Hi, i'm new to this board and would like to ask for some help. I'm a grad student at the University of Pittsburgh, PA and am working on a very large plasmid (~15 kbp).

The plasmid codes for a DNA repair protein which is ~360 kDa and maybe toxic if fully expressed. I've attempted expression in bacteria and via invitro transcription-translation - both methods have been unsuccessful. I'm currently attempting mammalian expression and am helping a post-doc with expression in Baculovirus.

I've used the Invitrogen Gateway system (pDEST47) and was unsuccessful in expressing any protein (no observable eGFP) in 293T cells. I have since moved on to using pcDNA3.1 and pT-SIGN (homegrown IRES/Neo-eGFP) constructs. I have assessed transfection efficiency based on eGFP expression while using the pT-SIGN construct. The pT-SIGN construct is bicistronic (promoter-my gene-IRES-Neo/eGFP) therefore, when I see green cells, I know that my gene's mRNA had to have been fully produced otherwise, I wouldn't see the Neo/eGFP production.

So, in order for me to observe green cells, I have found that I need to transfect 16 ug of my plasmid with 18 ul of Fugene6 in order to achieve approximately 10 - 12% green cells (this is starting with 1 million cells in a 10 cm dish). My Western results are inconclusive right now, however, it does not appear that I am producing my large protein (I've C-term flag tagged it and attempted a Flag-IP Western). In comparison, I have transfected pcDNA3/ATM-flag (4 ug DNA/16 ul of Fugene6) and have excellent results - there is tons of protein produced and it is easily visualized on Western.

Does anyone have experience working with very large plasmids and attempting to express very large proteins? Any help would be great. Let me know if you need any other information. Thank you in advance.

I forgot to add - I have also attempted Nucleofection (Amaxa) - those results were poor, but at the time I had only attempted 1 ug of DNA for 2 million cells. I observed about 100 green cells (thus I'm assuming my efficiency was either low because I used too little DNA or the protein produced was toxic).

I've also done a plasmid titration curve - 1 ug - 32 ug/300,000 cells transfected using 3 - 18 ul of Fugene6. At the low DNA concentrations, I observe a couple dozen cells lighting up green 2 days post-transfection. It's only when I begin going to 8 and 16 ug of DNA do I observe between 10 - 30% transfection.

Any recommendations in order to improve efficiency or help with protein expression would be greatly appreciated.

Firstly I think you need to ask yourself does this plasmid have to be this big? Are there any regions of unnecessary UTR, or other sequence cloned into the plasmid that you could excise? Secondly, if you have UTR in there is it really crucial for expression? There are many examples of genes that express better when the open reading frame alone is cloned into the expression vector.

Secondly, if it really is the your DNA repair protein which is causing toxicity, why dont you use an inducible expression plasmid like Stratagenes Complete Control Inducible Mammalian Expression System, or Invitrogens T-Rex system. This would allow you to transfect and still assess transfection efficiency using your reporter GFP, but you can then switch on your gene of interest at what ever level you wish once you have optimized transfection.

One thing I could not establish from you post is if the GFP reporter was on a separate vector or fused into your 15kb vector. If it is in a separate vector (which is probably or 1/3 the size of your 15kb plasmid) the optimal transfection parameter will probably be different for the reporter cDNA compared to your 15kb plasmid. Furthermore, if you transfect in too much reporter you are probably saturating the transcription/translation machinery and hardly any of your larger plasmid is expressed. I have had experience of such a phenomenon when I was a grad student working with large plasmids containing Sodium and Calcium channel subunits. Furthermore, if you over express GFP it can often become toxic. I would suggest trying to use the minimum possible amount of reporter plasmid possible. You really want the cell to be producing the same or a lesser amount of message for the reporter as your gene of interest but NOT more. You can usually get away with using hundreds of nanograms of GFP plasmids (with millions of cells) as reporters and still see bight enough expression to be an accurate marker of transfection.

RE your AMAXA trial I see you used 1ug of your plasmid in the trial. I would suggest upping the amount of your plasmid so that it is the molar equivalent of 1ug of the pmaxGFP that AMAXA provide with their kit. They optimize all their parameters using the pmaxGFP plasmid, which is small and is codon optimized for mammalian expression and while it may give 80-90% transfection efficiencies when expressed alone, when you are dealing with a plasmid 3 times larger, to get the same number of copies of plasmid in to each cell you probably need to add several micrograms per transfection.

A final suggestion for now is to culture at 27-30 degrees instead of 35-37 after transfection. This way the majority of cells will focus on protein production rather than cell division.

I hope I have been helpful. I have done lots of transfection optimization for many different genes, and quite frankly its a complete alchemy. If you are at all interested in complete optimization there is no other route apart from lots of trial and error with transfection reagents, cell numbers and quantities of cDNA. Good luck.

[quote=frasermoss]
I have had experience of such a phenomenon when I was a grad student working with large plasmids containing Sodium and Calcium channel subunits. ]

Hi frasermoss,
I am having problems getting good transfections with voltage-gated sodium channels in pcDNA3. I tagged the channels in the N-terminus with GFP and I judge by looking for green cells. By channel is 6kb with .7kb GFP. Any advice on improving the transfection and on getting stables.
Kind regards