Novel Macrocyclic Peptidomimetics Targeting the Polo-Box Domain of Polo-Like Kinase 1

Abstract

The polo-box domain (PBD) of Plk1 is a promising target for cancer therapeutics. We designed and synthesized novel phosphorylated macrocyclic peptidomimetics targeting PBD based on acyclic phosphopeptide PMQSpTPL. The inhibitory activities of 16e on Plk1-PBD is >30-fold higher than those of PMQSpTPL. Both 16a and 16e possess excellent selectivity for Plk1-PBD over Plk2/3-PBD. Analysis of the cocrystal structure of Plk1-PBD in complex with 16a reveals that the 3-(trifluoromethyl)benzoyl group in 16a interacts with Arg516 through a π-stacking interaction. This π-stacking interaction, which has not been reported previously, provides insight into the design of novel and potent Plk1-PBD inhibitors. Furthermore, 16h, a PEGlyated macrocyclic phosphopeptide derivative, induces Plk1 delocalization and mitotic failure in HeLa cells. Also, the number of phospho-H3-positive cells in a zebrafish embryo increases in proportion to the amount of 16a. Collectively, the novel macrocyclic peptidomimetics should serve as valuable templates for the design of potent and novel Plk1-PBD inhibitors.

Graphical Abstract

INTRODUCTION

Anti-mitotic drugs, such as the taxanes, have drawn much attention as effective anti-cancer therapeutics.¹ However, the development of specific anti-mitotic agents has proven to be difficult because only a limited number of mitosis-specific targets exist. Polo-like kinase 1 (Plk1) is an essential serine/threonine kinase that regulates various mitotic processes, including mitotic entry, centrosome maturation, bipolar spindle assembly, chromosome segregation, and cytokinesis.² Commensurate with the critical role it plays in promoting mitotic progression and cell proliferation, Plk1 is overexpressed in a broad range of human cancers, and the level of its overexpression closely correlates with the aggressiveness and poor prognosis associated with these cancers.^2,3 These observations suggest that antagonizing Plk1 could serve as an attractive strategy for designing new anti-cancer drugs.

Plk1 possesses a conserved N-terminal kinase domain and a noncatalytic C-terminal polo-box domain (PBD), consisting of two tandem polo-box motifs (i.e., PB1 and PB2).⁴ Plk1-PBD functions as a high-affinity phospho-Ser/Thr-binding module, with an optimal recognition sequence containing Ser-pSer/pThr-Pro/X (where X denotes any amino acid residue). In addition to its substrate targeting and subcellular localization properties, the PBD also regulates the kinase activity of Plk1 via its mutual auto-inhibitory interaction with the kinase domain.^5–7

The availability of well-characterized recognition sequences for the PBD and X-ray crystal structural information^4,7,8 makes the design of PBD-targeting agents a promising approach to developing new anti-Plk1 drugs. Indeed, recent studies examining the cellular consequences of targeted disruption of the function of the PBD have demonstrated that this domain is an attractive target for anticancer agent discovery.^5,8,9 For example, it was observed that a phosphopeptide PLHSpT, 74-PLHSpT-78 residue⁸ of Plk1 binding to polo-box-interacting protein 1, possesses an IC₅₀ value of 36 μM against Plk1 and a binding affinity (K_D) of 450 nM. Introduction of an alkyl-substituted phenyl ring at N-3 of a histidine imidazole in PLHSpT, which interacts with the hydrophobic cryptic pocket, provides PLH*SpT, which has an improved IC₅₀ value (17 nM) and binding affinity (K_D = 1–2 nM) against Plk1.¹⁰ Moreover, the proline and leucine residues of the phosphopeptide PLHSpT have been substituted with aromatic moieties to improve potency and plasma stability.¹¹ Also, affinity is enhanced by replacement of the N-terminal Pro and Leu by a N-ethylbenzamide moiety owing to the formation of a pair of H-bonds with Asp416.¹² Furthermore, it was reported that KBJK557, a barbiturate-fused pyrazole derivative derived by virtual screening, possesses a comparable IC₅₀ value (16.35 μM) to that of PLHSpT (29.02 μM) and exhibits a significant in vivo antitumorigenic effect upon intravenous administration.¹³ Also, 1-thioxo-2,4-dihydro-[1,2,4]triazolo[4,3-a]-quinazolin-5(1H)-one derivatives have been reported to potently and selectively inhibit Plk1-PBD.¹⁴ Interestingly, it was recently reported that abbapolins, nonpeptidic Plk1 binding inhibitors, induce Plk1 degradation in prostate cancer cells.¹⁵ However, extensive high-throughput screening has failed to identify PBD-targeting compounds that have high potency and specificity.¹⁶

Another representative phosphopeptide, PoloBoxtide-optimal (PMQSpTPL, Figure 1), was identified using peptide library screening⁹ to possess a high affinity for Plk1-PBD with a dissociation constant (K_D) of 280 nM. This peptide disrupts the ability of recombinant PBD protein to localize to the centrosomes of permeabilized U2OS cells. To investigate localization, metabolism, elimination, and biological effects, synthetic [¹⁸F]FB-MQSpTPL was used to obtain positron emission tomography (PET) images in rats.¹⁷ However, due to the presence of a negatively charged phosphate moiety, [¹⁸F]FB-MQSpTPL displays poor cellular uptake. Thus, installation of a cell-penetrating peptide on the phosphopeptide was carried out to improve cell permeability.¹⁸

Figure 1.

(A) Macrocyclization strategy in the design of 1 based on the linear analogue PMQSpTPL. (B) The distance between amide nitrogen of the Asn and C-4 carbon of the Pro in the X-ray cocrystal structure (PDB code 1UMW) is 4.3 Å. Protein residues of Plk1-PBD are represented by green cartoon and white stick and PMQSpTPL is depicted in violet.

Macrocyclization is one of the most effective strategies to rigidify conformations of large molecules and to improve their potencies, physical properties, cellular permeabilities, and stabilities.¹⁹ For example, macrocyclization of pepstatin A results in a substance that has enhanced inhibitory activity against CatD²⁰ and ATSP-7041. Also, the MDM-2 inhibitor, designed using an olefin metathesis-based staple strategy, has improved activity, cell permeability, stability, and PK properties.²¹ The macrocyclization strategy has also been adopted in studies aimed at developing unique Plk1-PBD inhibitors. Both C-terminal macrocyclization²² of PLH*SpT bearing a N(π),N-(τ)-bisalkylated His moiety and a thioether bridged macrocyclic peptomer²³ have been reported to be Plk1-PBD inhibitors. In this investigation, we utilized the macrocyclization strategy to increase the potency of PMQSpTPL against Plk1-PBD.¹⁹ The X-ray cocrystal structure of Plk1-PBD complexed with PMQSpTPL⁹ was employed to design novel macrocyclic phosphopeptide ligands. Guided by the bioactive conformation of PMQSpTPL (Figure 1A), we proposed to introduce the macrocyclic ring by installing an ethanolamine linker between the glutamine and proline residues of PMQSpTPL.

Based on the analysis (Figure 1B) that the distance between amide nitrogen atom of the Asn side chain and C-4 carbon of the Pro in the X-ray cocrystal structure (PDB code 1UMW) of Plk1-PBD and PMQSpTPL is 4.3 Å, we anticipated that the ethanolamine linker (4.4 Å) would be suitable to form a desired macrocycle. Thus, the target of our initial effort became the macrocyclic phosphopeptide 1. It is worth noting that macrocyclization by linking glutamine and proline residues of Plk1-PBD targeting phosphopeptides has not been reported.

In the effort described below, we prepared this substance and demonstrated that it has increased potency and isoform specificity relative to those of the linear analogue PMQSpTPL. Studies aimed at optimization of the biological properties of 1 led to identification novel cyclic phosphopeptides such as 16e that possess excellent potencies and high isotype specificities against Plk1-PBD, along with in vitro and in vivo antitumor activities.

