Beyond Nature or Nurture: Bringing the Noise

As an experimental biologist I like to think of myself as the god of my cultures. I am sovereign over my empire of petri plates and test tubes, I can make the cultures isogenic and the environment they are exposed to identical, but I can not and I will never get them to all fall in line and behave identically. There is always an inherent and extremely significant variation between cells. This is a long observed phenomenon called cellular noise AKA stochastic gene expression that has only become critically examined within the last decade or so. Noise is seen everywhere – genetically identical cells act disparately in homogeneous environments. Why?

Noise is observed to be influenced by two discrete catagories, extrinsic and intrinsic factors. In 2002 Elowitz et al inserted two copies of the same promotor into E. coli, one expressing cyan fluorescent protein and one yellow fluorescent protein. Whenever gene expression varies in concert it’s attributed to extrinsic factors, such as variation in RNA polymerase or ribosome avaibility, or fluctuations in other cellular components or general cell state. When the expression of the two genes varied independantly it’s due to intrinsic factors, variation due to the biochemical properties of transcription and translation, which often depend on molecules found in such small quantities that stochastic effects are prominent. Later experiments revealed that variations due to extrinsic factors flucutate at a time scale roughly equivelent to a cell cycle, while intrinsic variation degrades rapidly.

Here’s where it gets really cool: This noise can’t be thought as just some affliction of a cold and otherwise deterministic universe on cells, some property washing chaotic randomness over helpless and captive life.  It has a function, it’s been shaped by the same forces of natural selection that created life itself, and it contributes to the fitness of organisms.

First of all the noiseness of a gene correlates with its function. In the experiment I described above if  using different promotors results in different magnitudes of variation. Genes encoding dose-dependent proteins or proteins involved in multicomponent complexes don’t vary much, while genes responsible for pathogenicity, responding to changes in the enviroment or stress are very noisy

Noise enables a population of isogenic cells to create phenotypic diversity, which can be useful. Bacteria have two basic strageties for dealing with change or stress; 1) Sensing and adapting to the changes in their enviroment and 2) Maintaining a constant pool of phenotypic variation at all times. Hard times come? There are bacteria already in a low metabolic state. A sudden introduction of a specific sugar or virus? Some lucky cell has already been building up the enzymes for it. Empirical evidence shows that this stragety is superiour in changing enviroments. I wonder if in our lab, by constantly exposing strains to alternatively high nutrient or starvation medium, we have selected for a level of metabolic switching(or even -80 preparation) noise that is different than WT?

Now some of you may be thinking, ‘that previous statement makes no sense, in an isogenic population there’s nothing for selection to act on(caveats about mutation), so inheritable change is impossible’, but I say not so fast asshole(although my ponderance is flawed for other reasons). Noise has an epigenetic component. Yes, it just got that awesome. Genes can flip into an on or off state stochastically, and this state can be inherited. The relationship of epigenetics with noise is a pretty new idea and there wasn’t much info I could find on this.

Noise is useful in devolpment aswell. Consider a fetus. Let’s say it aspires to become a big strong lifeform, with many sharp senses with which to perceive the world. If it wants to smell the challenge is that while it may have thousands of genes encoding olfactory receptors, for specificty a neuron can only express one of these receptors. How can the fetus ensure that it builds a brain full of neurons expressing all those different olfactory receptors? Using stochastic gene expression each neuoron can randomly expresses a single olfactory receptors, the expression of which blocks the expression of all other receptors. By having a larger pool of neurons than receptors high coverage can be acheived. While we usually think of our brains as only being able to have been created by a highly complex deterministic process it turns out atleast a part of who we are is a total crapshoot. Granted we’re only talking about smell, but still, take from that what you will.

Randomness can be a problem, there are many cell processes that need to be kept steady. The circadian rhythm is a great example. Recent studies have found that as we get older the general level of noise in our cells increases, even of housekeeping genes. What does that mean? People speculate that this increase in noise may contribute to why we deteriorate as we age, but it’s all pretty bleeding-edge science at this point

Usually we can ignore all this noise business in our lab because we work with such large numbers the effect averages away, but there is one instance I know of where the effect appears. The bacteria we mainly work with, H. influenzae, is naturally competent. Under certain conditions it takes up DNA and uses it for recombination or nutrition(tangent: this last point is controversial, maybe I’ll make a post about that when I’m feeling braver). However when we put genetically identical clones in a homogenous solution only a tiny fraction of the cells will become competent and transform. This is most readily apparent when studying ‘congression’, the insertion of two unlinked markers at the same time. If lets say the transformation frequency of marker A is 10% and the transformation frequency of B is 10%, and all our cells had a noiseless phenotype we would expect to see cells that transformed both markers at a frequency of 1%

In this scenario the real observed frequency of double transformants would be somewhere around 6% instead of the 1% we predict if noise didn’t exist. In our culture of identical bacteria grown under identical conditions, some don’t transform at all and others are highly transformable. Very Noisy. What does that speak to about the purpose of competence? Thanks to our labs post-doc for tipping me off on this subject, I’m already learning so much.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: