A stapled peptide mimetic of the CtIP tetramerization motif interferes with double-strand break repair and replication fork protection

A stapled peptide targets CtIP tetramers to inhibit DNA repair and exert synthetic lethality in BRCA1-mutant cancer cells.


INTRODUCTION
DNA double-strand breaks (DSBs) are the most cytotoxic lesions induced by ionizing radiation and certain chemotherapeutic agents. In contrast to normally dividing cells, hyperproliferating cancer cells are commonly exposed to higher loads of DSBs due to oncogeneinduced replication stress and inherited or acquired defects in components of the DNA damage response machinery (1). Consequently, efficient DSB repair is pivotal for cancer progression and frequently associated with therapy resistance (2). DSBs are mainly repaired by classical nonhomologous end joining (c-NHEJ) and homologous recombination (HR) in an error-free manner. Otherwise, DSBs with flanking homologous sequences can be subjected to mutagenic pathways, including alternative end joining (a-EJ) (3).
The human CtIP protein, in conjunction with the MRE11-RAD50-NBS1 (MRN) nuclease complex, plays a critical role in DSB repair pathway choice through promoting DNA end resection, the initial step in HR (4). A-EJ, which is frequently found up-regulated in HR-deficient tumors associated with BRCA1 mutations, equally relies on CtIP-dependent end resection to expose microhomology regions used for annealing and repair of DSBs (5). Intriguingly, a study using mouse models of human breast cancer proposed that CtIP could promote tumorigenesis by facilitating a-EJ-dependent chromosomal instability (6). We have recently shown that CtIP protects nascent DNA at stalled replication forks from nucleolytic degradation by DNA2 (7). Moreover, we revealed that CtIP acts independently from BRCA1 in fork protection and that loss of CtIP in BRCA1deficient cells aggravates replication stress-induced genomic instability, ultimately leading to increased cell death. On the basis of these findings, targeted inhibition of CtIP in BRCA1-mutated breast and ovarian cancers might represent a promising therapeutic approach (7).
Over the past decade, CtIP has emerged as a polyvalent adaptor protein containing several short linear sequence motifs implicated in protein-protein interactions. The overall sequence similarity between CtIP counterparts of different species is extremely poor, with the most conserved regions located at the N and C terminus. The N-terminal domain (NTD) harbors a coiled-coil region implicated in CtIP oligomerization (8)(9)(10); meanwhile, the C-terminal domain (CTD) facilitates DNA binding and MRN interaction (4,11). A short, structurally defined -helical motif in the CtIP-NTD is required for proper tetramerization, conveying an architecture essential for effective DSB repair by HR. CtIP mutation L27E blocks tetramerization and completely abrogates CtIP recruitment to DSB sites and HR (8). Overall, a DNA bridging mechanism was proposed, whereby CtIP helps to link DNA ends, thereby positioning its CTD close to the breaks to promote DNA end resection by the MRN complex (12,13). Accordingly, recombinant CtIP adopts a dumbbell-like structure, in which two globular DNA binding domains are held apart by a flexible "rod" corresponding to the coiled-coil domain of CtIP (14). Notably, Schizosaccharomyces pombe Ctp1 can form synaptic filamentous complexes on double-stranded DNA (dsDNA) and bridges dsDNA chains intra-or intermolecularly (15). Here, to establish proof of principle for therapeutic targeting of CtIP, we developed a cell-permeable constrained peptide inhibiting CtIP functions in genome maintenance by interfering with CtIP tetramer assembly.
approach. Precisely, we aimed to design a cell-permeable constrained peptide mimicking the -helical tetramerization motif. Hydrocarbon stapling constitutes the most successful and most widely used strategy to reinforce an -helical conformation (16). Two olefin-bearing non-natural amino acids are incorporated in place of nonessential residues and cross-linked using rutheniumcatalyzed ring-closing olefin metathesis (17). It is noteworthy that hydrocarbon stapling of peptides increases both their proteolytic stability and membrane penetrance (18,19).
In addition, endogenous CtIP was also prone to accumulate in focal structures upon SP 18-28 treatment ( fig. S5B). These findings suggested that cellular treatment with SP 18-28 promotes the aggregation of CtIP tetramers, further corroborating our in vitro data. To examine whether these nonfunctional CtIP aggregates are more prone to being cleared by protein degradation, we performed Western blot analysis of time-course lysates from SP 18-28 -treated cells. However, we only observed a minor impact on steady-state levels of CtIP in a cycloheximide (CHX) chase experiment ( fig. S5, C and D). DNA end resection generates 3' single-stranded DNA tails that are immediately coated by replication protein A (RPA) and subsequently replaced by RAD51 to form recombinogenic nucleoprotein filaments (3,4). To study the impact of SP 18-28 on DNA end resection, we quantified RPA accumulation at H2AX-decorated laser tracks. Intriguingly, upon treatment with SP 18-28 , U2OS cells displayed a pronounced decrease in RPA-positive stripes that was comparable to that observed in CtIP-depleted cells (Fig. 4C). As DNA end resection predominantly occurs during the S and G 2 phases of the cell cycle (23), it was important to demonstrate that SP 18-28 treatment neither interferes with bulk DNA replication nor causes any significant changes in cell cycle distribution and cell proliferation (fig. S5, E and F). From these analyses, we concluded that incubation of cells with SP 18-28 compromised CtIP accumulation and resection at DSB sites, most likely due to the induction of dysfunctional higher-order CtIP complexes.

