c Indian Academy of Sciences

ONLINE RESOURCES

Microsatellite markers for the tramp , obscurior (Formicidae: )

CHRISTINE V. SCHMIDT∗, ALEXANDRA SCHREMPF, ANDREAS TRINDL and JÜRGEN HEINZE

Zoologie/Evolutionsbiologie, Universität Regensburg, 93053 Regensburg, Germany

[Schmidt C. V., Schrempf A., Trindl A. and Heinze J. 2016 Microsatellite markers for the tramp ant, Cardiocondyla obscurior (Formicidae: Myrmicinae). J. Genet. 95, e1–e4. Online only: http://www.ias.ac.in/jgenet/OnlineResources/95/e1.pdf]

Introduction disperse, wingless males stay inside their maternal nests throughout their lives and mate with young nestmate queens Cardiocondyla obscurior is a successful but unobtrusive (Kinomura and Yamauchi 1987;Stuartet al. 1987). This tramp ant which is widely distributed throughout the trop- results in considerable inbreeding: from fixation coefficients ics and subtropics. To understand its population structure in single-queen colonies of noninvasive Cardiocondyla,from and biology, we established six polymorphic microsatellite Mediterranean Europe and Madagascar it has been estimated markers. We determined the general variability of the six that up to 90% of the mating are between brothers and sis- loci by analysing individuals from four different popula- ters (Schrempf et al. 2005, 2014; Lenoir et al. 2007;Heinze tions. In addition, we analysed samples from a Japanese pop- et al. 2014). ulation and revealed first insights into the population and Due to the lack of variable genetic markers, nothing is colony structure of this tramp , including high nest- known about the genetic structure of the tramp species. mate relatedness in multi-queen colonies and a relatively Their colonies typically contain multiple queens and, like in high inbreeding coefficient. other invasive species, individual nests might not be closed Numerous plants and have become established in against alien individuals (Heinze and Delabie 2005). Our areas far away from their native ranges due to human activ- study species C. obscurior was originally described from Tai- ities (Hulme 2009). Among these are several dozen tramp wan (Wheeler 1929) and has spread from its presumed origin species of . Some of them, such as the red imported fire in Southeast Asia as far as to the Ryukyus, the Caribbean, the ant, the Argentine ant, and the yellow crazy ant, quickly be- Canary Islands, and other places (Seifert 2003). In come dominant in new ecosystems and are among the worst this study, we report the sequences of primers for microsatel- invasive species worldwide, on a par with tiger mosquitoes, lites developed for C. obscurior, data on nestmate related- black rats, or water hyacinths (reviewed by Holway et al. ness and inbreeding in colonies of this ant from Okinawa, 2002; Tsutsui and Suarez 2003). The ecological impact, Japan. Our results show that the microsatellite primers are behavioural ecology and sociobiology of these eye-catching suited well to conduct more detailed analyses, which will invaders have been intensively studied, but not much is give further insights into the population structure of this ant. known about tramps that less dramatically integrate in novel environments. Among the most widespread of these incon- spicuous tramp ants are the ‘sneaking ants’, Cardiocondyla. Material and methods At least seven of its ∼100 species are tramps and a subset of two or three of these can be found reliably on most tropical For the development of microsatellite primers, we fol- beaches, and in plantations and parks, around the tropics and lowed a method based on selective hybridization (Tenzer subtropics (Seifert 2003;Heinzeet al. 2006; Wetterer 2012a, et al. 1999; Gautschi et al. 2000). In detail, 50 individ- b, 2014). uals from a colony of Okinawa (collected in 2007) and Cardiocondyla exhibits a bizarre male diphenism with 200 individuals from three Brazilian colonies (collected at so-called ‘ergatoid’ wingless males in addition to the typical CEPLAC, Ilhéus, Brazil in 2004 and 2009) were pooled winged male of ants. While winged males are peaceful and and the genomic DNA was subsequently extracted using a modified CTAB (cetyltrimethylammonium bromide) proto- col (Sambrook and Russell 2001). According to the proto- ∗ For correspondence. E-mail: [email protected]. col of Kautz et al. (2009), we ligated the adapters onto the Keywords. relatedness; inbreeding; microsatellites; Cardiocondyla obscurior.

