How would you use readthrough as part of a genetic circuit design?

Question

Traditionally, in synthetic biology, researchers tried to avoid some transcription phenomena (like roadblocking of tandem promoters or readthrough of weak terminators) since they are not in line with the discipline's strive to design modular, digital genetic circuits. However, some researchers have started using these "problems" as part of the intended design (see Bordoy et al., for example).

How do you think transcriptional readthrough could be used as part of a genetic circuit design? When would you want to have a readthrough? What kind of logic or function would this emulate?

Answer 1

Readthrough has already been used to implement transcriptional NOR gates, wherein tandem input promoters express a repressor that represses an output promoter. (First time I recall here: 21150903, used widely here: 27034378, roadblocking modeled here: 32141239.) An advantage of this design is that the repressor need only be encoded in DNA once, which can significantly reduce the DNA footprint of the design and the likelihood of recombination if the repressor's DNA sequence is long (e.g., protein-based repressors). OR gates have similarly been implemented as two tandem promoters expressing a common output gene. (An alternative NOR design—which is common with repressors with short DNA sequences (e.g., sgRNAs)—is to duplicate the repressor and express it separately from both input promoters. This approach circumvents the roadblocking and input asymmetry that can occur with tandem promoters.)

Readthrough has also been used with antisense promoters to tune gene expression (see 26769567). In these designs, an input promoter must read through a downstream antisense promoter, and the strength of the antisense promoter can be used to tune the response of the input promoter.

Another application that comes to mind is encoding the relative ratio of expression of multiple genes by separating them with weak terminators (e.g., to achieve X-fold expression of the first gene relative to the second). This begins to evoke an operon, as jakebeal notes. The closest example I can think of comes from this paper: 23727987, although they actually used readthrough to address the inverse problem of characterizing terminator strength.

Answer 2

I think this is an interesting proposal to try to turn a problem into a benefit.

What you are considering is conceptually similar to the organization of an operon, in which multiple genes are controlled by a single promoter, with multiple ribosome binding site (RBS) entry points for translation, and read-through can certainly occur in these cases too. Structurally, this would be the same behavior, except that the entry points are promoters instead of RBSs.

At a logical level, in effect the second promoter ends up implementing one of two functions:

• If it is positively regulated, then it would be an OR, activated either by its own transcription factor or read-through from the 5' sequence.
• If it is negatively regulated, then it would a NOT IMPLY, activated by the 5' sequence, but dominated by its own repression, since steric repression mechanisms will tend to block the read-through as well.

There are lots of potential concerns regarding signal strength of the read-through, but if one wanted to make use of this mechanism there is no reason to think they would not be susceptible to engineering.