that the loss of Myc-335 did not have a majorimpact on intestinal cell differentiation. This re-sult is consistent with the lack of effect of Mycloss on homeostasis of intestinal epithelium(15–17).

A significant difference inMyc RNA expres-sion was not observed by qPCR from the duo-denum of the Myc-335−/− mice at p1. However,we detected a decreased expression of Myc tran-scripts in the colon of the Myc-335−/− mice bothby qPCR and exon array analysis at p1 (Fig. 2, Aand B). Chromatin immunoprecipitation and se-quencing (ChIP-seq) analysis for Tcf7l2 in adultmouse colon confirmed that, also in mice, thehighest peak of Tcf7l2 binding within 1 Mb ofMyc is located at Myc-335. Binding within thisregion was completely abolished in the colon ofMyc-335−/− mice, and no obvious compensatorychanges in binding pattern of Tcf7l2 were ob-served at other regions within 1 Mb ofMyc (Fig.2C). These results indicate thatMyc-335 containsa major Tcf7l2 binding site and that its loss re-sults in a modest decrease of Myc expression inthe mouse colon. However, the Myc-335−/−miceare viable and fertile, indicating that theMyc-335element is dispensable for function of the mouseintestine under standardmouse housing conditions.

TheMyc-335 element was first identified as aconserved element containing a cancer-associatedSNP that affects binding of TCF7L2 (5, 8), thetranscription factor that is activated in most casesof colorectal cancer (18, 19). Thus, even thoughMyc-335 is not critical for normal intestinal func-tion, it could still be required for tumorigenesis.To test this, we crossed the Myc-335−/−mice tothe Apcmin mouse strain that spontaneously de-velops tumors in the small intestine and colon.These tumors are dependent on the activity ofMyc (20–22) and the T cell factor/lymphoid en-hancer factor (TCF/LEF) family of transcriptionfactors (23–25). We generated Apcmin/+; Myc-335−/− mice and scored the number of polyps inthe small intestine and colon of these mice at4 months of age. We found a lower total numberof polyps in the intestines ofApcmin/+;Myc-335−/−

(n = 9) mice compared with that in the controlApcmin/+mice (n=9;P<0.00038; Fig. 3). The effectwas observed primarily at the level of frequencyof occurrence of tumors. From these results, weconclude that, in laboratory mice, Myc-335 func-tions as a tumor-specific enhancer element that isdispensable for normal intestinal function but re-quired for tumorigenesis.

Most associations identified by using GWAstudiesmap to noncoding regionswhose functionis largely unknown. This has been used as an ar-gument against the validity of the GWA approach[for example, (26)]. GWA studies have implicatedthe region containing the SNP rs6983267 as beingresponsible for more human cancer-associatedmorbidity than any other known inherited variantor mutation. Our results presented here validatethis region by showing that theMyc-335 enhancerhas a critical role in intestinal tumorigenesis inmice. Because MYC-335 does not contain other

SNPs with similar P values to rs6983267 (5), ouruse of theMyc-335 deletion allele does not affectthe conclusion that rs6983267 is the causativeSNP in humans.

Our results also highlight the fact that, al-though a disease-associated polymorphism typ-ically has a relatively modest effect size, theelement that it affects can be critically importantfor the underlying pathological process. In par-ticular, MYC is a promising target for cancertherapy (27) that has been difficult to target bydirect inhibitors (28). Our results suggest thatMYC expression could be decreased by inhib-itors targeting MYC-335. In a broader context,our finding that the phenotypic effect of Myc-335 loss is tumor-specific suggests that normalgrowth control and pathological growth inducedby cancer can use different mechanisms.

References and Notes1. F. R. Schumacher et al., Cancer Res. 67, 2951 (2007).2. L. A. Kiemeney et al., Nat. Genet. 40, 1307 (2008).3. L. T. Amundadottir et al., Nat. Genet. 38, 652

(2006).4. B. W. Zanke et al., Nat. Genet. 39, 989 (2007).5. S. Tuupanen et al., Nat. Genet. 41, 885 (2009).6. N. F. Wasserman, I. Aneas, M. A. Nobrega, Genome Res.

20, 1191 (2010).7. N. Ahmadiyeh et al., Proc. Natl. Acad. Sci. U.S.A. 107,

9742 (2010).8. M. M. Pomerantz et al., Nat. Genet. 41, 882 (2009).9. J. B. Wright, S. J. Brown, M. D. Cole, Mol. Cell. Biol.

30, 1411 (2010).10. M. M. Pomerantz et al., Cancer Res. 69, 5568 (2009).11. N. C. Dubois et al., Development 135, 2455 (2008).12. A. Trumpp et al., Nature 414, 768 (2001).

13. T. Sato et al., Nature 469, 415 (2011).14. T. Sato et al., Nature 459, 262 (2009).15. M. D. Bettess et al., Mol. Cell. Biol. 25, 7868

(2005).16. J. C. Cohen et al., BMC Dev. Biol. 4, 4 (2004).17. V. Muncan et al., Mol. Cell. Biol. 26, 8418 (2006).18. H. Clevers, Cell 127, 469 (2006).19. J. Taipale, P. A. Beachy, Nature 411, 349 (2001).20. D. Athineos, O. J. Sansom, Oncogene 29, 2585

(2010).21. N. A. Ignatenko et al., Cancer Biol. Ther. 5, 1658

(2006).22. K. Yekkala, T. A. Baudino, Mol. Cancer Res. 5, 1296

(2007).23. T.-C. He et al., Science 281, 1509 (1998).24. V. Korinek et al., Science 275, 1784 (1997).25. P. J. Morin et al., Science 275, 1787 (1997).26. J. McClellan, M. C. King, Cell 141, 210 (2010).27. L. Soucek et al., Nature 455, 679 (2008).28. G. Evan, Science 335, 293 (2012).

Acknowledgments: We thank S. Miettinen and A. Zetterlundfor technical assistance. This work was supported by theAcademy of Finland grant 140753, postdoctoral researcher’sproject 134073, Finnish Center of Excellence in CancerGenetics Research 250345, European Union FrameworkProgram 7 project SYSCOL, European Research Council projectGROWTHCONTROL, the Sigrid Juselius Foundation, the FinnishCancer Society, the Swedish Cancer Society, the Center forBiosciences at Karolinska Institutet, and Vetenskapsrådet.

Supplementary Materialswww.sciencemag.org/cgi/content/full/science.1228606/DC1Materials and MethodsFigs. S1 to S3Tables S1 and S2References (29–32)

9 August 2012; accepted 15 October 2012Published online 1 November 2012;10.1126/science.1228606

Evolution of an MCM Complex in FliesThat Promotes Meiotic Crossoversby Blocking BLM HelicaseKathryn P. Kohl,1 Corbin D. Jones,2,3 Jeff Sekelsky1,2,4*

Generation of meiotic crossovers in many eukaryotes requires the elimination of anti-crossoveractivities by using the Msh4-Msh5 heterodimer to block helicases. Msh4 and Msh5 have beenlost from the flies Drosophila and Glossina, but we identified a complex of minichromosomemaintenance (MCM) proteins that functionally replace Msh4-Msh5. We found that REC, an orthologof MCM8 that evolved under strong positive selection in flies, interacts with MEI-217 and MEI-218,which arose from a previously undescribed metazoan-specific MCM protein. Meiotic crossoverswere reduced in Drosophila rec, mei-217, and mei-218 mutants; however, removal of the Bloomsyndrome helicase (BLM) ortholog restored crossovers. Thus, MCMs were co-opted into a novelcomplex that replaced the meiotic pro-crossover function of Msh4-Msh5 in flies.

Crossovers (COs) between homologouschromosomes can be beneficial or detri-mental, depending on their context (1).

Meiotic COs increase genetic diversity and pro-mote accurate chromosome segregation, where-as mitotic COs can lead to loss of heterozygosity,potentially triggering tumorigenesis. Mitotic COsare prevented by “anti-CO” proteins. A key anti-CO protein is the Bloom syndrome helicase BLM,which generates non-CO products by unwinding

recombination intermediates that might other-wise be processed into COs (2). In meiosis, COformation is encouraged through inhibition ofanti-CO proteins. The budding yeast Msh4-Msh5heterodimer antagonizes the BLM ortholog Sgs1(3). Msh4 and Msh5 are found in all metazoansfor which sequence is available, except Drosoph-ila species and their fellow schizophoran Glos-sina morsitans, the tsetse fly (figs. S1 and S2).The lack of recognizable orthologs of these

www.sciencemag.org SCIENCE VOL 338 7 DECEMBER 2012 1363

REPORTSCORRECTED 29 MARCH 2013; SEE LAST PAGE

proteins suggests that these species evolved an-other protein or complex to block the anti-COactivity of BLM.

Like Saccharomyces cerevisiae Msh4 andMsh5mutants, the only defects in Drosophila rec, mei-217, and mei-218 mutants are in meiotic recom-bination (4–9). REC is orthologous to MCM8(6); MCMs have properties reminiscent of Msh4-Msh5. MCM2 through MCM7, which are essentialfor replication in eukaryotes, form a heterohexamerthat encircles DNA (10). Similarly, Msh4-Msh5is thought to encircle recombination intermedi-ates (11). In both cases, this activity is regulatedby adenosine triphosphate (ATP) binding and hy-drolysis (10, 11).

