Researchers at the University of Montreal have been studying how the protein Histone PARylation Factor 1 (HPF1) interacts with Poly (ADP-ribose) polymerases 1 and 2 (PARP1 and PARP2), two proteins that respond when there is damage to DNA1 and which may serve as clinical targets for several types of cancer2. Researchers were particularly interested in looking at the mechanism of the interaction because the abundance of HPF1 is so much less than would be stoichiometrically predicted. In their current research, the dynamic interaction between HPF1 and PARP1 was studied using surface plasmon resonance (SPR) analysis. Researchers used a Reichert4SPR outfitted with an additional valve to determine rate constants for relevant interactions which helped confirm the mechanism they were postulating. The valve allowed them to do a seamless second injection while the first compound which had started dissociating was still near its maximum response. Research presented shows how HPF1 works to encourage Ser modification over Glu/Asp modification even at sub stoichiometric ratios relative to PARP1.1

Despite being less abundant, HPF1 regulates PARP1/2 catalytic output. Researchers postulated a "hit and run" mechanism whereby each HPF1 regulates the serine production of multiple PARP molecules.1 The SPR results were particularly useful in confirming their proposed mechanism because it allowed them to quantitate this interaction, which involves fast on and fast off kinetics, which SPR is well suited to measuring. The rapid association/dissociation of HPF1 with multiple PARP1 molecules initiates serine modification with ADP-ribose before glutamate/aspartate modification can start.1

There are 17 proteins in the PARP family. In general, PARPs exhibit catalytic activity via covalent attachment of ADP-ribose to target proteins with NAD+ as a source of ADP-ribose. PARP1 is one of the best characterized PARPs and the one that accounts for about 90% of the PARP DNA damage response.3 PARP1 is a modular protein with 6 domains including 3 zinc domains and a catalytic domain (CAT) which in turn has 2 domains, a helical domain (HD) and an ADP- ribosyl transferase domain (ART).1


Researchers were interested in determining how PARP1 and 2 and HPF1 interact. 


 Sensor Chip: (1) Streptavidin coated Carboxymethyl Dextran Reichert P/N 13206071
(2) Carboxymethyl Dextran P/N 13206066
 Temperature:  25 C
 Flow Rate:  25 μL/min
 Target: (1) captured DNA SSB (20 nM) bearing a biotin group
(2) amine coupled HPF1
 Analyte: (1) PARP1 was captured at a concentration of 40 nM, then HPF1 was injected by itself, or in the presence of EB47 (5 μM), or BAD (300 μM), added to the running buffer
(2) PARP1 (800 nM) in the absence or presence of DNA (800 nM) and EB47 (5 μM)
Running Buffer 25mM HEPES pH 7.4, 250mM NaCl, 0.1mM TCEP, 1 mM EDTA, and 0.05% Tween 20 (with additives for some experiments)


Researchers used the results acquired from SPR to verify the binding mechanism of HPF1 to PARP1 and PARP2. Single stranded DNA was captured on a streptavidin dextran sensor chip and then PARP1 was bound and allowed to dissociate. While the response was still near a maximum, researchers used an additional valve to inject HPF1 and study its binding to PARP1. Binding of HPF1 alone, HPF1 in the presence of BAD, the non- hydrolyzable NAD+ analog, and HPF1 in the presence of EB47, a NAD+ mimic (Figure 1(a)) was studied. Results indicate that EB47 is a compound that maintains the HD of PARP1 in an open stable conformation and that BAD reacts similarly but not as strongly. Once it was decided that the PARP1-EB47 complex represented the most active form of PARP1 (Figure 1(a)), as well as the most stable species, EB47 was added to the running buffer, prior to running a full concentration series for HPF1, and the kinetics for the PARP1-DNA complex binding to HPF1 were determined (Figure 1(b)). Sensorgrams were fit to a 1:1 binding model in TraceDrawer, resulting in an association rate (ka) of 1.12 x 105 M−1s−1 and a dissociation rate (kd) of 0.071 s−1, which corresponds to a half-life of about 10 seconds. Using these values, the equilibrium dissociation constant KD is 634 nM.1

Figure 1:
(a) Sensorgrams of PARP1 normal dissociation, and binding of PARP1-DNA complex to HPF1 alone and with BAD, or EB47, added to the running buffer are shown. The weakest interaction is with HPF1 alone (KD of 3.5 μM). Addition of BAD to the running buffer increased the affinity of the interaction to a KD of 1.5 μM, and addition of EB47 to the running buffer showed a further strengthening to a KD of about 617 nM.1
(b) HPF1 injections showing binding to PARP1-DNA complex with EB47 in the buffer are labelled as HPF1. Concentrations of HPF1 are 0, 62.5 nM, 125 nM, 250 nM, 500 nM, 1000 nM and 2000 nM. The affinity plot is also displayed in the inset.1


  • A separate additional valve supplied by Reichert used with the Reichert4SPR allowed the researchers to help unravel the mechanism by which HPF1 regulates PARP activity in DNA repair.1
  • Researchers found that the PARP1-EB-47 complex was both the most stable and most active form of PARP1 so EB-47 was added to the running buffer prior to kinetics determination.1
  • Researchers used their Reichert4SPR to confirm that HPF1 regulates PARP1 catalytic reaction via a "hit and run" mechanism by taking advantage of the ability of SPR to study fast on, fast off interactions.1