b banding were performed according to the methods of Wang u P

& Fedoroff (1972) and Sumner (1972) respectively. Nucleolus e

h organizer regions were silver-stained following Bloom & t

y Goodpasture (1976). Collection localities and the number of b specimens utilized for this study are presented below: d e t n A. namaquensis (n = 24): Pretoria (25°4O'S/28°2O'E) 10' + a r 29 9; Springbok (29°4O'S/I7°52'E) 20' 0' + 29 9; Deel­ g

e fontein (30059'S123°48'E) 20' 0' + 29 9; Calvinia (31 ° 28'S/ c n 19°50'£) 10' + 19; Karoo National Park (32°22'S/22°44'E) e c

i 10' + 29 9; Thabazimbi (24°38'S127°25'E) 10' + 19; l

r Hutchinson (31 °30'S123° 11 'E) 20' 0' + 29 9; Alldays e d (22°43'S129°IO'£) 10' + 19 . n u

y . A. granti (n = 10): Sutherland (32°23'S120°4O'E) 40' 0' + a 699. w e t a A. chrysophilus 2n = 44 (n = 16): Vaalkopdam (25° 18'S/ G

t 27°25'E) 60' 0' + 5 9 9 ; Pilanesberg National Park (25° 12'S/ e

n 27° 15'E) 10'; Durban (29°55'S/30055'E) 29 9; Satara Camp i o b (24°2I'S/3I 46'E) 10' + 19. a S

y A. chrysophilus 2n = 50 (n = 22): Messina (22°2I'S/30003'E)

b D_S. Visser· and T.J. Robinson

d Mammal Research Institute, University of Pretoria, Pretoria, 10' + 19; Letsitele (23°50'S130° 18'E) 20' 0' + 29 9 ; e c 0002 Republic of South Africa Boshoek (25°28'S127°09'E) 1 0' + 1 9 ; Pilanesberg National u d *To whom correspondence should be addressed Park (25° 12'S127° 15'E) 20' 0' + 19; Rooibokkraal (24° 15'S/ o r 26°50'£) 10' + 19; Thabazimbi (24°38'SI27°25'E) 30' 0' p e Received 10 March 1986; accepfed 29 April 1986 + 19; Brits (25°34'SI27°45'E) 20' 0' + 19. R S. Arc 1. Zool. 1986, 21()) 265

Animals examined in this study are available as voucher quenuy obseJW'ed between NOR bearing chromosomes (Figure specimens in the mammal collections of the Transvaal IC arrow). Museum, Pretoria, Republic of South Africa. Grant's rock mouse, A. grant; (2n = 32) Results The G-banded karyotype of a male Grant's rock mouse is Namaqua rock mouse, A. namaquensis (2n = 24) shown in Figure 2A. The diploid number confums Matthey's The G-bandcd karyotype of a male Namaqua rock mouse (1964) report. Three distinct chromosomal categories are is shown in Figure IA. The aULOsomes can be arranged into evident based on morphology with the karyotype comprising three distinct groups on the basis of cenlromere position with four pairs of large melacenLric autosomes (pairs I - 4), one the karyotype comprising four pairs of large metacentric pair of medium acrocemrics (pair 5) and ten pairs of small chromosomes (pairs I -4), one pair of large submetacentrics acrocentrics (pairs 6 - IS). The X chromosome is the largest (pair 5) and six pairs of small acrocentric chromosomes (pairs acrocentric in the genome and the Y, also acrocentric, is 6 - II). The X is the largest acrocentric cluomosome in the similar in size to pair 5. genome and the Y is similar in size to the larger acrocentric The C-banded chromosomes of A. gTant; are illustrated in chromosomes (6 - 8). The morphology of all bi-armed auto- Figure 2B. A single band of interstitial heterochromatin can . somes is distinctive, but the gradation in size of the small be seen in the proximal region of the long arms of pair 3 acrocentrics necessitates the use of G-banding in the pairing (FIgure 2B arrows) mirroring the situation in A. fIIJ17UNjUensis. of homologues. These shared intmtitial C-bands are absent in A. c}uysoph;!us The C-banded chromosomes of A. nomaquensis are shown (see below) and this provides supportive evidence for the close in Figure I B. Interstitial constitutive heterochromatin was relatedness of the former species. The largest amount of found in the proximal portion of the long anns of chromo­ autosomal heterochromatin is concentrated in pairs 12 - IS some pair 3 (arrows). The largest amount of heterochromatin which probably contributes to the poor G-band resolution of is pericenLrorneric in distribution and is located on pairs 6 and these autosomes. Telomeric heterocluomatin was detected in 9, while the smallest autosomal chromosomes (pair II) appear the distal ends of pa,irs I and 2 (Figure 2B arrow heads). The to be almost totally heterochromatic (Figure IB arrow heads). Y chromosome can be easily identified as it is entirely hetero­ A partial silver-stained metaphase cell of a male is shown pycnotic while the distal half of the large acrocentric X in Figure I C. The nucleolar organizer regions were detected chromosome is C-band positive: the centromeric region of this at the telomeric ends of the shon arms of eight of the chromosome also stains darldy. In some meta phases a small acrocentric chromosomes. Satellite association was infre- euchromatic regjon was frequently visible at the extreme distal

