We used a nuclear run-on assay as a novel approach to study the changes in transcriptional activity that take place in Drosophila melanogaster during heat shock. In response to a rapid temperature upshift, total transcriptional activity in cultured KC161 cells decreased proportionally to the severity of the shock. After extended stress at 37 degrees C (15 min or more), transcription was severely reduced, and at 39 degrees C most transcription was instantaneously arrested. However, strikingly different responses were observed for individual genes. Transcription of histone H1 genes was severely inhibited even under mild heat shock conditions. Transcription of the actin 5C gene decreased progressively with increasing temperature, while transcription of the core histone genes or of the heat shock cognate genes was repressed only under severe heat shock conditions. Transcriptional activation of the D. melanogaster heat shock genes was also investigated. In unshocked cells, hsp84 was moderately transcribed, while transcriptional activity at the other protein-coding heat shock genes was undetectable (less than 0.2 polymerases per gene). Engaged but paused RNA polymerase molecules were found at the hsp70 and hsp26 genes, but not at the other heat shock genes. The rates of transcription increased with increasing temperature with a peak of expression at around 35 degrees C. At 37 degrees C, induction was less efficient, and no induction was achieved after a rapid shift to 39 degrees C. Increased transcription of the heat shock genes was observed within 1-2 min of heat shock, and maximal rates were reached within 2-5 min. Despite very similar profiles of response, different heat shock genes were transcribed at strikingly different rates, which varied over a 20-fold range. The noncoding heat shock locus 93D was transcribed at a very high rate under non-heat shock conditions, and showed a transcriptional response to elevated temperatures different from that of protein-coding heat shock genes. An estimation of the absolute rates of transcription at different temperatures was obtained.

Seven heat shock genes are clustered within 15 kilobases of DNA at the Drosophila melanogaster chromosomal site 67B. They show a complex pattern of expression in the absence of external stress during normal development of this organism. In this paper, we quantitatively compare the abundance of the messenger RNAs for these seven genes at all major stages of Drosophila development and then focus on hsp23 and hsp27 for which available antibodies allow the comparison between the accumulation of the mRNAs and that of their corresponding polypeptides. Transcripts for both genes are maximally abundant in white prepupae. We observe that the amount of hsp23 message decreases more rapidly than that of hsp27 mRNA throughout the pupal period. The maximal abundance of the proteins occurs at the middle of the pupal stage, when their corresponding RNAs have almost completely disappeared. The peaks of expression of the proteins are also broader than those of their transcripts, indicating that the half-lives of the polypeptides are longer. These observations suggest that complex mechanisms regulate the expression of the small heat shock genes during Drosophila development.

The expression of gene 1, a member of the small heat shock gene family from the Drosophila melanogaster chromosomal locus 67B was studied. In contrast to the other heat shock genes, the response of gene 1 to stress was modulated during development. In the absence of stress, gene 1 was expressed at the beginning of pupation, and at a very low level in adult males. Expression of gene 1 was substantially increased by heat shock in pupae, but was one to two orders of magnitude lower in adults or in embryos. Under the same conditions, hsp70 or hsp26 were induced to similar levels in all stages. This developmental effect could be mimicked in cultured Drosophila cells: expression of gene 1 was stimulated by heat shock in the presence, but not in the absence, of the moulting hormone ecdysterone, while the level of expression of hsp26 and hsp70 in response to heat shock was independent of the presence of the hormone. Thus, the presence and activity of the heat shock transcription factor are not sufficient for the maximal response of gene 1 to stress. These results suggest that the heat shock activator protein requires additional factors, which are developmentally regulated, to activate transcription of gene 1. Furthermore, S1 nuclease mapping analysis revealed several gene 1 mRNA species, which are generated by the use of alternative polyadenylation sites and by the use of differentially regulated transcriptional initiation sites.

No abstract