Journal of Genetics Vol. 95, Online Resources e1 Christine V. Schmidt et al.

DNA after restriction with Tsp509I. Microsatellite sequences 6) 4) 4) 5) 6) 6) = = = = = were fished with the help of magnetic beads. PCR fragments = n n n n n were cloned using the TOPO TA Cloning kit (Invitrogen, n Carlsbad, USA) following the manufacturer’s protocol. Suc- cessful recombinants were easily identified by the white 30) 206 ( 30) 139 (

colour of colonies. Two hundred and eighty six of them were = = n n 30) 110 ( dot-blotted on nylon membranes (Hybond-N Amersham) and 10) 185 ( = = n probed with (GA)13 oligonucleotide labelled with fluorescein n (MWG Biotech, Ebersberg, Germany). For detection, we used 10)10) 156 ( 161 ( = the Gene Images CDP-Star Detection Module (Amersham = n Life Science, Little Chalfont, UK). In total, 104 clones yiel- n ded a positive signal, and for the strongest 92 of them, plasmids were purified and sequenced by GATC Biotech. The obtained sequences were edited in Sequencing Analysis 4) 156,160 ( 6) 163,203 ( 6) 159,161,183,185 ( 2) 212,216,218,222,226 ( 5) 157,161,163,165,171 ( 3.4.1 (PE Biosystems, Waltham, USA) and visually checked 6) 152,154,156,158 ( ======n n n n n for microsatellites. Primers flanking the core microsatellite n repeats were manually designed and tested for 38 loci.

Standard PCR was carried out in a final 20 μL reac- Alleles (bp) tion volume containing 2.5 ng DNA, 1× buffer, 2.5 mM MgCl2, 200 μM dNTP (MBI Fermentas, Amherst, NY), 0.5 148) 161 ( μM (Cobs8), 1 μM (Cobs9, Cobs13) and 1.25 μM (Cobs 137) 189 ( = =

P3, Cobs13.2, Cobs 42) unlabelled reverse primer, the same n n amounts of FAM-labelled, HEX- labelled, or TET-labelled 133) PIC: 0.55 206 ( = forward primer, and 0.5 U Taq DNA polymerase (MBI, n ◦ 140) PIC: 0.40 145 ( Fermentas). Cycling conditions were 94 C for 70 s, anneal- =

◦ n ing for 45 s, 72 C for 25 s for 34 cycles with an initial denat- 149) PIC: 0.04 110 ( 7) *163*203 ( 8) *148*152*156*160 156 ( ◦ = = = uration step at 94 C for 4 min and a final extension step at n n

◦ n 72 C for 1 min. PCR products were visualized on an ABI 8) *185*187*189 ( = n PRISM genetic analyzer (PE Biosystems). Allele size was 130) PIC: 0.46 =

analysed using a GeneScan 500 size standard (TAMRA) and n GeneScan 3.1 software (PE Biosystems). To determine the general variability of the six microsatel- lite loci for which primers were developed, we analysed individuals from four different populations: Ilhéus, Brazil (collected in 2004 and 2009); Lake Alfred, Florida (2004); Okinawa, Japan (2007 and 2011); Ishigaki, Japan (2007); for number of investigated individuals see table 1. Finally, . we analysed the genetic structure of eight colonies of C. obscurior from four different sites in and around Naha, Okinawa, by genotyping 2 to 9 queens and 7 to 15 workers per colony (table 2). Six colonies originate from two nearby trees C. obscurior in Onoyama Park, Naha (tree 1: OypC51, OypC32, OypC48, ◦  ◦  R: GACGTACGGCCAGATGTCA R: TTTCGACAATGACAAACCGAGC ( F: TTATCGTGAGGATTTTGAGGC 160,180 ( F: ACTCAGTGCCAATTCGAATAAACAGCR: TGAACCGGGTAGAATCAATTA F: TATCTTTTCAACCCTCTCGC 163, 203 ( F: ACTCTCACAATCGCTACGC 189 ( F: AATCGCGCCTGCGACGGCG *110*130 ( R: AGTTTCTCACTTTTGCTCG PIC: 0.09 F: TCAGAGAAGTAAATATCAG *200*204*206*208 ( *137*139*145 ( 26 12.255 N, 127 40.656 E; tree 2: OypE3, OypE5, OypE4, R: TATTCCGCGATAGCTTAAAT PIC: 0.12 ◦  ◦ 