MEI-217 initially appeared to be novel, be-cause BLAST searches failed to identify homologsoutside dipterans, and searches of the ConservedDomain Database (CDD) (12) did not detectany domains. BLAST searches with MEI-218identified a single putative ortholog in meta-zoans (figs. S1 and S2). A CDD search returneda hit to the MCM domain in the C terminus ofMEI-218, but the score was low and the matchcovered only one-third of the domain (fig. S3A).To verify the presence of this domain, we con-ducted structure-based searches with PHYRE(Fig. 1A) (13). This analysis revealed that theC terminus of MEI-218 has a structure similarto that of the AAA adenosine triphosphatase(ATPase) domain of MCMs (fig. S3B). Canon-ical MCMs have both an N-terminal MCMdomain and a C-terminal ATPase domain. TheN-terminal domain is present in vertebrate MEI-218 but not in Drosophila MEI-218. However,PHYRE with MEI-217 shows that its predictedstructure is similar to the MCM N-terminal do-main. Because MEI-217 and MEI-218 are en-coded by overlapping open reading frames on thesame transcript (7), we infer that they evolvedfrom an MCM-like protein represented by a sin-gle polypeptide in other metazoans.

The shared phenotypes and MCM domainssuggest that REC, MEI-217, and MEI-218 func-tion together in meiotic recombination. To distin-guish them from the replicative MCMs, we referto REC, MEI-217, and MEI-218 as “mei-MCMs.”Because MCM2 through MCM7 function togeth-er as a heterohexamer, we investigated whetherthe mei-MCMs form a complex. MEI-217 inter-acted strongly with both the C-terminal third ofMEI-218 and REC (Fig. 1, B and C), which sug-gests that the mei-MCMs form a complex. Thiscomplex likely also contains one or more replica-tive MCMs. A meiosis-specific mutation inMcm5causes the same phenotypes as mei-MCM mutants

(14), making MCM5 a strong candidate to be acomponent of the complex.

Noting the genetic and biochemical similar-ities between mei-MCMs and Msh4-Msh5, wehypothesized that the mei-MCMs antagonizeDrosophila melanogaster BLM (DmBLM) inlieu of Msh4-Msh5. This hypothesis predicts thatremoving DmBLM should compensate for mei-MCM mutations; in budding yeast, the CO defectin msh4 mutants is suppressed by removing Sgs1(3). Few COs were made in rec and mei-218 sin-gle mutants, resulting in high nondisjunction (NDJ)of meiotic chromosomes (Fig. 2A). In contrast,mutations in mus309, which encodes DmBLM,

caused only a mild reduction in COs and corre-spondingly low levels of NDJ. Strikingly, mus309mutations suppressed the high-NDJ phenotypeof rec and mei-218 mutants (Fig. 2A). Further-more, the low CO rate in rec mutants returned toan approximately wild-type rate in mus309 recdouble mutants (Fig. 2B, fig. S4, and tables S1to S4); this finding indicates that mei-MCMs arenot essential for generating meiotic COs if DmBLMis absent, thereby supporting our hypothesis thatmei-MCMs oppose the known anti-CO activitiesof DmBLM (15, 16).

mei-MCMs appear to functionally replaceMsh4-Msh5 in Schizophora, and presumably

CB

MCM N-terminal domain

AAA ATPase domain

MCM5 73329 298 338 643

A

REC 885135 406 440 788

MEI-217 27941 278

MEI-218 1186850 1116

HsMEI-218 6815 275 311 618

x

x

IP:+ - + + - +

- + + - + +

-100

-33

-trp-leu

MEI-217

MEI-217

MEI-217

MEI-217

MEI-217

empty

MEI-218

MEI-218

MEI-218

MEI-218

1-1186

1-395

396-791

792-1186

-trp-leu-hisBD AD

FLAG HAFLAG REC

HA MEI-217

Fig. 1. (A) Structural domains identified through PHYRE. “MCMN-terminal domain” corresponds to ProteinData Bank fold ID 3f9v and “AAA ATPase domain” to fold ID 3f8t. The “x” on Drosophila and human (Hs)MEI-218 symbolizes changes in the ATP binding and hydrolysis motifs predicted to abolish ATPase activity(fig. S3B). Red arrows on MEI-218 indicate segments used in yeast two-hybrid analysis. (B) Yeast two-hybridinteractions between MEI-217 and MEI-218. Cells expressing the indicated fusions to the GAL4 DNA bindingdomain (BD) or activating domain (AD) were streaked onto selective media. Growth on –trp –leu –hisindicates an interaction. (C) Coimmunoprecipitation of REC and MEI-217. Epitope-tagged mei-MCMs werecoexpressed in insect cells, immunoprecipitated with antibodies to the indicated epitope tags, blotted, andprobed with antibodies to REC and to the hemagglutinin (HA) tag (19).

WT

rec

mus309

rec mus309

Per

cent

ND

J

A

mei-218

mei-218 mus309

B

Map

Dis

tanc

e (m

ap u

nits

)

0

10

20

30

40

50

0

10

20

30

40

50

Fig. 2. (A) X chromosome nondisjunction (NDJ) across more than 1500 individuals for each genotypeexceptmei-218; mus309 (n= 383). ***P < 0.0001. (B) Summedmap distance in map units (equivalent tocentimorgans) across five intervals spanning ~20% of the genome for more than 1000 individuals foreach genotype (19). P < 0.0001 for all comparisons except wild-type versus mus309 rec, P = 0.0674.

1Curriculum in Genetics and Molecular Biology, University ofNorth Carolina, Chapel Hill, NC 27599, USA. 2Department ofBiology, University of North Carolina, Chapel Hill, NC 27599,USA. 3Carolina Center for Genome Sciences, University of NorthCarolina, Chapel Hill, NC 27599, USA. 4Program in MolecularBiology and Biotechnology, University of North Carolina, ChapelHill, NC 27599, USA.

*To whom correspondence should be addressed. E-mail:sekelsky@unc.edu

7 DECEMBER 2012 VOL 338 SCIENCE www.sciencemag.org1364

REPORTS

evolved to do so in response to natural selection.Several evolutionary scenarios could lead to thisresult (fig. S5), but most predict that there wouldbe evidence of adaptive divergence of mei-MCMgenes in Schizophora. REC was previously notedto be highly diverged in Drosophila (6, 17); wefound that GlossinaMCM8/REC is similarly di-vergent (Fig. 3). The presence or absence ofMCM8correlates with that of its functional partner MCM9throughout eukaryotes, except inDrosophila andGlossina, which retainedMCM8/RECwhile losingMCM9 (figs. S1 and S2). The loss of MCM9 sug-gests that MCM8 evolved a novel function in anancestor to Schizophora.

Divergence in rec and loss of MCM9 occurredafter the split between mosquitos and higherflies 200 to 250 million years ago, but prior tothe emergence of the Schizophora 65 million yearsago. To test whether patterns of sequence evolu-tion were consistent with positive selection lead-ing to the divergence of rec, we estimated theratio between the rate of base pair substitutions atnonsynonymous sites (dN) and the rate at synon-

ymous sites (dS) among dipterans in MCM8/rec.We compared 15 evolutionary models, rangingfrom conservation of dN/dS ratios across all taxasurveyed to allowing free evolution of dN/dS ratiosalong all branches, and including models testingspecific hypotheses about the evolution of recalong different branches of the insect phylogeny.The best-fitting model (P = 0.0002 versus thenext best model) supports the hypothesis that rapidprotein-coding divergence was driven by positiveselection prior to the split of tsetse flies from fruitflies (Fig. 3, fig. S6, and table S5). Thus, we inferthat natural selection likely drove the repurpos-ing of REC into its new role as an antagonist ofDmBLM. Recent evolution of rec shows muchlower levels of nonsynonymous changes, suggest-ing subsequent functional constraint (fig. S6).MEI-217 and MEI-218 have also diverged sub-stantially from the ancestral MCM structure: Theysplit into two polypeptides, and MEI-218 acquiredan N-terminal extension (figs. S7 and S8).

Our data show that flies evolved a novel MCMcomplex to antagonize the anti-CO functions of

BLM during meiosis—a role held by Msh4-Msh5in other organisms. Although we do not knowwhat evolutionary forces ultimately drove the lossof Msh4-Msh5 and the repurposing of mei-MCMs,it is tempting to speculate that these forces alsoled to another fundamental meiotic difference inDrosophila and Glossina relative to mosquitoes:the absence of recombination in males, which wasfirst noted inDrosophila byMorgan 100 years ago(fig. S5) (18). Resolving the conundrum of why themei-MCMs supplanted Msh4-Msh5 will requirea deeper understanding of both the evolutionaryorigins of the mei-MCMs and the functional dif-ferences between mei-MCMs and Msh4-Msh5.

References and Notes1. S. L. Andersen, J. Sekelsky, Bioessays 32, 1058 (2010).2. W. K. Chu, I. D. Hickson, Nat. Rev. Cancer 9, 644

(2009).3. L. Jessop, B. Rockmill, G. S. Roeder, M. Lichten, PLoS

Genet. 2, e155 (2006).4. P. Ross-Macdonald, G. S. Roeder, Cell 79, 1069

(1994).5. N. M. Hollingsworth, L. Ponte, C. Halsey, Genes Dev. 9,

1728 (1995).6. H. L. Blanton et al., PLoS Genet. 1, e40 (2005).7. H. Liu, J. K. Jang, J. Graham, K. Nycz, K. S. McKim,

Genetics 154, 1735 (2000).8. B. S. Baker, A. T. C. Carpenter, Genetics 71, 255 (1972).9. E. A. Manheim, J. K. Jang, D. Dominic, K. S. McKim, Mol.

Biol. Cell 13, 84 (2002).10. D. Remus et al., Cell 139, 719 (2009).11. T. Snowden, S. Acharya, C. Butz, M. Berardini, R. Fishel,

Mol. Cell 15, 437 (2004).12. A. Marchler-Bauer et al., Nucleic Acids Res. 39, D225

(2011).13. L. A. Kelley, M. J. Sternberg, Nat. Protoc. 4, 363

(2009).14. C. M. Lake, K. Teeter, S. L. Page, R. Nielsen, R. S. Hawley,

Genetics 176, 2151 (2007).15. M. D. Adams, M. McVey, J. J. Sekelsky, Science 299,

265 (2003).16. M. McVey, S. L. Andersen, Y. Broze, J. Sekelsky, Genetics

176, 1979 (2007).17. Y. Liu, T. A. Richards, S. J. Aves, BMC Evol. Biol. 9,

60 (2009).18. T. H. Morgan, Science 36, 719 (1912).19. See supplementary materials on Science Online.

