Serum-dependent and independent regulation of PARP2

Qizhi Sun; Mohamed I. Gatie; Gregory M. Kelly

Together, the results indicate that serum withdrawal does not lead to PARP2 degradation. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder. . https://doi.org/10.1101/483289 doi: bioRxiv preprint factor and/or mitogen stimulation. To investigate this, we cultured different cell lines in 418 the presence or absence of serum and then assayed PARP2 levels. The loss of the PARP2 419 signal in cells cultured in SF medium (Fig. 1A) , and its return when serum was replaced 420 (Fig. 1B) supported the notion that the PARP2 gene was serum responsive. Given this on-421 off appearance at the protein level, and the presence of the putative SRE, we had expected 422 to see changes in PARP2 mRNA expression under the different culturing conditions. This, 423 however, was not the case and the presence of a PARP2 amplicon in cells cultured in the 424 SF medium indicated that the message was available (Fig. 1C) . Furthermore, the fact that 425 the quantitative-RT-PCR results showed no significant differences in expression in cells 426 cultured under the different conditions (Fig. 1D) indicated that the regulation of PARP2 in 427 cells deprived of serum was not due to the direct regulation of the gene, at least within the 428 time frame of our investigation. Thus, the presence of PARP2 mRNA under the SF 429 conditions did not seem to contribute to the fast return (within one hour) of the PARP2 430 signal after serum was added to cells incubated in CHX to block new protein synthesis (Fig. 431 6A) . Also, the reappearance of the PARP2 signal when serum was added was dose-432 dependent (Fig. 2B) and not reliant on increased transcription that was dependent on the 433 putative SRE in the PARP2 promoter (Fig. 1D) . These results implied that PARP2 under 434 serum-free conditions was either being rapidly degraded (Fig. 2C) proteases at the forefront of candidates responsible for the serum-dependent changes seen 442 with the different cell types (Fig. 1) . Unfortunately, our analysis using a caspase-specific 443 and a broad-spectrum caspase inhibitor (Fig. 3A ) as well as a proteolytic inhibitor cocktail 444 (Fig. 3B ) ruled out the possibility that caspases were responsible for the proteolysis. 445 Subsequent experiments were designed to explore the ability of serine, cysteine, metallo-446 and aspartic proteases to alter PARP2 levels (Fig. 3B) . As with the caspase inhibitors, those 447 routinely employed to prevent proteolysis did not have an effect on the PARP2 signals 448 when cells were cultured in serum-free conditions (Fig. 3B) . Furthermore, the reports that 449 the cysteine protease cathepsin L is present in the nucleus, like PARP2, and is involved in Thus, despite the link between serum starvation to activate the DNA damage response 486 All rights reserved. No reuse allowed without permission.