A turn-on split-luciferase sensor for the direct detection of poly(ADP-ribose) as a marker for DNA repair and cell death

Abstract

Designed sensors comprising split-firefly luciferase conjugated to tandem poly(ADP-ribose) binding domains allow for the direct solution phase detection of picogram quantities of PAR and for monitoring temporal changes in poly(ADP-ribosyl)ation events in mammalian cells.

---

The design of turn-on sensors for the detection of biologically important analytes is of much current interest to both chemists and biologists.¹ There have been many recent advances in sensor design for detecting proteins, RNA, DNA, carbohydrates, metabolites, and metal ions.² However, to date there are few methods for the direct solution phase detection of the biologically significant poly(ADP-ribose) (PAR) polymer that plays a central role in the response to DNA damage (Fig. 1). Herein we report a split-protein enabled turn-on sensor for PAR that allows for monitoring PAR dynamics in human cells exposed to DNA damage agents.

Fig. 1 Split-luciferase enabled sensor for the detection of poly(ADP-ribosyl)ation. (A) Chemical structure and schematic representation of poly(ADP-ribose) (PAR). Damaged DNA elicits a repair response in which nuclear proteins are post-translationally modified with PAR as catalyzed by poly(ADP-ribose) polymerase (PARP). (B) Genetically fragmented firefly luciferase halves (CFluc and NFluc) are attached to APLF domains that bind PAR. In the absence of PAR, no signal is generated. In the presence of PAR, the APLF domains bind, allowing for split-luciferase reassembly and concomitant luminescence.

The human genome has been reported to typically accrue >10⁴ lesions daily on a per cell basis from a variety of endogenous and environmental insults.³ Direct chemical modifications include bulky DNA adducts, oxidized or hydrolyzed bases, alkylation products, and strand breaks. In order to survive, cells have evolved specific mechanisms to counter DNA damage, collectively termed the DNA-damage response.⁴ We sought to develop a sensor that can report on a chemical change within a cell, which is invoked in multiple DNA damage-associated pathways. Specifically, we intended to design a biosensor for monitoring the PAR polymer, which is generated upon recruitment of poly(ADP-ribose) polymerase-1 (PARP-1) to sites of DNA damage (Fig. 1A).⁵PARP-1 catalyzes the transfer of the ADP-ribose moiety of NAD⁺ to glutamate residues of nuclear proteins leading to PAR polymer tagging (Fig. 1A), ultimately resulting in chromatin relaxation and recruitment of repair-associated proteins. PAR catabolism by poly(ADP-ribose) glycohydrolase (PARG) results in a very short half life of the polymer during DNA repair.⁵ However, in response to excessive DNA damage, PAR can also serve as a mediator of apoptosis. Thus, the design of sensors for the temporal detection of the PAR polymer has the potential to aid in furthering our understanding of the chemical biology of the DNA-damage response.

For our turn-on PAR sensor design, we have employed split-protein reassembly (also called protein-fragment complementation), wherein initially non-functional fragments of a split-signaling protein are induced to reassemble through the direct interaction of attached domains.⁶ Several signal generating split-proteins have been validated in this regard, including ubiquitin,^7a green fluorescent protein (GFP),^7b β-lactamase,^7c and luciferases.^7d,e Most split-protein systems to date have primarily focused on the detection of protein–protein interactions, though more recently RNA, DNA, and DNA modifications have also been targeted.⁸ Furthermore, split-firefly luciferase has provided a homogeneous assay platform for analysis of protease activity,^9akinase inhibitor selectivity,^9b and RNA-templated assemblies.^9c However, to our knowledge there are no solution phase sensors for direct measurement of PAR polymer levels. In order to generate a bivalent PAR-specific split-protein sensor, we reasoned that the recently identified PAR-binding zinc finger (PBZ) modules of aprataxin PNK-like factor (APLF) (residues 376-441)¹⁰ could be attached to each half of split-firefly luciferase (split-Fluc), generating CLuciferase-APLF and APLF-NLuciferase with the expectation that a statistical distribution of APLF-luciferase halves bound to the PAR polymer would still permit ∼50% complementation and concomitant luciferase activity (Fig. 1B). Potentially, this approach would allow for the direct detection ofPAR polymer in complex aqueous environments without the need for the chemical derivatizations or separation steps that are commonly required. Herein we report the design and validation of our APLF-based split-luciferase biosensor for the PAR polymer.