RESULTS AND DISCUSSION

Synthesis of Macrocyclic Phosphopeptides.

A retrosynthetic analysis suggested that the preparation of the target phosphopeptide 1 and its analogues 16(a–h) could be achieved using a convergent pathway involving condensation of the tripeptide fragment 3 with a modified proline fragment 4 to generate the key intermediate 2 (Scheme 1). The synthetic route based on this analysis commenced with the transformation of l-threonine to tripeptide 3 bearing four orthogonal protection groups through three steps (Scheme 2). The 4-(2-aminoethoxy)proline derivative 4 was prepared from 4-hydroxyproline via a five-step pathway (Scheme 2).

Scheme 1.

Retrosynthesis of Macrocyclic Phosphopeptide 1

Scheme 2.

Synthesis of Fragments 3 and 4^a

^aReagent and condition: (a) Allyl bromide, Na₂CO₃, DMSO, rt, 12 h, 96%; (b) (i) 20% TFA, CH₂Cl₂, rt, 3 h; (ii) O-benzyl-N-(tert-butoxycarbonyl)-l-serine, HATU, DIPEA, DMF, rt, 1 h, 85%; (c) (i) 20% TFA, CH₂Cl₂, rt, 3 h; (ii) (S)-2-((((9H-fluoren-9-yl)methoxy)-carbonyl)amino)-5-(tert-butoxy)-5-oxopentanoic acid, HATU, DIPEA, DMF, rt, 1 h, 81%; (d) allyl bromide, silver oxide, DMF, rt, 3 h, 80%; (e) (i) LiOH, THF, rt, 3 h; (ii) l-leucine methyl ester hydrochloride, HATU, DIPEA, DMF, rt, 1 h, 83%; (f) (i) N-methylmorpholine N-oxide, OsO₄, THF/H₂O, rt, 12 h; (ii) NaIO₄, THF/H₂O, rt, 2 h, 83%; (g) (i) NaBH₄, MeOH, rt, 1 h; (ii) methanesulfonyl chloride, DIPEA, CH₂Cl₂, rt, 3 h; (iii) NaN₃, DMF, 50 °C, 3 h, 71%; (h) triphenylphosphine, THF/H₂O, rt, 12 h, 81%.

The tripeptide fragment 3 was prepared from Boc protected l-threonine (5) through allyl protection, Boc deprotection, and amide bond formation (Scheme 2). The carboxylic acid of 5 was protected by allyl group in 96% yield. Removal of Boc group in 6 followed by the coupling with O-benzyl-N-(tert-butoxycarbonyl)-l-serine using HATU and DIPEA led to the formation of amide 7 in an overall yield of 85%. The tripeptide fragment 3 was synthesized in 81% yield through Boc deprotection and amide bond formation.

The synthesis of 4-(2-aminoethoxy)proline fragment 4 was accomplished through (i) allylation; (ii) Lemieux–Johnson oxidation; (iii) reduction, mesylation, and azidation; and (iv) Staudinger reduction. The O-allylation on 8 was carried out using silver oxide widely utilized for the O-alkylation of carbohydrates.²⁴ Hydrolysis of 9 followed by the amide bond formation was carried out to give 10 in yields of 83%, which was submitted to Lemieux–Johnson oxidation reaction to afford the aldehyde 11.²⁵ The aldehyde group of 11 was transformed into the azido group of 12 in an overall yield of 71%, which was reduced to the corresponding amine by Staudinger reaction²⁶ to yield 4 in 81% yield.

Amide bond formation between the carboxylic acid generated by acidic hydrolysis of 3 and the 4-(2-aminoethoxy)-proline derivative 4 was accomplished using HATU and DIPEA (Scheme 3). This process formed the macrocyclization substrate 13. Consecutive deprotection of the allyl²⁷ and Boc protecting groups in 13, followed by HATU-mediated macrocyclization resulted in the formation of 2 in an overall yield of 62% from 13. Treatment of 2 with dibenzyl N,N-diisopropylphoshoramidite in the presence of 1H-tetrazole followed by oxidation with 3-chloroperbenzoic acid^28,29 formed the corresponding dibenzylphosphate 14 in 68% yield. The Fmoc group of 14 was removed using piperidine to give the corresponding amine 15, which was coupled with various carboxylic acids using HATU and DIPEA, followed by hydrogenolysis, yielding the desired macrocyclic phosphopeptides 1 and related derivatives 16(a–h).²⁹

Scheme 3.

Synthesis of Macrocyclic Phosphopeptides 1 and 16(a–h)^a

^aReagent and condition: (a) (i) 20% TFA, CH₂Cl₂, rt, 3 h; (ii) 4, HATU, DIPEA, DMF, rt, 1 h, 53%; (b) (i) Pd(PPh₃)₄, N-methylaniline, THF, rt, 2 h; (ii) 20% TFA, CH₂Cl₂, rt, 3 h; (iii) HATU, DIPEA, DMF, rt, 1 h, 62%; (c) (i) dibenzyl N,N-diisopropylphosphoramidite, 1H-tetrazole, THF, 0 °C – rt, 2 h; (ii) 3-chloroperbenzoic acid, THF, 0 °C, 1 h, 68%; (d) piperidine, DMF, 0 °C, 3 h, 90%; (e) (i) R₂-COOH, HATU, DIPEA, DMF, rt, 1 h; (ii) Pd(OH)₂, H₂, EtOH/CH₂Cl₂/H₂O, rt, 3 h, 54–68%.

In order to synthesize the linear 3-(trifluoromethyl)benzoyl derivative 21 and PMQSpTPL, the coupling reaction was conducted between N-Boc-Gln-OH and the linear tripeptide 17 prepared from methyl l-leucinate hydrochloride through three steps (Supporting Information) using HATU and DIPEA in 90% yield (Scheme 4). A phosphate group was introduced on the secondary alcohol moiety of 18 in 72% yield, and then the Boc group was deprotected to form the secondary amine of 20 in 79% yield. Amide bond formation between 20 and 3-(trifluoromethyl)benzoic acid using HATU and DIPEA, followed by hydrogenolysis using palladium hydroxide was performed to generate phosphopeptide 21 in a 75% yield over two steps. PMQSpTPL was synthesized starting from 17 in a yield of 36% using a similar pathway.

Scheme 4.

Synthesis of Linear Phosphopeptides 21 and PMQSpTPL^a

^aReagent and condition: (a) (i) 20% TFA, CH₂Cl₂, rt, 3 h; (ii) N-Boc-Gln-OH, HATU, DIPEA, DMF, rt, 1 h, 90%; (b) (i) dibenzyl N,N-diisopropylphosphoramidite, 1H-tetrazole, THF, 0 °C – rt, 2 h; (ii) 3-chloroperbenzoic acid, THF, 0 °C, 1 h, 72%; (c) 20% TFA, CH₂Cl₂, rt, 3 h, 79%; (d) (i) 3-(trifluoromethyl)benzoic acid or 27, HATU, DIPEA, DMF, rt, 1 h; (ii) Pd(OH)₂, H₂, EtOH/CH₂Cl₂/H₂O, rt, 3 h, 71–75%.

SAR Study for the Macrocyclic Phosphopeptides.

The Plk1-PBD binding activities of the macrocyclic phosphopeptides were evaluated using enzyme-linked immunosorbent assay (ELISA). These measurements gave IC₅₀ values for each phosphopeptide along with those for PLHSpT and non-phospho-PLHST⁸ serving as a respective positive and a negative control (Figure 2). The results show that the binding activity of PMQSpTPL (IC₅₀ = 6.27 μM) is 5.9-fold higher than that of PLHSpT (IC₅₀ = 36.79 μM). Also, the binding activity of the phosphomacrocylic derivative 1 (IC₅₀ = 0.68 μM) is 9.2-fold higher than that of its non-macrocyclic PMQSpTPL (IC₅₀ = 6.27 μM). This finding confirms the viability of the macrocyclization strategy in designing improved Plk1-PBD inhibitors. To develop a novel peptidomimetic derivative, based on the previous reports^11,12 that substitution of the proline and leucine residues of the phosphopeptide with aromatic moieties contributes to potency and plasma stability, we substituted the Pro-Met dipeptide moiety in 1 with the hydrophobic 3-(trifluoromethyl)benzoyl group (16a). Even though its activity (IC₅₀ = 2.22 μM) is lower than that of 1, 16a possesses a higher potency than the corresponding acyclic derivative 21 (IC₅₀ = 17.63 μM), again supporting the viability of the macrocyclization strategy.

