Mammalian Base Excision Repair: Functional Partnership between PARP-1 and APE1 in AP-Site Repair

Abstract

The apurinic/apyrimidinic- (AP-) site in genomic DNA arises through spontaneous base loss and base removal by DNA glycosylases and is considered an abundant DNA lesion in mammalian cells. The base excision repair (BER) pathway repairs the AP-site lesion by excising and replacing the site with a normal nucleotide via template directed gap-filling DNA synthesis. The BER pathway is mediated by a specialized group of proteins, some of which can be found in multiprotein complexes in cultured mouse fibroblasts. Using a DNA polymerase (pol) β immunoaffinity-capture technique to isolate such a complex, we identified five tightly associated and abundant BER factors in the complex: PARP-1, XRCC1, DNA ligase III, PNKP, and Tdp1. AP endonuclease 1 (APE1), however, was not present. Nevertheless, the complex was capable of BER activity, since repair was initiated by PARP-1's AP lyase strand incision activity. Addition of purified APE1 increased the BER activity of the pol β complex. Surprisingly, the pol β complex stimulated the strand incision activity of APE1. Our results suggested that PARP-1 was responsible for this effect, whereas other proteins in the complex had no effect on APE1 strand incision activity. Studies of purified PARP-1 and APE1 revealed that PARP-1 was able to stimulate APE1 strand incision activity. These results illustrate roles of PARP-1 in BER including a functional partnership with APE1.

Fig 1. Effect of purified BER factors on APE1-independent BER by the pol β complex.

(A) A schematic representation of the DNA substrate containing the AP-site and the reaction scheme is shown. (B) BER activity of the pol β complex was evaluated on an AP- site-containing DNA substrate by measuring incorporation of [α-³²P]dCMP as a function of different components in the reaction mixture and incubation time. Reaction conditions and product analysis are described under Materials and Methods. AP-site DNA was incubated with the pol β complex in the presence (+) or absence (-) of purified BER factors including PARP-1 XRCC1, PNKP, DNA ligase I, as indicated at the top of the phosphorimage. Lane 13 represents the result after incubation of the reaction mixture without the pol β complex or purified proteins. Incubation was at 37°C for 15 and/or 30 min. The reaction products were separated by electrophoresis in a 16% polyacrylamide gel containing 8 M urea. A Typhoon PhosphorImager was used for gel scanning and imaging. The positions of the unligated BER product and ligated BER product are indicated. (C) AP-site DNA was incubated with the pol β complex in the presence (+) or absence (-) of purified BER factors, as indicated below the histogram. The ligated and unligated BER products at 30 min incubation were quantified using ImageQuant software and plotted in a histogram. The grey and black bars represent unligated and ligated BER products, respectively. (D) A histogram illustrating the ratios of ligated BER product to total BER products (both ligated plus unligated BER products) is shown.

Fig 2. Effect of PARP-1 on APE1-dependent BER.

(A) A schematic representation of the DNA substrate containing the AP-site and the reaction scheme is shown. The BER reaction conditions and product analysis are described under Materials and Methods. (B) The BER reaction mixtures containing purified proteins XRCC1, PNKP, DNA ligase I and APE1 were supplemented either with PARP-1 (lanes 1–3) or dilution buffer (lanes 4–6). Repair was initiated by transferring the reaction mixtures to 37°C. Aliquots were withdrawn at 5, 10 and 20 min. The reaction products were analyzed as in Fig 1. The positions of the BER intermediate (unligated) and ligated BER products are indicated. (C) Quantification of the BER products was performed using ImageQuant software and data plotted as a function of incubation time (min). The plot demonstrates that BER product formation was linear during the 20 min incubation and that PARP-1 stimulated BER at least 2-fold as compared to the reaction without additional PARP-1.

Fig 3. Stimulation of APE1 activity by the pol β complex or purified PARP-1.

A schematic representation of the DNA substrate containing the AP-site analogue THF is shown at the top. The reaction conditions and product analysis are described under Materials and Methods. (A) APE1 incision reactions were assembled on ice either with increasing amounts of pol β complex (A) or with increasing amounts of purified PARP-1 (B) The incision reaction was initiated by addition of 0.1 nM APE1 and transferring the reaction mixtures to 37°C for 10 min. The reaction products were analyzed as in Fig 1. The positions of the ³²P-labeled substrate and the product of APE1 strand incision are indicated. A representative phosphorimage of two repeats is illustrated. (C) Quantification of the APE1 product formed at the highest amount of pol β complex (3 μl) and the highest concentration of PARP-1 (50 nM) reveal an approximately 3-fold increase in APE1 activity as compared to that of APE1 alone. The mean of two repeats is illustrated.

Fig 4. Effect of PARP-1 on the steady-state rate of AP-site incision catalyzed by APE1.

The DNA substrate with THF (100 nM) was preincubated with 25–500 nM PARP-1. After adding 0.5 nM APE1, the reaction mixture was incubated for 10 s to 5 min at 37°C. The reaction conditions and data analysis are described in Materials and Methods. The data representing the reaction products were fitted to an exponential equation to determine the steady-state rate of the APE1 incision reaction in the absence and presence of PARP-1. The average from three repeats is represented.

Fig 5. A model illustrating APE1-dependent and-independent mammalian BER coordinated by BER factors in the pol β complex.

AP-site lesions in DNA that are formed by spontaneous hydrolysis of the N-glycosylic bond or by removal of inappropriate bases by DNA N-glycosylases are recognized by PARP-1 [44,50,61]. By virtue of the presence of PARP-1 in the pol β complex, the complex is recruited to the AP-site in DNA. Upon binding to AP-site, PARP-1 is auto-poly(ADP-ribosyl)ated [44]. While the complex remains bound to the AP-site DNA strand, BER may proceed either by an APE1-dependent (left-hand side of the scheme) or APE1-independent (right-hand side of the scheme) pathway. In the case of the APE1-dependent pathway, APE1 incises the AP-site, while the complex is still bound to the AP-site. The dRP removal, DNA synthesis and ligation steps are conducted. On the other hand, in situations where APE1 is deficient, APE1-independent BER operates where PNKP plays a central role [60]. In this case, for example, the complex bound at the AP-site incises the DNA strand by its PARP-1’s lyase activity [50]. Tdp1 and/or PNKP trim or edit the 3′blocked group to generate the 3′-OH necessary for the DNA synthesis and ligation steps, respectively. PAPR-1 is depicted as the blue triangle in the pol β complex.