Bryony Jones

Gene expression: layers of gene regulation.

A complex relationship between DNA methylation and gene expression is beginning to emerge Getty A new study investigates genetic and epigenetic influences on genome regulation and alternative splicing, and highlights the tissue specificity of some of these interactions. In a previous study, the researchers examined the relationship between genetic variation, DNA methyation and gene expression using samples from the GenCord cohort that were derived from umbilical cords of 204 newborn children. They genotyped 2.5 million single-nucleotide polymorphisms (SNPs), assayed the methylation levels of 482,421 CpG sites and sequenced the transcriptomes from three cell types — fibroblasts, lymphoblastoid cell lines (LCLs) and T cells. Extending this work, the team now further characterize the effects of genetic and epigenetic variation on the tissue specificity of gene expression and alternative splicing. Gutierrez-Arcelus et al. find that genetic variation affecting gene expression and DNA methylation levels often overlap functional elements. Across cell types, expression quantitative trait loci (eQTLs) are enriched in DNase I-hypersensitive sites, whereas methylation QTLs (mQTLs) are enriched in enhancers and insulators. Their statistical analyses show that, generally, genetic variation has a more consistent effect on gene expression across cell types, although the strength of the effect can be variable. However, the effects of epigenetic variation on gene expression are more tissue-specific. Methylation sites that are associated with expression levels are found predominantly in enhancers, gene bodies and CpG shores, and the correlation with gene expression can be either positive or negative. Alternative splicing levels were measured to see whether there was a tissue-specific correlation with genetic variation and DNA G E N E E X P R E S S I O N

DNA methylation: Switching phenotypes with epialleles

Natural occurring epialleles determine vitamin E accumulation in tomato fruits. Nature Commun. 5, 4027 (2014) | Pignatta, D. et al. Natural epigenetic polymorphisms lead to intraspecific variation in Arabidopsis gene imprinting. Elife http://dx.doi. org/10.7554/eLife.03198 (2014) Getty Epiallelic variation could underlie important phenotypic differences among natural plant populations. Two studies now provide evidence for such an effect in secondary metabolism and reproduction. Quadrana et al. sought to better understand the factors underlying vitamin E content heritability in cultivated tomato plants by focusing on a previously detected vitamin E quantitative trait locus (QTL). They used a combination of reverse genetic approaches, expression analyses, DNA methylation assays and small interfering RNA (siRNA) profiling to show that methyl ation QTL9-2-6 is associated with differential methylation of a transposable element (TE) in the promoter of VTE3 (which encodes a protein involved in the biosynthesis of vitamin E). This DNA methylation can be spontaneously reverted, which directly increases VTE3 expression and vitamin E content in tomatoes. These data show that naturally occurring epialleles are linked with the regulation of a nutritionally important element and demonstrate the importance of epigenetics in determining such traits. In the second study, Pignatta et al. examined variations in imprinting (which arises from the differential DNA methylation of parental alleles) and epigenetic polymorphisms in Arabidopsis thaliana. The researchers looked at genome-wide DNA methyl ation patterns, as well as small RNA and gene expression patterns in the seeds, embryo and endosperm (that is, the tissue inside the seeds of D N A M E T H Y L AT I O N

Evolutionary genetics: Becoming human [mdash] identifying human accelerated regulatory DNA

et al. Comprehensive identification and analysis of human accelerated regulatory DNA. Genome Res. http://www.dx.doi.org/10.1101/gr.192591.115 (2015) Changes in gene regulation are thought to have had a role in the evolution of many human-specific traits such as bipedalism, speech and increased cognition. However, identifying the regulatory genetic changes that might be responsible for such human-specific traits is more complex than identifying coding variants. In a new study, comparative and functional approaches were combined to identify and characterize human accelerated regulatory sequences genome wide. DNase I cleaves regions of open chromatin and thus demarcates potentially active and regulatory regions, and recently large-scale studies have begun to create genome-wide catalogues of DNase I hypersensitive sites (DHSs). Here, the team used the previously defined maps of DHSs from the 130 cell types defined in the Encyclopedia of DNA Elements (ENCODE) and Roadmap Epigenomics Projects. DHSs were merged across cell types into over two million loci with a median length of 290 base pairs, and then arranged into a multiple alignment of six primate genomes from the EPO pipeline of Ensembl. To identify DHSs that are evolutionarily conserved across primates, a maximum likelihood test was used to locate E VO L U T I O N A RY G E N E T I C S