Figure 2.

Specific binding to Plk1-PBD by various macrocyclic phosphopeptides. (A) Structures of the macrocyclic peptides. (B) ELISA-based Plk1-PBD binding assays were conducted using various concentrations of the inhibitors. Graphs represent mean values with s.d. (bars), determined from three independent experiments. O.D., optical density measured at 450 nm.

To assess the effects of changing hydrophilic and lipophilic characteristics, we explored the Plk1-PBD binding activities of substances having a variety of substituents on the benzoyl group in 16a (Figure 3). The results show that the binding activity of 16b (IC₅₀ = 1.92 μM) bearing 3-(2-(dimethylamino)ethoxy)benzoyl group on Plk1-PBD is comparable to that of 16a. Notably, the 3-isobutoxybenzoyl groupsubstituted macrocyclic phosphopeptide derivative 16c exhibits a 3.5-fold increased potency (IC₅₀ = 0.64 μM) relative to 16a, which suggests that a lipophilic moiety at this location contributes to Plk1-PBD binding. The effect of di-substitution was also probed with 16d, which contains both isobutoxy and 2-(dimethylamino)ethoxy groups. The binding activity of 16d (IC₅₀ = 0.50 μM) is comparable to that of 16c. Interestingly, the introduction of 3,5-diisobutoxy substituents in 16e results in a 11-fold higher binding potency (IC₅₀ = 0.20 μM) relative to 16a.

Figure 3.

Specific binding to Plk1-PBD by macrocyclic phosphopeptides 16(b–h). (A) Structures of the macrocyclic peptides. (B) ELISA-based Plk1-PBD binding assays were conducted using various concentrations of the inhibitors. Graphs represent the mean values with s.d. (bars) determined from three independent experiments. O.D., optical density measured at 450 nm.

The results of a molecular docking study (Figure 6A) reveal that the 3,5-diisobutoxy substituents in 16e interact more effectively with the hydrophobic surface formed by Phe535 than does the 3-trifluoromethyl substituent in 16a (see below). Considering the fact that negatively (around Asp416) and positively (around Arg516) charged regions are present on the electrostatic surface of the binding site on PBD (Figure 6B,D), it was hypothesized that substitution with a moiety containing both amine and carboxylic acid functionality would facilitate PBD binding. In contrast to this expectation, we observed that the binding affinities of 16f (IC₅₀ = 1.65 μM) and 16g (IC₅₀ = 1.23 μM), substances containing this feature, are lower than that of 16e. A potential disadvantage of the novel PBD inhibitors is poor cell permeability caused by the presence of a phosphate group. To increase cell permeability, PEGylated side chains were introduced into the benzoyl moiety (16h). We observed that 16h possesses a high binding activity (IC₅₀ = 0.83 μM) and, as a result, it was selected for later cell-based experiments (see below).

Figure 6.

Docking model of macrocyclic phosphopeptide (A,B) 16e, (C) 16h, and (D) 21 in complex with Plk1-PBD (PDB code 7MSO). The electrostatic surface of the Plk1-PBD substrate-bound pocket is represented. The electrostatic surface color of PBD is indicated with red for electronegative charge potential and blue for electropositive charge potential.

Selectivity for Binding to Plk1-PBD versus Other Plk Family Isoforms.

In mammalian cells, the serine–threonine kinase Plk1 family is composed of five Plk isoforms (Plk1–Plk5). Plk1 to Plk3 consist of an N-terminal kinase domain and the C-terminal PBD with a high level of homology. In contrast, the structures of Plk4 and Plk5 are significantly diverged from Plk1.^30,31 Plk1 is associated with cancer progression, whereas Plk2 and Plk3 are implicated in the depression of cancer cell growth.³¹ Thus, specific Plk1 binding agents are considered to be promising cancer therapeutics.

In vitro fluorescence polarization (FP) assays were performed to determine the relative binding affinities of the macrocyclic phosphopeptides toward Plk1-PBD.²³ In this assay, a change in FP of a fluorophore-labeled PBD-binding peptide derived from the Plk-PBDs occurs in response to competitive binding of a macrocyclic phosphopeptide.³² The results of these measurements show that macrocyclic phosphopeptides 16a, 16e, and 16h selectively bind to Plk1-PBD, and not to Plk2 and Plk3-PBDs (Figure 4).

Figure 4.

Fluorescence polarization assays with Plk1, Plk2, and Plk3-PBDs were carried out in the presence of the indicated substances as described in the methods. Data represent mean values with s.d. (bars) from three independent experiments.

Crystal Structure of the PBD in Complex with Macrocyclic Phosphopeptide 16a.

To gain information about the binding mode of 16a, the X-ray costructure of Plk1-PBD in complex with 16a was elucidated (Figure 5 and Supporting Information, Table S1). Analysis of the structure along with that of the complex with an optimal linear phosphopeptide, PMQSpTPL shows that the binding mode of 16a to the PBD is remarkably similar to that of PMQSpTPL.^4,8,9 Specifically, interactions between the phosphate group and key amino acid residues in the phosphopeptide binding groove of the PBD, including contacts mediated by a network of ordered water molecules, are similar in the linear and macrocyclic phosphopeptides. The ethoxy link, bridging the −2 Asn and +1 Pro residues of 16a, does not make any contacts with the Plk1-PBD, but (as intended) it does preserve the overall conformation of the ligand peptide backbone and maintain the H-bond interaction with Tyr485. The key difference between 16a and the PMQSpTPL is the presence of the 3-(trifluoromethyl)benzoyl group in the former, which participates in a π-stacking interaction with Arg516 of the PBD, resulting in a conformational change the Arg516 side chain. Additionally, the benzoyl and trifluoromethyl groups in 16a make van der Waals contacts with the side chains of Asp416 and Phe535, respectively. These additional contacts between the 3-(trifluoromethyl)benzoyl group of 16a with Arg516 of the PBD are not present in the PMQSpTPL complex, and likely contribute significantly to the enhanced binding affinity of 16a relative to the PMQSpTPL.

Figure 5.

X-ray co-crystal structure of Plk1-PBD in complex with 16a (PDB code 7MSO). Protein residues are represented by blue cartoon and cyan sticks, and 16a is depicted in yellow. The PMQSpTPL and selected key Plk1-PBD side chains from structure 1UMW (chain A)⁹ are superimposed and shown in thin sticks colored in orange.

Docking Model of Macrocyclic Phosphopeptide 16e, 16h, and 21 Binding to Plk1-PBD.