SP 18-28 interferes with homology-directed repair and replication fork protection
CtIP-mediated resection promotes error-free repair of DSBs by HR but, under certain conditions, also facilitates mutagenic repair by a-EJ implicated in tumorigenesis (5,6). To evaluate the effect of SP 18-28 on HR and a-EJ, we took advantage of two established DSB reporter cell lines (10,24). We first analyzed HR efficiency of I-SceIinduced DSBs in U2OS cells following incubation with CtIP-mimetic peptides and observed a nearly threefold decrease in HR frequency in the presence of SP 18-28 that was comparable to CtIP depletion, while LP 18-28 had no measurable impact (Fig. 5A). Moreover, SP 18-28 treatment caused a significant reduction in a-EJ frequency (Fig. 5B), reminiscent of what has been reported for tetramerization-and dimerization-deficient CtIP mutants (8,10).
Besides its well-established role in the processing of DSB ends together with MRN, CtIP was recently reported to protect stalled replication forks from DNA2-dependent nucleolytic degradation (7). To first assess whether CtIP tetramerization itself is required for fork protection in response to DNA replication stress, we first performed dual-labeling DNA fiber assays in U2OS GFP-CtIP-wt and U2OS GFP-CtIP-L27E cells treated with hydroxyurea (HU) to induce fork stalling ( fig. S6). Consistent with a crucial function of CtIP tetramers in stabilizing stalled replication forks, we found that expression of CtIP-L27E failed to rescue nascent strand degradation   in CtIP-depleted cells ( fig. S6). We found that SP 18-28 treatment promotes extensive fork degradation in U2OS cells, comparable to that observed upon CtIP depletion (Fig. 5C). SP 18-28 treatment of U2OS GFP-CtIP-L27E cells did not result in any additive increase in fork degradation ( fig. S6), indicating that SP 18-28 mediates its cellular activity largely through selective CtIP inhibition. Collectively, these findings further substantiate the use of SP 18-28 as a potent and specific inhibitor of CtIP tetramerization and, consequently, of homologydirected DSB repair and fork protection.

SP 18-28 confers DNA damage hypersensitivity and selectively kills BRCA1-mutated cancer cells
Given that unrepaired DSBs are highly cytotoxic, CtIP-depleted cells are hypersensitive to various DNA-damaging agents, most prominently to drugs inducing replication-associated DSBs, such as DNA interstrand-cross-linking drugs, DNA topoisomerase poisons [e.g., camptothecin (CPT)], and PARP1 inhibitors (e.g., olaparib) (4,(25)(26)(27). Therefore, we next sought to examine the peptide's potency to enhance cell death following treatment with anticancer drugs using clonogenic survival assays. Continuous exposure of HeLa cells to 5 M SP 18-28 considerably reduced colony formation when combined with increasing concentrations of either CPT or olaparib (Fig. 6, A and B).
To more precisely determine the dose-dependent effects of acute versus chronic peptide treatment on DNA damage sensitivity, we performed clonogenic assays with HeLa cells grown in the absence or presence of olaparib ( fig. S7A). While an acute (24-hour) treatment with 10 M SP 18-28 is well tolerated and resulted in significant olaparib hypersensitivity, a prolonged chronic treatment with the same dose caused extreme cytotoxicity on its own ( fig. S7A). Consistent with data obtained from HeLa cells (Fig. 6A), chronic treatment with 5 M SP 18-28 significantly resensitized CtIP-wt-but not CtIP-L27E-expressing U2OS cells to CPT, indicating that peptide-mediated chemosensitization can be mainly attributed to the lack of functional CtIP tetramers ( fig. S7B).

DISCUSSION
Cancer cells strongly rely on efficient mechanisms sensing and repairing different types of DNA damage to survive and proliferate.   CtIP-mediated DNA end resection is required for DSB repair by HR to resist conventional DNA-damaging anticancer regimens (2). We reasoned that targeted inhibition of CtIP's resection activity may provide a suitable approach for enhancing the efficacy of radio-and chemotherapy and may also be applied as monotherapy in certain genetic contexts based on the concept of synthetic lethality. Recent findings have established that CtIP and Ctp1 tetramerization, which is mediated via a structurally defined short -helical motif in the N terminus, is crucial for its DNA bridging and repair function (8,12,14,15).
CtIP assembles as a dumbbell-like structure, whereby two C-terminal DNA binding domains are held apart by two coiled-coil domains (14). Intriguingly, the L27E point mutation not only impairs the dimerof-dimers arrangement but also heavily reduces DNA binding (14).
Consequently, targeting tetramerization may cause severe alterations in CtIP function. Biochemical evidence suggests that the tetramer seems to be the constitutive oligomeric state of CtIP (13). On the basis of our in vitro experiments, SP 18-28 specifically binds the CtIP tetramerization domain, but instead of disrupting the tetramers into its dimeric or monomeric counterparts, the stapled peptide promotes extensive CtIP aggregation. Sae2, the yeast counterpart of CtIP, forms an inactive soluble multimeric complex during G 1 phase and transitions to active oligomers upon extensive phosphorylation in S-G 2 (31). Similarly, SP 18-28 seems to trap CtIP in an inactive multimeric protein complex by an as-yet unknown mechanism. SP 18-28 neither bound the N-terminal coiled-coil domain nor aggregated tetramerizationdeficient mutants CtIP-NTD 18-145,L27E or CtIP-NTD  . Thus, we hypothesize that the CtIP-mimetic peptides can stably associate with the CtIP tetramerization motif and induce the formation of complex CtIP-peptide hetero-oligomers.
Several studies have shown that CtIP/Ctp1 promotes inter-and intramolecular DNA bridging (12,14,15). Notably, Andres and colleagues (15) reported that Ctp1-mediated DNA bridging relies on the formation of synaptic filaments involving, on average, 12 Ctp1 tetramers. These findings suggest that dynamic oligomerization states of Sae2/Ctp1 are critical for efficient DSB repair. However, no reports about the dynamic control of CtIP oligomerization in a biological context are available. It is conceivable that CtIP may be able to transit between different oligomeric conformations through distinct post-translational modifications or DNA binding as it was reported for Sae2/Ctp1 (12, 31, 32). Overall, our findings suggest that CtIP can adopt distinct conformational states, such as inactive higher-order multimers, to regulate DSB repair. Hydrocarbon stapling frequently increases membrane penetrance, whereas native peptides commonly do not cross the cell membrane barrier (33). By hiding the hydrophilic peptide backbone and presenting the hydrophobic residues on one side, stapling may alleviate membrane crossing (22). Penetration mostly occurs through endocytosis, which conforms with our observation of intracellular punctuate staining patterns upon incubation with FITC-labeled SP [18][19][20][21][22][23][24][25][26][27][28] (22,34). This uptake mode is significantly slower than direct transduction through the membrane, which is frequently observed with cell-penetrating peptides, and requires time to escape from endosomes (34). Although a major part seemed to be trapped in endosomes, a significant fraction of SP 18-28 reached the nucleus and was sufficient to trigger a biological response. We speculate that the overall positive net charge of SP 18-28 combined with an extended hydrophobic interface conferred the peptide a higher intracellular uptake rate than its longer counterpart SP [18][19][20][21][22][23][24][25][26][27][28][29][30][31] (18).