RNA: All sorts of mitochondrial RNA

et al. High-resolution genomic analysis of human mitochondrial RNA sequence variation. Science 344, 413–415 (2014) N PG High-resolution RNA sequencing of the mitochondrial genome has revealed a substantial level of sequence variation both within and between individuals, which has potential implications for human health. A new study now reports that mitochondrial RNA heterogeneity (that is, heteroplasmy) often occurs at the same functionally relevant position within tRNA genes and that the proportion of heteroplasmy within individuals is associated with a single variant in the nuclear genome. “We identified a remarkable amount of RNA sequence variation within each individual, which represents variation across mitochondrial transcriptomes,” says Alan Hodgkinson (University of Montreal), co-first author of the study. The researchers carried out high-throughput deep sequencing (>6,000×) of whole-blood RNA in a population sample of 708 individuals from the CARTaGENE project, led by Philip Awadalla (University of Montreal). An average of 14 heteroplasmic sites were recorded per individual, totalling 650 unique sites across all individuals. Thirteen of these sites were found to contain three or more nucleotides (that is, they were multiallelic) across most of the individuals sampled. The majority of these multiallelic sites are situated at the ninth position of tRNAs, and approximately half are known to be post-transcriptionally methlyated. Sequencing of the corresponding nuclear DNA revealed no variation, which indicates that these sites are candidates for RNA modification events. Genome-wide analysis of nuclear DNA identified a missense mutation (rs11156878) in the mitochondrial R N A

Complex traits: Predicting phenotypes

NPG This study is the first application of phenotype prediction using wholegenome-sequencing data in a higher eukaryote. Traditional genome-wide association studies (GWASs) identify single variants that associate with a particular phenotype by testing the significance of each variant. Genome-based prediction instead uses information across the whole genome simultaneously to explain the variability in the observed phenotype. Prediction methods may be useful in identifying the genetic causes of complex polygenic traits, although so far the application of genome-based prediction has been restricted by the number of SNPs available and the small proportion of SNPs that are typically included in such studies. The authors analysed ~2.5 million SNPs obtained from sequence data of 157 inbred lines from the recently published ‘Drosophila Genetic Reference Panel’ in an attempt to predict phenotypes of two complex traits: starvation stress resistance and startle-induced locomotor behaviour. Both traits respond rapidly to artificial selection and are genetically variable in natural populations. After first creating a genomic relationship matrix, the authors used a genomic best linear unbiased prediction (GBLUP) method to evaluate predictive ability by cross-validation. GBLUP approaches take into account the covariance structure inferred from the genomic data. The authors compared GBLUP with a Bayesian approach and evaluated the effects of SNP density and training set size on predictive accuracy. Although the predictive abilities obtained were moderate, the authors state that this study provides proof of concept for this approach. Bryony Jones Prediction methods may be useful in identifying the genetic causes of complex polygenic traits R E S E A R C H H I G H L I G H T S

Chromatin: Cracking the nucleosome code.