Acknowledgments: We thank K. McKim for helpful discussionsand for sharing unpublished results, and N. Crown and othermembers of the Sekelsky laboratory for helpful commentson the manuscript. Supported by NIH grant GM061252 ( J.S.),NSF grant MCB-0618691 ( J.S., C.D.J., and G. C. Copenhaver),and NIH grant 5T32GM007092 (K.P.K.). Alignments andtrees were submitted to TreeBASE (http://purl.org/phylo/treebase/phylows/study/TB2:S13435).

Supplementary Materialswww.sciencemag.org/cgi/content/full/338/6112/1363/DC1Materials and MethodsFigs. S1 to S8Tables S1 to S7References (20–43)

31 July 2012; accepted 5 October 201210.1126/science.1228190

Fig. 3. Maximum likelihood tree from an alignment of the conserved MCM domains of MCM8/REC andMCM5 from diverse taxa. Branch lengths indicate the number of substitutions per site (see scale).Numbers above branches show dN/dS estimates for selected branches; those with black backgroundhighlight branches with dN/dS estimates greater than 1, suggesting positive selection. See fig. S6 formore description and additional dN/dS estimates.

www.sciencemag.org SCIENCE VOL 338 7 DECEMBER 2012 1365

REPORTS

www.sciencemag.org/cgi/content/full/338/6112/1363/DC1

Supplementary Materials for

Evolution of an MCM Complex in Flies That Promotes Meiotic Crossovers by

Blocking BLM Helicase

Kathryn P. Kohl, Corbin D. Jones, Jeff Sekelsky*

*To whom correspondence should be addressed. E-mail: sekelsky@unc.edu

Published 7 December 2012, Science 338, 1363 (2012)

DOI: 10.1126/science.1228190

This PDF file includes: Materials and Methods

Figs. S1 to S8

Tables S1 to S7

References

Materials and Methods Drosophila Stocks and Genetics

Flies were maintained on standard medium at 25˚C. All experimental flies were heteroallelic for null mutations that have been described previously: rec1 and rec2 (6); mus309N1 and mus309D2 (16, 20); mei-2181 and mei-2186 (8, 21). mus309 mutant females produce few viable progeny due to a requirement for DmBLM in the early embryo (16). To overcome this maternal-effect lethality, the mus309 coding sequence was cloned into the P{UASp} (22) vector, creating P{UASp-mus309}. This transgene was injected using standard P-element transformation procedures (Best Gene Inc., Chino Hills, CA). mus309N1 P{UASp-mus309} / mus309D2 P{matα4-GAL4::VP16} females express DmBLM during later stages of oocyte development, after meiotic recombination has taken place, thereby creating a mus309 meiotic null. This genotype was used in all assays with mus309 mutants. Nondisjunction was scored by crossing mutant females to y cv v f / T(1:Y)BS males. The number of exceptional progeny indicative of nondisjunction (Bar-eyed females and wild-type-eyed males) was multiplied by two to account for triplo-X and nullo-X progeny, which do not survive to adulthood. This number was divided by the total number of progeny and expressed as a percentage. Crossovers were scored by crossing net dppd-ho dp b pr cn /+ virgin females of various genetic backgrounds to net dppd-ho dp b pr cn males. Identification of orthologs

Orthologs were found through searches of public databases. The amino acid sequence of the human protein was used in BLASTP searches of refseq_protein at NCBI (http://blast.ncbi.nlm.nih.gov/Blast.cgi), using the default parameters. In cases where orthologs were not immediately identified, searches were repeated using sequence from a more closely related species and searching the nr database. If this was unsuccessful, TBLASTN was done using the nr database. Finally, species-specific databases were searched, either from the NCBI genomic BLAST page (http://www.ncbi.nlm.nih.gov/mapview/) or from species-specific websites. These are described in table S6. Additionally, a sequence for rec from D. pseudoobscura was obtained from (23).

Drosophila melanogaster mei-217 and mei-218 are well-annotated based on

experimental data (cDNA sequences and RNA-seq data), but annotations for other species have less support. To generate better predictions of the amino acid sequences of MEI-217 and MEI-218 in other species, we used D. melanogaster sequences in TBLASTN searches. Apparent conservation of MEI-217 allowed high-confidence CDS (coding sequence) predictions, but divergence of MEI-218 made predictions for this gene less reliable. The CDSs we used for our analyses are given below; accession numbers are given in cases in which the annotated CDS matched our CDS prediction.

Drosophila sechellia scaffold_17; GB:CH480832; release r1.3

MEI-217 complement(join(876950..877066, 877198..877355, 877556..877623, 878148..878588, 878706..878767))

MEI-218 XP_002042345.1

2

Drosophila yakuba loc=X; GB:CM000162; release r1.3

MEI-217 complement(join(15645064..15645176, 15645241..15645396, 15645605..15645672, 15646258..15646698, 15646817..15646878))

MEI-218 complement(join(15640672..15640991, 15641104..15641336, 15641391..15642781, 15642846..15643070, 15643269..15643696, 15643745..15644913, 15644997..15645085))

Drosophila erecta scaffold_4690; GB:CH954180; release r1.3

MEI-217 complement(join(7382254..7382366, 7382416..7382571, 7382726..7382793, 7383378..7383818, 7383937..7383998))

MEI-218 complement(join(7378204..7378523, 7378669..7378901, 7378958..7380333, 7380399..7380580, 7380620..7381042, 7381136..7382076, 7382186..7382275))

Drosophila ananassae scaffold_13248; dbxref=GB:CH902625; release r1.3

MEI-217 complement(join(2273229..2273341, 2273399..2273554, 2273667..2273734, 2273924..2274364, 2275326..2275405)

MEI-218 complement(join(2267716..2268035, 2268107..2268339, 2268406..2270384, 2270451..2273034, 2273087.. 2273250))

Drosophila persimilis ID=scaffold_17; dbxref=GB:CH479196; ; release r1.3

MEI-217 XP_002023388.1 MEI-218 join(662538..662656, 662721..663387, 663449..

665127, 665184..665416, 665507..665823)

D. pseudoobscura pseudoobscura XL_group1e, NW_001589960, release r2.27 MEI-217 join(3724816..3724910, 3725029..3725469, 3725576..3725643,

3725735..3725890, 3725956..3726065) MEI-218 XP_001354736.2

Drosophila willistoni scf2_1100000004515; GB:CH963851; release r1.3

MEI-217 join(3221627..3221727, 3221861..3222301, 3222689..3222756, 3223353..3223508, 3223568..3223680)

MEI-218 join(3223659..3223680, 3223738..3223945, 3224093..3225230, 3225291..3227119, 3227190..3227425, 3227504..3227817)

Drosophila virilis scaffold_12970; GB:CH940651; release r1.2

MEI-217 complement(join(9766858..9766970, 9767033..9767188, 9767480..9767547, 9768179..9768631, 9768706..9768803))

MEI-218 complement(join(9763751..9764061, 9764137..9764369, 9764431..9765578, 9765638..9766685, 9766857..9766879))

3

Drosophila mojavensis scaffold_6473; GB:CH933810; release r1.3

MEI-217 join(11800198..11800664, 11800793..11800860, 11801099..11801254, 11801332..11801444)

MEI-218 join(11801423..11801448, 11801612..11802692, 11802753..11803930, 11803986..11804218, 11804300..11804613)

Drosophila grimshawi scaffold_14853; dbxref=GB:CH916371; release r1.3

MEI-217 join(3073042..3073139, 3073185..3073637, 3073800..3073867, 3074121..3074273, 3074343..3074455)

MEI-218 join(3074434..3074516, 3074621..3075635, 3075776..3076833, 3076899..3077131, 3077213..3077496)

Sequence alignments and phylogenetic analysis

Protein sequences were aligned in MEGA5 (24) using the MUSCLE algorithm (25). Maximum-likelihood trees were generated in MEGA5 using the method based on the Whelan and Goldman model (26). Initial trees for the heuristic search were obtained automatically. When the number of common sites was <100 or less than one fourth of the total number of sites, the maximum parsimony method was used; otherwise BIONJ method with MCL distance matrix was used. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+G, parameter = 2.8099)). The rate variation model allowed for some sites to be evolutionarily invariable ([+I], 0.0000% sites).