Last, we report that inhibition of CtIP tetramerization interferes with its recently established role in the protection of reversed forks (7). The observed synergism between CtIP and BRCA1 in alleviating replication stress-induced chromosomal instability could be exploited by treating BRCA1-mutated cells with the CtIP inhibitor SP [18][19][20][21][22][23][24][25][26][27][28] (7). We observed specific cell killing in a BRCA1-deficient background, whereas we did not perceive a negative impact on cell survival of BRCA1 wt cell lines MCF7, MCF10A, and RPE1. Whether the lack of toxicity in MCF10A and RPE1 cells can solely be attributed to the presence of BRCA1 or whether it can also be partially explained by an overall decreased cellular uptake rate in normal versus cancer cells requires further investigation (36). SP 18-28 seems to be a potent inhibitor of functional CtIP oligomerization with a relatively low dissociation constant in the nanomolar range. Nevertheless, we had to provide peptide concentrations of 5 M or higher to observe bioactivity. A major limitation of the stapled peptide is its predominant endocytic uptake resulting in high peptide amounts being trapped in endosomes. To transform this first-generation compound into a lead compound, diversification of staple type and position combined with fine-tuning core hydrophobicity and positive net charge is likely to increase cellular uptake and robust bioactivity (37,38). Prominent amino acid substitutions include the incorporation of arginines or arginine derivatives, namely homo-arginine and 4-guanidino-phenylalanine (38). Alternatively, complementary drug delivery strategies including the use of cellpenetrating peptides and nuclear localization signals could be used to improve cytoplasmic and nuclear uptake efficiencies (36,38).
Overall, SP 18-28 constitutes a potent peptide therapeutic whose bioactive concentration range is comparable to successful peptidomimetics, including various p53/MDM2-targeting stapled peptides and a proapoptotic BH3-mimetic peptide (19,21). Intriguingly, clinical trials of ALRN-6924, a stapled peptide reactivating p53 expression, are underway and propose promising future applications of constrained peptides in the clinic (39). Notably, a recent report suggested that the extended drug interface of peptide therapeutics entails a higher resilience to point mutations in the target protein and impedes the acquisition of drug resistance (40).
We speculate that the stapled CtIP peptide inhibitor could have diverse areas of application in cancer therapy. In addition to conferring increased sensitivity to conventional DNA-damaging agents, CtIPtargeting therapeutics could be used in combination with PARPi in HR-proficient tumors. HR-mediated repair of DSBs predominantly operates during S phase of the cell cycle and is thus more significant for rapidly dividing cancer cells than for neighboring, healthy cells, providing a broad therapeutic window. Inhibition of mutagenic DSB repair by a-EJ may prolong the response of HR-deficient tumors to PARPi by preventing resistance acquisition (41). In addition to HR reactivation, restoration of fork protection seems to be a major mechanism conferring PARPi-and chemoresistance (42,43). Consequently, administration of SP 18-28 would allow a two-pronged strategy for treating PARPi-resistant cancers by simultaneous disruption of HR and fork protection. Last, we provide evidence that SP [18][19][20][21][22][23][24][25][26][27][28] can be applied in BRCA1-deficient tumors and is not toxic to noncancerous cell lines, which would allow selective tumor killing.