26 12.334 N, 127 40.592 E). Two additional colonies came 11 6 10 31 29 34 from Urasoe Sports Park, Naha (7 km distance to Onoyama, 11 35 26◦15.134N, 127◦43.262E) and the campus of the University TT(TC) Repeat Primer sequences Alleles (bp) Japan, Alleles (bp) Alleles (bp) of the Ryukyus (11 km distance to Onoyama, 26◦14.782N, TT(TC) ◦  C) sequence (5’ to 3’) Japan, Okinawa Ishigaki Brazil Florida a

127 45.864 E). Relatedness within the colonies and with- ◦ T in the whole population was estimated using the software Relatedness 4.2 (Goodnight and Queller 1994) based on the algorithm by Queller and Goodnight (1989). Groups were weighted equally and standard errors were esti- Accession

mated by jackknifing over loci. The polymorphic infor- Primer sequences of microsatellite loci in mation content (PIC, Botstein et al. 1980) was calculated using the software Cervus 3.0 (Kalinowski et al. 2007). , annealing temperature a Cobs9 HE608812 55 (TC) Cobs 13.2 AJ784427 45 (AG) Table 1. Cobs8 HE608811 55 (AG) T Locus no. (EMBL) ( Cobs13 HE608813 60 (CT) Cobs 42 AJ784429 45 (AG) Cobs P3 AJ784426 50 (AG) To obtain information about the amount of inbreeding, *Alleles resulting from the genetic population analysis.

Journal of Genetics Vol. 95, Online Resources e2 Microsatellite markers for C. obscurior

Table 2. Relatedness (mean ± SE) over all individuals, and for queens and workers separately in C. obscurior colonies from Okinawa.

Number of Number of r (queen) (number of r (worker) (number of Colony queens workers analysed individuals) analysed individuals)

