When a mutation arises in an egg or sperm cell, it could be evolutionarily important. But if a mutation occurs in somatic tissue instead, the result could be cancer. Mutations in the germline and soma not only have contrasting consequences, they also arise at different rates that may reflect the balance of DNA damage and repair pathways in different tissue types. In the September issue of GENETICS, Chen et al. predict gene mutation rates in different tissues and find that high expression increases mutation rates in the germline, but not in somatic tissue.  

The first step was to obtain a reliable estimate of the mutation rate in both germ cells and somatic tissues. The researchers relied on a set of germline mutations, previously identified using exome data from thousands of sets of parents and children. Any variation that was unique to the children must be caused by germline mutation in either the father or mother. To identify somatic mutations, the researchers analyzed three different cancer samples that included whole exome sequence of both normal and malignant cells. Variation unique to either tissue type predates the tumor and should be due to somatic mutations.

A statistical model that evaluated how well various factors predict the mutation rate revealed a key difference. In germ cells, a high gene expression level was linked to a higher mutation rate, but the opposite was observed in somatic tissues. Though the magnitude of the effect varied in the three different cancer types, there was always a negative correlation with expression. Other factors also contributed differently to mutation in the germline and somatic tissues, including GC content for the germline and replication timing in the soma.

Gene expression level probably affects mutation rate because the DNA double helix unzips to accommodate transcription machinery, making the individual strands more vulnerable to mutagens, and because there is a dedicated repair mechanism to fix DNA damage that occurs in transcribed regions. The opposite effects of expression level on mutation rates suggests germline and somatic tissues have marked differences in the balance between damage and repair. For example, expression may be more mutagenic in the germline, or repair mechanisms may be more efficient in the soma. There could even be unidentified DNA damage repair processes that are unique to certain tissues. Though somatic mutations can’t be passed down to the next generation like germline mutations, they are the root cause of most cancers. Quickly and correctly repairing this DNA damage is vital for an organism’s survival.

CITATION:

Contrasting Determinants of Mutation Rates in Germline and Soma

Chen Chen, Hongjian Qi, Yufeng Shen, Joseph Pickrell, and Molly Przeworski

GENETICS September 1, 2017. 207 (1): 255-267

https://doi.org/10.1534/genetics.117.1114

http://www.genetics.org/content/207/1/255

Inheriting an extra chromosome can sometimes be disastrous, but in the September issue of G3, Linder et al. investigate a chromosome duplication that helps yeast survive harsh conditions. Yeast with an extra copy of chromosome IV better tolerate hydrogen peroxide exposure, largely thanks to an extra copy of a gene that detoxifies the chemical. This work shows how large spontaneous mutations can help organisms thrive in the absence of natural genetic variation.

To generate mutants tolerant of hydrogen peroxide, the researchers plated haploid yeast (which have one copy of each chromosome) from three different strains on agar dosed with hydrogen peroxide. They chose thirty-seven colonies that tolerated higher concentrations of peroxide than their parent strains. Whole genome sequencing of those tolerant colonies revealed nearly half of them carried a duplication of chromosome IV, making it the most common mutation overall.

One colony had a duplication of the right arm of chromosome IV, allowing researchers to limit their search for the beneficial genes to just this area. They further narrowed down the search by systematically deleting portions of the right arm of the chromosome. Eventually, they targeted a region containing only five genes. Knocking them out one by one revealed most of these colonies’ resistance to hydrogen peroxide was due to an extra copy of the peroxidase gene TSA2, which is an enzyme that detoxifies peroxide. Boosting levels of TSA2 in parental cells increased their hydrogen peroxide tolerance, though not as much as the mutants carrying the whole duplicate chromosome. This means other unidentified factors on chromosome IV must also contribute to the phenotype.

This research shows chromosome duplication can be beneficial rather than harmful when the dosage of critical genes is increased. It all depends on the environment; in normal conditions, the hydrogen peroxide resistant mutants grew more slowly than their progenitors. Other studies in yeast have shown similar results for different stressors, suggesting spontaneous chromosome duplication may be commonly used by these microbes to quickly bridge the survival gap in tough environments.

