Maybe I do not understand what you mean in its entirety.

If you use a myristylation signal, it will be tethered to the membrane

If you use a signal peptide it will be secreted

For a protein to be present at the plasma membrane, it has to be physically tethered there or has to interact with a plasma membrane protein. With that in mind, I would suggest fusing your protein to another, that interacts with a plasma membrane protein.

Hi Sanath,

Thank you for the response. Let me try to clarify. We'll call the subunit of my protein A and 12A as the functional, oligomerized form. 

My Myr-Palm-A construct traffics to the membrane just fine. However, it needs 11 other free A subunits (untethered) in order to be functional and generate the final MyrPalm-12A protein (would be too large to genetically encode and therefore needs to be assembled within the cell). My problem is in attempting to convert A (naturally cytosolic) into a protein that can traffic to the membrane. Once at the membrane it will be able to interact with the MyrPalm-A constructs and form the functional protein. 

Hope this clarifies things for you as well as others. 

Instead of a Myr signal, try adding a transmembrane domain with a flexible linker to the N-terminus of the protein. I would add something like M-Transmembrane domain-flexible linker-A to get your protein trafficked to the membrane. You could use the signal sequence from human proteins like CD4 or CD8

Thanks for the suggestion. The subunit will traffic to the membrane but I fear that the addition of the transmembrane domain might confound assembly. Now the final protein will be something like Myr-Palm-A+(11A_TM domain). 

Ideally, I was looking for some sort of cleaved signal peptide that will get the protein to the ER and further traffic cues to promote ER exit -> Golgi export -> membrane.