Molecular evolutionary analysis

We compared a suite of molecular evolutionary models of the evolution of rec using PAML, which provides a set of maximum likelihood-based tools for combining DNA sequence and phylogenetic data to test molecular evolutionary hypotheses (27). We used a phylogeny based on DNA polymerase alpha catalytic subunit and published trees of insects to identify good outgroups (28). We limited our analysis of rec to the region 3’ of position 135 (figure 1A) as this section was clearly homologous across taxa. There are three major steps to using PAML: (i) choice of appropriate model, (ii) parameterization of that model, and (iii) sequential comparison using log-likelihood ratio tests of simpler to more complex models to evaluate whether a more complex model provides a significantly better fit to the data. In several cases, convergence to maximum likelihood estimates was verified by changing the “small difference” parameter see (28), p. 19. Evolution of protein-coding regions of rec was analyzed independently using the codon model CODEML (29, 30). The difference in the log likelihood (Δlnl), for the relevant degrees of freedom, was used to infer a P-value. Unless noted otherwise, model comparisons involving multiple tests remained significant after Bonferroni corrections. The estimated codon table fit the rec data the best. When appropriate, κ, ω, and α fit the data had the best log-likelihood when estimated (29). For the analyses discussed, we a priori hypothesized that the lineage prior to the split of Glossina from Drosophila rec underwent more rapid protein coding evolution because rec was under strong selective pressure to functionally compensate for the loss of the Msh4–Msh5. As shown in table S5, we tested a variety of simple and complex models motivated by our understanding of

4

the evolutionary history of flies. Model 1 is the most conservative as it assumes a single evolutionary rate for all branches. Models 2 and 3 are controls that test the hypothesis that rec is “special” in D. melanogaster (Model 2) and that the patterns observed are specific to higher flies, not just any clade of the tree (Model 3). As expected neither of these models is different from Model 1. The remaining models grow increasingly complex as they step through a variety of evolutionary scenarios (e.g., Drosophila differ from others; Sophophora, which have reported evidence of male recombination in D. willistoni and D. ananassae, differ; Old World Sophophora differ, etc.). All fit the data better than Models 1-3 and generally improve the fit as their complexity grows. However, Model 15, which allows for dN/dS ratio to vary freely among all branches, has the best fit to the data and is a significantly better than any alternative. This model suggests strong adaptive evolution prior to the divergence of the Schizophora and prior to the radiation of the Old World Sophophora. Simpler models (e.g., 4-14) similarly support that the dN/dS rate of rec in higher flies is different compared to the other Dipteran and our outgroups.

Immunoprecipitation

rec was cloned into pFastBac with an N-terminal FLAG tag. mei-217 and mei-218 were cloned into pFastBacDual with N-terminal HA and Strep-II tags, respectively. As per the Bac-to-Bac Baculovirus System (Invitrogen, Carlsbad, CA) protocol, constructs were transformed into DH10Bac cells. Sf9 cells were transfected with bacmid DNA extracted from transformed DH10Bac cells. Following two rounds of viral amplification in Sf9 cells, High Five cells were infected with either a single virus or with both viruses. Cells were harvested 2.5 days after infection and were stored at -80˚C until needed. Pellets were resuspended in lysis buffer (50mM Tris-HCl pH 7.5, 150mM NaCl, 1mM EDTA, 1% Triton X-100), sonicated using a Bioruptor (Diagenode, Denville, NJ) on highest setting with 30s on/30s off cycles for 20min at 4˚C. Cell suspension was pelleted with centrifugation at 14k rpm for 5min. Clarified supernatant containing FLAGREC and HAMEI-217 was added to either anti-HA agarose (Sigma, St. Louis, MO) or anti-FLAG M2 agarose (Sigma) for 2h. Strep-MEI-218 was co-expressed with these two proteins but was insoluble and therefore not in the clarified lysate from which immunoprecipitation was carried out. Beads were washed in a cellulose acetate filter spin column (Pierce, Rockford, IL) with either TBS (FLAG purification) or PBS (HA purification) prior to boiling in SDS-PAGE sample buffer. Samples were run on a 4-15% SDS-PAGE gel, transferred to PVDF membrane, and probed with appropriate antibodies: anti-HA (Sigma) 1:20,000; anti-REC-C (raised to amino acids 875-885; Pacific Immunology, San Diego, CA) 1:20,000; HRP-conjugated anti-rabbit secondary (Santa Cruz Biotechnology, Santa Cruz, CA) 1:20,000. SuperSignal West Pico Chemiluminescent Substrate (Pierce) was used to detect proteins.

Yeast two-hybrid assay

mei-217 was cloned into pGBD-DEST, a Gateway-compatible derivative of pGBD-C1 (30) constructed with the Gateway Vector Conversion System (Life Technologies, Carlsbad, CA). Full-length or truncated mei-218 was cloned into pACT2.2gtwy (Addgene plasmid 11346 deposited by Guy Caldwell), a Gateway-compatible derivative of pACT2.2. Constructs were transformed into Saccharomyces cerevisiae strain PJ69-4A (30). Co-transformants were selected on plates of SD minimal medium containing

5

dropout supplements lacking leucine (-leu) and tryptophan (-trp) for 3 days at 30˚C. Single colonies were streaked onto fresh –trp –leu plates and grown for 3 days at 30˚. Colonies were then streaked onto –trp –leu –histidine plates containing 3mM 3AT. Interaction between proteins was scored 3 days later.

6

Fig. S1. Drosophila and Glossina uniquely lack MSH4, MSH5, and MCM9. The presence (filled black circles) or absence (open red circles) of MSH4, MSH5, MCM8, MCM9, and MEI-218 orthologs within representative genera spanning the metazoa is indicated. The topology of phylogenetic relationships was created from an alignment of DNA polymerase alpha catalytic subunit (this protein was not present in the Saccoglossus kovalevski sequences, so this species was inserted into its consensus location). Accession numbers for all database entries used to generate this figure and figure S2 are given in table S6. Protein absence was assumed when an ortholog could not be identified following multiple rounds of BLAST searches against several databases (see Materials and Methods).

7

Fig. S2 Drosophila and Glossina uniquely lack MSH4, MSH5, and MCM9. This figure is similar to fig. S1, but includes only arthropods. See fig. S1 for details.

8

Fig. S3 (A) Domains identified by a search of the Conserved Domain Database (CDD v3.05 – 42589 PSSMs) in Drosophila melanogaster REC, MEI-217, MEI-218, and human MEI-218. In database descriptions COG1241 is “MCM2”, while smart00350 and pfam00493 are both listed as “MCM”. To the right of each domain is the E value, percent identity/percent similarity between the domain and input sequences, and the percentage of the domain definition that the alignment spans. For REC only the top three domain hits are shown. (B) Alignment of the Walker A and B motifs in human and Drosophila melanogaster MCM5, MCM8/REC, and MEI-218. These motifs are involved in ATP binding and hydrolysis. Identical or conserved residues have a black background. Numbers in parentheses denote number of amino acids between motifs. The consensus sequences are given below the alignment. The changes in both Drosophila and human MEI-218 suggest that this protein does not bind or hydrolyze ATP.

9

Fig. S4 Crossovers in rec and mus309 mutants. Crossover distribution across five intervals on chromosome 2L and proximal 2R is shown. The markers used in mapping are shown above the graph. Hash marks between pr and cn indicate the position of the centromere. Solid lines depict the number of map units (m.u.) per megabase pair (Mb) in each interval for wild-type, rec, mus309, and mus309 rec for more than 1000 individuals for each genotype. Dashed lines indicate the mean CO rate across the entire region. Determining the cause of the increased COs in mus309 rec compared to mus309 is difficult, since we cannot distinguish between true meiotic COs and pre-meiotic mitotic COs occurring in the mus309 mutant background (2). The occurrence of pre-meiotic mitotic crossovers also complicates efforts to determine whether crossover interference is altered in mus309 genotypes (16); however, there are significantly more double crossover (DCO) and triple crossover (TCO) progeny in mus309 rec compared to mus309 (tables S1-S4) (P < 0.0001 for DCO, P = 0.046 for TCO, Fisher’s exact test) and to wild-type (P < 0.0001 for DCO, P = 0.0002 for TCO, Fisher’s exact test), suggesting that interference is disrupted in rec mutants even in the absence of DmBLM. Recent studies suggest that Sgs1 has, in addi-tion to its anti-CO functions, a pro-CO role in meiosis, perhaps directing recombination into the Msh4–Msh5 pathway (32, 33). If true in Drosophila, the COs in mus309 rec mutants may arise from a mei-MCM-independent CO pathway rather than the lack of need to block DmBLM anti-CO activities. However, this scenario cannot explain the finding that REC regulates CO distribution even in the absence of DmBLM.

10

Fig. S5 Possible evolutionary scenarios for the loss of Msh4–Msh5 in Drosophila and the evolu-tion of mei-MCM complex. Across metazoans Msh4–Msh5 antagonize the anti-CO activity of BLM to promote meiotic COs. Given the presumed evolutionary stability of this interaction over hundreds of millions of years, loss of Msh4–Msh5 and its replace-ment by mei-MCMs in the higher Dipterans is surprising. A range of evolutionary scenarios could explain this pattern. The four scenarios presented here depict archetypes of these possibilities (within a panel, top to bottom represents past to present). (A) Sudden loss of Msh4–Msh5 (green), followed by recruitment and evolution of mei-MCMs (red). This scenario would provide the strong signature of selection observed at REC (shift from blue, to purple, to red), as it would be needed to restore antagonism of BLM and promote the formation of crossovers. However, this scenario seems unlikely, as a sudden loss of Msh4–Msh5 would lead to meiotic chromosome instability and could likely occur only in small populations with substantial genetic drift. (B) Gradual recruit-ment and replacement of Msh4–Msh5 by mei-MCMs. MCM8 and MCM9 have a role in meiotic recombination and other DNA repair processes (34, 35). This may have been as a nascent BLM antagonist with a minor recombination role that gradually expanded and eventually supplanted that of Msh4–Msh5. This scenario also seems unlikely as it does not necessarily result in a strong signature of selection. The strong signature of selection we observed suggests that extensive evolution was needed in REC to replace Msh4–Msh5, but this also seems implausible given the evolutionary stability of Msh4–Msh5 in all other metazoans. (C) Selection against Msh4–Msh5 results in the repurposing and

11

remodeling of REC from an ancestral MCM. Either an endogenous force (e.g., transposon, driving chromosome, or other selfish element) or exogenous force (e.g., viruses integrating at breaks, highly consistent environment) selected for reduced crossover activity during meiosis. This could result in a gradual diminishing of Msh4–Msh5 activity. Dollo’s law of irreversibility suggests that once a trait is lost, it never re-evolves exactly as before (36, 37). It may have been evolutionarily simpler to evolve the mei-MCMs than to restore Msh4–Msh5 activity (and potentially face the same evolutionary pressure that drove this activity down). (D) Positive selection for increased CO activity or antagonism of BLM drives evolution of mei-MCMs. CO rates are genetically labile and increased recombination rate is thought to facilitate adaptation (38). Schizophora represents a recent rapid radiation of lineages that may have diversified along with flowering plants in an anciently tropical world (28, 39). Selection for increased crossover rate may have resulted in the repurposing of REC and its elevated rate of evolution. Once these new niches were filled, the need for elevated crossing over may have reduced. As Blanton et al. (6) suggested, REC may promote repair synthesis to generate more stable intermediates that are refractory to unwinding by DmBLM. Thus, in this reduced CO scenario selection would act to preserve REC at the cost of Msh4–Msh5.

