The GM-CSF gene has been reported to generate two mRNA transcripts as a result of a combination of alternative promoter usage and differential splicing. Several different sized GM-CSF mRNA transcripts have been reported, some have been reported to be associated with embryonic implantation in mammals, neutrophilic activation, and clonal bone marrow growth. We carried out a project designed to determine more rigorously whether alternative GM-CSF transcripts exist, and if they do, to bridge the gap of knowledge between the observation of alternative GM-CSF promoters and longer GM-CSF mRNA species. We began this project in September, 1995 and have recently concluded it (manuscript in preparation). Our findings can be summarized as follows: an RNA species of approximately 1.5 Kb was observed using a GM-CSF probe in Northern blot experiments. We were able to detect the fusion mRNA detected previously using reverse transcriptase and PCR. However, we suggest that this 5' fusion transcript results from mispriming of the mRNA transcript by a 9 nucleotide segment of RNA that is identical in both transcripts. This is suggested by the observation that using 5' Rapid Amplification of cDNA Ends (RACE), we were not able to detect the predicted product mRNA. We have identified a transcript for a minor form of tubulin that has several short regions of sequence similarity to GM-CSF by generating a fusion cDNA consisting of GM-CSF fused to this tubulin transcript. This tubulin transcript is the same size as the alternative form of GM-CSF mRNA (1.5 Kb), and we believe that it is the source of the additional band resulting from use of the GM-CSF probe in Northern blot experiments. Consequently, we are unable to observe convincing evidence for the longer forms of GM-CSF and believe their observation can result from short regions of sequence similarity with other transcripts, combined with the methods used in their detection.
