PARP-1 catalyses the covalent attachment of ADP-ribose units on to the γ carboxyl group of glutamate residues in acceptor proteins including PARP-1 itself [5
]. Each of these ADP-ribose units has an adenine moiety capable of base stacking and hydrogen bonding, along with two phosphate groups that carry negative charges. These polymers cause profound changes in the structure and function of key proteins that respond to DNA damage. For example, proteins like histones, topoisomerases I and II, and DNA helicases undergo poly (ADP-ribosylation) and help in regulation of chromatin structure and genomic integrity [3
]. However, a growing body of evidence suggests that in addition to its regulation of chromatin structure, PARP-1 also plays a key role in gene-specific transcription.
PARP-1 regulates gene-specific transcription by two possible mechanisms. Kim et al
. demonstrated that PARP-1 binds to specific nucleosomes and leads to formation of compact, transcriptionally repressed chromatin structures [33
]. Upon activation, PARP-1 is poly (ADP-ribosylated) and dissociates from chromatin leading to formation of decondensed, transcriptionally active euchromatin structures. It has also been proposed that PARP-1 is part of a nucleosome complex along with a histone variant macroH2A (mH2A). At baseline conditions, mH2A and an inactive PARP-1 are associated with the HSP-70 promoter. Upon heat shock, HSP-70 promoter bound PARP-1 is released to activate HSP-70 transcription [34
]. Hence, siRNA mediated knockdown of either PARP-1 or mH2A1 downregulated transcription of HSP-70 gene. This observation was complemented by another study that demonstrated chromosomal "puffing" with increased HSP-70 gene expression in Drosophila
salivary glands after heat shock [35
]. It was proposed that PARP dissociates chromatin proteins at induced chromosomal loci, thus allowing increased transcription of target heat shock genes. In contrast to our study, treatment with an inhibitor of PARP activity reduced puffing and consequently decreased transcription of heat shock genes following heat shock. Our data are consistent with our previous in vivo
study of myocardial ischemia-reperfusion injury, where we demonstrated that mice with genetic ablation of PARP-1 exhibit significant cardioprotection, associated with enhanced upregulation of HSF-1 DNA binding in the heart [26
]. The reasons for the different response of dipterans and mammals to PARP-1 inhibition are unknown. It is plausible that factors including signaling mediators have evolved different effector proteins that can affect the heat shock response and its regulation by PARP-1.
Another potential mechanism which may be more relevant to this study entails the role of PARP as a gene specific transcription enhancer/promoter binding cofactor activity that can enhance or inhibit gene expression. PARP-1 has been shown to interact with the transcription factors NF-κB, HTLV Tax-1 and RAR and this was associated with increased expression from dependent promoters [7
]. Similarly, it has been suggested that PARP-1 is a co-transcription factor for the mammalian achaete-scute
homologue (MASH) gene. PARP-1 is present in an inactive state as part of a co-repressor complex. Upon activation, PARP-1 is required for the dismissal of the co-repressor complex, and a second subsequent event leads to activation of the target MASH gene [36
Our experiments indicate that PARP-1 modulates the heat shock response by functioning as a repressing factor of HSP-70 gene expression. We provide two lines of evidence in this regard. First, in this study we demonstrated that knockdown of PARP-1 gene increased HSP-70 gene expression as evidenced by increased DNA-binding activity of HSF-1, HSP-70 mRNA and protein expression. Secondly, using co-immunoprecipitation we demonstrated that PARP-1 may also regulate HSF-1 activation through direct interaction with this transcription factor. Protein-protein interaction is recognized as a mechanism for PARP-1 to function as a specific transcriptional co-activator of NF-κB [37
Fossati et al
. similarly documented increased HSP-70 expression in murine PARP-1 deficient fibroblasts as compared to wild type fibroblasts [27
]. In contrast to our findings, this study was unable to detect PARP-1 and HSF-1 interaction by co-immunoprecipitation studies. While the cell type utilized in the two studies was remarkably similar, the duration of heat shock was different. In our study, the cells were subjected to 45 min of heat shock in comparison to 30 min in the study by Fossati et al.
]. Other differences that could lead to different results may be the antibody type and protocol design for immunoprecipitation studies.
Other studies have also proven that PARP-1 modulates transcription by direct interaction with AP2 [38
], YY-1 [40
] and TEF-1 [41
]. PARP has been also shown to alter RNA polymerase II dependent transcription [42
] and to effectively prevent and reverse p53 binding to the palindromic p53 consensus sequence [43
]. Before HSF-1 is activated there are a series of processes that involve phosphorylation, translocation from the cytosol to the nucleus, formation of a trimer, binding to heat shock elements (HSE), and initiating HSP-70 gene expression [44
]. It has been postulated that addition of long ADP-ribose tails to transcription factors can disable or dissociate the binding of transcription factors to their DNA recognition sites, also in part by electrostatic repulsion. This modification results in inhibition of transcription. Poly (ADP-ribosylation) of transcription factors prevents their binding to DNA. On the contrary, inhibition of PARP-1 enables the binding of the transcription factors to their specific DNA sites [5
]. Thus, it is possible that both physical interaction with PARP-1 and poly (ADP-ribosylation) of HSF-1 reduce the availability of HSF-1 to initiate transcription. The increase in HSF-1 content, albeit inactive with DIQ pretreatment further reinforces the notion that PARP-1 represses HSP-70 gene transcription. Further studies need to be conducted to understand the precise mechanism as to how and where PARP-1 regulates HSF-1 activation.
In conclusion, our results indicate that PARP-1 serves as a repressing factor of the heat shock response by regulating the expression of HSP-70. Both protein-protein interaction and catalytic activity of the PARP protein play a key role in modulation of the heat shock response.