Population Genomic Time Series Data of a Natural Population Suggests Adaptive Tracking of Fluctuating Environmental Changes

Natural populations are constantly exposed to fluctuating environmental changes that negatively affect their fitness in unpredictable ways. While theoretical models show the possibility of counteracting these environmental changes through rapid evolutionary adaptations, there have been few empirical studies demonstrating such adaptive tracking in natural populations. Here, we analyzed environmental data, fitness-related phenotyping and genomic time-series data sampled over 3 years from a natural Chironomus riparius (Diptera, Insecta) population to address this question. We show that the population's environment varied significantly on the time scale of the sampling in many selectively relevant dimensions, independently of each other. Similarly, phenotypic fitness components evolved significantly on the same temporal scale (mean 0.32 Haldanes), likewise independent from each other. The allele frequencies of 367,446 SNPs across the genome showed evidence of positive selection. Using temporal correlation of spatially coherent allele frequency changes revealed 35,574 haplotypes with more than one selected SNP. The mean selection coefficient for these haplotypes was 0.30 (s.d. = 0.68). The frequency changes of these haplotypes clustered in 46 different temporal patterns, indicating concerted, independent evolution of many polygenic traits. Nine of these patterns were strongly correlated with measured environmental variables. Enrichment analysis of affected genes suggested the implication of a wide variety of biological processes. Thus, our results suggest overall that the natural population of C. riparius tracks environmental change through rapid polygenic adaptation in many independent dimensions. This is further evidence that natural selection is pervasive at the genomic level and that evolutionary and ecological time scales may not differ at all, at least in some organisms.

Karyotypes and COI Gene Sequences of Chironomus sp. Le1 (Kiknadze and Salova, 1996), Ch. laetus (Belyanina and Filinkova, 1996) and Their Hybrid from the Yamal Peninsula, Arctic Zone of Russia

The study of the biological diversity of the Arctic zone yields intriguing results. Initial research on the lakes of the Yamal Peninsula resulted in the identification of Chironomus laetus and the hybrid Ch. laetus × Ch. sp. Le1. To avoid misidentification, we used morphological, cytogenetic, and molecular genetic approaches. By cytogenetics, in Ch. sp. Le1, seven banding sequences were found: Le1A1, Le1B1, Le1C1, Le1D1, Le1E1, Le1F1, and Le1G1. The karyotype of Ch. laetus was mapped for the first time; it is the first species with the arm combinations AE BC DF G. We propose the name of a new cytocomplex-"laetus". DNA-barcoding of the COI gene was carried out for Ch. laetus and Ch. laetus × Ch. sp. Le1 for the first time. The estimated genetic distance between the sequences of Ch. laetus and Ch. riihimakiensis is 2.3-2.5%. The high similarity in morphology, banding sequences, and the possibility of hybridization indicate a close relationship between Ch. laetus and Ch. sp. Le1, which is assumed to be the northern variant of Ch. riihimakiensis. Molecular genetic data suggests the presence of a subgroup with Ch. laetus.

Genomic divergence landscape in recurrently hybridizing Chironomus sister taxa suggests stable steady state between mutual gene flow and isolation

Divergence is mostly viewed as a progressive process often initiated by selection targeting individual loci, ultimately resulting in ever increasing genomic isolation due to linkage. However, recent studies show that this process may stall at intermediate stable equilibrium states without achieving complete genomic isolation. We tested the extent of genomic isolation between two recurrently hybridizing nonbiting midge sister taxa, Chironomus riparius and Chironomus piger, by analyzing the divergence landscape. Using a principal component-based method, we estimated that only about 28.44% of the genomes were mutually isolated, whereas the rest was still exchanged. The divergence landscape was fragmented into isolated regions of on average 30 kb, distributed throughout the genome. Selection and divergence time strongly influenced lengths of isolated regions, whereas local recombination rate only had minor impact. Comparison of divergence time distributions obtained from several coalescence-simulated divergence scenarios with the observed divergence time estimates in an approximate Bayesian computation framework favored a short and concluded divergence event in the past. Most divergence happened during a short time span about 4.5 million generations ago, followed by a stable equilibrium between mutual gene flow through ongoing hybridization for the larger part of the genome and isolation in some regions due to rapid purifying selection of introgression, supported by high effective population sizes and recombination rates.