CITATION

The Stress-Inducible Peroxidase TSA2 Underlies a Conditionally Beneficial Chromosomal Duplication in Saccharomyces cerevisiae

Robert A. Linder, John P. Greco, Fabian Seidl, Takeshi Matsui and Ian M. Ehrenreich

G3: Genes, Genomes, Genetics September 2017 7: 3177-3184;

https://doi.org/10.1534/g3.117.300069

http://www.g3journal.org/content/7/9/3177

It’s surprisingly common for babies to be born missing one or both kidneys; an estimated one in one thousand babies are born with a single kidney. Called renal agenesis, this condition is fatal if both kidneys are missing, and having just one can also lead to serious health problems like hypertension and early renal failure. In the September issue of GENETICS, Brophy et al. show for the first time that renal agenesis in humans can be caused by disruptions in the retinoic acid receptor pathway. They used whole exome sequencing in two affected families to identify a causal mutation and applied innovative CRISPR mutagenesis in mice to confirm their findings.

Two unrelated families from Iowa and Denmark each had multiple cases of renal agenesis. For both families, the researchers identified potentially causal mutations by comparing the whole exome sequences of several affected and unaffected family members. The gene GREB1L carried harmful mutations in both families: a missense SNV in one and a deletion interrupting a splice site in the other. Further sequencing confirmed that all affected individuals carried the mutated gene copies.

GREB1L is a cofactor for retinoic acid receptors that until now has never been implicated in mammalian kidney development—let alone renal agenesis. To confirm its effect, the researchers obtained a zebrafish mutant for the corresponding gene. Fish homozygous for the mutation showed abnormal early kidney development and died before reaching maturity. Knockdown treatment that decreased GREB1L expression in genetically normal fish had similar results, suggesting that GREB1L was indeed the gene causing the kidney developmental problems.

The final confirmation came from CRISPR-mediated mutations which were generated in F0 mice, eliminating the need for performing genetic crosses. Brophy et al. replicated the GREB1L mutation found in the Iowa family and generated mice with a variety of kidney development phenotypes spanning the range observed in the family. This suggests that there is developmental flexibility in how much GREB1L expression is needed to make one or two healthy kidneys. Furthermore, the use of CRISPR to generate mice that mirrored human phenotypes demonstrate how this technology can be used to quickly model idiosyncratic human mutations to better understand the causes of conditions like renal agenesis.

CITATION:

A Gene Implicated in Activation of Retinoic Acid Receptor Targets Is a Novel Renal Agenesis Gene in Humans

Patrick D. Brophy, Maria Rasmussen, Mrutyunjaya Parida, Greg Bonde, Benjamin W. Darbro, Xiaojing Hong, Jason C. Clarke, Kevin A. Peterson, James Denegre, Michael Schneider, Caroline R. Sussman, Lone Sunde, Dorte L. Lildballe, Jens Michael Hertz, Robert A. Cornell, Stephen A. Murray and J. Robert Manak

GENETICS September 1, 2017 207: 1 215-228; https://doi.org/10.1534/genetics.117.1125

http://www.genetics.org/content/207/1/215

There’s no such thing as an obese plant. But that doesn’t mean plants can’t teach us something about fat. In the September issue of GENETICS, Ducos et al. show that a protein that controls fat accumulation in humans has a similar function in Arabidopsis. They also find that the human and plant proteins may be regulated in similar ways, indicating that the pathways controlling fat deposition have deep evolutionary conservation.

In humans, the highly conserved gene WDTC1 has a well-established link with body fat content. It controls the number of fat cells present, and genetic variation in WDTC1 is associated with obesity. The encoded protein acts as a substrate adaptor protein for a ubiquitin ligase complex. It is made up of several characteristic repeat structures, which Ducos et al. noticed were similar to the structure of the Arabidopsis protein ASG2. This gene was known to regulate seed germination in some way, but it was not known whether it was an ortholog of WDTC1.

To confirm the relationship between these plant and animal genes, the researchers first looked for other highly similar proteins in existing sequence databases. A phylogenetic analysis showed WDTC1 and ASG2 cluster together among all the examined sequences from plant and animal, but rarely with fungi groups. ASG2 is also widespread among plants and is found in rice and other diverse species, supporting a very old origin for this gene. The structural similarity of the introns and exons of the plant and animal genes further suggest they arose from a shared common ancestor.