end of the long arm of the X (Figure 2B open arrow) . . ) 0 1 0 2 d e t a d ( r e h s i l b u P e h t y b d e t n a r g e c n e c i l r e d n u y a w e t a G t e n i M .~ __. __I ~ Y~ ______b • a figure 1 Karyolype of a male Namaqua rock mouse, AelnQmys S

y /UU11oquensis (2n = 24): (A) G-banding; (8) C-banding. The arrows figure 2 KaryOlype of a male Grant's rock mouse, Aelnomys granli b indicate a band of interstitial heterochromatin on the long arms of pair (2n == 32): (A) G-banding; (8) C-banding. The arrows indicate a band d e J. Arrow heads illustrate the acrocentric chromosomes (pair tl) !.hal of interstitial heterochromatin in !.he long arms of pair 3. The loca1ions c

u are almost totally heterochromatic. (q Silver-stained metaphase cell of lelomeri<: heterochromatin on pairs J and 2 are shown by arrow heads d showing (he presence of eighl NOR-bearing chromosomes. The arrow while the small C1Jchromatic section visible at !.he distal end of !.he X o r .indicates the satelli(e association sometimes observed between NOR· chromosome·s long arm (open arrow) is also indicated. (C) A panial p e bearing chromosomes. silver-Stained metaphase ceU showing \0 NOR-bearing chromosomes. R Io.,...w ...... _ '_ ... _._0.-'. _01 .. __ ''''' ___ ' 55 '."" rod _ ..._ .... io ~l$Yr

,I ~ 55 u" •• • • • " •. ," •, "• do .... ' IE ..... ""'"- ,>_ . 1:0 " -.10 If - ,_ of ' " - ...... ' . _... .01_ ...... • • • • " '/, ..... "'''"'low, -- 'Cta... __ • ..,.,.. _ .. 'fI' '"J .,

"if...... ) 10...... ____ ... "' .., , . ","'" ..... ~ 0 " , 1 -..(110 . .. , io -...... Ie, F"", NO .. __ ..,..., It! I. '11" 0 0 0

2 • • "' .... <1"00'" ...... -. .._ '"' "" __ 01 " d e • o' t a

d f ... It (

0 0 o • r " ' " , "I e . . h • s i l ., M ,'" • b • ., . u .. ., 0 0 0 ..

P - ...... ft • 0'0 \ e x ,"• • • • h ~-.~ \ t " - .. . .t y 0 -

b ._ .f o~ .... 0 d .. •• .. e " t - .. n .. •• . a .. .. • • r • • t{ " g • e . • .. c o • • n e c i l • • • 0 • r e " " " d " n . I u • y H o a 0 0 w • •" e " t '.

a - G t

e H n 0 i b a S y b d e c u d ' : " o , r ._. ..._ ...... "'" • .. ..,0' '. ,",C' 5.",,10. __ _ p -- e ... R _.. _ _.. _-_. 10 7 11 9 1210 13 11 1413 8 12 7 8 8 16

15 15 16 14 17 18 1817 te 12 9 10 101 1113

2018 2118 x x y y x X Y ..;..v _____---I

. • )

0 Figure S Haploid composilt: iUustraUng Ihe proposed G-band homo­ Figure 6 Comparison of half karyotypeS showing propo5ed G·band 1

0 logies belween Ihe 2n = 44 and 2n '= 50 Aelhomys chrysophilus homologies between Aelhomys nomoquensis (2n '= 24; N) and Aelh(}. 2 cylOIype; (indicated with Ihe symbols B and A respectively). Chromo­ mys granl; (2n '" 32; G). Chromosome identification numbers cor· d e some identification numbers correspond to those of their respective respond 10 those of their respective karyotypes. The first chromosome t a karyotypes. The rlrSl chromosome in each pair is thaI of A. chrysoplu'/us in each pair is that of A. namoquel!SlS unless otherwise indicated. d

