I had a similar problem with one of my proteins years ago. I don't know which expression vector you are using but the His-tag is usually already in frame with the start codon of the vector so that shouldn't be an issue. With my protein, I concluded that the His-tag was folded somehow inside the protein and was thus not exposed - could not bind to Ni2+ IMAC column or react with anti-His antibody on Western blot. You could try purifying your protein under denaturing conditions which will expose the hidden his-tag for easier purification using IMAC - but then you obviously need to re-fold which will take some time to optimise. Can you use another method of purification such as size-exclusion or the like?
The protein is very difficult to purify which is why I decided to go with his tag to begin with. The odd thing is that I attached the his-tag with all the different linkers in between including a long glycine polylinker and it won't bind. Also wouldn't protein be completely unfolded and thus histag exposed when you do SDS PAGE gel before you do Western blot?
thanks for the post. Misery loves company.