But does the plant version of this gene still function in fat regulation? To test this idea, the researchers knocked out ASG2. Seeds made by these mutant plants were heavier and contained higher levels of monounsaturated fats. They also had denser networks of oil bodies, the structures where fat is stored. These mutant seeds were essentially “obese,” showing that the animal and plant proteins are not only structurally similar, they have similar functions. Some other shared aspects of the structure of these two proteins even suggest that they share the same downstream binding partners, though more work is needed to confirm this possibility.

The remarkable functional conservation of this obesity-linked gene suggests the pathway plays a crucial role in physiology. But fat and oil accumulation are important for more than human health; increased seed fat levels could prove a major boost to crops being bred for biodiesel and food oils.

CITATION

Remarkable Evolutionary Conservation of Antiobesity ADIPOSE/WDTC1 Homologs in Animals and Plants

Eric Ducos, Valentin Vergès, Thomas Dugé de Bernonville, Nathalie Blanc, Nathalie Giglioli-Guivarc’h and Christelle Dutilleul

GENETICS September 1, 2017 vol. 207 no. 1 153-162; https://doi.org/10.1534/genetics.116.198382

http://www.genetics.org/content/207/1/153

When a group of microorganisms needs to stick together, they build a biofilm. The cells cement themselves together onto a surface, forming durable structures that are notoriously hard to remove. In a medical setting, biofilms can contribute to dangerous antibiotic resistance. In the August issue of G3, Cromie et al. use a yeast model to identify six genes involved in biofilm formation, providing new leads for methods to combat biofilm-forming pathogens.

When forming a biofilm, microorganisms have to transition from a solitary lifestyle to one anchored down and surrounded by companions. This requires substantial changes in gene expression. To identify genes whose regulation contributes to biofilm formation, the authors used a genome-wide overexpression screen in a yeast strain that naturally forms biofilms. They examined 4,600 colonies that had been transformed with plasmids carrying individual yeast genes, looking for a change from the “fluffy” biofilm colony structure to the smooth unstructured globs typical of laboratory yeast. In the end, they identified six genes whose overexpression disrupted biofilm formation, causing colonies to change from fluffy to smooth.

Interestingly, all six of these genes are regulatory proteins—and four of them are transcription factors. To investigate the gene network changes triggered by overexpression, the authors performed RNA-seq on the six transformed yeast strains. They found evidence that the same transcriptional network was activated by overexpression in five of the six strains, and this network contains some genes already known to affect colony morphology.

Expression levels of these regulators may act as a switch that yeast that can use to easily transition between growing in a solitary state and growing as a biofilm. Targeting these switches might be one way to prevent pernicious biofilms from forming in the first place.

CITATION

Transcriptional Profiling of Biofilm Regulators Identified by an Overexpression Screen in Saccharomyces cerevisiae

Gareth A. Cromie, Zhihao Tan, Michelle Hays, Amy Sirr, Eric W. Jeffery and  Aimée M. Dudley

G3: GENES, GENOMES, GENETICS August 1, 2017 vol. 7 no. 8 2845-2854; https://doi.org/10.1534/g3.117.042440

http://www.g3journal.org/content/7/8/2845

Even though domestic plants usually appear radically different from their wild relatives, they are often still able to interbreed. For transgenic crops carrying traits like herbicide resistance, this flexibility could pose a problem if they were to pollinate weedy relatives nearby. In the July issue of GENETICS, Adamczyk-Chauvat et al. examine the extent to which alleles from cultivated oilseed rape introgressed into the genome of wild radish after several generations of cross-pollination. They found that certain regions of the genome were more likely to be passed from domestic to wild plants, suggesting that targeting transgenes into specific genomic locations could limit their putative escape into weeds.

Oilseed rape, also known as rapeseed, is a brilliant yellow flowering plant cultivated around the world for its oil-rich seeds, which are used for animal feed, biodiesel, and edible vegetable oil. Oilseed rape is one of the main transgenic crops worldwide, carrying genes introduced mainly for herbicide resistance. The crop has a very low rate of natural hybridization with wild radish, a weedy relative, but the transfer of herbicide resistance to weeds could pose a major economic problem for farmers.

