An antibiotic selection marker for nematode transgenesis
Rosina Giordano-Santini1, Stuart Milstein2–4, Nenad Svrzikapa2–4, Domena Tu5, Robert Johnsen5, David Baillie5, Marc Vidal2–4 & Denis Dupuy1
We have developed a nematode transformation vector carrying the bacterial neomycin resistance gene (NeoR) and shown that it
could confer resistance to G-418 on both wild-type Caenorhabditis elegans and C. briggsae. This selection system allows hands-off maintenance and enrichment of transgenic worms carrying non-integrated transgenes on selective plates. We also show that this marker can be used for Mos1-mediated single-copy insertion in wild-type genetic backgrounds (MosSCI-biotic).
Antibiotic-resistance genes are commonly used as markers to monitor the introduction of exogenous genetic material into cells. Although they are widely used for genetic manipulation of cultured eukaryotic cells, yeast and bacteria1, antibiotic selec- tion systems have not so far been used for nematode transgene- sis. Most genetic markers for nematode transgenesis have been based on easy-to-score phenotypes2–5. With a few exceptions3,5, these markers do not provide a selective advantage, making the and prokaryotes and that is commonly used as a selective agent for eukaryotic cells1. Young larvae were more sensitive to G-418 than adults, probably because they are more dependent on their protein synthesis machinery for development. For each species, we determined a critical G-418 concentration, corresponding to the optimal selective conditions, at which young larvae were unable to develop and died after a few days, whereas young adults lived and were able to lay eggs (Fig. 1a).
We developed nematode transformation vectors carrying NeoR under the control of the ubiquitous promoter of C. elegans ribosomal protein gene rps-27 (Fig. 1b,c). NeoR is widely used to confer resistance to G-418 to eukaryotic cells1. We injected wild-type C. elegans and C. briggsae with pDD04Neo also car- rying the fluorescent reporter C.e.pmyo-2::gfp. We then placed single injected P0 parents onto nematode growth medium plates containing G-418 at the critical concentration (selective plates) and allowed them to lay eggs at 20 °C. After 3 or 4 d, we screened plates for F1 adults and we isolated them onto selective plates. We obtained several independent extrachromosomal G-418-resist- ant lines for both species. It was possible to identify at a glance plates from the F1 generation in which transmission of the array occurred, as only transgenic worms could give rise to mixed- stage populations of worms. These results showed that pDD04Neo confers resistance to G-418 on wild-type C. elegans and C. briggsae and allows hands-off selection of transgenic worms in the presence of the drug. To facilitate cloning experiments, we designed pdestDD04Neo (Fig. 1b) as a destination vector in the populations, on the basis of GFP expression, using a COPAS Profiler (a fluorescence-assisted nematode sorter, Union Biometrica) (Fig. 2). When worms were grown on nonselective plates, transgenic worms represented 6.82% of the total popu- lation on average owing to the instability of the array3 (Fig. 2c). When worms were grown at the critical G-418 concentration, the mean enrichment percentage reached 37.68% (three independ- ent experiments with six C. elegans and two C. briggsae stable lines each). As the sorter also counted arrested or dead wild-type larvae, this value underestimates the actual enrichment. When we calculated enrichment percentages excluding L1 larvae on the basis of size, worms expressing GFP represented 93.09% of worms from L2 to adults, on average (Fig. 2d). The percentage of transgenic worms could reach up to 99.5% and was independent of species and transmission rate. We also carried out preliminary selection experiments in liquid medium (M9 buffer), in which we obtained 25–125-fold enrichment of transgenic worms, after a 1:10,000 dilution with wild-type worms, in one generation (4 d) (Supplementary Table 1). To our knowledge, this is the first evidence of a nematode transformation marker allowing near- perfect enrichment of non-integrated transgenic populations.
Figure 1 | NeoR as a genetic marker for nematode transgenesis. (a) Sensitivity of wild-type nematodes to G-418. Gravid adults (A) or hatchlings (H) were placed on selective plates with the indicated drug concentrations and growth was monitored for 1 month. Black frames indicate critical concentrations that were lethal for larvae but not for adults. (b) The pdestDD04Neo vector carries the bacterial neomycin resistance gene (NeoR) under the control of the C. elegans rps-27 promoter.