can identify and map concurrent modifications on individual histones. The team behind this approach were able to identify truly bivalent nucleosomes that harboured both repressive and activating marks, and map their genomic location. Nucleosomal histones can be chemically modified at multiple positions, and it has been hypothesized that various combinations of these marks could direct cell-type-specific regulatory outcomes by altering interactions between DNA and chromatin. Established methods to investigate combinatorially modified histones, such as chromatin immunoprecipitation (ChIP) and mass spectrometry, are limited in their ability to differentiate whether marks exist on the same individual histone, and to precisely localize chemical modifications genome wide. The novel experimental approach is based on first isolating individual mononucleosomes from cells and ligating fluorescent adaptors to their free DNA ends. Ligated mono nucleosomes are then separated on a glycerol gradient, spatially captured on slides and incubated with fluorescent antibodies against various histone modifications. Total internal reflection (TIRF) microscopy is used to identify the position and modification state of each nucleosome. The team used this approach to probe the existence of marks on nucleosomes from pluripotent embryonic stem (ES) cells, embryoid bodies (EBs), fully differentiated lung fibroblasts and cancer cells. The authors found that the repressive mark trimethylated histone H3 lysine 27 (H3K27me3) was increasingly prevalent during cell differentiation, whereas the activating H3K4me3 mark remained relatively constant. In ES cells, 0.5% of nucleosomes carried both marks, and were thus truly ‘bivalent’. Notably, bivalent nucleosomes were far less prevalent in EBs and fibroblasts, and 94% of bivalent nucleo somes were asymmetric — their opposing marks usually occurred on histones at opposite ends of a nucleosome. Interestingly, bivalent nucleo somes were also prevalent in lymphoma cells, with about half of all H3K4me3-marked nucleosomes also carrying H3K27me3. Using successive steps of antibody incubation and imaging, the simultaneous presence of four histone marks were then investigated (H3K27 acetylation (H3K27ac), H3K27me2, H3K4me3 and H3K27me3). The proportions of these four modifications were similar between ES cells and lung fibroblasts, although there were notable differences regarding the combination of these marks between cell types — for example, ES cell chromatin was enriched for active marks — which the authors attribute to the different chromatin environments. Finally, after capturing the nucleosomes and probing their modifiction state, direct single-molecule sequencing-by-synthesis was used to investigate the DNA associated with each nucleosome. About 1,000 reads from concomitantly marked molecules were generated and aligned to the genome, thus identifying the genomic locations of individual bivalent nucleosomes. The single-molecule assay developed by Shema et al. enables millions of nucleosomes to be decoded and sequenced in a single automated imaging run. This approach may prove important to better understand the nature and effect of combinatorial chromatin modifications. Bryony Jones C H R O M AT I N

Evolution: Duplicate gene co-regulation slows evolution

Gene duplications can drive evolutionary processes and contribute to the development of new biological functions. In mammals, however, new gene duplicates are often degraded into non-functional pseudogenes. Now, researchers from Stanford University describe the evolutionary forces that control the fate of young gene duplicates and the processes that contribute to the long-term survival in some duplicate pairs. By first analysing RNA sequencing (RNA-seq) data from 46 human tissues and replicating their findings in 26 mouse tissues, the team sought to identify whether the survival of duplicate genes in the mammalian genome is due to subfunctionalization (in which each gene retains a part of the ancestral gene’s function) and neofunctionalization (in which one gene takes on a new function after the duplication event), or whether gene dosage-sharing models can better explain the persistence of gene duplicates. A total of 1,444 duplicate gene pairs were identified in the human genome and then classified according to their co-expression patterns. A gene pair was classified as possibly subfunctionalized or neofunctionalized if each gene was significantly more highly expressed than (especially promoter–promoter links) of tandem duplicates than between singleton genes. By contrast, Hi-C showed no evidence of linkage between duplicates on different chromosomes. The authors hypothesize that tandem duplicates are highly co-regulated and that the process of genomic separation is important for enabling independent evolution. Using RNA-seq data from six human and macaque tissues, the researchers analysed the expression levels of young gene duplicates that arose since the human–macaque evolutionary split. They found that human copies usually evolve reduced expression (with the average summed expression in the human duplicates similar to that of the singleton orthologues in the macaque). They propose that downregulation first achieves dosage balance between tandem duplicates and enables early survival of both genes. If a balance is achieved, the expression levels then evolve slowly, owing to constraints on their combined expression. However, if genomic separation occurs, the duplicate genes are able to evolve independently, resulting in long-term persistence and potentially “true functional differentiation”.