Wild radish flowers. Photo Brenda Dobbs via Flickr.

The authors grew a population of oilseed rape plants in a natural field environment and pollinated the crops with wild radish. They continued pollinating the hybrid offspring with wild radish for five generations. To trace how the genomes of these two species had merged in the fifth-generation plants, the team chose 307 hybrids with a chromosome number close to the one of wild radish. They identified oilseed rape alleles in each of these hybrids by genotyping a set of markers spanning the genome and found around half the individuals had a detectable oilseed rape allele. However, nearly 70% of markers carried mainly the radish allele—there was a very low probability of introgression (p₁=0.003) at those sites.

In contrast, a few markers were frequently introgressed; interestingly, these markers were not found in random locations but were grouped together in specific regions of the genome. The two most highly introgressed markers were located together on one chromosome and adjacent to several other less frequently introgressed markers. They also found that 30% of introgressed markers were located on the ends of chromosomes rather than in the middle.

These patterns likely arise from the differences in genome structure between oilseed rape and wild radish. Though they can hybridize, they have different chromosome numbers. This striking pattern suggests that using a precisely targeted genome editing method like CRISPR to insert transgenes in select genomic regions could be a possible strategy for keeping them confined to the cultivated species, though more work is needed to see how widely applicable this finding might be.

CITATION:

Gene Introgression in Weeds Depends on Initial Gene Location in the Crop: Brassica napus–Raphanus raphanistrum Model

Katarzyna Adamczyk-Chauvat, Sabrina Delaunay, Anne Vannier, Caroline François, Gwenaëlle Thomas, Frédérique Eber, Maryse Lodé, Marie Gilet, Virginie Huteau, Jérôme Morice, Sylvie Nègre, Cyril Falentin, Olivier Coriton, Henri Darmency, Bachar Alrustom, Eric Jenczewski, Mathieu Rousseau-Gueutin and Anne-Marie Chèvre

GENETICS July 1, 2017 vol. 206 no. 3 1361-1372; https://doi.org/10.1534/genetics.117.201715

http://www.genetics.org/content/206/3/1361

Epilepsy is characterized by recurrent seizures, often with no immediately obvious cause. In the August issue of G3, Ferland et al. use a genome-wide association study in mice to show that after multiple seizures, the genetic basis of seizure variation shifts from previously identified genomic regions to new ones. This research shows that the genetic causes of initial seizures and the progression of epilepsy may be distinct.

Having a seizure changes brain physiology and alters its cellular environment, which can affect how genes are activated. To identify genetic regions that are involved in repeated seizures, the authors used fifty-eight genetically variable mouse strains. For eight consecutive days, they exposed these mice to a seizure-inducing chemical and measured how long it took the mice to have a generalized seizure. As expected, this time decreased as the days went on. They then let the mice rest for twenty-eight days before inducing a final seizure.

The team then identified variants associated with the time to seizure onset. Mice with the shortest time before seizure onset on the first day carried variants at Szs1 and Szs6 that had been previously associated with seizure susceptibility. However, on each passing day, these associations grew weaker. On day three, a new association with time to seizure began to appear, growing stronger until it reached statistical significance on day six. This locus was also associated with seizure susceptibility after twenty-eight days. Clearly, the genetic factors involved change along with brain physiology after repeated seizures.

They named the newly identified locus Epileptogenesis susceptibility factor 1, or Esf1. It may have a function related to calcium signaling in the brain, which is an important molecular mechanism in epilepsy progression. After repeated seizures, the brain tries to adjust itself to increase neuronal stability. Understanding this plastic response and the genes that control it may one day lead to novel epilepsy treatment and management options.