Based on our crystal structure of Plk1-PBD in complex with 16a, we conducted studies of molecular docking of 16e, 16h, and 21 on Plk1-PBD (Figure 6). Analysis of the results reveal that the phosphate group in 16e, 16h, and 21 is engaged in stabilizing contacts with Plk1-PBD, including water-mediated H-bonds to His538/Asn533 residues and direct H-bonds interaction with the side chain of Lys540. The serine residue in 16e and 16h interacts with the backbone Trp414 via H-bond, and the hydroxyl group of Tyr485 forms a H-bond with the glutamine backbone carbonyl group. The benzoyl ring of 16e and 16h participates in π-stacking interactions with Arg516 of the protein. Additionally, the 3,5-diisobutoxybenzoyl group in 16e seems to bind more favorably to the hydrophobic surface created by Phe535 than was observed for the 3-(trifluoromethyl)benzoyl group in 16a, providing a rationale for the enhanced activity of 16e. It is notable that both the π-stacking and hydrophobic interactions make 16e a more potent binding agent even though it does not engage in interactions with the tyrosine-rich channel of the protein. Finally, the leucine residue in the C-terminus of 16e and 16h appears to function as a clasp on the hydrophobic channel composed of Leu490 and Leu491. The results of molecular docking studies reveal that 21 are also capable of π-stacking interaction with the Arg516 residue and forming H-bonds with Tyr485 residues, Asp416 backbone, Trp414 backbone, and Leu491 backbone. However, 21 is inferior to 16a in terms of the distance at which 21 interacts with Arg516 residue, Tyr485 residues, and Asp416 backbone, which might be the reason why 21 is less potent than 16a.

Phenotypes and Phospho-H3 Staining of Zebrafish Embryos Treated with 16a and 16e.

We next sought to evaluate the biological activity of 16a in zebrafish embryos (Figure 7). Zebrafish embryos serve as a powerful system to perform chemical genetic screening in the context of a living organism³³ and have been used to evaluate agents that target the cell cycle.³⁴ In addition to the advantages of being optically transparent and small, foreign substances can also be introduced into zebrafish embryos using microinjection. Because of its poor cell permeability, 16a was microinjected into zebrafish embryos 24 h post fertilization. Cell cycle progression was then monitored using whole-mount antibody immunostaining for serine-10 phosphorylated histone H3 (phospho-H3). Histone H3 is phosphorylated during the late G₂ and M phases of the cell cycle³⁵ and this process is an indicative mitotic marker in zebrafish embryos.³⁶ The results of this assay show that the number of phospho-H3-positive cells in zebrafish embryos increases in proportion to the amount of injected 16a. Phospho-H3 level per unit area was increased in a dose-dependent manner (Figure 7B). This trend was also observed in western blot analysis (Figure S1). Thus, disruption of Plk1 function by 16a appears to lead to mitotic arrest. Moreover, inhibition of Plk1 induces mitotic infidelity and embryonic growth defects in development of zebrafish embryos³⁷ (Figure S2). The extension of embryo yolk increases as embryos develop³⁸ and is observed between 16 hpf (hours post fertilization) and 24 hpf.³⁹ It was observed yolk extension defect upon treatment of 16a (Figure 7C). As a result, the yolk extension area was decreased by 19% (12.2 ng of 16a) and 31% (24.5 ng of 16a) as compared to control.

Figure 7.

Representative images of phenotypes and phospho-H3 staining of the embryos. (A) Embryos injected with 16a (12.2 or 24.5 ng) at 24 hpf (hours post fertilization) and (D) embryos injected with 16a (17.1 ng) and 16e (18.6 ng) at 24 hpf and then stained for phospho-H3 as described in the Materials and Methods section. Photomicrographs are representative of triplicate experiments. (B,E) Quantification of phospho-H3 per unit area at 24 hpf (n = 8). The numbers of phospho-H3 were detected using ImageJ. (C,F) Quantification of yolk extension area at 24 hpf (n = 5). The yolk extension area was evaluated using ImageJ. Statistical significance was evaluated using a one-way ANOVA analysis (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

In addition, in vivo efficacy of 16e using zebrafish embryos was also evaluated compared with that of 16a (Figure 7D–F). The results show that quantified phospho-H3 level per unit area, upon treatment of 16a and 16e, was increased by 1.48-fold and 1.60-fold, respectively, compared with control. Also, reduction in yolk extension area was observed upon the treatment of 16a and 16e by 37 and 57%, respectively, relative to control. Taken together, 16a and 16e led to the cell mitotic arrest and zebrafish embryonic growth retardation.

Assessment of Anti-Plk1-PBD Activity in a Cell-Based Assay.

To determine whether the phosphomacrocyclic derivatives described above induces anti Plk1-PBD activity in cultured cells, asynchronously growing HeLa cells were treated with the indicated compounds for 2 days and viable cells were quantified using the MTS assay. For comparison, parental PLHSpT and its respective nonphospho form (PLHST) were also included. The results showed that unlike both PLHSpT and the nonphospho-PLHST control peptide, which failed to show any detectable level of cell cytotoxicity, the PEGylated derivative 16h markedly inhibited cell proliferation in a dose-dependent manner (Figure 8A,B). Under the same conditions, however, other phosphomacrocyclic derivatives (16a and 16e) showed only a low degree of anti-proliferative activity. It is envisioned that either membrane permeability or intracellular stability of these derivatives (16a and 16e) may not be optimal in cultured cells.

Figure 8.

Compound 16h induces Plk1 delocalization and mitotic failure in HeLa cells. (A) An MTS assay was performed using asynchronously growing HeLa cells treated with the indicated compounds for 2 days. Samples obtained from three independent experiments were quantified. Bars, mean relative MTS signals with s.d. (B) The data for PLHSpT and 16h in (A) were replotted to clearly display the level of concentration-dependent cell proliferation inhibition. (C) HeLa cells released from double thymidine block (G1/S) were treated with 16h 7 h after the release. Five hours after treatment, cells were stained with Hoechst 33342, fixed, and analyzed. Quantification was performed with cells obtained from three independent experiments (>930 cells/sample/each experiment). Examples of cells showing abnormal chromosome morphologies are shown in Figure S3. ***p < 0.001 (unpaired t test), **p < 0.01 (unpaired t test). Bars, mean % cells with s.d. (D, E) Asynchronously growing HeLa cells were treated with control DMSO or 200 μM of 16h for 4 h and immunostained with the indicated antibodies and DAPI to decorate chromosomal DNA. Confocal images in (D) show representative mitotic cells for each treatment. A centrosomal protein, Cep192,⁴² serves as a marker for centrosomes. Asterisks, centrosomes; arrowed bracket, kinetochores. Note that, as observed previously,⁴⁴ ectopic Plk1 dot signals (arrows) are apparent. The fluorescence signal intensities of endogenous Plk1 localized at centrosomes (asterisks) or kinetochores (arrowed bracket) were quantified from >97 cells for control DMSO and >153 cells for 16h obtained from three independent experiments (E) ****p < 0.0001 (unpaired t test). Bars, mean signal intensities with s.d. (F) Confocal images of cells prepared as in (D) and immunostained with anti-α-tubulin antibody and DAPI. Among mitotic cells acquired from three independent experiments, the fraction of cells exhibiting misaligned and multipolar mitotic chromosomes (>520 cells) (right, top) or lagging chromosomes (>218 cells) (right, bottom) were quantified. Asterisks, centrosomes; red arrowhead, misaligned chromosome; yellow arrowhead, lagging chromosome. **p < 0.01 (unpaired t test). Bars, mean % cells with s.d.

