Thinking outside the gene

  Our DNA contains the code that builds the bodies we call ourselves. These days, we are used to hearing about genes: phrases of DNA, read out by cellular machinery to construct the components of our bodies. We are used to the idea that mutations in our genes, changes or mistakes in the code, can make people sick. But the code written into our DNA is not as static or inflexible as we might imagine and it is not only your genetic sequence that has an effect on your physical traits (phenotype). Cells have layer upon layer of processes that control when and how much a gene is expressed, introducing complexity at multiple levels. Not only (as it often seems) to frustrate scientists, but rather to confer the redundancy, flexibility and robustness that allow development and survival to continue in the face of environmental change. One group at Columbia University is now looking at the role played by these extra levels of regulation in age-related memory loss. The reason some people experience memory loss in old age and others don’t, may have nothing to do with which genes you have. Rather, the difference may lie in how and when your cells express those genes.


We have a storage problem. At the risk of repeating a decades-old factoid, the DNA contained within a single cell is around 2 metres long. The average diameter of a human cell is 10 micrometres giving a shortfall of space in the order of two hundred thousand. Somehow all that DNA has to fit inside the cell, and histone proteins are the contortionists that make that possible. By winding DNA around itself, then around histone proteins, then winding those around each other, then winding that again a few more times, our cells can cram in all the DNA necessary to code for everything that makes us human. But now we have a new problem. If the code we need is in the middle of a tangled mess of other code and wrapped around bulky proteins, which are then crammed together even further, how can that code be accessed? This is where epigenetics comes in. Epigenetics is a rather vague term used to describe a whole host of strategies used by cells to regulate the expression of genes. But why do cells have to regulate gene expression? And how does that relate to the problem of genetic storage? Every single cell in your body* contains all of your genetic information. In other words, a single cell in your skin (or anywhere else, for that matter) contains all the information necessary to make any other cell in the body and, theoretically, could be reprogrammed to become any other cell. But a skin cell has no use for, say, proteins used to send nervous impulses, and can exploit this position of limited need to tackle the problem of genetic storage. Cells don’t need to access the entire genetic code all of the time. There are things in there, for example, that are used when we’re developing in the womb, but have no function once we’re out in the wide world. These genes, then, can be archived – set aside to be passed on to our offspring for their in utero development. By selecting which genes are buried away and which are close to the surface, ready to be decoded, the cell can perform efficiently and still house the entire human genome.


Figure from


The amount of control imparted by epigenetic mechanisms is only just beginning to be appreciated. Perhaps from fear of a return to Lamarckism, there was a reluctance in the scientific community to attribute heritable changes to anything other than mutations in DNA. However, we now know that differences in phenotype can be the result of processes other than changes in genetic sequence. These epigenetic mechanisms have been shown not only to influence an organism’s phenotype, but also to have the capacity to be inherited by offspring. That is to say that two organisms can have a different phenotype, not because their genetic sequence is different, but because their parents regulated the expression of that gene in different ways. One highly visual example of this is the Agouti mouse, in which the coat colour of the offspring can be influenced by supplements given to the mother during pregnancy. Expose the mother to bisphenol A (BPA) and her offspring are more likely to be yellow. Without BPA, they come out brown [1].


Figure modified from reference 1.


In this recent paper on memory loss [2], the authors wanted to look at what causes age-related memory loss and how it differs, if indeed it does differ, from Alzheimer’s disease. Previous studies have suggested that Alzheimer’s primarily affects an area of the hippocampus called the entorhinal cortex. In contrast, normal ageing (which is also associated with memory loss) involves changes in a different part of the hippocampus – the dentate gyrus [3]. With this in mind, the authors took brain tissue from post-mortem samples of healthy people to look for differences between the entorhinal cortex and the dentate gyrus. They looked at changes in gene expression that were associated with age by measuring how much of each gene was being expressed in each brain region and matching expression level to the age of the person. One difference they saw was in the dentate gyrus, which showed a large, age-related decrease in the expression of an enzyme (RbAp48) that modifies histone proteins. These, remember, provide a scaffold for DNA and help to determine which genes are accessible and which are archived. This finding suggested that age-related memory loss may not be the result of a person having a defective gene, but rather the result of incorrect genetic archiving. As is usual in this kind of study, they turned to a mouse model to look at this enzyme in more detail. By breeding mice unable to make RbAp48, they were able to show that this enzyme is necessary for normal memory: mice lacking RbAp48 performed worse on memory tests (navigating a maze or recognising an unfamiliar object). As mice get older, their memory appears to deteriorate based on tests like this, and mice lacking RbAp48 experienced this deterioration at a younger age than mice with normal levels of RbAp48. When looking at human brains, the decrease in RbAp48 wasn’t seen in the area of the brain associated with Alzheimer’s disease, suggesting that age-related memory loss has a unique starting point and is not just an early sign of Alzheimer’s. This could have important consequences in the future for diagnostics.


The more we learn about epigenetics, the more obvious it becomes that there is more to go wrong than we thought. You not only need the right genes, but you need the right control mechanisms in place to make sure have the right amount of each gene in each cell at all times throughout life. At the same time, we know that most people manage this, reflecting the amazing robustness of the system. Increasing our understanding of these control mechanisms has implications for treatment too. By looking at the underlying cause of a disease, we can treat it more effectively. This has been going on for decades in infection research, but may be applied more to other diseases in the future. For example, two patients presenting with fever and breathing difficulties will be tested for pneumonia. One may have a fungal infection and the other a bacterial infection. These need to be treated very differently, but only a knowledge of the underlying cause can tell us how to treat each patient. Similarly, treatment may be very different for someone lacking a gene completely compared with someone who has the gene in an inaccessible place. Both patients would have the same symptoms, but an analysis of the underlying causes could completely change the nature of the treatment. It is this sort of personalised diagnosis that could help to provide the right treatment for a patient; which would not only help the patient recover more quickly, but could also help to reduce the amount of money wasted on ineffective treatments.


*There are a few notable exceptions. Red blood cells have no nucleus and contain no genetic DNA. Egg/sperm cells have half the amount of DNA as the rest of your cells to make sure an embryo has the correct amount after fusion.


Jargon box

Histone: a type of protein used as a scaffold for DNA. DNA molecules wind themselves around histones to reduce the amount of space needed to house the genome.

Phenotype: observable characteristics of an organism from visible traits e.g. hair colour to cellular traits e.g. cell shape or structure.



1)    Dolinoy et al. Maternal nutrient supplementation counteracts bisphenol A-induced DNA hypomethylation in early development. Proc Natl Acad Sci U S A. (2007) 104 (32): 13056–13061. Link. OPEN ACCESS!

2)    Pavlopoulos et al. Molecular Mechanism for Age-Related Memory Loss: The Histone-Binding Protein RbAp48. Science Translational Medicine (2013) 200 (5): 200. Link.

3)    Small et al. A pathophysiological framework of hippocampal dysfunction in ageing and disease. Nat. Rev. Neurosci. (2011) 12: 585–601. Link. OPEN ACCESS!