Promoter strength isn’t something that I recall being discussed in the elegans field (mainly because of arrays, not many single-copy trans genes out there - or because most labs are interested in cell specificity rather than strength), but on the off chance I’m wrong (and my non-nematode co-PI is right…), what’s the strongest ubiquitous promoter?
eft-3 doesn’t seem that strong based on the level of GFP in the miniMos strains.
sur-5 seems fairly strong, but most of the strains are multi-copy arrays or integrants…
I’ve been thinking about this question with respect to cell-type specific promoters, trying to find the strongest promoter to drive TIR for the auxin degron system in specific tissues to get the best knockdown. I’ve been trying to look at the FPKM data on the expression page of a gene of interest as a proxy for promoter strength. Caveats are that one would need to figure out the promoter sequence required to drive that expression, and that other elements like the 3’UTR could impact the FPKM.
That said, eft-3p seems pretty decent (in the 2-10FPKM range). sur-5p does seem better (4-120 FPKM; a bit low in the early embryo, otherwise very good). smu-1p and smu-2p are in the 8-75 FPKM range.
Do you need germline expression? If not, it looks like act-1 is crazy high (444.6-3100 FPKM). You might poke around other really strong housekeeping genes to see what they look like.
Interesting observation Jordan. The trick is to indeed get the right regulatory elements.
It would be nice if anyone had done a comparison of known/used promoters in the same way, and at single copy. Too bad the promoterome existed before the revolution of single-copy transgenes…
I did a little mini-screen of a dozen or so promoters at one point, using single-copy MosSCI insertions. I did not succeed in finding anything truly ubiquitous. You’d think ribosome/translation genes ought to be pretty ubiquitous; in practice they are probably expressed in most every cell, but at much higher levels in the intestine than other tissues. his-72 is quite strong and is expressed in (at least nearly) every cell, but it is much stronger in embryos than in larvae and adult tissues. I had poor luck with sur-5and let-858. I didn’t test actin and tubulin promoters, which might work although I wouldn’t be surprised if their expression is also non-uniform (you might see a lot more tubulin in neurons, for example). Ultimately I think it depends on what you’re trying to do; if you just want blasting high expression and don’t care so much about uniformity, I’d use eft-3 if you are most interested in somatic tissues or his-72 if you’re more interested in embryos.
I tried sur-5 once (in a multicopy array) and would agree that it’s highly overrated as “ubiquitous” - expression in neurons was extremely weak or absent.
Dan - I’d be interested to know which promoters you screened. A few years ago I embarked on a wee effort to insert the rpl-10 promoter and 3’ UTR into pDD122 in an effort to get better Cas9 editing efficiency by increasing germline expression… but, like most things I do in the lab, it never got finished (and then we started using Cas9 protein instead).
I put ubiquitous in the title, but for our studies it’s actually not important that it be in all cells. Highly expressed in intestine and other things would be ok. Going for a full-body–ish fluorescence that would be easy to spot in a fluor stereo or a sorter or plate reader.
I looked at eft-3, let-858, sur-5, his-72 and rpl-3 as single-copy MosSCI insertions. I also looked at his-72, nmy-2, rap-1 and cdc-42 as endogenous promoter fusions (non-floxed mNG^SEC^3xFlag insertions). I cloned a couple of other rpl promotors but it doesn’t look like I ever injected and imaged them. I concluded that his-72 was the most nearly ubiquitous (and more importantly for my purposes, the most highly expressed in embryos while still being present in other tissues). eft-3 was brighter but strongly enriched in the intestine as mentioned (and absent in the germline in the animals I looked at, probably due to silencing).
@ Steve. Xiao Liu’s 2009 paper (MID: 19879847) with automated imaging in L1s might be a resource to look at. Figure 2 shows the cell by cell expression level in all L1 cells, including Peft-3. Those are not single-copy strains but biolistic integrants so you’ll have copy-number and position effects. Xiao used the same imaging on some of our single-copy Peft-3 integrants (PMID: 27374334). Perhaps some of the Murray and Bao labs’ embryonic cell resolution data could be used?
@ Dan. When not silenced, Peft-3 germline expression is the brightest I have seen, also compared with his-72 endogenous tagged germline expression. However, it’s also the most silencing-prone promoter we’ve worked with…
@steve It’s germline-specific, but the glh-1 coding sequence replaced with GFP is the brightest single-copy reporter I’ve seen so far. GLH-1::GFP is bright, but GLH-1 autoregulates so when it is entirely replaced with GFP it gets about 3-5X brighter. The loss of GLH-1 is partially compensated by a 2-3X fold increase of GLH-2 expression so the worms are still quite healthy at 20C and just super bright. I attempted to drive TIR1::mRuby expression with this, but it was toxic (too much expression?) as red F1 hets were all sterile.
Dustin - that’s interesting about the sterility of pGLH-1::mRuby:TIR1. Did you characterize the sterility in any detail? I’m wondering because we had concerns that TIR1 could potentially sequester SKP-1/2, but have never seen that using other promoters. Did you try more than one construct?