Population genetics: Bursts of male-lineage expansions.

Edward Kinsman/Oxford Scientific/Getty A large demographic study of 1,244 human Y-chromosome sequences reveals several independent, extreme bursts in male population numbers over the past 55,000 years. The research, published in Nature Genetics, indicates that these male-specific population expansions occurred during known migrations and periods of technological innovation. Owing to its patrilineal inheritance pattern and the absence of recombination with the X chromosome over most of its length, variation in Y-chromosomal DNA can be used to build robust phylogenies, and thereby inform the study of male human evolution and demographic patterns. The researchers analysed over 65,000 Y-DNA variants in more than 1,200 sequenced individuals from 26 global populations that were ascertained in phase 3 of the 1000 Genomes Project. After identifying the Y-chromosome haplogroup of each individual, the authors built a maximum-likelihood phylogenetic tree using 60,555 biallelic singlenucleotide variants (SNVs). The tree maps how the individuals relate to one another, with the most recent common male-line ancestor reported to have lived ~190,000 years ago. Strikingly, the branching pattern suggested the existence of several rapid increases — within just a few generations — in the number of men carrying certain Y-chromosome haplogroups, beginning about 50,000–55,000 years ago, when an increase in the number of male lineages outside of Africa could be observed. The authors posit that this burst may reflect the settling of the Eurasian continent around this time period. The phylogeny also supports the theory that the predominant African haplogroup originated P O P U L AT I O N G E N E T I C S

Microbiomes: Symbionts — in it for the long run

Cospeciation of gut microbiota with hominids. Science http://dx.doi.org/10.1126/science. aaf3951 (2016) Research into the co-evolutionary history of hominids and their gut bacteria shows that multiple symbiotic associations arose in a common ancestor to all African great apes. Over the past 15 million years, these ancient symbionts have speciated in parallel (‘co-speciation’) with the nuclear and mitochondrial genomes of humans and African apes. Communities of gut bacterial symbionts in mammals are necessary for development, immunity and health. Bacterial community profiles are influenced by genetics and evolutionary history, as well as by external factors, such as diet and antibiotics. However, little is known about the evolutionary origins of the bacterial lineages that exist in the human gut, including whether some lineages of gut bacteria have persisted in host hominid lineages over periods of time long enough to permit co-speciation. To see whether the gut bacteria present in modern-day humans are descended from ancient bacterial symbionts that subsequently co-speciated with hominids, Moeller et al. first amplified a region of the DNA gyrase subunit B gene (gyrB) from M I C R O B I O M E S

Symbionts — in it for the long run: Microbiomes

Cospeciation of gut microbiota with hominids. Science http://dx.doi.org/10.1126/science. aaf3951 (2016) Research into the co-evolutionary history of hominids and their gut bacteria shows that multiple symbiotic associations arose in a common ancestor to all African great apes. Over the past 15 million years, these ancient symbionts have speciated in parallel (‘co-speciation’) with the nuclear and mitochondrial genomes of humans and African apes. Communities of gut bacterial symbionts in mammals are necessary for development, immunity and health. Bacterial community profiles are influenced by genetics and evolutionary history, as well as by external factors, such as diet and antibiotics. However, little is known about the evolutionary origins of the bacterial lineages that exist in the human gut, including whether some lineages of gut bacteria have persisted in host hominid lineages over periods of time long enough to permit co-speciation. To see whether the gut bacteria present in modern-day humans are descended from ancient bacterial symbionts that subsequently co-speciated with hominids, Moeller et al. first amplified a region of the DNA gyrase subunit B gene (gyrB) from M I C R O B I O M E S