bonjour!

jf.allemand et p.desbiolles.Physioque et biologie;edp sciences

constituent des analyses structurales.

h.schmitt.pharmacologie.flammarion

concerne les régulations métaboliques et la diffusion des cinétique

=Structure/Ecologie

Hi Nadia,

Just to know, when you said that the tagged GFP remained cytosolic you meant all of it? Could you see at least some GFP at mitochondria?

That would be a starting point to analyse your problem. The tag and linker between tag and GFP was used successfully by others or is it your design? If you trust the design an expression of the tagged GFP (I guess so since you see GFP expression), then you have to think about three possibilities: 1) the overexpressed tagged GFP might be saturating/collapsing de machinery used to target proteins with that tag motif, therefore only a fraction of the GFP gets targeted while the rest remain cytosolic. This is way I asked about GFP location. 2) The tag somehow causes the expressed GFP to get stuck at other endomembrane like the Golgi. 3) Your cells cannot sort proteins properly (toxicity of some kind). If you suspect of this last one, expressing and following GFP with tags for a different compartment may give you the info you need.

Best,

Hi Nadia,

Are you over-expressing MTS-GFP without a promoter? GFP is known to get truncated. What you see might be an artifact of over-expression. If you are using a promoter that can be regulated, it is possible to adjust the expression level and identify the most suitable expression condition to express GFP only in the mitochondria.

I study yeast mitochondria and have been using a lot of MTS-conjugated proteins and tags. Using strong promoters have resulted in cytoplasmic and nuclear localization of my tagged proteins (detected using Western Blots). However, weak promoters showed only mitochondrial localisation (cytoplasmic and nuclear localisations were not detected in Western Blots).

Suggestion 1:

My suggestion is to use a promoter controlled MTS-GFP. At minimal expression, proteins conjugated to MTS can successfully localise into mitochondria.

Suggestion 2:

I would also suggest to run a Western Blot using fractionated protein samples of mitochondria, cytoplasm and nucleus to inspect GFP expression in each of these compartments.

Suggestion 3:

I know GFP is an amazing tool for imaging purposes but is it possible to switch to a different tag? 3XFLAG worked the best in my case. I have tried FLAG, HA and GFP with MTS and I would recommend 3XFLAG.

Suggestion 4:

Are you sure the MTS sequence is present in your plasmid in the first place? I’m sure you made sure of that by plasmid sequencing but I just wanted to make sure. MTS is a really short sequence and you might get into trouble properly conjugating it with your gene of interest.

I’d be happy to answer if you have more questions. It would also be helpful if you can tell us the cell lines that you’re working with and the purpose of this MTS-GFP expression.

Goodluck.

Best regards,

Thev

Is your fluorescent protein a rapidly folding variant?

Modern fluorescent proteins are designed to mature more rapidly than older generations of GFP and DsRed. This can be a problem for mitochondrial targeting, because mitochondria can only import unfolded fluorescent proteins.

See the following paper for more details about this topic.

DOI: 10.1007/s11033-007-9073-7

Thanks everyone!

@Sebastian, it is purely cytosolic/diffuse, indistinguishable from the control plasmid. @Theventhiran, its a standard CMV promoter, which is fairly strong, but all my other mito plasmids have this promoter. I can't use another tag for this kind of experiment. We are trying to localise the ATP sensor Perceval (https://www.nature.com/articles/nmeth.1288#methods) to mitochondria. In terms of the linker, it was just based on the restriction sites, so I didn't give it much thought beyong making it in frame... I attach an image of the sequence.

@Ben, thanks for that paper, very useful. I couldn't find anything about its folding, beyond that it is modified from Venus, and Venus can be targetted to mitos fine, from what I gather. But obviously the protein has been changed again, into Perceval, so I guess that's just the problem.

Hi Nadia,

Just a couple of thoughts, do you really need the tandem of 2 mito-target sequences? wouldn't be worth to try just with one? Do you know whether these epitopes can interact with, and some how inhibit/interfere, each other?

I'll try inserting a longer linker to separate a bit better the mito-target sequence from the GFP, just to avoid weird interaction that may hide the targeting sequence.

Best,

Hello, Nadia. How is everything going?

It is highly possible that the author in Nature method paper have tried that already (may not working) . I would contact them for suggestion.

Second, it is possible that the MTS at the N-terminal of Perceval is structurally unaccessible for mitochondria importing machinery. Although most of the MTS is N-terminal, there are still a few C-terminal ones you could try. eg: the last 14 residues of Hmi1 (Ref: http://www.jbc.org/content/274/30/20937.long); or the last 30 residues of APE1 (Ref: http://www.jbc.org/content/285/20/14871.full.pdf). Goodluck!

https://journals.plos.org › plosone › article