That "eukaryotic cells are more complex" and "compartmentalized" are used to justify the need of more level of control of gene expression. I get the basic idea but can't convince myself why complexity or compartmentalization leads to more complex control (i.e. multiple levels) of gene regulation!

In control theory, we often use multiple level of control to minimize the variance (there are few other benefits as well from design perspective); probably that is what eukaryotic cell would also do. Moreover, simple combinatorics suggest that two level of control with $n_1$ and $n_2$ molecules involved at each level respectively would give me $n_1n_2$ possibilities of output which is higher than $n_1 + n_2$ distinct output I might get if I were to use these molecules with 1 level of control/regulation.

I was wondering if such simple reasoning are correct and can be used in the context of evolution? Are there other reasons as well?

  • $\begingroup$ There are additional levels of regulation in eukaryotes such as alternate splicing, miRNAs, long distance chromosomal interactions (enhancers/silencers), epigenetic regulation etc. Please add details as your question seems both a bit unclear and broad. $\endgroup$ – WYSIWYG Jun 1 '15 at 9:17
  • $\begingroup$ @WYSIWYG I rephrased the question to make is less broad. I may still be unclear since I am poor in Biological vocabulary/terms. $\endgroup$ – Dilawar Jun 1 '15 at 9:21
  • 1
    $\begingroup$ Some gene expressions need to response quickly by stimuli. In this case, cells may degrade mRNAs or proteins rather than inhibit gene promoters. Even if cells need proteins in some small areas of cells, cells could not turn down the promoter activity of the gene. Instead cells control translocation or degrade mis-localized proteins. Space and time are also important. Multiple ways of regulations make cellular activities possible. $\endgroup$ – 243 Jun 1 '15 at 10:21
  • $\begingroup$ You don't want all genes expressed at all timepoints. Some need a tight control, while other shall react quick upon stimuli. This is only possible with a complex regulation. $\endgroup$ – Chris Jun 1 '15 at 11:40

I think you need to consider the overall amount of information that is needed to implement a proper control system in living organisms.

In prokaryotic cells, in most of the cases, you will find that one gene code for one enzyme, so if a simple organisms need 500 enzymes to stay alive you can expect 500 genes that code for that enzymes. A more complex organism may need 100 times more enzymes to stay alive and so theoretically 100 times more genes. Now, evolution wise, to have a linear increase of DNA vs complexity is not a good option for very complex organisms. To keep everything compartmentalised in one single membrane is also not a good idea if you need to run millions of chemical reactions using thousands of enzymes that can cross react and need different optimal conditions to work properly, so multi-compartimentalization is a must. Linked with it arises the problem of sending the right enzyme in the right place and this has been solved by adding tags, transporter enzymes, etc... in the pot... So, lots of more information is needed to make everything run smoothly.

It turned out that an efficient way to code for all that enzymes, tags, promoters and other regulators without increment the size of the DNA too much is by multi level control of the expression of a gene i.e. the same gene can be used to produce different enzymes (splicing etc...). The tradeoff to pay to have a more compact genome is to increase the complexity of its control system.

Lastly, consider that differently from normal electronic controllers, the biological one do affect each other. A biological controller system may produce an output molecule that interact with the system under control but also with other reactions going on all over... that is another burden that require an increase of complexity in the control system to make it work.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.