In a second experiment, we also used HeLa cells synchronously released from a G1/S blockade (a double thymidine blockade), an experimental set-up that permits an accurate assessment of inhibitor’s anti-mitotic capacity by treating the cells prior to mitotic entry. Under these conditions, after 7 h release from the second thymidine blockade, treatment of cells with 16h for 5 h significantly induced aberrant mitotic cells with improperly congressed chromosomes at a concentration as low as 100 μM (Figure 8C and Figure S3). The fraction of cells with aberrant mitotic chromosomes was increased greater than threefold at 100 μM of 16h. Aberrant mitotic progression is expected when PBD-dependent Plk1 function becomes disrupted.^5,8

To directly examine whether 16h can interfere with PBD-dependent Plk1 localization to centrosomes and mitotic kinetochores, asynchronously growing HeLa cells were treated with 200 μM of the PEGylated derivative 16h for 4 h and then subjected to immunostaining analysis. It was confirmed that 16a, 16e, and 16h are noncytotoxic in HeLa cells, even at 300 μM. PEGylation⁴⁰ is regarded as a strategy to enhance the pharmaceutical properties of inhibitors and has been successfully used to improve cellular activity of Plk1 inhibitors bearing the phosphate group.^10,41 The results of immunostaining showed that, while fluorescence microscope images of control DMSO-treated cells exhibited distinctively localized Plk1 fluorescence signals at mitotic centrosomes (asterisks in Figure 8D) and kinetochores (punctate signals indicated by the arrowed bracket), those treated with 16h displayed significantly diminished Plk1 signals at both of these sites (Figure 8D,E). Notably, the induction of cells with multipolar centrosomes⁴² was apparent. This is likely due to cytokinetic failure, as was previously reported to take place upon inhibiting the PBD of Plk1.⁴³ Interestingly, images of cells treated with 16h also contained additional Plk1 fluorescent dots not associated with centrosomes (arrows in Figure 8D). A fraction of Plk1 molecules delocalized from their native location could have formed ectopic aggregates after being delocalized from their native subcellular localization sites by PBD inhibition. Similar ectopic Plk1 aggregates were observed after being delocalized from kinetochores.⁴⁴ As a consequence of their failure to undergo proper PBD-dependent localization, a significant number of 16h-treated cells showed aberrant chromosomal morphologies, including misaligned and multipolar chromosomes (Figure 8F, left) and lagging chromosomes (Figure 8F, right).

CONCLUSIONS

In an effort to increase the potency of PMQSpTPL, a known acyclic phosphopeptide targeting Plk1-PBD, we utilized an analysis of the crystal structure of Plk1-PBD complexed with PMQSpTPL to design a novel macrocyclic phosphopeptide 1, which contains an ethanolamine linker between the Glu and Pro residues of the PMQSpTPL. We observed that 1, synthesized through a 26-step sequence, binds to Plk1-PBD with an activity (IC₅₀ = 0.68 μM) that is 9.2-fold greater than that of the corresponding acyclic phosphopeptide, PMQSpTPL (IC₅₀ = 6.27 μM). The finding shows that the macrocyclization strategy is effective in designing potent Plk1-PBD inhibitors. Based on previous reports,^11,12 we replaced the Pro-Met dipeptide moiety of 1 with 3-(trifluoromethyl)-benzoyl group to generate 16a. We believed that the changes made in generating this substance, associated with reducing the natural peptide moiety and increasing the small molecule character, would be beneficial in terms of removing disadvantages of natural peptides including low metabolic stability and intrinsic flexibility. Even though the binding activity of 16a (IC₅₀ = 2.22 μM) is slightly lower than that of 1, analysis of the crystal structure of its complex with Plk1-PBD reveals that the 3-(trifluoromethyl)benzoyl group induces a conformational change in the side chain of Arg516 through a π-stacking interaction. It is worth noting that this π-stacking interaction has not been reported in previous studies, and, as a result, it provides novel insights for designing potent Plk1-PBD inhibitors. In the course of optimization studies, the 3-(trifluoromethyl)benzoyl group in 16a was replaced by a 3,5-diisobutoxybenzoyl group to generate 16e. We found that 16e (IC₅₀ = 0.20 μM) has a 11-fold more potent binding activity than 16a and possesses excellent selectivity for Plk1-PBD over Plk2- and Plk3-PBD. Analysis of the docking study of 16e on Plk1-PBD reveals that the 3,5-diisobutoxybenzoyl group may interact more favorably with Phe535 of the PBD than was observed for the 3-(trifluoromethyl)benzoyl group in 16a. In addition, we observed that Plk1-PBD selective 16h, a PEGlyated derivative of 16e, induces Plk1 delocalization and mitotic failure in HeLa cells. Furthermore, we also observed that phospho-H3-positive cells in zebrafish embryo are increased with the microinjection of 16a or 16e. Also, reduction in yolk extension area of zebrafish embryo was observed upon the treatment of 16a and 16e by 37 and 57%, respectively, relative to control, which suggests that both 16a and 16e are capable of inhibiting Plk1-PBD in in vivo. Collectively, our novel macrocyclic phospho-peptidomimetics will serve as valuable templates in designing potent and novel Plk1-PBD inhibitors.

The results of the investigation described above provide an experimental verification for a novel strategy for allosteric inhibition of Plk1, which utilizes rationally designed macrocyclic phosphopeptides as PBD binding agents. Future efforts in this area will focus on the synthesis of pro-drugs that mask the negatively charged phosphate group in the macrocyclic phosphopeptides and exploration of noncharged phosphatemimetics. Ultimately these efforts may lead to a new class of anti-mitotic agents that could be used as drugs to treat cancer.

EXPERIMENTAL SECTION

Chemistry.

General Information.

Unless otherwise described, all commercial reagents and solvents were purchased from commercial suppliers and used without further purification. All reactions were performed under a N₂ atmosphere in flame-dried glassware. Reactions were monitored using TLC with 0.25 mm E. Merck precoated silica gel plates (60 F254). Reaction progress was monitored using TLC analysis using a UV lamp, ninhydrin, or p-anisaldehyde stain for detection purposes. All solvents were purified using standard techniques. Purification of reaction products was carried out using silica gel column chromatography with Kieselgel 60 Art. 9385 (230–400 mesh). Purities of all compounds were ≥95%, and mass spectra and purities of all compounds was accessed using Waters LC/MS system (Waters QDA Detector, Waters 2998 Photodiode Array Detector, Waters SFO System Fluidics Organizer, Water 2545 Binary Gradient Module, Waters 2767 Sample Manager) using SunFire C₁₈ column (4.6 × 50 mm, 5 μm particle size): solvent gradient = 30% B at 0.00 min, 30% B at 1.00 min, 100% B at 7.00 min, 100% B at 8.00 min, 30% B at 8.01 min, 30% B at 10.00 min. Solvent A = 0.1% HCOOH in H₂O; Solvent B = 0.1% HCOOH in MeOH; flow rate = 0.8 mL/min. ¹H and ¹³C NMR spectra were obtained using Bruker 400 MHz FT-NMR (400 MHz for ¹H, and 100 MHz for ¹³C) spectrometer. Standard abbreviations are used for denoting the signal multiplicities.

(E)-Prop-1-en-1-yl(tert-butoxycarbonyl)-L-threoninate (6).

To a solution of 5 (2.20 g, 10.03 mmol) in DMSO (20 mL), sodium carbonate (1.60 g, 15.05 mmol) and allyl bromide (1.56 mL, 18.06 mmol) were added at room temperature. The mixture was stirred at room temperature for 12 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (20 to 30% EtOAc/hexane) to afford 6 (2.50 g, 96%) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 6.57 (d, J = 8.0 Hz, 1H), 5.94–5.85 (m, 1H), 5.34 (dd, J = 16.0, 4.0 Hz, 1H), 5.20 (dd, J = 8.0, 4.0 Hz, 1H), 4.74 (d, J = 8.0 Hz, 1H), 4.64–4.53 (m, 2H), 4.10–4.04 (m, 1H), 4.02–3.99 (m, 1H), 1.39 (s, 9H), 1.09 (d, J = 4.0 Hz, 3H); ¹³C NMR (100 MHz, DMSO-d₆) δ 170.69, 155.67, 132.41, 117.43, 78.44, 66.49, 64.68, 59.78, 28.09, 20.11; LRMS (ESI) m/z 260 [M + H]⁺.

(E)-Prop-1-en-1-yl O-benzyl-N-(tert-butoxycarbonyl)-L-seryl-L-threoninate (7).

To a solution of 6 (1.50 g, 5.78 mmol) in CH₂Cl₂ (15 mL), trifluoroacetic acid (4.43 mL, 57.85 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. To a solution of crude residue (0.89 g, 5.61 mmol) in DMF (10 mL), O-benzyl-N-(tert-butoxycarbonyl)-l-serine (1.99 g, 6.73 mmol), HATU (6.40 g, 16.83 mmol), and DIPEA (4.89 mL, 28.06 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (30 to 40% EtOAc/hexane) to afford 7 (2.15 g, 85% overall yield of 2 steps) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 7.85 (d, J = 8.0 Hz, 1H), 7.35–7.26 (m, 5H), 7.04 (d, J = 8.0 Hz, 1H), 5.92–5.83 (m, 1H), 5.32 (dd, J = 16.0, 4.0 Hz, 1H), 5.18 (dd, J = 12.0, 4.0 Hz, 1H), 5.04 (d, J = 4.0 Hz, 1H), 4.57 (d, J = 4.0 Hz, 2H), 4.49 (d, J = 4.0 Hz, 2H), 4.34 (dd, J = 8.0, 4.0 Hz, 2H), 4.17–4.14 (m, 1H), 3.67–3.56 (m, 2H), 1.39 (s, 9H), 1.06 (d, J = 8.0 Hz, 3H); ¹³C NMR (100 MHz, DMSO-d₆) δ 170.37, 170.06, 155.27, 138.22, 132.35, 128.15, 127.47, 127.38, 117.54, 78.34, 72.00, 69.87, 66.36, 64.85, 57.70, 54.31, 28.17, 20.07; LRMS (ESI) m/z 437 [M + H]⁺.

(E)-Prop-1-en-1-yl(5S,8S,11S)-8-((benzyloxy)methyl)-5-(3-(tert-butoxy)-3-oxopropyl)-1-(9H-fluoren-9-yl)-11-((R)-1-hydroxyethyl)-3,6,9-trioxo-2-oxa-4,7,10-triazadodecan-12-oate (3).

To a solution of 7 (3.40 g, 7.79 mmol) in CH₂Cl₂ (35 mL), trifluoroacetic acid (5.96 mL, 77.89 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. To a solution of crude residue (2.49 g, 7.40 mmol) in DMF (25 mL), (S)-2-((((9H-fluoren-9-yl)methoxy)carbonyl)amino)-5-(tert-butoxy)-5-oxopentanoic acid (3.78 g, 8.88 mmol), HATU (8.44 g, 22.20 mmol), and DIPEA (6.44 mL, 37.00 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (30 to 40% EtOAc/hexane) to afford 3 (4.69 g, 81% overall yield of 2 steps) as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.15–8.12 (m, 2H), 7.89 (d, J = 8.0 Hz, 2H), 7.73 (t, J = 8.0 Hz, 2H), 7.61 (d, J = 8.0 Hz, 1H), 7.41 (t, J = 8.0 Hz, 2H), 7.34–7.23 (m, 7H), 5.93–5.83 (m, 1H), 5.33 (dd, J = 16.0, 4.0 Hz, 1H), 5.18 (d, J = 8.0 Hz, 1H), 5.04 (d, J = 4.0 Hz, 1H), 4.76–4.72 (m, 1H), 4.58 (d, J = 4.0 Hz, 2H), 4.51 (s, 2H), 4.38 (dd, J = 8.0, 4.0 Hz, 1H), 4.32–4.22 (m, 3H), 4.19–4.11 (m, 2H), 3.69–3.62 (m, 2H), 2.27 (t, J = 8.0 Hz, 2H), 1.95–1.91 (m, 1H), 1.82–1.75 (m, 1H), 1.39 (s, 9H), 1.09–1.02 (m, 3H); ¹³C NMR (100 MHz, DMSO-d₆) δ 171.73, 171.41, 170.01, 169.93, 155.96, 143.90, 143.73, 140.72, 138.14, 132.35, 128.14, 127.64, 127.49, 127.35, 127.07, 125.32, 120.09, 117.57, 79.62, 72.12, 69.84, 66.33, 65.73, 64.84, 57.88, 53.81, 52.47, 46.67, 31.42, 27.74, 27.50, 20.09; LRMS (ESI) m/z 744 [M + H]⁺.

1-(tert-Butyl) 2-methyl (2S,4R)-4-hydroxypyrrolidine-1,2-dicarboxylate (9).

To a solution of 8 (2.10 g, 8.56 mmol) in DMF (20 mL), silver oxide (2.98 g, 12.84 mmol) and allyl bromide (1.33 mL, 15.41 mmol) were added at room temperature. The mixture was stirred at room temperature for 3 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (15 to 20% EtOAc/hexane) to afford 9 (1.95 g, 80%) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 5.92–5.82 (m, 1H), 5.24 (dd, J = 16.0, 4.0 Hz, 1H), 5.13 (dd, J = 12.0, 4.0 Hz, 1H), 4.21–4.15 (m, 1H), 4.11–4.08 (m, 1H), 3.96–3.95 (m, 2H), 3.66–3.63 (m, 3H), 3.48–3.39 (m, 2H), 2.35–2.25 (m, 1H), 2.01–1.92 (m, 1H), 1.39–1.32 (m, 9H); ¹³C NMR (100 MHz, DMSO-d₆) δ 173.01, 152.85, 135.07, 116.51, 79.14, 75.49, 68.99, 57.57, 51.86, 51.45, 35.69, 27.81; LRMS (ESI) m/z 286 [M + H]⁺.

tert-Butyl (2S,4R)-4-(allyloxy)-2-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)carbamoyl)pyrrolidine-1-carboxylate (10).

To a solution of 9 (1.50 g, 5.26 mmol) in THF (15 mL), lithium hydroxide monohydrate (2.21 g, 52.57 mmol) was added at room temperature. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. To a solution of crude residue (1.35 g, 4.99 mmol) in DMF (15 mL), l-leucine methyl ester hydrochloride (1.09 g, 5.99 mmol), HATU (5.70 g, 14.98 mmol), and DIPEA (4.35 mL, 24.97 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (20% EtOAc/hexane) to afford 10 (1.74 g, 83% overall yield of 2 steps) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 8.30–8.24 (m, 1H), 5.92–5.82 (m, 1H), 5.24 (dd, J = 16.0, 4.0 Hz, 1H), 5.13 (dd, J = 12.0, 4.0 Hz, 1H), 4.27–4.23 (m, 1H), 4.21–4.17 (m, 1H), 4.10–4.02 (m, 1H), 3.95–3.93 (m, 2H), 3.61–3.60 (m, 3H), 3.46–3.36 (m, 2H), 2.26–2.20 (m, 1H), 1.91–1.83 (m, 1H), 1.70–1.62 (m, 1H), 1.60–1.53 (m, 1H), 1.50–1.43 (m, 1H), 1.38–1.31 (m, 9H), 0.90–0.84 (m, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 172.88, 172.53, 153.39, 135.18, 116.40, 78.64, 75.56, 69.01, 58.10, 51.80, 50.27, 36.47, 30.66, 27.85, 24.10, 22.80, 21.11; LRMS (ESI) m/z 399 [M + H]⁺.

tert-Butyl (2S,4R)-2-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)carbamoyl)-4-(2-oxoethoxy)pyrrolidine-1-carboxylate (11).

To a solution of 10 (3.00 g, 7.59 mmol) in THF/H₂O (40 mL, 3/1), N-methylmorpholine N-oxide (2.67 g, 22.76 mmol) and 4.0 wt % osmium tetroxide solution in tert-butanol (4.79 mL, 0.76 mmol) were added at room temperature. The mixture was stirred at room temperature for 12 h, diluted with EtOAc, and quenched with saturated sodium sulfite solution. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. To a solution of crude diol (2.95 g, 6.83 mmol) in THF/H₂O (40 mL, 3/1), sodium periodate (4.83 g, 20.48 mmol) was added at room temperature. The mixture was stirred at room temperature for 2 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (3 to 5% MeOH/CH₂Cl₂) to afford 11 (2.32 g, 83% overall yield of 2 steps) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 9.55 (s, 1H), 8.31–8.25 (m, 1H), 4.27–4.18 (m, 2H), 4.14–4.08 (m, 1H), 3.61–3.60 (m, 3H), 3.47–3.40 (m, 2H), 2.33–2.18 (m, 1H), 1.91–1.82 (m, 1H), 1.70–1.62 (m, 1H), 1.60–1.53 (m, 1H), 1.50–1.44 (m, 1H), 1.38–1.32 (m, 9H), 0.90–0.84 (m, 6H); LRMS (ESI) m/z 401 [M + H]⁺.

tert-Butyl (2S,4R)-4-(2-azidoethoxy)-2-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)carbamoyl)pyrrolidine-1-carboxylate (12).