CITATION

Multidimensional Genetic Analysis of Repeated Seizures in the Hybrid Mouse Diversity Panel Reveals a Novel Epileptogenesis Susceptibility Locus

Russell J. Ferland, Jason Smith, Dominick Papandrea, Jessica Gracias, Leah Hains, Sridhar B. Kadiyala, Brittany O’Brien, Eun Yong Kang, Barbara S. Beyer and Bruce J. Herron

G3: Genes, Genomes, Genetics August 2017 7: 2545-2558

https://doi.org/10.1534/g3.117.042234

http://www.g3journal.org/content/7/8/2545

Genes involved in male reproduction tend to evolve rapidly. This has been observed in many different species and is thought to be due to sexual selection as males compete over mating opportunities. But in the August issue of GENETICS, Whittle and Extavour present results that flip this paradigm upside down. They find that in the yellow-fever mosquito, female-biased genes expressed in the ovaries evolve faster than their male counterparts. This fascinating break from the trend could be due to increased competition between females for mates, adaptive evolution during egg-sperm attraction, and/or limited sperm competition in this species.

First, the authors identified genes more highly expressed in either male or female gonads using whole transcriptome data. They found that ovarian-biased expression was typically due to elevated expression in females, not just reduced expression in males as has been observed in other species. They then identified nucleotide changes that altered the protein composition in these genes to compare the rates of protein evolution. Although a small subset of testis-biased genes were evolving rapidly, on average transcripts with ovary-biased expression showed a significantly higher protein evolution rate than those with testis-biased expression. Genes expressed only in the ovaries had the fastest protein evolution rate of all. They determined that the rapid evolution of some of these genes is most likely due to positive selection using a phylogenetic analysis including two other mosquito species.   

Interestingly, members of this set of rapidly-evolving, ovary-specific genes have functions preferentially related to the mosquito’s olfactory system, including odor molecule binding and smell receptor activity. Olfactory signaling appears to be important for mosquito mating; groups of males will gather together and swarm females, who are lured over by their scent. These types of chemical cues may also be important for guiding the sperm to the egg or directing females to store sperm after mating. Like some other insects, female mosquitoes have special storage organs that allow them to keep enough sperm from a single mating to fertilize all their eggs throughout their entire lives. There may be strong selective pressure on proteins that drive evolution of these critical reproductive functions.

The yellow-fever mosquito’s mating system is likely behind its unusual rapid evolution of ovary-biased genes. There might be competition between females to attract males or male mate choice, which could result in strong sexual selection on ovary-expressed genes involved in chemical sensing. These females usually mate once, and the male deposits a “copulation plug” in the female’s reproductive tract. This physical and chemical barrier prevents the sperm of another male from passing through, ensuring the first male will father all her offspring. The plug cuts off nearly any chance of competition between the sperm of multiple males, in contrast to many other organisms where the rapid evolution of testis-biased genes could be due to the pressure of this sperm competition arms race.

A combination of these diverse factors likely influences the rapid protein evolution of ovary-biased genes in yellow-fever mosquitoes. These results offer a fascinating glimpse into how ecology and reproductive lifestyle can affect genome evolution and illustrate how there are notable exceptions for every trend observed in nature.

CITATION:

Rapid Evolution of Ovarian-Biased Genes in the Yellow Fever Mosquito (Aedes aegypti)

Carrie A. Whittle and Cassandra G. Extavour

Genetics August 2017 206: 2119-2137

https://doi.org/10.1534/genetics.117.201343

http://www.genetics.org/content/206/4/2119

In the 1940s, C. H. Waddington discovered a peculiar phenomenon in fruit flies: traits could appear in response to environmental stress in an individual’s lifetime and then be passed down to future generations. Waddington proposed that this wasn’t the inheritance of acquired traits, but actually due to pre-existing genetic variation that had no effect until the flies were stressed. In the August issue of GENETICS, Fanti and Piacentini et al. revisit Waddington’s famous experiments with modern sequencing technology to show that this phenomenon can also be driven by newly arising DNA mutations.  

The authors followed Waddington’s fairly simple experimental framework: take flies from natural populations and expose them to high temperatures for a short time during pupation. Observe the adult phenotypes, and then do it all again the next generation. Repeat this process until strange phenotypes emerge following the heat shock treatment.

The authors focused on four phenotypes caused by well-studied mutations in fruit flies, including sepia eye color and forked bristle mutations. These phenotypes began to appear after heat shock treatment between four and twelve generations after the experiment began. The scientists then selected for a phenotype by crossing the flies displaying it. They repeated this procedure until the phenotypes appeared not only after heat stress but were stably maintained in regular conditions.

They confirmed the presence of genetic mutations in the heat shocked mutant stocks using DNA sequencing. In the fixed mutant stocks, they found clear genetic causes of the mutant phenotypes. In two cases, a deletion mutation disrupted the protein coding sequence, and the other two genes carried transposable element insertions. None of the mutations were present in the genomes of the parental flies; these mutations were new, arising during the course of the experiment. Clearly, they were what allowed the phenotypes to be maintained stably without heat stress.

These results show that Waddington was wrong: inheritance of the abnormal post-heat shock phenotypes was not due to cryptic variation present in the parent lines. A different mechanism must be responsible. Heat shock stress may cause double-stranded DNA breaks, which can lead to deletions like the ones observed here. It may also activate transposable elements since the authors found higher levels of transposable element transcripts in flies that had been heat shocked.

But how could these new mutations appear in the same genes that were initially disabled by heat shock? How can a stress-dependent phenotype become genetically encoded? The authors suggest epigenetic changes may be involved. Heat stress could result in epigenetic alteration of particular loci to change their expression, leading to the observed heat-dependent phenotypes. This same epigenetic activity may make this stretch of DNA more susceptible to mutation, which could be very likely in the face of heat shock-induced double-stranded breaks or activated transposable elements.

If this model holds true, it could have serious evolutionary implications. In the wild, plastic traits like these post-heat shock phenotypes are often adaptive and help organisms survive in difficult, changing environments. It has been proposed that under natural selection such environmentally-induced traits can eventually become genetically encoded. Called the Baldwin Effect, this process illustrates how heritable behaviors like the human capacity for language might evolve. The results of this study provide a viable mechanism for this powerful evolutionary phenomenon.

CITATION

Canalization by Selection of de Novo Induced Mutations

Rice is one of the most important food crops on earth. Like many other plants, the genome of this critical global species is dominated by transposable elements—selfish genes that multiply themselves to the detriment of their host. In the June issue of G3, Zhang and Gao analyze the genomic long terminal repeat (LTR) retrotransposon content of the two species of cultivated rice and their six closely related wild relatives. They find that these LTRs have multiplied in a lineage specific manner and suggest that the unique activity of retrotransposons in these rice species has contributed to their diversification and isolation.

The eight rice species included in this study present an excellent opportunity to investigate the dynamics of LTR retrotransposons during speciation. The common ancestor of this group lived only 5 million years ago, and each member has whole genome sequence data available. For each species, Zhang and Gao extracted the LTR retrotransposon sequences from each genome and sorted them into 790 families based on homology. For very highly similar LTRs, they were able to estimate genome abundance from the sequencing read count. They then used homology between LTR retrotransposon families to estimate the time of their origin, placed in the phylogenetic context of the rice species group.

Overall, they found that the LTR retrotransposon content of the genomes varied greatly among different rice species and that genome size varied due to repeat content. However, almost all of the families with very high copy numbers were found in all of the species, indicating they were present in the most recent common ancestor. It is possible that even the newer LTR retrotransposon families were present in the ancestor, but as these sequences are very fast evolving, they may have diverged past the point of recognition. Extremely highly amplified LTR retrotransposon families also generally have much shorter periods of activity than families with fewer overall sequences.

There is clear evidence that lineage specific LTR retrotransposon activity has shaped the genomes of these groups. Certain families of transposable elements have high copy numbers in only one or a few closely related species, indicating that bursts of retrotransposon activity occurred after the split between these groups. Notably, there is also a difference in LTR retrotransposon content and activity between domesticated African rice and its wild progenitor, which split about 260,000 years ago. Such differences between lineages may indicate that LTR retrotransposon activity is associated with changes in environment or life history. Certainly, they have helped rapidly shape distinct genomes in these diverging species.

CITATION:

Rapid and Recent Evolution of LTR Retrotransposons Drives Rice Genome Evolution During the Speciation of AA-Genome Oryza Species

Qun-Jie Zhang and Li-Zhi Gao

G3: Genes, Genomes, Genetics June 2017 7: 1875-1885

https://doi.org/10.1534/g3.116.037572

http://www.g3journal.org/content/7/6/1875