They can. Constructs are also already available. The first one was called eNPAC, and was using the 2A peptide strategy as said above. If you don't need identical stoichiometry, two separate virus injections work as well.

Hi Sunil,

I agree you can put both ChR2 and NpHR into the same cells and get functional and practical silencing and activation using dual wavelengths of light. Moreover, in this case silencing is basically perfect when using yellow or red light wavelengths (of suitable high power) since ChR2 is not activated effectively above 580nm. There is an important caveat in that the activation part only really works well with very narrow light centered around 405nm, because there is a shoulder on the NpHR activation spectrum at the blue end, and thus blue light is pretty effective at activating NpHR (and Arch, ArchT, etc). I see about a 10-15% of peak photocurrent at saturating power with 400-405nm with NpHR alone, but this is sufficiently low to allow titration of the light down to a level that does not produce appreciable NpHR photocurrent. Since most ChR2 variants and especially the new Chronos are very sensitive to blue light, this low level will be more than enough to get reliable activation. Just be warned that fast kinetic variants such as ChETA are much less sensitive to light and are slightly red shifted in activation spectrum and are thus not a good choice for this application.

All of the info I have provided is based on direct experiments using transgenic mice developed in Guoping Feng's lab while I was a postdoc there. I did slice recordings with VGAT-ChR2 mice, Thy1-NpHR2.0 mice, and double transgenic mice. In the cortex of double transgenic mice the pyramidal neuron subsets have functional NpHR and nearly all interneurons have functional ChR2. Thus, I could patch onto non-labeled pyramidal cells and carefully examine both light-evoked GABAergic inputs and light-evoked Glutamatergic inputs to define the optimal wavelengths and power densities required for true dual-wavelength bidirectional control. For getting expression in the same cells I could cross our Thy1-NpHR2.0 mice with Thy1-ChR2 mice and get some pyramidal cells in cortex or hippocampus with co-expression of both.

I also had good luck using 2A plasmid designs with eArchT and ChR2 (e.g. eArchT3.0-P2A-ChR2) for equal levels of photocurrents for in vitro experiments. I should mention I am not fully convinced the Kleinlogel design (designed for equal stochiometry) will be adequate (especially in vivo), since the original paper only showed very small/modest photocurrent amplitudes in vitro. Many things that worked OK in vitro have failed in vivo due to low expression, especially when making transgenic mice.

can you express them both in the same cell using Cre/lox system?or there might be homologus recombination between the 2 lox sites?

You can express both in the same cell using Cre/loxP. You would ideally have both genes in a single double-floxed cassette e.g. viral vector with DIO design. You could then inject that virus into the brain of a mouse or rat with cell type specific cre recombinase expression. That would be the obvious way. If you have ChR2 and NpHR on separate expression vectors, e.g. two independent DIO viral vectors, this could in principle cause some issues with undesirable recombination events when virus is co-injected into a Cre mouse, but I have not tried this.

Sorry to dig this up, but is it possible to transgenically express Arch and ChR2? We've been trying to breed Arch/ChR2 mice via first crossing floxed Arch and ChR2 mice together, then crossing that with Emx-Cre, but with about 30 mice, we've had no mice that express both Arch and ChR2.

Hi Kevin,

I know this is a few months old, but we are looking into doing a similar thing (a Cre-lox Arch+ChR2 mouse).

I am not an expert in this, but I have looked into this and I believe that both the Cre-lox ChR2 and Arch mice have their transgenes inserted at the Rosa26 locus (and therefore you could only have 1 or the other on a single chromosome).

The best you could do, would be to have a Arch-het/ChR2-het mouse (1 copy of a chromosome with ChR2 at locus Rosa26, and 1 copy with Arch at Rosa26). If you crossed it with your Emx-Cre mice, you would then have half Emc-ChR2 and half Emx-Arch.

A possible solution may be to first cross the Emx-Cre mice with one of the opsin mice (for example ChR2) and get hets with Emx-Cre and ChR2. You then cross the hets to get the Emx-Cre and ChR2 homo mice. Once you have those, you could cross them again with (homo) Arch mice and use the offspring (which should now be hets on Emx, Arch and ChR2).

Did you ever get this to work? I am looking to see if there are any mice that have both Cre-lox Arch and ChR2 transgenes inserted in the same cassette (or on different loci).

Hi, Jonathan T Ting

In case of co-expression of opsin pair, the expression level of individual opsin will be similar to that when they were expressed individually?