co-injection marker advice

I am trying to detect a fusion protein I have introduced by microinjection containing an exogenous tag. When I injected initially, I used rol-6 and myo-2:GFP as my markers. Unfortunately, the expression level is not very high and I am unable to visualize it by Western blot (only IF), presumably because the % of the protein I am looking for out of the total protein is so low. I want to re-inject these constructs using a marker that rescues either a lethal or growth arrest phenotype, like pha-1, so that all or almost all of the worms on my plate have the array and I don’t have to pick rollers to separate plates before extracting protein.

The other big problem is that for the assays I am doing, I need to grow the worms at 15C, which is the permissive temperature for almost all of the commonly used ts mutations that can be rescued by microinjection that I could find, which does not solve my problem of too many non-array worms being present in my protein extracts. Are there any cold-sensitive mutations that are easy to rescue by microinjection and that would likely not interfere with (too much anyway) with normal cellular functioning after the rescue gene has been injected (or other types of conditional sensitivity - I am not that familiar with the different phenotypes these mutations can take).

I know I can also try to integrate the array to solve this problem, but I have had problems integrating arrays which contain these proteins in the past so I am uncertain if that is even feasible for these experiments. Any suggestions would be very helpful. Thanks!