You can’t always get what you want…

Our goals for this project include understanding how a typical gene found on the Drosophila melanogaster fourth chromosome has adapted to allow expression in the context of heterochromatin.  Generally, we consider heterochromatic regions of chromosomes to be hostile towards gene expression, maintaining a compact state that prevents necessary transcription factors and/or transcriptional machinery to access genes.  We have selected the Rad23 gene, located with in a stereotypically heterochromatic region on the fourth chromosome, but known to be expressed.  This gene should allow us to answer questions regarding how a gene adapts to heterochromatin.  We want to know if we move Rad23 to another heterochromatic region on the fourth, will it behave as it does in its normal location?  In order to do this, we have to make a version of Rad23 that we can analyze.  To do this, we planned to link Rad23 to the enzyme LacZ (also known as beta-galactosidase).  In enzymatic assays using X-gal, LacZ produces a product that can be visualized and quantitated.  This technique is commonly used in molecular biology research, and Drosophila researchers have used it in the past (including our collaborators: Lu BY, Ma J, and Eissenberg JC.  Development 1998).

Cloning strategy for creating the Rad23-LacZ fusion.

For this technique to work, we attempted to create a protein fusion between Rad23 and LacZ based on the predicted protein structure of Rad23.  Using standard molecular cloning techniques, we attempted to bring together the DNA of Rad23 with the DNA of LacZ, and a linker region that would form a flexible hinge between the two proteins, in an effort to ensure that the LacZ protein would retain its enzymatic function.  This is where we ran into problems.  The LacZ DNA cloning did not go well, but we were able to grow it in our desired vector if we only grew our bacteria with the DNA on plates and not in liquid culture.  However, once we attempted to add the Rad23 to the LacZ DNA, nothing would grow.  We tried different primer sets, different restriction enzymes.  We checked the ligations to ensure that the DNA pieces were being linked together.  Everything was working as it should, but once the DNA was put in the bacteria so we could grow it up and work with it, the bacteria wouldn’t grow. 

Ligations of 5'UTR and Rad23
with LacZ and the 3'UTR
*Indicate successful ligation products
of expected sizes.

So why didn’t our strategy work?  It’s hard to know exactly.  There are a couple possibilities.  It could be that LacZ was toxic to the cells we are using.  This seems unlikely, at least on it’s own, as LacZ is commonly used in this way.  However in the context of our particular DNA vector, it is possible that the LacZ gene was not stable or caused some other problems.  When we started having problems we inquired of other labs to see if they had experience with this sort of construct and were told they no longer used LacZ in the way we were trying to because they’d encountered similar problems.  After a lot of work trying to create the Rad23-LacZ fusion we decided it was time to cut our losses and find a different system to mark Rad23.