It is likely that the actual events encompassed elements of one or more of these scenarios. In three of these scenarios, the strong pressure to find a means to segregate chromosomes accurately may have driven the development of recombination-independent segregation in male Drosophila and Glossina (18, 40). Consistent with this speculative hypothesis, male recombination is retained in mosquitoes (41-43), which also have conserved Msh4–Msh5 (Fig. S1 and S2).

12

Fig. S6 Rates of evolution of REC/MCM8 in Dipteran insects and select outgroups. The tree represents a topology of phylogenetic relationships among the insect species shown. Branch lengths are proportional for ease of reading and do not reflect divergence, which is shown in Fig. 3. dN/dS ratios along each branch for the best fitting model (Model 15) are shown. Values less than one suggest purifying selection, whereas values greater than one suggest positive selection along that branch. Consistent with our prediction of rapid evolution of REC before the emergence of Schizophora, the branch leading to Glossina and Drosophila has a high positive value (red). The biological significance of the other high positive value, on the branch leading to Old World Sophophora (blue), is unknown, since our data do not suggest any hypotheses concerning evolution of REC along this branch.

13

Fig. S7 Conservation of MEI-217 and MEI-218 as separate polypeptides throughout Drosophila. D. melanogaster MEI-217 and MEI-218 are predicted to be separate polypeptides encoded on a single transcript (7). This figure shows an alignment of the sequence coding for the C-terminus of MEI-217 (purple) and the N-terminus of MEI-218 (orange) based on TBLASTN searches with D. melanogaster MEI-217. Positions that are identical (or similar, for amino acid residues) in at least nine of the twelve species shown are in white text on a dark background. This pattern is also seen in D. similans, D. elegans, D. rhopaloa, D. ficusphila, D. takahashii, D. biarmipes, and D. bipectinata. Apparent conservation at the C-terminus of MEI-217 suggests that the coding sequences are accurately represented here. Notably, the ATG predicted to begin the MEI-218 coding sequence is absolutely conserved, even though the A could have mutated to any other base without affecting the MEI-217 sequence. These findings strongly support the conclusion that MEI-217 and MEI-218 are separate polypeptides and that there has been selection to maintain this configuration. We could not detect conservation in this region in Glossina morsitans, so that sequence cannot be aligned to these sequences.

14

Fig. S8

Structural changes in MEI-218. These schematics illustrate the structures of MEI-218 orthologs from humans and several insects. Purple cylinder represents the N-terminal MCM domain, orange the C-terminal AAA ATPase-like domain. The human and Bombyx mori orthologs are similar in structure (as are other vertebrates and arthropods other than Drosophila and Glossina) to canonical MCMs, whereas Drosophila and Glossina are substantially longer. In Drosophila the two domains are found on separate polypeptides.

15

Table S1. Meiotic crossovers from wild-type females. Each row lists the number of progeny from one vial for non-crossover (NCO), single crossover (SCO), double crossover (DCO), and triple crossover (TCO) classes. The bottom row lists the total number of progeny in each class, summed over all vials. For NCO and SCO, the + symbol indicates wild-type for a marker, while the gene name indicates mutant for a marker, in the order along the chromosome (net ho dp b pr cn). Intervals for DCOs and TCOs are given in the columns on the right, with interval I being net to ho, etc.

Table S1 - WT

CO.xlsx

16

Table S3. Crossovers from wild-type females.

Vial ne

t h

o d

p b

pr

cn

+

+

+

+

+

+

+

ho

dp

b p

r cn

ne

t

+

+

+

+

+

+

+

dp

b p

r cn

ne

t h

o

+

+

+

+

+

+

+ b

pr

cn

ne

t h

o

dp

+

+

+

+

+

+

+ p

r cn

ne

t h

o

dp

b +

+

+

+

+

+

+ c

n

ne

t h

o

dp

b p

r +

I

an

d I

I

I

an

d I

II

I

an

d I

V

I

an

d V

I

I a

nd

III

I

I a

nd

IV

I

I a

nd

V

I

II a

nd

IV

I

II a

nd

V

I

V a

nd

V

1 0 0 62

2 3 0 78 1 1 1

3 2 1 74 1 1 III, IV, V

4 0 0 93

5 1 0 88 1

6 3 0 89 1 1 1

7 4 0 70 1 1 1 1

8 2 0 91 1 1

9 3 0 74 1 1 1

10 2 0 99 1 1

11 0 0 105

12 1 0 79 1

13 3 0 96 2 1

14 2 0 75 1 1

15 6 0 99 2 1 1 1 1

16 0 0 88

17 4 0 93 1 2 1

18 0 0 105

19 1 0 80 1

20 0 0 76

21 1 0 96 1

22 1 0 81 1

23 4 0 95 1 1 1 1

24 1 0 94 1

25 0 0 99

26 0 0 47

27 0 0 94

Total 44 1 2320 0 7 3 2 5 2 5 9 10 1

non-CO

(NCO)

Tri

ple

Cro

sso

ve

rs (

TC

Os)

Double Crossovers (DCOs)Single Crossovers (SCOs)

27

24

4

1

3

I II III

2

3

DC

O f

lies

TC

O f

lies

To

tal

flie

s

IV V

5 5 20 5 0

5 8

2112

3 04 20

25

6 13 3 1

6 18 3 1

4 20 3 0

01219

2 31 1 0

33

37

44

36

50

49

4 0

012162

51

36

48

6 4 21

27 3 0

10

55

39

7

1

49

52

43

53

48

6

57

60

6

2

5

2

1

0

10 23 3

4

10

48

6

2

5

6

2

2

42

42

56

52

62

2

3

61

2

5

7

20

26

32

0

4

4

2

38

8

1

7

17

24

20

17

10

6

27

58 24

3

2

1

16

5

1

6

5

4

5

4

26

25

12

1323 106 163 602 65 16

0

0

2

1

0

1

0

1

2

0

1

0

1

0

1

55

60

28

Table S2. Meiotic crossovers from rec mutant females. Each row lists the number of progeny from one vial for non-crossover (NCO), single crossover (SCO), double crossover (DCO), and triple crossover (TCO) classes. The bottom row lists the total number of progeny in each class, summed over all vials. For NCO and SCO, the + symbol indicates wild-type for a marker, while the gene name indicates mutant for a marker, in the order along the chromosome (net ho dp b pr cn). Intervals for DCOs and TCOs are given in the columns on the right, with interval I being net to ho, etc.

Table S2- rec CO.xlsx

17

Table S2. Crossovers from rec mutant females.

Vial net

ho d

p b

pr

cn

+

+

+

+

+

+

+

ho d

p b

pr

cn

net

+

+

+

+

+

+

+

dp b

pr

cn

net

ho

+

+

+

+

+

+

+ b

pr

cn

net

ho

dp +

+

+

+

+

+

+ p

r cn

net

ho

dp b

+

+

+

+

+

+

+ c

n

net

ho

dp b

p

r +

I

and

II

I

and

III

I

and

IV

I

and

V

I

I a

nd

III

I

I a

nd

IV

I

I a

nd

V

I

II a

nd I

V

I

II a

nd V

I

V a

nd

V

1 0 0 63

2 0 0 68

3 0 0 76

4 0 0 62

5 0 0 75

6 0 0 68

7 0 0 63

8 0 0 45

9 0 0 46

10 0 0 53

11 0 0 56

12 0 0 67

13 0 0 66

14 0 0 43

15 0 0 90

16 0 0 93

17 0 0 74

18 0 0 89

19 0 0 60

20 0 0 60

21 0 0 69

22 0 0 71

23 0 0 85

24 0 0 62

25 0 0 49

26 0 0 79

27 0 0 89

28 0 0 89

29 0 0 49

30 0 0 77

Total 0 0 2036 0 0 0 0 0 0 0 0 0 0

44

2

non-CO

(NCO)

Trip

le C

rossove

rs (

TC

Os)

Double Crossovers (DCOs)

2

2

0

0

0

I II III

0

0

DC

O f

lies

TC

O f

lies

To

tal

flie

s

IV V

0 0 1 1 2

Single Crossovers (SCOs)

21 0

62

1

1 0 0 2

1 1 0 0

0 0 0 0

1020

0

0

41 1 1

0

1 0 2

10

00100

65

62

0 0

84

90

60

66

72

58

73

0

1 0

0 2 049

53

1

45

43

0

0

59

1 3 0

0

1

0

2

1 6

1

2010 0

0 10

0

1

0

0

0

5

1

1

83

73

0

0 0

0

1

1

1

3

0 0

0

2

2

2

2

3

0

53

59

67

65

77

0

0

000

1

0

0

1

3

0

0

1923 10 14 51 13 25

0

1

0

0

1

1

1

0 1

0

0

1

2 1

1

0 0

1

1

0

2

0

0

2

0

57

47

76

71

84

85

2

1

3

2

1

0

1

2

0

2

1

Table S3. Meiotic crossovers from mus309 mutant females. Each row lists the number of progeny from one vial for non-crossover (NCO), single crossover (SCO), double crossover (DCO), and triple crossover (TCO) classes. The bottom row lists the total number of progeny in each class, summed over all vials. For NCO and SCO, the + symbol indicates wild-type for a marker, while the gene name indicates mutant for a marker, in the order along the chromosome (net ho dp b pr cn). Intervals for DCOs and TCOs are given in the columns on the right, with interval I being net to ho, etc.

