Flagella Regeneration – Experimental Design

Obviously, for microtubules to assemble the cell must either have a "stash" of free tubulin subunits or else must be ready to make tubulin on demand. We can learn a bit about how cells regulate microtubule assembly, then, by asking if tubulin MUST be synthesized fresh. If so, then we might focus on the trigger – what specific condition causes initiation of tubulin synthesis, and how? If not, then we might think of an explanation (a model to be tested) based upon what we already know about the mechanisms of microtubule assembly and maintenance. How do we find out if protein synthesis is essential to a process? Prevent it from taking place, and see whether or not the process in question still takes place.

Cycloheximide inhibits protein synthesis by preventing the translocation step after the first two amino acids are linked by the ribosome. The messenger RNA is unable to index, the next transfer RNA cannot be attached, the next amino acid cannot be lined up, and protein synthesis stops. Cycloheximide added to a freshly deflagellated culture to a final concentration of 10 micrograms/ml is sufficient to block nearly all (> 99%) protein synthesis. If we amputate flagella in a culture of Chlamydomonas and quickly expose them to cycloheximide at the required concentration, we can answer the question by taking periodic measurements to determine whether or not the flagella grow back. We can learn even more by studying the time course of regeneration in cultures that grow their flagella back.

Controls

Suppose that you treat (with cycloheximide) a culture of cells with flagella amputated, and the flagella don't grow back. Does this mean that the cells must conduct protein synthesis to regenerate flagella? Maybe the result simply means that when you amputate flagella under these conditions they won't grow back, whether you treat them or not.

An experimental control is a standard to which to compare an experimental group. Controls are important to ensure that we have accounted for all variables. We must use controls to ensure that we correctly identify causal relationships. For this experiment we need a positive control to ensure that the deflagellated Chlamydomonas can grow back in the first place. If flagella grow back in an untreated culture and if they do not grow in the treated culture then you can draw the conclusion that the lack of growth is due to lack of protein synthesis. Without the untreated culture we have no way of knowing if some other factor, such as the amputation procedure itself, prevented growth.

Colchicine was originally identified in extracts of the plant Cochicum autumnale, known as the autumn crocus. It is highly poisonous, however the substance has traditionally been used in the treatment of gout, and in fact is still the drug of choice for that ailment. Extreme care is used in its administration, of course. Colchicine binds the soluble tubulin heterodimers. After binding with colchicine a heterodimer is still capable of assembing to the growing end of a microtubule. However, the colchicine molecule 'caps' the end, so that no additional heterodimers can be added. All microtubule growth, including flagellar growth, stops, however tubulin is still free to detach from assembled microtubules. A culture in a medium that includes 3 mg/ml colchicine serves as an effective negative control group.

Controls for stability of intact flagella

One of the most difficult tasks in designing experiments is to ensure that all variables have been carefully controlled. Without adequate controls a reviewer might reject an otherwise meritorious study, leaving someone else to publish the results and to be credited with the discovery.

For all we know, the length of flagella of our model species may vary as part of some circadian (daily) rhythm. Either of the poisons used in our experiment could have an effect on assembled flagella. This would complicate matters, of course. To control for possible effect of poisons on intact flagella and to control for possible variation in assembled flagella length, e might check samples of treated and untreated cultures that have not had flagella removed. It is not necessary to determine a time course, but we should ensure that the parent cultures didn't change between the beginning and end of the experiment.

Notice that these control cultures are neither positive controls (demonstrating an effect) nor negative controls (demonstrating lack of an effect).