As an initial validation of our biosensor, we tested an in vitro generated PAR polymer with a chain length between 2–300 units. We translated our split-proteins in a cell-free system using 10 μM ZnCl₂ typically used for Cys2His2 zinc finger domains (Fig. 2A, inset).^8e Incubation of the sensor halves with 50 picograms PAR resulted in a 4.5-fold increase in signal, indicating that the proximate binding of both APLF fusions was feasible (Fig. 2A). We furthermore investigated the use of our sensor for following PAR degradation over 2 h by exposure to the glycohydrolase, PARG. We observed a time dependent decrease in luminescence following incubation of presumably hydrolyzed PAR with our split-luciferase biosensors, again demonstrating that this sensor can potentially detect PAR metabolism (Fig. 2B). We next sought to optimize the effect of ZnCl₂ concentration on luminescent signal output in the presence of PAR, since the PBZ domains of APLF require Zn²⁺ for binding PAR, but the optimum levels are not known.¹⁰ The titration revealed that a concentration of 50 μM ZnCl₂ present during translation provided the maximal signal over background, likely assisting in the proper folding of the Zn²⁺ dependent APLF domain (Fig. 2C). Using these optimized conditions, the lowest quantity of PAR detectable over background was determined to be below 12 picograms (Fig. 2D). Thus, this new sensor provides a rapid and convenient method for directly detecting PAR with high sensitivity in a homogeneous solution, as well as for following PARG dependent PAR polymer hydrolysis.¹¹

Fig. 2 Split-luciferase sensor for detection ofPAR polymers. (A) CLuciferase-APLF and APLF-NLuciferase were translated with 10 μM ZnCl₂, followed by addition of PAR or no PAR, and luminescence was recorded. (inset) CLuciferase-APLF and APLF-NLuciferase were translated in the presence of ³⁵S-methionine and analyzed by SDS-PAGE. (B) PAR was treated with poly(ADP-ribose) glycohydrolase (PARG) for 5, 30, 60, or 120 min, followed by direct analysis with translated CLuciferase-APLF and APLF-NLuciferase to monitor PAR catabolism. (C) CLuciferase-APLF and APLF-NLuciferase were translated in the presence of varying concentrations of ZnCl₂, followed by PAR addition, and the luminescence was recorded. (D) CLuciferase-APLF and APLF-NLuciferase were translated with 50 μM ZnCl₂ and incubated with varying concentrations of PAR, followed by luminescence readings showing a linear response.

Encouraged by the sensitivity of detection ofin vitro generated PAR, we next sought to induce PAR formation in mammalian cell cultures by addition of a genotoxic agent for induction of single-strand breaks. N-methyl-N′-nitro-N-nitrosoguanidine (MNNG) is a known DNA alkylating agent that is reported to primarily generate 7-methylguanine, 3-methyladenine, and O⁶-methylguanine lesions.¹² Although MNNG does not induce direct scission of the DNA backbone, formation of these methyl adducts has been proposed to weaken the N-glycosidic bond, leading to depurination, which is subsequently processed by apurinic endonucleases to yield single-strand breaks. PARP-1 recognizes these breaks and is activated to catalyze induction of PAR. To confirm PAR detection in cell lysates, we treated exponentially growing cultures of HeLa cells with 100 μM MNNG for 3 min, followed by immediate removal of the compound and cell lysis (Fig. 3A). The total protein content of each sample was normalized using the BCA assay, and PAR was analyzed by addition of the sensor halves, CLuciferase-APLF and APLF-NLuciferase (Fig. 3B, 0 min). Since poly(ADP-ribosyl)ation is a dynamic event in cell signaling, with PAR undergoing rapid catabolism by PARG, we anticipated that we could monitor the temporal nature of this modification with our PAR sensor. We challenged HeLa cells with 100 μM MNNG for 3 min, followed by a recovery period of 0, 5, or 10 min. Upon exposure of cell lysates to our split-protein sensors, we observed a 3-fold increase in luminescent signal compared to untreated cells with a 0 min recovery time, followed by a maximal signal at 5 min (5-fold increase) and a return to near basal levels of PAR after 10 min recovery (Fig. 3B). To establish generality, we next evaluated PAR dynamics in a second cell type, MCF7, which were treated with 100 μM MNNG for 3 min, followed by 0, 5, or 10 min recovery. A similar trend in PAR dynamics was observed as compared to HeLa cells. At 0 min recovery a 5-fold increase in signal was observed. A maximum of 8-fold signal increase occurred at 5 min, and PAR levels dropped to 6-fold by 10 min (Fig. 3C). These results likely reflect differences in PAR turnover between the two cells types, which may be a function of the previously reported higher basal PAR levels observed for MCF7 cells.¹³ Importantly, this new PAR sensor allows for a simple means to directly monitor PAR induction and recovery, which has conventionally been limited to traditional immunological probes.¹⁴