Table S3- mus309 CO.xlsx

18

Table S3. Crossovers in mus309 mutant females.

Vial net

ho d

p b

pr

cn

+

+

+

+

+

+

+

ho d

p b

pr

cn

net

+

+

+

+

+

+

+

dp b

pr

cn

net

ho

+

+

+

+

+

+

+ b

pr

cn

net

ho

dp +

+

+

+

+

+

+ p

r cn

net

ho

dp b

+

+

+

+

+

+

+ c

n

net

ho

dp b

pr

+

I

and I

I

I

and I

II

I

and I

V

I

and V

I

I and I

II

I

I and I

V

I

I and V

I

II a

nd I

V

I

II a

nd V

I

V a

nd V

1 12

2 3 46 1 1 1

3 1 51 1

4 64

5 1 42 1

6 1 54 1

7 2 45 1 1

8 1 61 1

9 34

10 1 51 1

11 55

12 56

13 1 79 1

14 2 57 1 1

15 34

16 54

17 1 50 1

18 3 43 1 1 1

19 7 66 6 1

20 4 57 1 1 1 1

21 3 62 1 1 1

22 1 57 1

23 1 46 1

Total 33 0 1176 1 4 2 1 1 2 2 11 7 2

non-CO

(NCO)

Triple

Cro

ssovers

(T

CO

s)

Double Crossovers (DCOs)

1 2

8

29

38

44

28

2

34

29

2

0

1

0 1 1

5

7

3

3

4

I II III

0

2

DC

O f

lies

TC

O f

lies

To

tal

flie

s

IV V

2 0 5 4 1

3

2

2 3 3 4

7 0

4422

0

1 11

10 3 4

08

4 8 1

1412

45

25

1

1

45

51

47

22

42

1 1 6 8

1450 0

34

45

0

2

6 3

9 7

4

12

3

4

3

0

4

1

2

1

0 2

1 6

1

0

2

3

4

2

1

4

4

5

38

0

0

1

2

42

44

50

45

30

1

0

29

0 1

Single Crossovers (SCOs)

2

1

0

5

3

2

2

844 23 22 89 104 61

5

5

4

1

4

40

0

0

1

Table S4. Meiotic crossovers from mus309 rec double mutant females. Each row lists the number of progeny from one vial for non-crossover (NCO), single crossover (SCO), double crossover (DCO), and triple crossover (TCO) classes. The bottom row lists the total number of progeny in each class, summed over all vials. For NCO and SCO, the + symbol indicates wild-type for a marker, while the gene name indicates mutant for a marker, in the order along the chromosome (net ho dp b pr cn). Intervals for DCOs and TCOs are given in the columns on the right, with interval I being net to ho, etc.

Table S4- mus309 rec CO.xlsx

19

Table S4. Crossovers in mus309 rec double mutant females.

Vial net ho d

p b

pr

cn

+ +

+

+

+

+

+ h

o d

p b

pr

cn

net +

+

+

+

+

+ +

dp b

pr

cn

net ho +

+

+

+

+ +

+

b p

r cn

net ho dp +

+

+

+ +

+

+

pr

cn

net ho dp b

+

+

+ +

+

+

+

cn

net ho dp b

pr

+

I a

nd II

I a

nd III

I a

nd IV

I a

nd V

II and III

II and IV

II and V

III a

nd IV

III a

nd V

IV

and V

1 6 1 80 1 1 2 1 1

2 7 93 1 1 3 2

3 13

4 2 18 1 1

5 5 1 73 1 1 1 2

6 1 38 1

7 7 73 1 1 1 1 1 1 1

8 6 1 68 1 4 1

9 4 67 1 1 2

10 3 2 48 1 1 1 I,II,IV

11 19

12 2 43 1 1

13 1 1 28 1

14 1 30 1

15 3 1 34 1 1 1

16 24

17 1 1 45 1

18 6 43 1 3 2

19 3 65 2 1

20 1 34 1

21 1 10 1

22 2 36 1 1

23 3 23 1 2

24 2 40 1 1

25 5 1 51 2 2 1

26 3 29 1 1 1

27 4 33 2 1 1

28 1 1 23 1

80 10 1181 0 11 5 2 5 7 2 19 21 8

I,III,IV

Single Crossovers (SCOs)

I,III,IV

5

3

I II III

4

0

DC

O flie

s

TC

O flie

s

To

tal fl

ies

IV V

I,II,IV

705

0

non-COs

Triple

Cro

ssovers

(T

CO

s)

Double Crossovers (DCOs)

I,III,V

III,IV,V &

I,III,V

118 2

1

1 6 1

5

2

1

I,II,IV

I,III,V

2 1 0 3 1

0 0

161

5 4

5 2

4

2

III,IV,V

6

1 9 5 1

2 15 10 9

2 5 4 7

6482

3 7 3 6

3

53

50

6

13

50

4 6

751002

16

42

33

0 2 9

2 0 1

32

43

39

0

4

30

15

20

18

16

0

23

8

1

0

2

4

1

1

3

0

2

0

2

1

0 2

0 2

2

0

5

3

0

1

31

0

0

1

0

54

20

5

20

12

0

2

23

1

31 29 136 103 87

1

2

4

3

1

0

2

2

3

1

2

0

5

5

2

5

5

2

1

25

27

16

12 4

4

1

3

5

1

1

0

35

5

4

2

5

20

0

0

7

2

Table S5. Models for the evolution of MCM8/rec in Diptera. Please see Materials and Methods for additional information.

Model Type1 Para-meters

Log-Likelihood Description

1 0 33 -29353.66955 One dN/dS rate for all taxa

2 2 34 -29353.13881 One rate for D. melanogaster; one rate for all other taxa

3 2 34 -29352.42575 One rate for mosquitoes; one rate for all other taxa

4 2 34 -29348.27369 One rate for Old World Sophophora; one rate for all other taxa

5 2 34 -29340.13854 One rate for Sophophora; one rate for all other taxa

6 2 34 -29322.82126 One rate for higher flies one rate for all other taxa

7 2 34 -29313.19077 One rate for Drosophila; one rate for all other taxa

8 2 35 -29341.74626 One rate for Drosophila; one rate for Glossina; one rate for all other taxa

9 2 35 -29313.18287 One rate for ancestor to higher flies; one rate for Drosophila; one rate for all other taxa

10 2 35 -29313.15944 One rate for ancestor to higher flies; one rate for Sophophora; one rate for all other taxa

11 2 36 -29313.15172 One rate for ancestor to higher flies; one rate for Drosophila; one rate for Glossina; one rate for all other taxa

12 2 36 -29310.98226 One rate for ancestor to higher flies; one rate for Drosophila; one rate for Sophophora; one rate for all other taxa

13 2 37 -29310.95229 One rate for ancestor to higher flies; one rate for Drosophila; one rate for Sophophora; one rate for Glossina; one rate for all other taxa

14 2 51 -29300.10477 dN/dS rate varies for all higher flies

15 1 63 -29280.42216 dN/dS rate varies for all taxa

1 Model-type: 0, common rate all lineages; 1, lineage-specific rates; 2, two or more rates assigned to different lineages (detailed in description).

20

Table S6.

Accession numbers for proteins discussed in this report. Accession numbers in blue are

protein entries in GenBank (NP and XP) or species-specific databases and are

hyperlinked to the relevant record. Numbers in green are from mRNA (cDNA sequence)

databases. Text in orange is from species-specific database searches (see Table S7 for

details); “not found” means no hits found in any searches. In some cases this is probably

due to gaps in the current sequence coverage.

Genus and species Common Name / Group Msh4 Msh5 MCM8 MCM9 MCM5 MEI-217 MEI-218 DNA polα

Homo sapiens human NP_002431.2 NP_002432.1 NP_115874.3 NP_060166.2 NP_006730.2 NP_058633.2

Monodelphis domestica opossum XP_001368667.2 XP_003340425.1 XP_001382134.1 XP_001379759.1 XP_001366505.1 XP_001365174.2

Xenopus tropicalis western clawed frog XP_002936134.1 XP_002939511.1 NP_001072344.1 XP_002932153.1 NP_001017327.2 XP_002933658.1

Danio rerio zebrafish XP_688406.1 ENSDARP00000097595 XP_002665161.2 XP_683173.1* NP_848523.2 XP_001922054.2

Branchiostoma floridae lancelet XP_002597456.1 XP_002601387.1 XP_002610069.1 XP_002603732.1 XP_002596449.1 XP_002593091.1

Ciona intestinalis sea squirt XP_002122852.1* XP_002119672.1 XP_002129364.1 XP_002131219.1 XP_002128725.1 XP_002123526.1*

Strongylocentrotus purpuratus sea urchin XP_780878.2 XP_791267.1* XP_797782.2 XP_785765.2 XP_801948.1 XP_003728068.1

Saccoglossus kowalevskii acorn worm scaffold30423 XP_002730576.1 XP_002739234.1 XP_002732221.1 XP_002739225.1 not found

Drosophila melanogaster fruit fly not found not found NP_732072.1 not found NP_524308.2 NP_001027069.3 AAB61476.1 NP_536736.2

Drosophila sechellia fruit fly not found not found XP_002031013.1 not found GM23897-PA † XP_002042345.1 XP_002038769.1

Drosophila erecta fruit fly not found not found XP_001980104.1 not found XP_001980641.1 † † XP_001979261.1

Drosophila yakuba fruit fly not found not found XP_002097692.1 not found XP_002097129.1 † † XP_002096581.1

