Many genes not associated previously with salivary gland PCD are differentially expressed

Read Also

We conducted pairwise comparisons between SAGE libraries and associated the differentially expressed genes with biological annotations.   In the 16 hour versus 23 hour comparison, we found 522 (12.1%) transcripts upregulated significantly (P £ 0.05) and 331 (7.7%) transcripts downregulated significantly (P £ 0.05).  Together, these transcripts account for almost 20% of all transcripts expressed in the salivary glands during these two stages.   In the 16 hr versus 20 hr and 20 hr versus 23 hour comparisons, we found 288 (7.0%) and 459 (11.2%) transcripts significantly upregulated as well as 255 (6.2%) and 287 (7.0%) transcripts significantly downregulated, respectively.  A complete list of differentially expressed transcripts associated with biological annotations is available in the Supplementary Material.   Table 1 and the description below includes a subset of the differentially expressed genes annotated previously [9], sorted here by functional category and annotated using controlled vocabulary Gene Ontology (GO) molecular function terms [10]. 

Protein Synthesis: Our findings are consistent with the notion that autophagic cell death, though a degradative cellular process, requires active protein synthesis.  We found significant upregulation of one translational elongation factor, one translation termination factor and at least six different translation initiation factors (Table 1).  

Ecdysone-induced and hormone-related genes: We detected multiple ecdysone-induced genes, in addition to those already described, which were differentially expressed prior to salivary gland death.  Abundantly expressed were members of the L71, or Eig71E, late gene family.  The function of the L71 genes has not been established but they are reported to be induced in late third instar larvae [11].  Their abundance at 16 hr APF and decline by 23 hr APF is consistent with a role during the early larval ecdysone pulse.  Eip63F-1, a calcium binding EF-hand family member, and Eip71CD (or Eip28), a protein-methionine-S-oxide reductase, both peaked in gene expression at 20 hr APF, similar to the profile we observed for E74 and E75.  While Eip63F-1 has been implicated in calcium-dependent salivary gland glue secretion during earlier stages of salivary gland development [11], a role for Eip63F-1 or Eip71CD in salivary glands at the prepupal –pupal stage transition has not been described.  Similarly, a role for Hormone-receptor-like in 78 (Hr78) at this stage has not been characterized.

Transcription factors: Our findings indicate that transcriptional regulators other than the known ecdysone-induced factors, may be involved in autophagic cell death regulation.  Transcription factors with an expression profile similar to E74 and E75 include bunched (bun), a RNA polymerase II, and EP2237, a transcriptional activator implicated previously in sensory organ development [9].  Also upregulated was Drosophila maf-S, a gene similar to a v-maf musculoaponeurotic fibrosarcoma oncogene family member in humans [9].  Another upregulated transcription factor, CG3350, has no previous associated function.

Signal Transduction:  We detected increased expression of genes implicated in multiple different signal transduction pathways, emphasizing the likely complex interplay of signaling pathways in autophagic cell death.  One gene highly induced was A kinase anchoring protein 200 (akap200).  In general, Akaps function in cyclic AMP-dependent protein kinase (PKA) signal transduction, targeting bound PKA to docking sites in organelles or the cytoskeleton [12].  Redistribution of the cytoskeleton is a feature of autophagic cell death [13], and it is possible that Akap200 plays a role in cytoskeleton remodeling [12].  Genetic studies in Drosophila have also implicated akap200 as a negative regulator of Ras pathway signaling [14] and thus it may regulate PCD via this pathway.   Another gene significantly upregulated was Darkener of apricot (Doa), a dual specificity LAMMER kinase [15] that is involved in the differentiation of a wide variety of cell types.  Our findings indicate that Doa, in addition to several other differentially expressed kinases and phosphatases identified (Table 1), may also be involved in regulating autophagic cell death. 

Detection of members of the Drosophila defense response pathways (i.e. Toll pathway and imd/TNFα-like pathway; Table 1 and Supplemental Material) suggests that these pathways or some of their components may play a role in developmentally regulated autophagic cell death.  In mammals, TNFα signaling can lead to NFκB activation or to apoptosis [16], and has been linked to a possible autophagic type of cell death in T-lymphoblastic leukaemic cells [17].  In Drosophila, the TNFα-like pathway functions in both apoptosis and the immune response [16], and our results indicate that it may also be involved in autophagic cell death. 

Cell Death related genes:

 Our analyses showed that multiple genes involved in apoptotic cell death are also expressed during autophagic cell death, supporting the notion that these two processes can utilize common pathways or pathway components [3, 6].    In addition to the cell death genes described previously in the salivary gland, we identified additional genes associated, in other tissues, with apoptotic cell death.   Besides dronc, a second caspase, dcp-1, was upregulated transcriptionally in pre-death stage salivary glands.  In addition to the CD36-related scavenger receptor crq, we detected upregulation of three other CD36-related scavenger receptor genes whose function has not yet been characterized.   The expression of additional cell death-related genes, death executioner Bcl-2 homologue (debcl or dborg-1), buffy/dborg-2, iap-1, dredd and sickle, was detected in salivary glands and showed low level changes or no changes in expression levels (Table 1 and Supplementary Material).  It is possible that these genes play a role in salivary gland death but are regulated primarily at the protein level. Given the overlap of genes involved in autophagic and apoptotic cell, it is reasonable to expect that some of the novel autophagic cell death associated genes identified in this study may also be associated with apoptotic cell death.

Autophagy associated genes:  Our results suggest that genes associated with the process of autophagy (i.e. bulk cellular degradation) can be regulated transcriptionally and that this regulation is likely integral to the mechanism of autophagic cell death.  Known genes involved in autophagy have been defined largely by genetic screens in yeast and include at least 16 autophagy-defective (apg) genes and 6 autophagy (aut) genes, with overlap between the two groups [18, 19].  Based on other reports [9, 20] and our own analyses (Experimental Procedures), we identified putative Drosophila orthologs of at least 10 of the apg/aut genes, and found evidence of expression for at least nine of these (Experimental Procedures and Table 1).  Strikingly induced prior to cell death was CG6194 (Table 1), one of two Drosophila genes similar to apg4/aut2, a yeast gene encoding a novel cysteine endoprotease required for autophagy [in 18, 19].  Recently, CG6194 was demonstrated to encode a functional homolog of APG4/AUT2 and shown to interact genetically with several members of the Notch signaling pathway [20].  Results of real-time RT-PCR analyses indicated upregulated expression of other apg/aut-like genes including CG1643 (apg5-like), CG10861 (apg12-like) and CG5429 (apg6-like) (Table 1).  In addition to apg/aut-like genes, we found evidence for upregulated expression of Drosophila rab-7, one of several rab gene family members implicated in autophagy in yeast and humans [18].   

          The terminal phase of autophagy involves autolysosome formation by fusion of the autophagosome with a lysosome, and subsequent degradation of sequestered cellular components.  Lysosomal components with upregulated transcripts in pre-death stage salivary glands include lysozyme, beta-galactosidase, and cathepsins B, D, E, F, and L.  Our analyses indicate that multiple components involved in autophagy are conserved in Drosophila and likely play a role in ecdysone-induced autophagic cell death in the salivary glands. 

Unknowns: In addition to assigning a possible new role to genes already annotated functionally, we have implicated in the autophagic cell death process more than 732 differentially expressed genes with unknown function (Supplementary Material).  377 of these genes were unpredicted and 48 of these genes are represented solely by our salivary gland ESTs.  A major challenge is to identify which genes, both previously described and newly discovered, are likely to play an important role in the autophagic cell death process.  We have initiated mutant expression analyses (below) as one strategy to help prioritize candidate genes for further studies.