We obtained G-418-resistant lines by injection of either circular pDD04Neo pmyo-2::gfp (yielding transgenic worms with repetitive extrachromosomal arrays) or a complex mixture of linear plasmid and digested conspecific genomic DNA (yielding transgenic worms with complex extrachromosomal arrays). Some progeny of C. elegans lines with repetitive extrachromosomal arrays showed germline morphological defects and lower fertility even in the absence of the drug (data not shown), although the lines could still be efficiently propagated on selective plates over many generations without manual maintenance. This phenotype is unlikely to be a side effect of the expression of NeoR itself as extra- chromosomal arrays are silenced in the germline3. Conversely, silencing of the pDD04neo array might lead to a co-suppression effect11. This hypothesis is supported by the observation that both C. elegans and C. briggsae NeoR transgenic lines carrying complex arrays did not show this phenotype. Array composition did not have an effect on the efficiency of G-418 selection (Fig. 2c,d).
In the course of this work we have maintained more than 12 strains on selective media for several months without observing any adverse effect on the resistant worms other than the accumu- lation of dead eggs and arrested larvae due to the instability of the extrachromosomal array. There is no indication that G-418 causes harm to non-mosaic transgenic worms that are properly protected by the transgene. However it remains possible that the presence of G-418 in the medium affects the outcome of certain types of experiment. For such cases the use of integrated lines or preselected resistant individuals should overcome the problem.
To test whether this selection system was compatible with the recently described Mos1-mediated single-copy insertion method (MosSCI)12, we selected two strains with an intergenic Mos1 insertion (EN5271, EN5273). For each of them we built a repair template vector containing the NeoR cassette and a pmyo-2::gfp transgene between ~1.4 kb of homologous chromosomal DNA from each flanking side of the Mos1 element insertion site, as described12 (Fig. 1c). We co-injected these vectors together with a vector encoding a transposase under the control of a heat-shock pro- moter (pJL44) and a vector carrying prgef-1::DsRed2 (pCB101). As expected, the obtained extrachromosomal array strains
were resistant to G-418 and expressed GFP in the pharynx and DsRed2 in the nervous system. After exposing the worms to heat shock13, we allowed their progeny to proliferate and screened for integration events by identifying G-418-resistant individuals that expressed GFP in the pharynx and did not express DsRed2 (ref. 12). We confirmed insertion events by PCR (Supplementary Fig. 1) and checked that the worms carried a single copy of the inserted transgene by quantitative PCR (qPCR) (Supplementary Figs. 2 and 3). Thus, we demonstrated that our antibiotic selection system could be used in the context of MosSCI (Fig. 2b). This combined MosSCI-biotic method can be used directly on any strain from the NemaGENETAG14 collection without the need to introduce a mutant unc-119, which is typically used as a co-insertion marker for MosSCI, into the genetic background12 (Supplementary Table 2).
Another commonly used method for C. elegans transformation is microparticle bombardment3. We did not succeed in using G-418 selection with this technique as clear-cut selection cannot be obtained in excessive crowding conditions and ballistic trans- formation requires a very large population of target worms.
Antibiotic selection has several advantages over commonly used nematode transformation markers. First, rescue of nonlethal mutations such as unc-119 or dpy-5 requires the use of specific mutant backgrounds3,4 that might not be appropriate for some biological studies; by contrast, antibiotic resistance can be used with any genetic background including transgenic or mutant strains of interest. Second, maintenance and enrichment of non- integrated transgenic lines using existing markers is done by man- ually picking transgenic worms. This process is time consuming and may limit experiments that need many transgenic worms.
