Transcriptional gene fusion, aka promoter fusion or RNA fusion is used to measure mRNA stability.
Transcription produces a single mRNA with the genes of interest and a reporter gene, but a translational stop codon between the gene of interest and reporter gene leads to only the reporter protein being produced.
Transcription encodes two polypeptides but translation only encodes one.
This type of fusion construct allows you to determine whether sequences other than the promoter affect gene expression.
For instance, do Amastin 5` or 3` UTRs affect mRNA levels? LacZ is a good reporter gene, but there can be other ones. Trypamastigotes vs. amastigotes used to describe mRNA degradation. LacZ is only bacterial, which is why it can be used in eukaryotes as a good reporter gene. lacZ will go in front of the promoter and before the stop codon and amastin.
The stop codon is a translation stop codon not transcriptional. So transcription ends much more upstream of the amastin gene. The stop codon does not come into play until the translation. So then the amount of B-gal produces corresponds to the stability of the mRNA and.
Transcription produces one mRNA with the reporter gene (e.g. gfp) and gene of interest. Translation gives one polypeptide the reporter protein (e.g. GFP).
Why use this type of construct? It allows you to measure activity of the sigma-responsive promoter and therefore of sigma protein activity by determining the location of GFP as it is expressed from different "sigma-responsive" promoters.
For example, is sigma protein activity cell-type specific?