siRNA and antibodies
Small interfering RNA (siRNA) sequences are listed in table S2. Transfections were performed with a final concentration of 10 nM using Lipofectamine RNAiMAX (Thermo Fisher Scientific) according to the manufacturer's instructions. A detailed list of primary and secondary antibodies is provided in table S3.

Fluorescence polarization
CtIP-NTD  or CtIP-NTD  was serially diluted in black, flat-bottom 96-well plates (Greiner) in 20 mM tris (pH 8.0) and 150 mM NaCl. FITC-SP 18-28/18-31 stock (1 mg/ml) was diluted at 1:10,000, and 50 l of protein dilutions was mixed with 50 l of FITC-labeled peptides before incubating for 10 min at room temperature (RT). Fluorescence polarization was recorded with an excitation wavelength of 470 nm, emission wavelength of 527 nm, 20-nm emission band-width, and 100 reads per well using a Tecan Safire 2 spectrometer. K d values were calculated by nonlinear regression of dose-response curves.

Protein cross-linking
Recombinant proteins (1 or 2 g) were mixed with peptides in 1× phosphate-buffered saline (PBS) and incubated at RT for 30 min with gentle shaking. Chemical cross-linking was carried out with 100 M disuccinimidyl suberate (Sigma-Aldrich) at RT for 30 min with gentle shaking. Cross-linking reactions were quenched by the addition of 50 mM tris (pH 7.5) for 5 min at RT before boiling in 1× SDS sample buffer [5 mM tris (pH 6.8), 10% glycerol, 1.6% SDS, 100 mM DTT, and 0.02% bromophenol blue] for 5 min. Protein samples were separated by SDS-PAGE, and the gels were stained with InstantBlue (Expedeon).

Confocal and immunofluorescence microscopy
Cells were seeded into eight-well chamber imaging slides (Ibidi) and grown overnight. Upon treatment with 10 M FITC-labeled peptides for 24 hours, cells were imaged in Live Cell Imaging Medium (Gibco, Thermo Fisher Scientific) containing Hoechst 33342 (0.5 g/ml; Thermo Fisher Scientific) using CLSM SP5 Mid UV-VIS Leica with 63× objective at 37°C and ambient CO 2 concentrations. Nuclear peptide uptake was evaluated by washing cells twice with cold PBS before preextraction for 5 min on ice [25 mM Hepes (pH 7.4), 50 mM NaCl, 1 mM EDTA, 3 mM MgCl 2 , 300 mM sucrose, and 0.5% Triton X-100], fixation with 4% formaldehyde (w/v) in PBS for 15 min at RT, and imaging with Leica DM6, 63× objective.
Laser microirradiation was carried out as described previously (25). Briefly, cells were grown in 10 M 5-bromo-2′-deoxyuridine (BrdU) for 24 hours before irradiation. Laser microirradiation was performed using a MMI CELLCUT system containing an ultraviolet (UV) A laser of 355 nm. Energy output was set to 50%, and each cell was exposed to laser beam for <300 ms. Cells were released for 30 min before fixation in 4% formaldehyde (w/v) in PBS for 15 min and permeabilization with 0.5% Triton X-100 (w/v) in PBS for 5 min at RT. After blocking with 3% FCS (w/v) in PBS for 1 hour, cells were stained with primary antibodies (table S3) for 2 hours. Staining with  secondary antibodies (table S3) was performed for 1 hour. Coverslips were mounted with Vectashield supplemented with DAPI (4′,6-diamidino-2-phenylindole; Vector Laboratories), and images were acquired on a Leica DM6, 63× objective. For CtIP foci analyses, cells were fixed in 4% formaldehyde (w/v) in PBS for 15 min before permeabilization with 0.5% Triton X-100 (w/v) in PBS for 5 min at RT. A blocking step with 3% FCS (w/v) in PBS for 1 hour was followed by primary (2-hour) and secondary (1-hour) staining at RT.

Immunoblotting
Cells were lysed in Laemmli buffer [4% SDS, 20% glycerol, and 120 mM tris (pH 6.8)], resolved by SDS-PAGE, and transferred to nitrocellulose membranes. Immunoblotting was performed with indicated primary antibodies (table S3) overnight at 4°C and horseradish peroxidaseconjugated secondary antibodies (table S3) for 1 hour at RT. Proteins were visualized using the Advansta Western Bright enhanced chemiluminesence reagent (Advansta) and Fusion Solo S imaging system. For the CHX chase assay, U2OS cells were either mock-treated or incubated with the indicated peptides for 24 hours and afterward treated with CHX (50 g/ml; Sigma-Aldrich) for 0, 2, or 6 hours before cell lysis.

Flow cytometry analysis
HeLa cells were seeded into six-well plates and incubated with peptides in a volume of 2 ml for varying time points. Cells were harvested by trypsinization, washed, and resuspended in PBS before subjecting them to flow cytometry analysis. FITC intensity was measured with an Attune Nxt flow cytometer equipped with a 488-nm laser and 530/30 band-pass filter. Analysis of 5-Ethynyl-2′-deoxyuridine (EdU) incorporation was carried out using the Click-iT EdU technology (Thermo Fisher Scientific) as described in the manufacturer's instructions. A minimum of 20,000 events were recorded.