( (2n "" 44) unless otherwise indicated. The box contains the unmatched r small mctacenlric chromosomes remaining from each genome. e h s i l and the corresponding chromosome 5 in the 2n = 50 cyto­ b

u type. Other matches involving repeated tandem fusions could P

sequences of these small metacentric chromosomes. In ad­ be made but the homologous segments could, aJ best, onJy e h dition, noticeable differences were also apparenJ in the G-band be regarded as tentative. It is interesting to nOle tbat the X t

y patterns of both the X and the Y chromosomes. chromosomes of all four taxa show conservation of the b Half karyotypes of A. namaquensis (2n == 24) and A. primitive X chromosome banding pattern (pathak & Stock d e t granl; (2n = 32) are compared in Figure 6. The banding 1974). n

a patterns of the four large metacentric pairs (I - 4) of these r g species show good concordance. However, owing to the Discussion e c indistinctive G·bands and similarity in size of the acroce:n trics , The unbanded chromosomes of A. gran/i, A. ~maqlitnsis n e some difficulty was experienced in matching the homologous and A. chrysophilus were first reported by Matthey (1958, c i l chromosomes. The submetacentric chromosome pair 5 present 1964). The diploid nwnbers for A. namaquensis (2n = 24) r e in A. namaquensis is thought to have originated through a and A. granli (20 = 32) are confumed in the present investi· d pairs n series of tandem fusions, involving four of the small gation. More recently, Gordon and Rautenbach (1980) pro­ u

A. granti acrocentrics (nwnbers 5, 14, 8 and 7), and one vided cytogenetic data on A. chrysophilus from Zimbabwe y a centric fusion involving cru-omosome 11. The euchromatic which indicate the presence of two cytOtypes (2n = 44, 2n = w

e portions of the X chromosomes of A. namaquensis and A. 50) Ln this region. AJthough sympatric populations were t a granri are similar. The discrepancy in size of the X chromo­ identified, no evidence of hybricfuat.ion was found and they G somes of these species is attributable to the presence of a large suggested that the two fonm represent.different species. This t e block of heterochromatin in the X chromosome of A. granD. n observation is confirmed and extended by the present investi· i b No apparent band homology was evidem in the comparisons gation. The distinctness of the karyotypes of the two A. a

S chrysophilus cytOtypes, the absence of hybrids in areas of

of the Y chromosomes. y Only a very small portion of !.he genomes of A. nama­ sympalry as as pronounced differences in the morphology b well

d quensis and A. granti on one hand, and the two A. chryso­ of their spermatozoa (Gordon & Wat5()n 1986) provide clear e c philuscytorypes on the OIher, are directly homologous Ln band evidence of the absence of gene flow between the two groups u d sequence. The long arm of chromosome 3 in both A. noma­ and suppan the recognition of the cytotypes as two distinct o r quensis and A. granti have good banding homology with the sibling species. Importantly in the absence of diagnostic criteria p e long arms of pair 3 in the 2n = 44 A. chrysophilus cytotype other than lhe chromosomes, only the A. chrysophilus R 268 S.-Afr. Tydskr. Dierk. 1986, 21(3)