To a solution of 11 (4.40 g, 10.99 mmol) in MeOH (30 mL), sodium borohydride (0.62 g, 16.48 mmol) was added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. To a solution of crude alcohol (3.41 g, 8.46 mmol) in CH₂Cl₂ (40 mL), methanesulfonyl chloride (0.98 mL, 12.69 mmol) and DIPEA (4.42 mL, 25.38 mmol) were added at 0 °C. The mixture was stirred at room temperature for 3 h, diluted with CH₂Cl₂, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. To a solution of crude mesylate (3.70 g, 7.70 mmol) in DMF (40 mL), sodium azide (1.50 g, 23.10 mmol) was added at room temperature. The mixture was stirred at 50 °C for 3 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (30 to 40% EtOAc/hexane) to afford 12 (3.02 g, 71% overall yield of 3 steps) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 8.30–8.20 (m, 1H), 4.28–4.24 (m, 1H), 4.22–4.18 (m, 1H), 4.14–4.05 (m, 1H), 3.63–3.55 (m, 2H), 3.61–3.60 (m, 3H), 3.47–3.42 (m, 2H), 3.39–3.35 (m, 2H), 2.28–2.15 (m, 1H), 1.91–1.83 (m, 1H), 1.70–1.62 (m, 1H), 1.60–1.53 (m, 1H), 1.50–1.43 (m, 1H), 1.38–1.31 (m, 9H), 0.90–0.82 (m, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 172.88, 172.52, 153.41, 78.59, 76.36, 67.33, 58.07, 51.77, 50.25, 49.97, 36.53, 30.65, 27.83, 24.10, 22.80, 21.08; LRMS (ESI) m/z 428 [M + H]⁺.

tert-Butyl (2S,4R)-4-(2-aminoethoxy)-2-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)carbamoyl)pyrrolidine-1-carboxylate (4).

To a solution of 12 (3.00 g, 7.02 mmol) in THF/H₂O (40 mL, 20/1), triphenylphosphine (3.68 g, 14.04 mmol) was added at room temperature. The mixture was stirred at room temperature for 12 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (7 to 10% MeOH/CH₂Cl₂) to afford 4 (2.28 g, 81%) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 8.31–8.24 (m, 1H), 7.76 (s, 2H), 4.29–4.24 (m, 1H), 4.22– 4.18 (m, 1H), 4.14–4.05 (m, 1H), 3.62–3.61 (m, 3H), 3.58–3.48 (m, 4H), 2.99–2.96 (m, 2H), 2.31–2.15 (m, 1H), 1.93–1.86 (m, 1H), 1.70–1.61 (m, 1H), 1.58–1.54 (m, 1H), 1.52–1.45 (m, 1H), 1.38–1.31 (m, 9H), 0.90–0.82 (m, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 172.89, 172.52, 153.34, 78.70, 76.40, 64.67, 58.11, 51.80, 50.22, 38.62, 36.12, 30.42, 27.84, 24.11, 22.80, 21.11; LRMS (ESI) m/z 402 [M + H]⁺.

tert-Butyl (2S,4R)-4-(((7S,10S,13S,E)-7-((((9H-fluoren-9-yl)-methoxy)carbonyl)amino)-10-((benzyloxy)methyl)-13-((R)-1-hydroxyethyl)-4,8,11,14-tetraoxo-15-oxa-3,9,12-triazaoctadec-16-en-1-yl)oxy)-2-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)carbamoyl)pyrrolidine-1-carboxylate (13).

To a solution of 3 (2.0 g, 2.69 mmol) in CH₂Cl₂ (20 mL), trifluoroacetic acid (2.06 mL, 26.89 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. The residue was dissolved in CH₂Cl₂, solidified by adding Et₂O, and filtered to afford a solid. To a solution of the solid (1.79 g, 2.61 mmol) in DMF (20 mL), 4 (1.26 g, 3.13 mmol), HATU (2.98 g, 7.82 mmol), and DIPEA (2.27 mL, 13.04 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (3 to 4% MeOH/CH₂Cl₂) to afford 13 (1.53 g, 53% overall yield of 2 steps) as a white solid. ¹H NMR (400 MHz, CD₃OD) δ 7.80 (d, J = 8.0 Hz, 2H), 7.66 (t, J = 8.0 Hz, 2H), 7.39 (t, J = 8.0 Hz, 2H), 7.32–7.20 (m, 7H), 5.97–5.87 (m, 1H), 5.33 (dd, J = 16.0, 4.0 Hz, 1H), 5.21 (dd, J = 12.0, 4.0 Hz, 1H), 4.71–4.69 (m, 1H), 4.63 (d, J = 4.0 Hz, 2H), 4.54–4.51 (m, 3H), 4.46–4.39 (m, 1H), 4.36–4.29 (m, 4H), 4.22–4.17 (m, 2H), 4.10–4.07 (m, 1H), 3.86–3.82 (m, 1H), 3.77–3.73 (m, 1H), 3.67 (s, 3H), 3.60–3.43 (m, 4H), 3.34–3.32 (m, 2H), 2.41–2.36 (m, 1H), 2.32–2.28 (m, 2H), 2.11–1.94 (m, 3H), 1.75–1.67 (m, 1H), 1.61–1.58 (m, 2H), 1.43–1.39 (m, 9H), 1.18 (d, J = 4.0 Hz, 3H), 0.95–0.89 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.66, 175.11, 174.38, 174.16, 172.39, 171.43, 158.38, 156.01, 145.28, 145.15, 142.56, 139.08, 133.27, 129.35, 128.95, 128.79, 128.71, 128.19, 126.27, 120.94, 118.67, 81.61, 78.92, 78.17, 74.35, 70.72, 68.47, 68.11, 66.93, 60.12, 59.46, 55.71, 54.72, 52.64, 52.25, 41.31, 40.49, 37.94, 33.08, 29.24, 28.73, 28.61, 25.89, 23.28, 21.85, 20.45; LRMS (ESI) m/z 1071 [M + H]⁺.

Methyl ((3S,6S,9S,17R,19S)-9-((((9H-fluoren-9-yl)methoxy)-carbonyl)amino)-6-((benzyloxy)methyl)-3-((R)-1-hydroxyethyl)-2,5,8,12-tetraoxo-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (2).