Drosophila ananassae fruit fly not found not found XP_001954977.1 not found XP_001953721.1 † † XP_001953530.1

Drosophila persimilis fruit fly not found not found XP_002017197.1 not found XP_002016737.1 XP_002023388.1 † XP_002017478.1

Drosophila pseudoobscura fruit fly not found not found XP_001359378.1 not found FBpp0281774* † XP_001354736.2 XP_001359561.2

Drosophila willistoni fruit fly not found not found XP_002074122.1 not found XP_002072999.1 † † XP_002070466.1

Drosophila mojavensis fruit fly not found not found XP_001998663.1 not found XP_001998246.1 † † XP_002000500.1

Drosophila virilis fruit fly not found not found XP_002056376.1 not found XP_002055797.1 † † XP_002056674.1

Drosophila grimshawi fruit fly not found not found XP_001990349.1 not found XP_001990801.1 † † XP_001994525.1

Glossina morsitans tsetse fly not found not found TMP011185-PA not found TMP008325-PA TMP014156-PA

Mayetiola destructor Hessian fly AEGA01027097 AEGA01023846 AEGA01015366 AEGA01030763 AEGA01000382 AEGA01019368

Culex quinquefasciatus southern house mosquito XP_001847132.1 XP_001851012.1 XP_001846901.1 XP_001849859.1 XP_001850246.1 XP_001848722.1 XP_001848723.1 XP_001865668.1

Aedes aegypti yellow fever mosquito XP_001652421.1 XP_001650955.1 XP_001648699.1 XP_001659014.1 XP_001662466.1 XM_001660047.1 supercont1.399 XP_001653143.1

Anopheles gambiae malaria mosquito XP_320297.4 XP_312286.5 XP_312356.3 XP_550917.4 XP_313694.2 XP_321134.5

Anopheles darlingi American malaria mosquito EFR23955.1 EFR21053.1 EFR21220.1 EFR26227.1 EFR19390.1 EFR24380.1

Bombyx morii silkmoth BGIBMGA009815-TA BGIBMGA013481-PA BGIBMGA003507-PA¥ BGIBMGA006347-PA BGIBMGA005151-PA BGIBMGA003396-TA

Manduca sexta tobacco hornworm JH668311 JH668327 JH668592 JH668285 JH668674 JH668790

Heliconius melpomene postman butterfly HMEL015322-PA HMEL009167-PB HMEL005943-PA¥ HMEL002456-PA HMEL009914-PA HMEL015249-PA

Danaus plexippus monarch butterfly DPGLEAN10058-PA* EHJ77334 DPGLEAN08583-PA DPGLEAN01761-PA DPGLEAN09814-PA EHJ65949

Tribolium castaneum red flour beetle XP_973276.1 XP_969800.1 XP_974075.2 XP_971324.1 XP_974797.1 XP_967510.2

Apis mellifera western honey bee GB17569-PA GB11065-PA XP_395500.4 XP_003249672.1¥ XP_624292.1 XP_001121438.2

Apis florea dwarf honey bee XP_003693226.1 AEKZ01010091.1 XP_003692876.1 XP_003692643.1¥XP_003696706.1 XP_003693786.1

Bombus terrestris buff-tailed bumblebee XP_003398685.1 not found Contig6150 XP_003400070.1¥ XP_003395810.1 XP_003398401.1

Nasonia vitripennis parasitoid wasp NP_001153243.1 XP_003426092.1* XP_003426698.1 XP_001600103.2 XP_001600627.1 XP_001603716.2

Acromyrmex echinatior fungus-eating ant EGI61704.1 EGI66859.1 EGI61642.1 EGI59971.1 Aech_00346 EGI70263.1

Atta cephalotes leaf-cutter ant ACEP_00009299-RA* ACEP_00017651-RA* ACEP_00011243-RA ACEP_00006565-RA ACEP_00006097-RA ACEP_00004003-RA

Solenopsis invicta red imported fire ant SI2.2.0_10715 EFZ22081.1 EFZ15107.1* AEAQ01007991.1 EFZ22728.1 EFZ15596.1

Pogonomyrmex barbatus red harverster ant PB24966-RA PB17509-RA* PB16997-RA PB11098-RA PB11133-RA PB17252-RA

Harpegnathos saltator Jerdon's jumping ant EFN83847.1 EFN76321.1 EFN80018.1 EFN88385.1 EFN79794.1 EFN85393.1

Camponotus floridanus Florida carpenter ant EFN66386.1 EFN61172.1 EFN64021.1 EFN73196.1 EFN70560.1 EFN60202.1

Linepithema humile Argentine ant LH24499-RA LH12032-RA LH16983-RA LH21716-RA LH22204-RA LH23420-RA

Acyrthosiphon pisum pea aphid XP_001943834.2 ABLF01023252 XP_003245200.1 XP_001948467.2 XP_001950445.2 XP_003244454.1

Rhodnius prolixus kissing bug RPTMP06549-PA RPTMP11338-PA* RPTMP07262-PA RPTMP08860-PA RPTMP12015-PA RPTMP05810-PA

Pediculus humanus corporis body louse XP_002428327.1 XP_002425413.1 XP_002425072.1 XP_002427257.1 XP_002431159.1 XP_002425588.1

Daphnia pulex water flea EFX81651.1 EFX62143.1 EFX69314.1* EFX76708.1 EFX72666.1 EFX79893.1

Strigamia maritima centipede Smar_temp_005851-PA Smar_temp_004379-PA Smar_temp_007699-PA Smar_temp_002855-PA¥Smar_temp_009055-PA Smar_temp_006773-PA

Metaseiulus occidentalus spider mite JH621002 AFFJ01003132.1 not found not found JH621049 JH620983*

Caenorhabditis elegans nematode NP_495451.1 NP_502531.1 not found not found NP_497858.1 NP_499437.2

Caenorhabditis briggsae nematode XP_002630506.1 XP_002634125.1 not found not found XP_002642710.1 XP_002641336.1

Hydra magnipapillata hydra Contig37161 XM_002168106.1* Contig37595 Contig38910 XP_002159900.1 XP_002158235.1

Trichoplax adhaerens placozoa XP_002108721.1 XP_002114319.1 XP_002115374.1 XP_002117105.1 XP_002113208.1 XP_002107594.1Amphimedon queenslandica sponge XP_003384311.1 XP_003388269.1 NW_003546263.1 XP_003386352.1 AMPQUscaffold_13501 XP_003383251.1

Dictyostelium discoideum slime mold XP_638826.1 XP_003289889.1 XP_639313.1 XP_637904.1 XP_629372.1 XP_640277.1

Arabidopsis thaliana thale cress NP_193469.2 NP_188683.3 NP_187577.1 NP_179021.3 NP_178812.1 NP_201511.2

† Most of the Drosophila MEI-217 and MEI-218 annotations appear to be incorrect. The open reading frames predicted by our analyses are given in Materials and Methods.

not found

not found

not found

RPTMP01162-RA

PHUM401830-PA

not found

not found

XP_002112941.1*

JH668728

HMEL007883-PA

EHJ74933

XP_003383251.1

* Partial sequence only.

not foundnot found

not found

not found

not found

not found

not found

not found

not found

not found

not found

not found

not found

not found

¥ Annotation appears to have fused coding sequences for two proteins.

NP_775789.3

XP_001923762.1

XP_001378952.2

not found

not found

not found

XP_002589555.1

not found

not found

not found

TMP005003-PA

BGIBMGA008428-TA

XENTRscaffold_189

scaffold63640*

scaffold47446*

1

Table S7. Search details for the species-specific database searches (text in orange in Table S6).

Species Protein Site Genome Query Hit E value

Aedes aegypti MEI-218 http://aaegypti.vectorbase.org/Tools/BLAST/ AaegL1 XP_001848723.1 supercont1.399 3 x 10-76

Apis florea Msh5 http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=insects WGS GB11065-PA AEKZ01010091.1 (contig 10091) 0.0

Amphimedon queenslandica MCM8 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi v1.0 XP_002115374.1 AMPQUscaffold_13501 3 x 10-31

Amphimedon queenslandica MEI-218 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi v1.0 XP_001923762.1 AMPQUscaffold_12883 0.83

Atta cephalotes MCM9 http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=insects WGS EGI59971.1 ADTU01012255.1 (contig 12255) 0.0

Hydra magnipapillata Msh4 http://hydrazome.metazome.net/ Hydra magnipapillata XP_002108721.1 Contig37161 6 x 10-13

Hydra magnipapillata MCM8 http://hydrazome.metazome.net/ Hydra magnipapillata XP_002115374.1 Contig37595 2 x 10-22

Hydra magnipapillata MCM9 http://hydrazome.metazome.net/ Hydra magnipapillata XP_002117105.1 Contig38910 2 x 10-43

Manduca sexta* COSA-1 http://blast.hgsc.bcm.tmc.edu/blast.hgsc?organism=MandSexta Scaffolds, 2012-06-11 BGIBMGA012301-PA 1 x 10-10

Metaseiulus occidentalus* Msh5 http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=insects WGS GB11065-PA AFFJ01003132.1 (ctg7180000074512) 3 x 10-57

Saccoglossus kowalevskii Msh4 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi Skow_1.1 XP_002108721.1 scaffold30423 2 x 10-13

Saccoglossus kowalevskii MEI-218 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi Skow_1.1 XP_001923762.1 scaffold47446 1 x 10-04

Solenopsis invicta MCM9 http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi?organism=insects WGS EGI59971.1 AEAQ01007991.1 (contig 11736) 2 x 10-138

Strongylocentrotus purpuratus MEI-218 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi v2.1 XP_001923762.1 scaffold63640 0.009

Xenopus tropicalis MEI-218 http://www.ncbi.nlm.nih.gov/blast/Blast.cgi v4.2 XP_001923762.1 XENTRscaffold_189, E 3 x 10-14

All searches were TBLASTN with default parameters. E value is the highest "expected" parameter. High E values are due to short matches, usually from incomplete sequence.

