I am facing a problem cloning my ATAC fragments into a STARR-sequencing vector to build my enhancer screening library. The main issue: I have two STARR screening vectors- one containing RFP reporter, and the other containing GFP. Besides these reporter genes, the plasmids are entirely identical. However, the GFP one has far worse cloning efficiency than the RFP one, and I have no idea why. I have included some images/files to help explain the situation. They are numbered and named accordingly.

More detailed explanation…

I obtained the STARR-seq luciferase validation vector_ORI_empty (Plasmid #99297) from Addgene (file 1), prepared a maxiprep, then made various alterations to the construct for the purposes of my study. Specifically, I removed the luciferase gene and replaced it with either RFP or GFP, inserted beta-globin minimal promoter, and inserted a special cloning site downstream from RFP/GFP. This site contains the EcoRV cut site and some surrounding sequences, which are compatible with the Tn5-tagmented and amplified ATAC fragments which can be cloned into this site by In-Fusion (file 2 illustrates this). SnapGene files of the final STARR-RFP and STARR-GFP vectors containing no insert are attached (files 3 and 4).

To prepare the vector and insert for In-Fusion cloning:

1. The STARR plasmid is linearized by EcoRV digestion. It is digested for 2-3 hours. Then I use a QIAGEN Gel Extraction kit to gel-purify. Sometimes I use additional purification steps.

2. The ATAC fragments are amplified using special primers that will be compatible with this cloning site. The fragments are then bead-purified and eluted in low-EDTA TE buffer.

To see the principle and procedure for cloning, please take a look at the attached PDF (file 2)! Various sequences and some additional background info are also included in the PDF.

After cloning, STARR-RFP-ATAC plates have many colonies, while STARR-GFP-ATAC have few. Attached are typical examples of what this looks like (jpg files 5 and 6). Please disregard the different molar ratios I was testing. Normally I stick with 2:1. The main point is that RFP has significantly more colonies than GFP in general. Lately I have not been using a negative control because it always looks the same. The negative control for STARR-RFP I normally get 0-2 colonies on the plate. For the STARR-GFP one, unfortunately the negative control often contains 5-10 colonies, so many of the colonies in the picture are probably false anyway.

Downstream, I will most likely utilize other methods of transformation for better library cloning, such as an electroporation approach, but for the time being this is the primary method I am able to use. I'd just like to understand why the RFP and GFP results are so dissimilar. I need both of them for my study. Did I miss something obvious? Please let me know if any more information is needed.

Similar questions and discussions