Fig. 3 Split-luciferase detection of PAR cellular dynamics. (A) Mammalian cells treated with a DNA alkylating agent, N-methyl-N′-nitro-N-nitrosoguanidine (MNNG), will produce PAR during the DNA-damage response. (B) HeLa cells were treated with MNNG and then allowed to recover for 0, 5, or 10 min. Poly(ADP-ribosyl)ation was detected by CLuciferase-APLF and APLF-NLuciferase, followed by luminescence readings. (C) MCF7 cells were treated as described in (B), followed by detection with CLuciferase-APLF and APLF-NLuciferase.

In conclusion, we have developed a turn-on sensor for the direct detection of post-translational modifications accrued in the DNA-damage response. Specifically, we have adapted APLF as a useful modular domain for the determination of PAR levels involved in single-strand break repair and caspase-independent apoptosis. By employing the genetically encoded split-luciferase as the signaling domain, we are able to synthesize our sensor in vitro in 1.5 h without further purification, providing ease of access to the biosensor. Importantly, this split-protein sensor is capable of reporting on the presence of this transient protein modification from mammalian cells and allows for monitoring temporal changes in PAR levels, which may ultimately aid in the elucidation of the kinetics of DNA repair and identification of differences in PAR dynamics upon exposure to various environmental conditions. This sensor adds to the tool-box for studying the chemical biology of DNA repair and cell death.

This research was supported by the NIH (R21CA143661) and NSF (CHE-0548264). We thank D. Piwnica-Worms for the luciferase constructs. J.L.F. thanks the University of Arizona TRIF Imaging Fellowship for financial support.