It is also possible that antibiotic resistance markers might prove useful for nematodes other than C. elegans and C. briggsae. Commonly used markers rely on mutant strains or dominant alleles, which are not always available for other nematode species, although some genes such as C.e.rol-6(su-1006) and C. elegans fluorescent reporters have been shown to work in other Caenorhabditis species8. Although many efforts have been made to apply C. elegans transgenesis methods to other species, the lack
of a convenient selection system remains a limitation for com- parative and evolutionary studies6–8. Here we have shown that five nematode species are sensitive to G-418; it should therefore be possible to use a common G-418 resistance marker for trans- genesis in other species as well.
BRIEF COMMUNICATIONS
Finally, the use of antibiotic resistance markers for nematode transgenesis can be expanded to other drugs15, which should enable the development of a wide range of powerful applications.
METHODS
Methods and any associated references are available in the online version of the paper at http://www.nature.com/naturemethods/.
Note: Supplementary information is available on the Nature Methods website.
ACKNOWLEDGMENTS
Supported by the Program INSERM “Avenir” (D.D.), la Fondation Bettencourt- Schueller (D.D.), le Conseil Régional d’Aquitaine (D.D.), la Fondation pour la Recherche Médicale (D.D.), Natural Science and Engineering Research Council of Canada (D.B.) and le Ministère Français de l’Enseignement de la Recherche et des Technologies (R.G.-S.). We thank I.A. Hope, J. Ewbanks and J.L. Bessereau for discussions and access to facilities; and T. Leste-Lasserre and G. Drut for discussion about qPCR. EN5271 and EN5273 and MosSCI related plasmids were provided by J.L. Bessereau (INSERM U1024, Institute of Biology of the École Normale Supérieure). The Caenorhabditis Genetics Center, which is funded by the National Center for Research Resources of the US National Institutes of Health, provided the nematode species.
AUTHOR CONTRIBUTIONS
S.M. and N.S. performed preliminary experiments under the supervision of M.V. D.D. designed and supervised the project and constructed the pDD04neo vector with N.S. Microinjections were performed by R.G.-S. and D.T. under the supervision of D.D., D.B. and R.J. R.G.-S. constructed MosSCI-biotic related vectors, characterized the transgenic worms and wrote the manuscript with D.D.
COMPETING FINANCIAL INTERESTS
The authors declare no competing financial interests.
Published online at http://www.nature.com/naturemethods/.
Reprints and permissions information is available online at http://npg.nature. com/reprintsandpermissions/.
1. Colbere-Garapin, F., Horodniceanu, F., Kourilsky, P. & Garapin, A.C. J. Mol. Biol. 150, 1–14 (1981).
2. Mello, C.C., Kramer, J.M., Stinchcomb, D. & Ambros, V. EMBO J. 10, 3959–3970 (1991).
3. Praitis, V., Casey, E., Collar, D. & Austin, J. Genetics 157, 1217–1226 (2001).
4. Thacker, C., Sheps, J.A. & Rose, A.M. Cell. Mol. Life Sci. 63, 1193–1204 (2006).
5. Granato, M., Schnabel, H. & Schnabel, R. Nucleic Acids Res. 22, 1762–1763 (1994).
6. Schlager, B., Wang, X., Braach, G. & Sommer, R.J. Genesis 47, 300–304 (2009).
7. Zhao, Z. et al. Genetics 184, 853–863 (2010).
8. Streit, A. et al. Genetics 152, 1573–1584 (1999).
9. Dupuy, D. et al. Genome Res. 14, 2169–2175 (2004).
10. Fire, A., Kondo, K. & Waterston, R. Nucleic Acids Res. 18, 4269–4270 (1990).
11. Dernburg, A.F., Zalevsky, J., Colaiacovo, M.P. & Villeneuve, A.M. Genes Dev.
14, 1578–1583 (2000).
12. Frokjaer-Jensen, C. et al. Nat. Genet. 40, 1375–1383 (2008).
13. Robert, V.J., Katic, I. & Bessereau, J.L. Methods 49, 263–269 (2009).
14. Bazopoulou, D. & Tavernarakis, N. Genetica 137, 39–46 (2009).
15. Semple, J.I., Garcia-Verdugo, R. & Lehner, B. Nat.Geneticin Methods advance online publication, doi:10.1038/nmeth.1495 (22 August 2010).