All hits were validated by reciprocal BLAST searches using the subject predicted amino acid sequence as a query against the human reference protein set.

*All Manduca sexta and Metaseiulus occidentalus searches were done at the BCM site. Only one example result is given above.

1

References

1. S. L. Andersen, J. Sekelsky, Meiotic versus mitotic recombination: Two different routes for

double-strand break repair. Bioessays 32, 1058 (2010). doi:10.1002/bies.201000087

Medline

2. W. K. Chu, I. D. Hickson, RecQ helicases: Multifunctional genome caretakers. Nat. Rev.

Cancer 9, 644 (2009). doi:10.1038/nrc2682 Medline

3. L. Jessop, B. Rockmill, G. S. Roeder, M. Lichten, Meiotic chromosome synapsis-promoting

proteins antagonize the anti-crossover activity of sgs1. PLoS Genet. 2, e155 (2006).

doi:10.1371/journal.pgen.0020155 Medline

4. P. Ross-Macdonald, G. S. Roeder, Mutation of a meiosis-specific MutS homolog decreases

crossing over but not mismatch correction. Cell 79, 1069 (1994). doi:10.1016/0092-

8674(94)90037-X Medline

5. N. M. Hollingsworth, L. Ponte, C. Halsey, MSH5, a novel MutS homolog, facilitates meiotic

reciprocal recombination between homologs in Saccharomyces cerevisiae but not

mismatch repair. Genes Dev. 9, 1728 (1995). doi:10.1101/gad.9.14.1728 Medline

6. H. L. Blanton et al., REC, Drosophila MCM8, drives formation of meiotic crossovers. PLoS

Genet. 1, e40 (2005). doi:10.1371/journal.pgen.0010040 Medline

7. H. Liu, J. K. Jang, J. Graham, K. Nycz, K. S. McKim, Two genes required for meiotic

recombination in Drosophila are expressed from a dicistronic message. Genetics 154,

1735 (2000). Medline

8. B. S. Baker, A. T. C. Carpenter, Genetic analysis of sex chromosomal meiotic mutants in

Drosophilia melanogaster. Genetics 71, 255 (1972). Medline

9. E. A. Manheim, J. K. Jang, D. Dominic, K. S. McKim, Cytoplasmic localization and

evolutionary conservation of MEI-218, a protein required for meiotic crossing-over in

Drosophila. Mol. Biol. Cell 13, 84 (2002). doi:10.1091/mbc.01-06-0318 Medline

10. D. Remus et al., Concerted loading of Mcm2-7 double hexamers around DNA during DNA

replication origin licensing. Cell 139, 719 (2009). doi:10.1016/j.cell.2009.10.015 Medline

2

11. T. Snowden, S. Acharya, C. Butz, M. Berardini, R. Fishel, hMSH4-hMSH5 recognizes

Holliday junctions and forms a meiosis-specific sliding clamp that embraces homologous

chromosomes. Mol. Cell 15, 437 (2004). doi:10.1016/j.molcel.2004.06.040 Medline

12. A. Marchler-Bauer et al., CDD: A Conserved Domain Database for the functional annotation

of proteins. Nucleic Acids Res. 39, D225 (2011). doi:10.1093/nar/gkq1189 Medline

13. L. A. Kelley, M. J. Sternberg, Protein structure prediction on the Web: A case study using the

Phyre server. Nat. Protoc. 4, 363 (2009). doi:10.1038/nprot.2009.2 Medline

14. C. M. Lake, K. Teeter, S. L. Page, R. Nielsen, R. S. Hawley, A genetic analysis of the

Drosophila mcm5 gene defines a domain specifically required for meiotic recombination.

Genetics 176, 2151 (2007). doi:10.1534/genetics.107.073551 Medline

15. M. D. Adams, M. McVey, J. J. Sekelsky, Drosophila BLM in double-strand break repair by

synthesis-dependent strand annealing. Science 299, 265 (2003).

doi:10.1126/science.1077198 Medline

16. M. McVey, S. L. Andersen, Y. Broze, J. Sekelsky, Multiple functions of Drosophila BLM

helicase in maintenance of genome stability. Genetics 176, 1979 (2007).

doi:10.1534/genetics.106.070052 Medline

17. Y. Liu, T. A. Richards, S. J. Aves, Ancient diversification of eukaryotic MCM DNA

replication proteins. BMC Evol. Biol. 9, 60 (2009). doi:10.1186/1471-2148-9-60 Medline

18. T. H. Morgan, Complete linkage in the second chromosome of the male of Drosophila.

Science 36, 719 (1912).

19. See supplementary materials on Science Online.

20. J. B. Boyd, M. D. Golino, K. E. S. Shaw, C. J. Osgood, M. M. Green, Third-chromosome

mutagen-sensitive mutants of Drosophila melanogaster. Genetics 97, 607 (1981).

Medline

21. K. S. McKim, J. B. Dahmus, R. S. Hawley, Cloning of the Drosophila melanogaster meiotic

recombination gene mei-218: A genetic and molecular analysis of interval 15E. Genetics

144, 215 (1996). Medline

3

22. P. Rørth, Gal4 in the Drosophila female germline. Mech. Dev. 78, 113 (1998).

doi:10.1016/S0925-4773(98)00157-9 Medline

23. S. E. McGaugh, M. A. Noor, Genomic impacts of chromosomal inversions in parapatric

Drosophila species. Philos. Trans. R. Soc. London Ser. B 367, 422 (2012).

doi:10.1098/rstb.2011.0250 Medline

24. K. Tamura et al., MEGA5: Molecular evolutionary genetics analysis using maximum

likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28,

2731 (2011). doi:10.1093/molbev/msr121 Medline

25. R. C. Edgar, MUSCLE: Multiple sequence alignment with high accuracy and high

throughput. Nucleic Acids Res. 32, 1792 (2004). doi:10.1093/nar/gkh340 Medline

26. S. Whelan, N. Goldman, A general empirical model of protein evolution derived from

multiple protein families using a maximum-likelihood approach. Mol. Biol. Evol. 18, 691

(2001). doi:10.1093/oxfordjournals.molbev.a003851 Medline

27. Z. Yang, PAML 4: Phylogenetic analysis by maximum likelihood. Mol. Biol. Evol. 24, 1586

(2007). doi:10.1093/molbev/msm088 Medline

28. D. A. Grimaldi, M. S. Engel, Evolution of the Insects (Cambridge Univ. Press, Cambridge,

2005).

29. Z. Yang, PAML: A program package for phylogenetic analysis by maximum likelihood.

Comput. Appl. Biosci. 13, 555 (1997). Medline

30. N. Goldman, Z. Yang, A codon-based model of nucleotide substitution for protein-coding

DNA sequences. Mol. Biol. Evol. 11, 725 (1994). Medline

31. P. James, J. Halladay, E. A. Craig, Genomic libraries and a host strain designed for highly

efficient two-hybrid selection in yeast. Genetics 144, 1425 (1996). Medline

32. A. De Muyt et al., BLM helicase ortholog Sgs1 is a central regulator of meiotic

recombination intermediate metabolism. Mol. Cell 46, 43 (2012).

doi:10.1016/j.molcel.2012.02.020 Medline

4

33. K. Zakharyevich, S. Tang, Y. Ma, N. Hunter, Delineation of joint molecule resolution

pathways in meiosis identifies a crossover-specific resolvase. Cell 149, 334 (2012).

doi:10.1016/j.cell.2012.03.023 Medline

34. M. Lutzmann et al., MCM8- and MCM9-deficient mice reveal gametogenesis defects and

genome instability due to impaired homologous recombination. Mol. Cell 47, 523 (2012).

doi:10.1016/j.molcel.2012.05.048 Medline

35. K. Nishimura et al., Mcm8 and Mcm9 form a complex that functions in homologous

recombination repair induced by DNA interstrand crosslinks. Mol. Cell 47, 511 (2012).

doi:10.1016/j.molcel.2012.05.047 Medline

36. R. Collin, R. Cipriani, Dollo’s law and the re-evolution of shell coiling. Proc. Biol. Sci. 270,

2551 (2003). doi:10.1098/rspb.2003.2517 Medline

37. R. Collin, M. P. Miglietta, Reversing opinions on Dollo’s Law. Trends Ecol. Evol. 23, 602

(2008). doi:10.1016/j.tree.2008.06.013 Medline

38. M. D. Adams et al., The genome sequence of Drosophila melanogaster. Science 287, 2185

(2000). doi:10.1126/science.287.5461.2185 Medline

39. B. M. Wiegmann et al., Episodic radiations in the fly tree of life. Proc. Natl. Acad. Sci.

U.S.A. 108, 5690 (2011). doi:10.1073/pnas.1012675108 Medline

40. R. H. Gooding, B. M. Rolseth, Genetics of Glossina palpalis palpalis: Designation of linkage

groups and the mapping of eight biochemical and visible marker genes. Genome 38, 833

(1995). doi:10.1139/g95-109 Medline

41. B. M. Gilchrist, J. B. S. Haldane, Sex linkage and sex determination in a mosquito, Culex

molestus. Hereditas 33, 175 (1947). doi:10.1111/j.1601-5223.1947.tb02800.x

42. G. A. McClelland, Sex-linkage at two loci affecting eye pigment in the mosquito Aedes

aegypti (Diptera: Culicidae). Can. J. Genet. Cytol. 8, 192 (1966). Medline

43. L. Zheng, M. Q. Benedict, A. J. Cornel, F. H. Collins, F. C. Kafatos, An integrated genetic

map of the African human malaria vector mosquito, Anopheles gambiae. Genetics 143,

941 (1996). Medline