Notes and references

1. (a) J. Zhang, R. E. Campbell, A. Y. Ting and R. Y. Tsien, Nat. Rev. Mol. Cell Biol., 2002, 3, 906–918; (b) E. V. Anslyn, J. Org. Chem., 2007, 72, 687–699; (c) J. L. Reymond, V. S. Fluxa and N. Maillard, Chem. Commun., 2009, 34–46.
2. (a) T. D. James, H. Shinmori and S. Shinkai, Chem. Commun., 1997, 71–72; (b) G. Klein, D. Kaufmann, S. Schurch and J. L. Reymond, Chem. Commun., 2001, 561–562; (c) M. D. Shults and B. Imperiali, J. Am. Chem. Soc., 2003, 125, 14248–14249; (d) M. J. McDonough, A. J. Reynolds, W. Y. G. Lee and K. A. Jolliffe, Chem. Commun., 2006, 2971–2973; (e) J. N. Wilson and E. T. Kool, Org. Biomol. Chem., 2006, 4, 4265–4274; (f) S. C. Zondlo, F. Gao and N. J. Zondlo, J. Am. Chem. Soc., 2010, 132, 5619–5621.
3. O. D. Scharer, Angew. Chem., Int. Ed., 2003, 42, 2946–2974.
4. S. P. Jackson and J. Bartek, Nature, 2009, 461, 1071–1078.
5. (a) D. D'Amours, S. Desnoyers, I. D'Silva and G. G. Poirier, Biochem. J., 1999, 342, 249–268; (b) V. Schreiber, F. Dantzer, J. C. Ame and G. de Murcia, Nat. Rev. Mol. Cell Biol., 2006, 7, 517–528.
6. (a) S. W. Michnick, P. H. Ear, E. N. Manderson, I. Remy and E. Stefan, Nat. Rev. Drug Discov., 2007, 6, 569–582; (b) J. Muller and N. Johnsson, ChemBioChem, 2008, 9, 2029–2038.
7. (a) N. Johnsson and A. Varshavsky, Proc. Natl. Acad. Sci. U. S. A., 1994, 94, 8405–8410; (b) I. Ghosh, A. D. Hamilton and L. Regan, J. Am. Chem. Soc., 2000, 122, 5658–5659; (c) A. Galarneau, M. Primeau, L. E. Trudeau and S. W. Michnick, Nat. Biotechnol., 2002, 20, 619–622; (d) R. Paulmurugan and S. S. Gambhir, Anal. Chem., 2003, 75, 1584–1589; (e) K. E. Luker, M. C. P. Smith, G. D. Luker, S. T. Gammon, H. Piwnica-Worms and D. Piwnica-Worms, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 12288–12293.
8. (a) C. I. Stains, J. R. Porter, A. T. Ooi, D. J. Segal and I. Ghosh, J. Am. Chem. Soc., 2005, 127, 10782–10783; (b) A. T. Ooi, C. I. Stains, D. J. Segal and I. Ghosh, Biochemistry, 2006, 45, 3620–3625; (c) C. I. Stains, J. L. Furman, D. J. Segal and I. Ghosh, J. Am. Chem. Soc., 2006, 128, 9761–9765; (d) J. R. Porter, C. I. Stains, D. J. Segal and I. Ghosh, Anal. Chem., 2007, 79, 6702–6708; (e) J. R. Porter, C. I. Stains, B. W. Jester and I. Ghosh, J. Am. Chem. Soc., 2008, 130, 6488–6497; (f) J. L. Furman, A. H. Badran, S. Y. Shen, C. I. Stains, J. Hannallah, D. J. Segal and I. Ghosh, Bioorg. Med. Chem. Lett., 2009, 19, 3748–3751.
9. (a) S. S. Shekhawat, J. R. Porter, A. Sriprasad and I. Ghosh, J. Am. Chem. Soc., 2009, 131, 15284–15290; (b) B. W. Jester, K. J. Cox, A. Gaj, C. D. Shomin, J. R. Porter and I. Ghosh, J. Am. Chem. Soc., 2010, 132, 11727–11735; (c) J. L. Furman, A. H. Badran, O. Ajulo, J. R. Porter, C. I. Stains, D. J. Segal and I. Ghosh, J. Am. Chem. Soc., 2010, 132, 11692–11701.
10. (a) I. Ahel, D. Ahel, T. Matsusaka, A. J. Clark, J. Pines, S. J. Boulton and S. C. West, Nature, 2008, 451, 81–85; (b) S. Eustermann, C. Brockmann, P. V. Mehrotra, J.-C. Yang, D. Loakes, S. C. West, I. Ahel and D. Neuhaus, Nat. Struct. Mol. Biol., 2010, 17, 241–243.
11. (a) K. S. Putt, G. J. Beilman and P. J. Hergenrother, ChemBioChem, 2005, 6, 53–55; (b) A. C. Nottbohm, R. S. Dothager, K. S. Putt, M. T. Hoyt and P. J. Hergenrother, Angew. Chem., Int. Ed., 2007, 46, 2066–2069.
12. M. D. Wyatt and D. L. Pittman, Chem. Res. Toxicol., 2006, 19, 1580–1594.
13. T. Zaremba, P. Ketzer, M. Cole, S. Coulthard, E. R. Plummer and N. J. Curtin, Br. J. Cancer, 2009, 101, 256–262.
14. (a) G. Cipriani, E. Rapizzi, A. Vannacci, R. Rizzuto, F. Moroni and A. Chiarugi, J. Biol. Chem., 2005, 280, 17227–17234; (b) N. I. Ryabokon, R. I. Goncharova, G. Duburs, R. Hancock and J. Rzeszowska-Wolny, Mutat. Res., Fundam. Mol. Mech. Mutagen., 2008, 637, 173–181.

---

Footnotes

† This article is part of the ‘Emerging Investigators’ themed issue for ChemComm.

‡ Electronic supplementary information (ESI) available: Details of experimental procedures. See DOI: 10.1039/c0cc02229b

---