2n = 50 cytotype occurs at Mazoe (Gordon, D.H. 1986, isolated demes. A. chrysophilus, on the other hand, occurs pers. comm.), the type locality of this species. in open savanna and may be less prone to isolation. Matthey (1964) proposed that A. namaquensis and A. The of the genus Aethomys has in the past been granti are closely related. In his comparison of the karyotypes the subject of much debate (De Graaff 1981). Of relevence of these two species, he suggested that Robertsonian trans­ here is that Davis (1965) treated Aethomys as a full genus locations (or centric fusions) and pericentric inversions could which included the subgenera Stochomys and Michaelamys. have played a significant role in their chromosomal evolution. He later removed Stochomys from the genus and divided it Although a close chromosomal relationship between these into the subgenera Aethomys and Michaelamys (Davis 1975). species is indicated by our data the difference in diploid This treatment is currently generally accepted although some number was traced to tandem fusions and a single centric uncertainty still exists as to the status of some species and fusion, with no evidence of pericentric inversions being found. their taxonomic affinities. In this respect, the present results Gordon & Rautenbach (1980) refer to a report by Matthey provide supportive evidence for Davis's later classification. (1954) of a submetacentric X chromosome in A. chrysophilus There are evidently closer karyologicaI affInities between A. (2n = 44), which is in accordance with their own findings. granti and A. namaquensis on the one hand, and the two However, in a subsequent paper Matthey (1964) documented A. chrysophilus cytotypes on the other, than between the two the X chromosome as being a large acrocentric. This morpho­ subgenera. logy is confirmed by the present results. Unfortunately Matthey's original report (Matthey 1954) and somewhat Acknowledgements surprisingly, Gordon & Rautenbach (1980), provide only Financial support from the University of Pretoria and the schematic representations of the species chromosomes thereby Council for Scientific and Industrial Research is gratefully confounding comparisons of the data sets. The acrocentric acknowledged. X chromosome evident in the test material could have arisen through a pericentric inversion, or alternatively, the meta­ References centric morphology reported by Gordon & Rautenbach (1980) BLOOM, S.E. & GOODPASTURE, c. 1976. An improved and the earlier Matthey report (1954) may have resulted from technique for selective silver-staining of nucleolar organizer the addition of heterochromatin to the short arms of a regions in human chromosomes. Hum. Genet. 34: 199-206. DAVIS, D.H.S. 1965. Classification problems of African Muridae. previously acrocentric chromosome. Since banded karyotypes Zool. Afr. I: 125 -145. were not presented by these authors, it is impossible to DA VIS, D.H.S. 1975. Genus Aethomys. Part 6.6, pp. 1 - 5. In: determine which of these mechanisms were responsible for The of Africa: an identification manual. (Eds) this structural change. However, the X chromosomes of the Meester, J. and Setzer, H.W. Smithsonian Institution Press,

.

) two A. chrysophilus cytotypes appear to be homologous. Washington D.C. 0 DE GRAAFF, G. 1981. The rodents of Southern Africa. 1 Furthermore, the X chromosomes of all four taxa retained 0 Butterworth & Co., South Africa. 2

the two major bands representing the primitive G-banding GORDON, D.H. & RAUTENBACH, I.L. 1980. Species d e pattern characteristic of most mammals (Figure 7; Pathak & complexes in medically important rodents: Chromosomes of t a Stock 1974). Aethomys, Tatera and Saccostomus (Rodentia: Muridae, d

( J. Of particular interest is the contrast in modes of karyotypic Cricetidae). S. Afr. Sci. 76: 559 - 561. r e change followed by the species of each SUbgenus. Representa­ GORDON, D.H. & WATSON, C.R.B. 1986. Identification of h cryptic species of rodents

s (Mastomys, Aethomys, Saccostomus) i tives of the subgenus Aethomys, A. bocagei, A. kaiseri and l in the Kruger National Park. S. Afr. J. Zoo!_ 21: 95-99. b A. chrysophilus are all characterized by a diploid number of u MATIHEY, R. 1954. Nouveaux documents sur les chromosomes

P 2n = 50 (Matthey 1954), the 2n = 44 cytotype of A. chryso­ des Muridae. Caryologia 6: 1 - 44. e

h philus being the only exception. Should the chromosomal MATIHEY, R. 1958. Les chromosomes et la position t constitutions of the species comprising this subgenus closely systematique de quelques africains (Mammalia­ y b Rodentia). Acta tropica 15: 97 - 117. reflect the ancestral condition for the two subgenera then it d MATIHEY, R. 1964. La formule chromosomique chez sept e is not unreasonable to argue that the constituent species of t especes et sous-especes de Murinae africains. Mammalia 28: n the subgenus Michaelamys have undergone what amounts to a 403-418. r a rapid and extensive reorganization of their genomes since g PAUL, J. 1975. Cell and tissue culture. Longman Group Ltd.,

e diverging from a common ancestor. An obvious question that Edinburgh. c n arises is what factors may have contributed to the differential PATHAK, S. & STOCK, A.D. 1974. The X chromosomes of e c mammals: karyological homology as revealed by banding i rates of karyotypic change between the two subgenera? In l techniques. Genetics 78: 703 -714. r this respect it is interesting to note that both A. namaquensis e SUMNER, A.T. 1972. A simple technique for demonstrating d and A. granti inhabit rocky and mountainous situations whose

n centromeric heterochromatin. Expl Cell Res. 75: 304 - 306.

u disjunct distribution may facilitate the fIXation of structural

WANG, H.C. & FEDOROFF, S. 1972. Banding in human y rearrangements through fragmentation of the species into chromosomes treated with trypsin. Nat. New Bioi. 235: 52 - 54. a w e t a G t e n i b a S y b d e c u d o r p e R J