Mia Levine, Ph.D.

The evolution of young genes via de novo- and duplication- based mechanisms
Evolutionary mechanisms of innovation at the molecular level are numerous. Codons diverge, regulatory elements arise and degenerate, and new genes are born. These signatures of adaptive evolutionary change are frequently species-restricted. My PhD research identified very young genes that harbor no homology to exons in related genomes. In contrast to classic mechanisms of novel gene formation like gene duplication, these de novo genes arise instead from fortuitous sequence evolution at noncoding DNA. These genes exhibited testis-biased expression and signatures of adaptive evolution, implicating male germline processes as potent agents of selection of these rare mutations. This publication was the first to describe such de novo genes that have since been documented in a wide array of taxa, including humans. My postdoctoral research focused instead on gene duplication as a potent mechanism of adaptive diversification. A shared domain structure between parent and daughter proteins facilitated my goal to identify lineage-specific innovations in proteins that package DNA. Prior to my research, the Heterochromatin Protein 1 (HP1) gene family was thought to encode between 2 and 5 members across eukaryotes. I discovered 22 new HP1 members in Drosophila. These 22 paralogs were all born less than 20 million years ago. Nevertheless, the number of HP1 genes per species remains relatively constant. This revolving door of gene replacement implicates conserved, currently undefined chromatin functions encoded by unconserved components recurrently generated by gene duplication.

Levine, M.T., C. D. Jones, A. D. Kern, H. A. Lindfors, and Begun, D.J. (2006) Novel genes derived from noncoding DNA in Drosophila melanogaster are frequently X-linked and exhibit testis-biased expression.Proceedings of the National Academy of Sciences 103: 9935-9939. PMCID: PMC1502557.
Levine, M.T., McCoy, C. Vermaak. D., LeeY.C.G, Hiatt, M.A., Matsen, F.A., and H.S. Malik (2012) Phylogenomic analysis reveals dynamic evolutionary history of the Drosophila Heterochromatin Protein 1 (HP1) gene family.  PLOS Genetics8(6): e1002729. PMCID: PMC3380853.

Adaption to novel environments at DNA packaging proteins

melanogaster, like humans, evolved in Africa but more recently invaded the New World and established populations from tropical to temperate climates. These geographically structured populations are panmictic (randomly mating), so any genotypic or phenotypic differences observed are likely the product of fitness variation across environments, i.e., natural selection. This spatially-structured system therefore presents a unique opportunity to elucidate the molecular basis of adaptation. I have taken both whole genome- and single locus- approaches. In one report, genomic DNA fromD. melanogastertropical and temperate populations from both Australia and the US were hybridized to whole-genome tiling arrays from which we inferred geographic sequence divergence (allele frequency variation) based on geographically structured differences in probe intensities. Our analysis demonstrated that a remarkably large fraction of the D. melanogaster genome has been targeted by spatially-varying positive selection. Our whole-genome analysis also uncovered many previously unsuspected biological functions associated with adaptation to novel environments. One of the most intriguing of these functions was chromatin binding. In light of the extensive data on the environment sensitivity of chromatin dynamics, I was especially interested in the unexplored role that chromatin-remodeling factors play in adaptation to novel habitat. Under one model, chromatin-remodeling factors evolve to maintain chromatin structure that is perturbed by environmental fluctuations. I focused primarily on the Polycomb Group genechameau. I discovered a linear relationship between latitude and allele frequency at several SNPs in both the US and Australian populations, which represent independent colonization events. Moreover, an amino acid-changing SNP predicted variation in tolerance to freezing temperatures. These data strongly implicated the action of natural selection and introduced chromatin-remodeling factors as a potentially rich source of adaptive genetic variation. Inspired by the observation that chromatin-based gene regulation can span more than one promoter, I also tested the hypothesis that adaptive expression variation across latitudinal gradients spans physically linked genes. I found that gene “neighborhoods” (of up to 15 genes), rather than single genes, exhibit adaptive transcriptional profiles, consistent with the notion that chromatin factors regulate adaptive expression variation across space.
Turner, L.T., Levine, M.T., and Begun, D.J. (2008). Genomic analysis of adaptive differentiation in Drosophila melanogaster.Genetics 179: 475-485. PMCID: PMC2390623.
Levine, M.T. and Begun, D.J. (2008). Evidence of spatially varying selection at four chromatin-remodeling loci inDrosophila melanogaster. Genetics 179: 455-473. PMCID: PMC2390624.
Levine, M.T., Eckert, M., and D.J. Begun (2011) Whole genome expression plasticity across tropical and temperateDrosophila melanogaster populations from eastern Australia. Molecular  Biology and Evolution 28: 249–256. PMCID: PMC3002243.

Evolutionary and functional diversification of essential DNA packaging proteins
Conserved nuclear proteins support conserved nuclear processes. Yeast and humans, for example, share essential, homologous chromatin proteins that package eukaryotic DNA and support shared, essential functions like chromosome segregation and telomere integrity. These cellular processes, however, also rely on unconserved molecular machinery. A surprisingly large fraction of essential genes that encode chromatin proteins evolve rapidly. My dissertation documented early evidence of this paradoxical phenomenon. The Dosage Compensation Complex (DCC) is responsible for equalizing X-linked gene dosage via chromatin remodeling of the single male X chromosome.  Loss of function at DCC genes is lethal. I discovered population genetic evidence of positive selection at four of the five DCC complex components. Continuing this theme during my postdoctoral research, I uncovered the essential function of the Heterochromatin Protein 1 paralog, HP1E. HP1E is required for faithful segregation of paternal DNA during the first embryonic mitosis. Nevertheless, a subset of Drosophila species apparently persists without HP1E. I discovered that in D. melanogaster not all paternal chromosomes are equally vulnerable to chromatin bridging during the first embryonic mitosis—the heterochromatin-rich sex chromosomes are more likely to mis-segregate than the large autosomes. Intriguingly, over evolutionary time major rearrangements of these same sex chromosomes co-occur with the pseudogenization of HP1E in the obscura group of Drosophila. These data support a model under which karyotype evolution rendered dispensable a once-essential gene. My findings thus provided a neat hypothesis to resolve the apparent paradox of HP1E’s essentiality in D. melanogaster together with its loss in related species.

Levine, M.T., Holloway, A.K., Arshad, U., and Begun, D.J. (2007) Pervasive and largely lineage-specific adaptive protein evolution in the dosage compensation complex of Drosophila melanogaster.Genetics 177: 1959–1962. PMCID: PMC2147993.
Levine, M.T. and H.S. Malik (2013) A rapidly evolving genomic toolkit of Drosophila heterochromatin.Fly 7: 137-141. PMCID: PMC4049844.
Levine, M.T., Vander Wende, H., and H.S. Malik (2015) Mitotic fidelity requires transgenerational action of a testis-restricted HP1.eLife 4: e07378. PMCID: PMC4491702