Final submission 15 August 2006

Supplemental on-line material for Russell et al.

Helpers improve the reproductive potential of offspring in cooperative meerkats

Methods for Genetic Techniques

Tissue samples were obtained by cutting (under licence) a small (<3mm) piece of tail-tip with a pair of scissors from pups around 3 weeks of age. Animals were sampled without anesthesia, returned to normal behaviour within seconds of the cut, and suffered no apparentill-effects. Samples were stored in DSMO or 100mM EDTA 95% EtOH at -20oC. DNA was extracted using standard chelex or phenol/chloroform methods and stored in ddH2O at -20oC (Griffin et al. 2003).

We ran 14 polymorphic microsatellite markers to assess paternity (AHT130, Hg810, Ssu7.1, Ssu8.5, Ssu10.4, Ssu11.12, Ssu13.8, Ssu13.9, Ssu14.14, Ssu14.18 (cloned from meerkats, Griffin et al. 2003) and Fca045, Fca077, Fca232 (cloned from domestic cat, Menotti-Raymond et al. 1999). Touchdown 10μl PCR multiplexes contained three to four markers labelled with fluorescent dyes HEX™, TET™ and FAM™ (PE Applied Biosystems®). All PCR reactions contained 1.5μl Applied Biosystems® Gold Buffer™, 0.8μl 25mM MgCl2, 1μl 10mM dNTP mix, 0.2ml Taq Gold™. The amount of marker and annealing temperature varied between markers and multiplexes, but ranged from 0.25 to 0.8μl 10mM and from 48 to 58oC. PCR products were pooled and run on a PE Applied Biosystem’s® ABI 377XL™ using TAMRA™ 500 size standard and cellulose combs to prevent spillover. Results were extracted and analysed blindly with Genotyper™ software.

Duplicate runs allowed us to assess marker error rates. Found error rates were accounted for in the analysis of paternity, and no attempt was made to rectify errors by additional runs, since such an approach may bias the data set if ‘expected’ alleles are searched. The overall error rate was below 3% including both dropouts (creating false homozygotes) and misprintings. We assessed paternity with the software NewPat (v. 5) (available at Previous work has shown that maternity (except in cases of simultaneous births which were excluded from this study) can be assigned with confidence using observational data (Griffin et al. 2003). Thus, for all individuals we ran the analyses with the maternal information in the background, improving our ability to distinguish the father. Runs always included every male alive in the population to avoid biasing by inadvertently omitting unexpected fathers. Data were excluded where we were less than 80% certain about the true father. It is important to note that this is not a study of paternity levels, and there is no reason why small numbers of mismatches in paternity would be biased with regards to their helper-mediated weights at independence.

Griffin, A. S.,Pemberton, J. M., Brotherton, P. N. M., McIlrath, G. M., Gaynor, D.,

Kansky, R. & Clutton-Brock, T. H.2003 A genetic analysis of breeding success in

the cooperative meerkat. Behav.Ecol. 14, 472-480.

Menotti-Raymond, M., David, V. A., Lyons, L. A., Schaffer, A. A., Tomlin, J. F., Hutton, M. K. & O’Brian, S. J. 1999 A genetic linkage map of microsatellites in the

domestic cat Felis catus. Genomics57, 9-23.

1