Cell proliferation assay
A total of 125 and 250 U2OS cells were seeded in triplicates in 96-well plates in a volume of 90 l. Twenty-four hours after seeding, 10 l of medium only or medium with the indicated peptides was added to the wells reaching a final peptide concentration of 5 M. Cell proliferation was monitored using a CellTiter-Blue-based (Promega) approach at specific time points (0, 2, 5, and 7 days after peptide treatment). In brief, 20-l CellTiter-Blue reagent (Promega) was added to the wells, incubated for 4 hours, and fluorescence intensity was measured at 560/590 nm using a SpectraMax M5 microplate reader (Molecular Devices).
HR and a-EJ reporter assay HR reporter assay was performed as described previously (45,46). Briefly, U2OS EGFP-HR were seeded into a 12-well plate and the day after transfected with pCBA I-SceI expression plasmid using jetPRIME transfection reagent (Polyplus). Four hours later, medium was exchanged, and cells were incubated with peptides for 24 hours in 1-ml total volume. Twenty-four hours later, medium was replaced, and cells were harvested 48 hours after I-SceI transfection.
A-EJ reporter assay was performed according to (10) with some minor modifications. Specifically, U2OS EGFP-aEJ were seeded into six-well plates and siRNA transfected. Six hours later, cells were transfected with I-SceI expression plasmid (pCBA) using Fugene 6 (Promega). Medium was exchanged 24 hours after and replaced with 2-ml fresh medium and the peptides (10 M). Twentyfour hours later, medium was again replaced with fresh medium without peptides, and 72 hours after I-SceI transfection, cells were harvested. GFP expression (readout for HR and a-EJ frequency) was measured by flow cytometry using the Attune Nxt flow cytometer equipped with a 488-nm laser and 530/30 band-pass filter. A minimum of 20,000 events were recorded.
DNA fiber analysis DNA fiber analysis was carried out according to (7). U2OS cells were plated into six-well plates and incubated with 2 ml of medium/peptide mix. Twenty-four hours later, U2OS cells were labeled with 33 M CldU (Sigma-Aldrich) for 30 min, 340 M IdU (5′-iododeoxyuridine; Sigma-Aldrich) for 30 min, and followed by treatment with 2 mM HU (Sigma-Aldrich) for 4 hours. Cells were lysed [200 mM tris (pH 7.4), 50 mM EDTA, and 0.5% SDS], DNA was stretched onto glass slides, and fibers were fixed in MeOH:acetic acid (3:1). Rehydration with PBS was followed by denaturation with 2.5 M HCl for 1 hour and a PBS wash. DNA fibers were blocked in 2% bovine serum albumin and 0.1% Tween 20 (w/v) in PBS for 40 min. CldU/IdU tracks were immunostained using anti-BrdU primary and corresponding secondary antibodies for 2.5 hours each (see table S3). Coverslips were mounted using ProLong Gold Antifade Mountant (Life Technologies), and images were acquired on a Leica DM6 microscope, 63× objective. DNA fiber lengths were analyzed using Fiji software.

Clonogenic survival assay
Two hundred cells per well were seeded into poly-l-lysine (Sigma-Aldrich)-coated 24-well plates and the next day treated with respective drugs (CPT, Sigma-Aldrich; olaparib, Selleck Chemicals) and peptides (total volume: 300 l per well). For details, see figure legends. After 10 days of growth, cells were fixed with crystal violet solution [0.5% crystal violet and 20% ethanol (w/v)]. Plates were scanned, and survival was analyzed with the ImageJ plugin Colony Area using the parameter colony intensity as readout (47).

Quantification and statistical analysis
Statistical analyses were performed using GraphPad Prism. Statistical tests are reported in the figure legends. If not indicated otherwise, each experiment was repeated at least three times. If the data conformed to a normal distribution, then an unpaired two-tailed t test was used. One-way analysis of variance (ANOVA) and Tukey's multiple comparison test were used when comparing multiple groups with each other. Two-way ANOVA and Sidak's multiple comparison test were applied to compare multiple groups of two factors. Fiber experiments were performed twice (n = 2), and representative experiments are depicted. Samples were subjected to a Mann-Whitney analysis. P values of <0.05 were considered statistically significant. ns, not significant; *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001; ****P ≤ 0.0001. Fluorescence polarization was fitted with nonlinear regression using the model Y = B max * X h / (K D h + X h ), where binding at equilibrium by B max is the maximum specific binding, K D is the ligand concentration needed to achieve a half-maximum binding at equilibrium, and h is the Hill slope.