No this is not 'homework'. I'm a PhD student, and this was something brought up at the end of a journal club recently, and one of the PIs posed this question and told us to think about it. It sounds simple, but it sort of has me baffled.

  • Western blot: No. Need a known antigen and so, since we don't know what we're after, this won't work.

  • Total protein gel: Collect cell lysates from different cell populations that are in different parts of the cell cycle, run it on a gel and observe any cyclic expression. This seems sort of ok but I feel like there are a few problems with it: 1) identifying the cell populations that are in the different cycle stages; 2) other proteins of same/similar size that cannot be differentiated.

  • Some sort of RNAi, knockdown, etc.: Like the western blot, we don't know what we're dealing with and so can't target anything.

  • Pulldown assay with mass spec: I don't know a lot about it aside from that it looks at protein-protein interactions. Unsure if it would be appropriate here.

As far as a model, I was thinking either yeast or some immortalized mammalian cell line.

  • 1
    $\begingroup$ One can fairly easily do genome-wide knockdown experiments nowadays, so don't discount RNAi/siRNA right off the bat. Which tool(s) you pick is going to be dependent on the model you're using. If it's yeast, you have many validated tools at your disposal, including mutants/knockouts for every single gene in the S. cerevisiae genome, two-hybrid systems for looking at protein-protein interactions, reporter constructs that indicate by fluorescence or enzymatic activity when the gene of interest is being expressed, and others. $\endgroup$
    – MattDMo
    Dec 12, 2016 at 20:39
  • 1
    $\begingroup$ For yeast, people used to rely on mutants/knockouts which are supposedly easy to produce using HR, compared to other organisms. For animal cells people extensively use high throughput RNAi/CRISPR-Cas screens to identify the functions of a gene. If you have the protein sequence then you can express it and raise antibodies against it. With the antibodies, you can Co-IP-MS for identifying protein-protein interactions and ChIP-Seq (for TFs) or CLIP-RNAseq (for RBPs) to identify protein-DNA and protein-RNA interaction respectively. Overall, your question is quite broad. You should narrow it down. $\endgroup$
    Dec 13, 2016 at 5:43

1 Answer 1


I've actually been involved in a hunt for checkpoint proteins in diatoms (a type of eukaryotic algae), the cyclin family has been greatly expanded in these organisms and we were trying to figure what these extra cyclins were sensing/signalling[1].

In a nutshell, the first author looked at different conditions which were likely to affect cyclin expression. Most of these conditions prevented cell division (e.g. lack of nutrients, G2/M chemical inhibitors or light) and samples were taken before and after the blocking conditions were removed. Afterwards mRNA's were quantified of suspected cell cycle genes. As the checkpoint genes' mRNA and protein cycles extremely transiently this took some doing. Cyclins that reacted to the conditions were characterised using complementation assays and other techniques.

Afterwards yeast two hybrid assays were used to identify interacting proteins to build up the network[2]. In general, these experiments were not that straightforward and messing with essential genes causes a lot of trouble but it still nothing compared to the original yeast cyclin work where they were slogging through mountains of temperature sensitive mutants defective in their cell cycle (see the work of Lee Hartwell and Paul Nurse who got the Nobel prize for this in 2001 with Tim Hunt).


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .