If the vector is under a strong promoter, we could imagine that both parts of eYFP are over produced, making possible the reconstruction of the protein. Maybe you should analyse your interaction earlier, in order to avoid this situation.
Then, the main problem with BiFC is the irreversible complementation/reconstruction of the fluorophore.
I recommend to you a really nice publication about PPI in vivo, dealing with both good and bad side of every method.
Thank you Hervé Begue , and if I use one of the empty vector (pSPYNE or p SPYCE) and the other (pSPYCE or pSPYNE ) recombined into my gene, then I co-transformed tobacco followed this method: Agrobacterium tumefaciens strain GV3101 was transformed
with the indicated expression clones and grown in YEB medium
supplemented with kanamycin and hygromycin. Bacterial cul-
tures were precipitated and resuspended in infiltration medium
containing 10 mM MgCl 2 , 10 mM MES (pH 5.6), and 150 mM
acetosyringone. To enhance the transient expression in tobacco,
cultures were mixed with an Agrobacterium culture that allows
expression of the silencing suppressor p19 (a gift from Professor
Sir David Baulcomb) (Voinnet et al. 2003), giving an optical
density at 600 nm of 0.3 for each strain. The mixed bacterial
suspensions were then infiltrated into young, but fully expanded,
leaves of N. benthamiana plants using a needleless syringe. After
infiltration, plants were grown further for 48–72 h at 25°C and
collected for further biochemical assays.
After 48h I also can see fluorescence from this control, So I don't know whether it is real when I see the fluorescence of my two proteins?
A clue to know if the interaction is "true" or not, could be the localisation of the fluorescence. In my case, I was lucky, because the interaction was detected only in a specific compartment, and one of my partner is ubiquitous.
Another problem could be the system of detection.
Even with two-photon microscopy (really powerfull method), a yellow autofluorescence can be detected... actually, yellow is a little green, a little red... so chloroplasts and a lot of compounds interfer.
That's why we decided to change the method and tried FRET-FLIM with CFP and YFP. And this one is amazing and really reproducible!
I would not recommend you to use empty vectors as negative control. As mentioned previously, the YFP-termini can complement each other at times if they are expressed to frequently. Try a fusion with a protein that does not interact with ur protein-of-interest instead as a negative control.
I have used them, however unsuccessfully. BiFC is a technique full of artifacts as far as I am concern. Cells around the infiltration point and around damaged tissue will be always fluorescent. Same is the case for cells next to guard cells and trichomes. Look for your interaction far away from the infiltration point and only trust it if you can see it more often than your negative control.