OypC32 21 76 0.838 ± 0.202 (n =9) 0.619 ± 0.176 (n =15) OypC48 15 61 0.331 ± 0.197 (n =9) 0.582 ± 0.216 (n =15) OypC51 26 34 0.817 ± 0.155 (n =9) 0.784 ± 0.133 (n =15) OypE3 14 17 0.227 ± 0.170 (n =9) 0.636 ± 0.201 (n =15) OypE5 7 11 -0.014 ± 0.229 (n =7) 0.448 ± 0.208 (n =11) OypE4 9 25 0.297 ± 0.276 (n =9) 0.357 ± 0.149 (n =15) Urasoe Sports Park 2 9 1.000 (n =2) 0.778 ± 0.119 (n =9) Univ. of the Ryukyus, Campus 10 8 0.913 ± 0.109 (n =7) 0.411 ± 0.206 (n =7) we calculated Weir’s F (inbreeding coefficient) with the soft- father (unrelated queens having mated with the same male). ware GDA (Lewis and Zaykin 2001) for the six colonies of Half-sisters in societies usually have the same mother the two nearby trees. Even though sample size is low for but different fathers (as in honeybees or other species with these six colonies, we also give the P-value, which gives multiply mated queens). In contrast, the special mating biol- information on population structure. Confidence intervals ogy of Cardiocondyla with non-dispersing, wingless males were calculated by bootstrapping 5000 times over loci. that monopolize mating raises the possibility of paternal half- sibs. Even though we found significant inbreeding in the pop- ulation (six colonies: F = 0.293, 95% CI 0.145–0395; P: 0.260), sib-mating appears less common than in monogy- Results and discussion nous Cardiocondyla species. This indicates that the likely From the 38 loci which yielded a good amplification prod- adoption of alien queens allows the opportunity for outbreed- uct, six were polymorphic with three to nine alleles per locus ing despite mating in the nest in C. obscurior. The same holds (see table 1). Interestingly, there is no overlap in allele length for the tramp ant, Linepithema humile. Although mating between Brazilian and Japanese, and Florida populations at takes place in the nest, inbreeding coefficients have also been three primers, respectively, which suggests a long separa- found to be very low as foreign queens are adopted in the tion of the Brazilian population, and/or alternatively differ- colonies (Krieger and Keller 2000). ent source populations during colonization. A recent study could show that populations differ strongly in several aspects Acknowledgements (Schrader et al. 2014). Our preliminary analysis confirms that there is sufficient genetic variation for detailed stud- Supported by Deutsche Forschungsgemeinschaft (SCHR 1135/1-1). ies, and hence, a future analysis will allow gaining further We thank M. Suefuji for providing the specimens used in the genetic insights into possible differences concerning the colony and analysis of Japanese C. obscurior. population structure between the two populations. Our analysis of a small number of colonies from Okinawa References yielded some important insights into the breeding structure of C. obscurior. Generally, relatedness varied greatly among Botstein D., White R. L., Skolnick M. and Davis R. W. 1980 colonies and in some cases was surprisingly low. Overall, Construction of a genetic linkage map in man using restriction worker relatedness and queen relatedness are not different fragment length polymorphisms. Am.J.Hum.Genet.32, 314– from each other (worker relatedness: 0.580 ± SE 0.127 (102 331. individuals), queen relatedness: 0.612 ± SE 0.366 (61 indi- Gautschi B., Tenzer I., Muller J. P. and Schmid B. 2000 Isolation =− = and characterization of microsatellite loci in the bearded vulture viduals); t-test for paired samples: t 0.22, P 0.83; see (Gypaetus barbatus) and cross-amplification in three Old World table 2). Relatively low relatedness among queens in some of vulture species. Mol. Ecol. 9, 2193–2195. the colonies (colonies OypC48, OypE3, OypE5 and OypE4, r Goodisman M. and Ross K. 1997 Relationship of queen number <0.35) might indicate that alien queens can be easily adopted and queen relatedness in multiple-queen colonies of the fire ant in the colonies as has already been suggested by Kinomura Solenopsis invicta. Ecol. Entomol. 22, 150–157. Goodnight K. F. and Queller D. C. 1994. Relatedness 4.2b. Good- and Yamauchi (1987). Queen adoption appears to be the rule night Software, Houston, USA. in other tramp species such as Solenopsis invicta and Linep- Heinze J. and Delabie J. H. C. 2005 Population structure of the ithema humile, resulting in low relatedness values close to male-polymorphic ant Cardiocondyla obscurior. Stud. Neotrop. zero (e.g. Krieger and Keller 2000; Goodisman and Ross Fauna Environ. 40, 187–190. 1997). The occasionally higher worker relatedness in com- Heinze J., Cremer S., Eckl N. and Schrempf A. 2006 Stealthy invaders: the biology of Cardiocondyla tramp ants. Insect. Soc. parison to queen relatedness in some colonies might be either 53, 1–7. the result of reproductive skew, i.e. some queens produc- Heinze J., Schrempf A., Rakotondrazafy H., Rakotondranaivo T. ing more worker offspring than others, or due to a common and Fisher B. L. 2014 Polygyny, inbreeding, and wingless

Journal of Genetics Vol. 95, Online Resources e3 Christine V. Schmidt et al.

