no insert found - (Feb/21/2014 )

Yes, that's right.

Do you have any idea why a lot of background colonies on experimental plates while only a few colonies on negative control plate ( cut vector only,  no insert, with ligase )?

Thanks again.

No, I have observed similar things before, and also did not understand the cause. When you get the DNA fragments going into the ligation correct, you will get a great many more correct clones. A possible issue not yet mentioned is that your insert is toxic in E. coli. You could try a low copy number plasmid. Outgrowth and incubation at a lower temperature might also help.

Here's a possible answer to your question. Your PCR could be producing primer-dimers (hetero) which will have both cut sites. They will be cut and clone more efficiently than the longer correct fragment. They will be nearly invisible on a gel, since they are so short. Depending on how you are purifying your PCR reaction, some may be eliminated, but there will be lots more where they came from. Another possibility is that your insert has an unexpected restriction site close to one of the ends, which is leaving a short fragment that has correct ends, and a longer fragment that will be visible to a gel, but will not clone. Both of these could be checked by sequencing some of your 50 colonies "without" inserts to see if they really have short inserts.

For the primer-dimers, you could check your primers with the IDT tools. You might be able to fix this by adjusting PCR conditions (likely raising the anealing temperature). There isn't much to be done about an unexpected site, but it could be found by sequencing your PCR product.

I analyzed the primers with IDT tools and found there're several possibilities of hetero primer-dimers. I think they might be the problem and can explain why low background on control plate whereas dozens of transformants on experimental plate. The short fragments generated by hetero primer dimers were clones into the vector and those inserts were not visible on gel analysis either single cut or double cut.

Since my PCR product was purified with Qiagen kit not with gel elusion, it is highly probable that the kit did not eliminate efficiently the primer dimers. One of my primer ( the one with XmaI RE site ) is over than 40 base pairs long. In my experience, Qiagen kit does not get rid of oligos or DNA fragments longer than 40mers efficiently. The reason I use such long primer is that

I added His tag as well as linker sequence that is crucial in my further purification procedure. I have no antibody to prepare the column.

I will run the PCR again and gel elute the product. I will let you know the result.

Thanks for your advice. 

After selecting colonies with colony PCR, 4 out of 36 colonies are positive for insert. I think the issue here is due to preferential annealing of primer-primer heterodimer into the vector thus generating more transformants in experimental plate than in control plate.

Thanks a lot for your opinions. I finally got my clone.