Idea to determine order of assembly of spliceosome

[nukeman]

Hi everyone,

I've got this assignment where I'm supposed to present a suggestion of a series of methids to use in order to determine the order of binding of snRNP:s which eventually assemble to a spliceosome.

My initial thought was to use in vitro splicing assay with a couple of extra steps.

1. Ligate a traget gene to a vector with some kind of bacteriophage RNA-pol promotor
2. In vitro transcription with radioactive nucleotides
3. Extraction of pre-mRNA
4. Run agarose-gel where I can identify and roughly determine the size of my marked RNA
5. Add RNA to mammalian cell extracts, e.g. HeLa
6. While the RNA undergoes maturing, I collect samples with regular intervals of time
7. Use phenol to extract proteins from each sample
8. Run MALDI-MS starting with 0 minutes sample. After that, continue with the next sample in chronological order. Somewhere along the line, one of the bars should start to increase in mass (we assume that we don't know the mass or other physical properties of the snRNP:s, thus the "target bar" is unknown to us prior to the appearance of incremental increase of mass). The mass should continue to increase with specific amounts during the maturation. Re-running this test should show a similar pattern of mass increase at specific points in time if it is the snRNP:s mass increase we are detecting.

What do you think of this? Keep in mind I'm studying on the most basic level, so I've probably got something grossly wrong

Any and all suggestions are more than welcome.

[JackBean]

The in vivo transcription will incorporate radionucleotides into all molecules of RNA, so you won't be able to discriminate your transcript. But you could purify it on affinity column. However, that would be quite laborious.
Better would be probably to induce your RNA right in the system you want to study and do pull down MS experiment.

[JackBean]

However, you asked for order of binding, right? In such I case I suppose you already know the biding partners. Thus you can make knock-outs and see which proteins will be bound to pre-mRNA

[nukeman]

The thing is that the questions are made so that we will pretend that we don't know that much about the spliceosome as we actually do, and from there proceed to map its function and so on. I should have written this more clearly.

In essence, that means I "don't know" the binding partners, only that there is some kind of complex performing the splicing. I want to know how this complex forms, and how many subunits comprise it.

Do you propose I run a pull down and then use MS? When doing the pull down, how do I go on about getting antibodies for a completely unknown protein (we're pretending its unknown)?

I could use knock out, but that would require me to know the corresponding genes, which I don't (pretending, yet again). Also, wouldn't knock down of genes of such high importance cause all kinds of trouble in the cell; probably death before the spliceosome even starts to assemble?

[JackBean]

You could do pull down with binding to antisense RNA (or DNA would be better). With this you should get at least some proteins. These you can use subsequently for further identifications.

For such important proteins you can use inducible knock-out.

[nukeman]

How would I differentiate my target proteins from the rest of the proteins? Do you mean that I should use DNA/RNA that matches the pre-mRNA and that way isolate the target pre-mRNA along with the associated snRNP:s?

[nukeman]

Oh and forgive me if I'm asking stupid questions, like I wrote in my initial post, I'm only studying on a basic level.

[JackBean]

That's OK. You're just trying to learn something (and at least you're trying opposite to others

Anyway, yes, using either specific or unspecific NA probes will show you which proteins bind to nucleic acids. The question is, how to discriminate those bound in spliceosome from others. One way could be specific binding to introns (although that could show proteins bound to gDNA), but with that you could have false results because of one-case-only binding/non-binding.
To explain - imagine you will design probes for intron in gene for actin. Generaly there are five proteins in spliceosome, let's call them Spliceosome Protein1 - 5, but in actin is not present SP3, but are present other proteins, let's call them SP6 and SP7. Thus if you try to generalize your result, you will get wrong number AND wrong proteins. To avoid this you probably would have to try several different introns in several genes.

[nukeman]

Isn't it possible to precipitate samples during different times? Like I had planned initially. If the spliceosome assembles stepwise, then wouldn't we be able to detect different masses at different points in time?

[JackBean]

Theoretically yes, but I think the speed is quite high, so you've had to stop it after like each 5 minutes and even that could be too much.

[nukeman]

Aha ok.

Thank you very much, I just received a mail from my teacher and he actually suggested I do the precipitation, just as you did.

From one thing to another though.
What is the purpose of a bacteriophage promoter being ligated in to a vector along with your gene? Is it just to ensure that you get strong expression if you use some kind of phage RNA-pol?

[nukeman]

Forgot one thing:

Say I precipitate the samples at different points in time, how would I go on about analysing them? Is it feasible to use MS here or are the differences (when the next snRNP joins in) in snRNP mass enough for me to be able to detect change in mass on a western?