Generations of biology students have been taught that the instructions for creating an organism are completely contained in its DNA. The sequence of our DNA, the order of letters representing its chemical constituents, is what we inherit and in turn pass down as the template our cells use for creating the tools they need to carry out their functions. It is reassuring, really, to know that modifications we make to ourselves that don’t change our DNA sequence won’t be passed on. For example, our children will not, thankfully, be born with any of our tattoos, piercings, or other physical manifestations of our youthful indiscretions.
However, our DNA sequence is only a tiny part of what makes us who we are. What may be even more important is how our cells’ use of DNA is regulated, a process that is still mysterious. For example, at conception an embryo receives genetic information from both parents, some of which may give conflicting instructions. How does it decide which to use during development? Additionally, since almost all cells in an organism have the same DNA, how do some turn into fingernails and others into a spleen? Finally, many organisms (including humans) have vast stretches of DNA that either have no known function or that can be harmful (some of those harmful bits are called transposons, elements that hop around the genome indiscriminately, potentially destroying important genes in the process). So, how does a cell find and express only the DNA relevant to its role in an organism?
The answers to these questions lie in epigenetics (meaning “above genetics”)—the study of phenomena that regulate gene expression without altering the underlying DNA sequence. One crucial component of this kind of regulation is methylation, the attachment of a methyl group (represented as “CH3,” one carbon and three hydrogen atoms) to cytosine, one of the chemical constituents of DNA. DNA methylation was originally thought primarily to turn genes off, a process key to development (choosing between genes from mom and dad), cell differentiation (turning off fingernail genes in the spleen), and healthy cell function (inactivating those pesky transposons). Levels of methylation can change over time under the influence of environmental factors. If methyl groups are added to the wrong bits of DNA (or are absent in tumor suppressor genes), they can lead to cancer.
The pattern of DNA methylation found in a genome is known to be heritable in plants, and there are tantalizing hints that the same is true for humans. Methylation is thus like a genetic tattoo, but one that actually does get passed on to offspring—an acquired characteristic that is heritable. However, this seemingly essential form of DNA regulation has some mysterious properties. For example, it does not always act to silence genes—it can also be found in the middle of genes that are turned on and being actively expressed, where it causes muta- tions. It is also stunningly absent in some of biology’s most beloved model organisms, including the worm Caenorhabditis elegans, the fly Drosophila melanogaster, and the yeast Saccharomyces cerevisiae.
Because of these puzzling inconsistencies, the general importance of DNA methylation tended, until quite recently, to be discounted. The data, “seemed like a real mess,” says Daniel Zilberman, professor of plant and microbial biology at UC Berkeley. “The roots of these phenomena are not understood. We know they happen. How they happen is a mystery.” He himself had found DNA methylation in the actively expressed genes of his favorite model organ- ism, the mustard weed Arabidopsis thaliana, which “suggested that DNA methylation had a function we didn’t know about.” But because of the phenomenon’s lack of uniformity throughout the different kingdoms of life, he didn’t know if he could extrapolate his results to other organisms. Thus, the Zilberman lab set out to untangle the roles and patterns of DNA methylation, attempting to address two major problems. First, there was no big picture understanding of how the use of methylation is related between different organisms, or if it is related at all. Second, there was no explanation for why methylation is present both in genes that need to be silenced and in active genes.
The team’s key idea was that in order to properly investigate the pattern of DNA methylation between organisms, they would have to unravel the evolutionary context of the phenomenon. Therefore, they decided to study methylation in organisms at different points along the evolutionary tree. They evaluated 17 selected organisms with sequenced genomes: five plants, seven animals, and five fungi. Their menagerie included anemones, moss, green algae, puffer fish, honeybees, and rice. To quantify the methylation in each genome, the lab used a technique called deep bisulfite sequencing, which converts unmethylated cytosine into uracil (a genetic building block normally only found in RNA), while leaving methylated cytosine untouched. The methylated sites can then be read from the altered sequence. DNA methylation was correlated to gene expression by looking at messenger RNA levels (the presence of messenger RNA for a gene indicates that it is actually expressed). This correlation was important for investigating whether methylation has a role in turning genes off in a particular region of DNA.
Even though they only needed less than one microgram (about two billionths of a pound) of tissue from each subject, obtaining samples of all these organisms was “a real adventure,” says Assaf Zemach, a postdoctoral fellow in the lab, whose face lights up when describing his new collection. Learning to grow new organisms like algae was challenging but fun, whereas the animals, unfortunately, had to be sacrificed. “Some I had to kill as soon as I received them,” he says, referring to the bags of swimming fugu fish and the boxes of honeybees that showed up in the mail, “but the rest…[now] I just grow them for fun in lab.” Investigating such a diverse array of organisms paid off, as their results showed a coherent pattern of DNA methylation for the first time. While methylation is mostly absent in some classes of organisms (invertebrates and fungi), it is surprisingly consistent in others (plants and animals). Invertebrates and fungi show no Rice was chosen for study because itis only distantly related to Arabidopsis, providing a useful point of comparison, and because of its importance as a food crop. It also has the smallest genome of any cereal, simplifying the analysis. methylation of transposons (those bits of harmful DNA), but some methylation of active genes. On the other hand, plants and animals consistently show methylation of both transposons and active genes.
One of the most surprising results was that methylation of active genes appears to be an ancient property that can be traced all the way back to anemones and down the plant lineage. It was most likely present in the last common ancestor of plants and animals. Genes transcribed into mRNA at a modest level are most likely to be methylated while genes transcribed at high or low levels are least likely to be methylated, a parabolic relationship. Essentially the same kinds of genes tend to be methylated across plants and animals. “We are looking at a conserved phenomenon,” says Zilberman, “so we can use Arabidopsis to look at DNA methylation.” What they learn from this plant is relevant to other organisms, including humans.
While the role of DNA methylation in active genes remains a major unsolved problem, its function in gene silencing is becoming clear, and it all comes down to sex. Those nasty bits of harmful DNA, transposons, require sexual reproduction with another individual in order to spread, so organisms that reproduce asexually don’t need to worry about them. Plants and animals are obligate sexual outcrossers, which means that they have to reproduce through sex with another individual, whereas fungi mostly reproduce asexually. Invertebrates do sexually outcross; however, Zilberman postulates that invertebrates lost the ability to use DNA methylation to silence transposons early in evolution when they were still single-celled organisms that reproduced asexually. Ultimately, this study underscores the importance of studying complicated biological phenomena through the lens of evolution. As Zilberman says, “The trend toward using a few model organisms was good and important, but organismal biology that provides an evolutionary context is also very important because biology doesn’t make sense if you don’t understand its evolution.”