males in the Malagasy ant Cardiocondyla shuckardi Forel Schrader L., Kim J. Y., Ence D., Zimin A., Klein A., Wyschetzki K. (, Formicidae). Sociobiology 61, 300–306. et al. 2014 Transposable element islands facilitate adaptation to HolwayD.A.,LachL.,SuarezA.V.,TsutsuiN.D.andCaseT.J. novel environments in an invasive species. Nat. Comm. 5, 5495. 2002 The causes and consequences of ant invasions. Annu. Rev. Schrempf A 2014 Inbreeding, multiple mating and foreign sexu- Ecol. Syst. 33, 181–233. als in the ant Cardiocondyla nigra (Hymenoptera: Formicidae). Hulme P. E. 2009 Trade, transport and trouble: managing invasive Myrmecol. News 20, 1–5. species pathways in an era of globalization. J. Appl. Ecol. 46, Schrempf A., Reber C., Tinaut A. and Heinze J. 2005 Inbreed- 10–18. ing and local mate competition in the ant Cardiocondyla batesii. Kalinowski S. T., Taper M. L. and Marshall T. C 2007 Revising Behav. Ecol. Sociobiol. 57, 502–510. how the computer program CERVUS accommodates genotyping Seifert B. 2003 The ant genus Cardiocondyla (: error increases success in paternity assignment. Mol. Ecol. 16, Hymenoptera: Formicidae) – a taxonomic revision of the C. 1099–1106. elegans, C. bulgarica, C. batesii, C. nuda, C. shuckardi, C. Kautz S., Schmid V. S., Trindl A., Heinze J., Ballhorn D. J. and stambuloffii, C. wroughtonii, C. emeryi and C. minutior species Heil M. 2009 Isolation and characterization of 22 microsatellite groups. Ann. Naturhist. Mus. Wien. 104, 203–338. loci in the plant-ant Pseudomyrmex ferrugineus (Formicidae: Stuart R. J., Francoeur A. and Loiselle R. 1987 Lethal fighting Pseudomyrmecinae) and cross-testing for two congeneric among dimorphic males of the ant Cardiocondyla wroughtonii. species. Mol. Ecol. Res. 9, 1016–1019. Naturwissenschaften 74, 548–549. Kinomura K. and Yamauchi K 1987 Fighting and mating behav- Tenzer I., Degli Ivanissevich S., Morgante M. and Gessler C. 1999 iors of dimorphic males in the ant Cardiocondyla wroughtoni. J. Identification of microsatellite markers and their application to Ethol. 5, 75–81. population genetics of Venturiainaequalis. Phytopathology 89, Krieger M. J. B. and Keller L. 2000 Mating frequency and genetic 748–753. structure of the Argentine ant Linepithema humile. Mol. Ecol. 9, Tsutsui N. D. and Suarez A. V. 2003 The colony structure and 119–126. population biology of invasive ants. Cons. Biol. 17, 48–58. Lenoir J. C., Schrempf A., Lenoir A., Heinze J. and Mercier J. Wetterer J. K. 2012a Worldwide spread of the moorish sneaking L. 2007 Genetic structure and reproductive strategy of the ant ant, Cardiocondyla mauritanica (Hymenoptera: Formicidae). Cardiocondyla elegans: strictly monogynous nests invaded by Sociobiology 59, 985–997. unrelated sexuals. Mol. Ecol. 16, 345–354. Wetterer J. K. 2012b Worldwide spread of Emery’s sneaking ant, Lewis P. O. and Zaykin D. 2001. Genetic data analysis: com- Cardiocondyla emeryi (Hymenoptera: Formicidae). Myrmecol. puter program for the analysis of allelic data.Version 1.0 (d16c) News 17, 13–20. available at http://en.bio-soft.net/dna/gda.html. Wetterer J. K. 2014 Worldwide spread of the lesser sneaking Queller D. C. and Goodnight K. F. 1989 Estimating relatedness ant, Cardiocondyla minutior (Hymenoptera: Formicidae). Fla. using genetic markers. Evolution 43, 258–275. Entomol. 97, 567–574. Sambrook J. and Russell D. W. 2001. Molecular cloning: a labora- Wheeler W. M. 1929 Ants collected by Professor F. Silvestri in tory manual, 3rd edition. Cold Spring Harbor Laboratory Press, Formosa, the Malay Peninsula and the Philippines. Boll. Lab. New York, USA. Zool. Gen. Agr. R. Scuola. Agric. Portici. 24, 27–64.

Received 9 April 2015, in revised form 16 June 2015; accepted 21 July 2015 Unedited version published online: 23 July 2015 Final version published online: 15 February 2016

Journal of Genetics Vol. 95, Online Resources e4