To a solution of 13 (1.40 g, 1.31 mmol) in THF (20 mL), tetrakis(triphenylphosphine)palladium(0) (0.76 g, 0.65 mmol) and N-methylaniline (0.35 g, 3.27 mmol) were added at room temperature. The mixture was stirred at room temperature for 2 h and concentrated under reduced pressure. The residue was dissolved in CH₂Cl₂, solidified by adding Et₂O, and filtered to afford a solid. To a solution of the solid (1.21 g, 1.18 mmol) in CH₂Cl₂ (15 mL), trifluoroacetic acid (0.90 mL, 11.76 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. The residue was dissolved in CH₂Cl₂, solidified by adding Et₂O, and filtered to afford a solid. To a solution of the solid (1.03 g, 1.11 mmol) in DMF (10 mL), HATU (1.26 g, 3.32 mmol) and DIPEA (0.96 mL, 5.53 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (5% MeOH/CH₂Cl₂) to afford 2 (0.74 g, 62% overall yield of 3 steps) as a white solid. ¹H NMR (400 MHz, CD₃OD) δ 7.77 (d, J = 8.0 Hz, 2H), 7.63 (t, J = 8.0 Hz, 2H), 7.37 (t, J = 8.0 Hz, 2H), 7.32–7.21 (m, 7H), 4.71–4.68 (m, 1H), 4.57 (s, 2H), 4.54–4.50 (m, 1H), 4.44–4.39 (m, 1H), 4.36–4.29 (m, 4H), 4.19–4.16 (m, 2H), 4.12–4.10 (m, 1H), 3.99–3.92 (m, 1H), 3.86–3.82 (m, 1H), 3.78–3.70 (m, 2H), 3.68 (s, 3H), 3.64–3.60 (m, 2H), 3.50–3.42 (m, 1H), 3.14–3.08 (m, 1H), 2.59–2.52 (m, 1H), 2.43–2.37 (m, 1H), 2.31–2.23 (m, 1H), 2.11–1.97 (m, 2H), 1.95–1.83 (m, 1H), 1.79–1.72 (m, 1H), 1.62–1.57 (m, 2H), 1.37–1.29 (m, 3H), 0.96–0.90 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.56, 175.46, 174.52, 174.35, 174.00, 172.18, 171.67, 157.96, 145.28, 145.17, 142.54, 142.53, 139.12, 129.41, 129.07, 128.78, 128.74, 128.18, 126.23, 120.91, 79.66, 74.38, 70.57, 69.54, 68.60, 68.00, 60.22, 59.80, 55.50, 55.12, 54.34, 52.62, 52.28, 41.37, 40.62, 40.50, 38.87, 37.00, 32.76, 30.56, 25.79, 23.32, 21.90, 20.00; LRMS (ESI) m/z 913 [M + H]⁺.

Methyl ((3S,6S,9S,17R,19S)-9-((((9H-fluoren-9-yl)methoxy)-carbonyl)amino)-6-((benzyloxy)-methyl)-3-((R)-1-((bis(benzyloxy)phosphoryl)oxy)ethyl)-2,5,8,12-tetraoxo-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (14).

To a solution of 1H-tetrazole (0.42 g, 6.02 mmol) in MeCN (13.39 mL), dibenzyl N,N-diisopropylphosphoramidite (90% tech. Grade, 2.31 g, 6.02 mmol) was added at 0 °C. The mixture was stirred at room temperature for 1 h. To the mixture was added a solution of 2 (0.55 g, 0.60 mmol) in MeCN (5 mL) at 0 °C. The mixture was stirred at room temperature for 1 h. To the mixture was added 3-chloroperbenzoic acid (1.04 g, 6.02 mmol) at 0 °C. The mixture was stirred at 0 °C for 1 h, diluted with EtOAc, and quenched with saturated sodium bicarbonate. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (4 to 5% MeOH/CH₂Cl₂) to afford 14 (0.48 g, 68% overall yield of 2 steps) as a white solid. ¹H NMR (400 MHz, CD₃OD) δ 8.37–8.25 (m, 2H), 7.79 (d, J = 4.0 Hz, 2H), 7.64 (t, J = 8.0 Hz, 2H), 7.39–7.24 (m, 17H), 5.08–5.01 (m, 4H), 4.72–4.60 (m, 2H), 4.55 (s, 2H), 4.52–4.47 (m, 1H), 4.44–4.38 (m, 1H), 4.32–4.31 (m, 3H), 4.20–4.15 (m, 2H), 4.11–4.09 (m, 1H), 3.75–3.73 (m, 2H), 3.68 (s, 3H), 3.64–3.58 (m, 2H), 3.54–3.43 (m, 1H), 3.14–3.09 (m, 1H), 2.57–2.49 (m, 1H), 2.43–2.37 (m, 1H), 2.35–2.27 (m, 1H), 2.09–2.01 (m, 2H), 1.99–1.91 (m, 1H), 1.79–1.71 (m, 1H), 1.61–1.54 (m, 2H), 1.44 (d, J = 8.0 Hz, 2H), 1.32–1.21 (m, 3H), 0.93–0.88 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.43, 174.56, 174.05, 173.97, 173.89, 172.15, 169.64, 157.97, 145.27, 145.16, 142.53, 142.52, 139.51, 139.43, 139.03, 137.10, 137.03, 129.74, 129.70, 129.55, 129.46, 129.28, 129.26, 129.22, 129.17, 129.11, 128.77, 128.66, 128.62, 128.47, 128.17, 126.21, 120.91, 79.65, 76.69, 76.63, 74.40, 71.14, 71.09, 71.03, 70.25, 69.64, 68.78, 68.71, 68.66, 68.47, 68.42, 67.99, 60.27, 55.63, 55.07, 54.51, 52.60, 52.35, 52.26, 48.31, 47.41, 41.40, 41.34, 40.43, 36.90, 32.73, 30.41, 25.75, 23.32, 22.80, 21.91, 19.29, 18.49; LRMS (ESI) m/z 1173 [M + H]⁺.

Methyl ((3S,6S,9S,17R,19S)-9-amino-6-((benzyloxy)methyl)-3-((R)-1-((bis(benzyloxy)phosphoryl)oxy)ethyl)-2,5,8,12-tetraoxo-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (15).

To a solution of 14 (0.65 g, 0.55 mmol) in DMF (6 mL), piperidine (0.16 mL, 1.66 mmol) was added at 0 °C. The mixture was stirred at 0 °C for 3 h and concentrated under reduced pressure. The residue was dissolved in CH₂Cl₂, solidified by adding Et₂O, and filtered to afford 15 (0.47 g, 90%) as a white solid. ¹H NMR (400 MHz, CD₃OD) δ 7.44–7.22 (m, 15H), 5.09–4.98 (m, 4H), 4.67–4.40 (m, 8H), 4.38–4.12 (m, 2H), 4.00–3.79 (m, 2H), 3.79–3.66 (m, 2H), 3.68 (s, 3H), 3.65–3.60 (m, 1H), 3.60–3.46 (m, 1H), 3.23–3.03 (m, 1H), 2.64–2.47 (m, 1H), 2.43–2.30 (m, 1H), 2.23–2.12 (m, 1H), 2.10–1.91 (m, 3H), 1.82–1.72 (m, 1H), 1.63–1.55 (m, 2H), 1.47–1.39 (m, 3H), 1.34–1.22 (m, 3H), 0.97–0.87 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 174.78, 174.56, 174.26, 173.99, 171.62, 171.17, 170.51, 169.63, 139.58, 139.36, 139.03, 137.12, 137.06, 129.80, 129.74, 129.61, 129.50 129.37, 129.34, 129.26, 129.19, 129.13, 129.08, 129.01, 128.87, 128.67, 128.27, 79.95, 79.75, 76.72, 74.48, 74.30, 71.17, 71.11, 70.92, 70.23, 69.74, 68.52, 60.28, 58.48, 55.69, 54.70, 52.62, 52.29, 41.42, 41.34, 40.50, 37.08, 32.11, 30.76, 30.33, 25.79, 25.73, 23.35, 21.90, 18.71, 18.43; LRMS (ESI) m/z 951 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₇H₆₄N₆O₁₃P⁺ [M + H]⁺: 951.4263. Found: 951.4269.

Methyl O-benzyl-N-((tert-butoxycarbonyl)-L-glutaminyl)-L-seryl-L-threonyl-L-prolyl-L-leucinate (18).

To a solution of 17 prepared from methyl l-leucinate hydrochloride through three steps (Supporting Information) (0.84 g, 1.35 mmol) in CH₂Cl₂ (10 mL), trifluoroacetic acid (1.04 mL, 13.53 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. To a solution of crude residue (0.68 g, 1.31 mmol) in DMF (10 mL), (tert-butoxycarbonyl)-l-glutamine (0.39 g, 1.58 mmol), HATU (1.50 g, 3.94 mmol), and DIPEA (1.14 mL, 6.56 mmol) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (3 to 5% MeOH/CH₂Cl₂) to afford 18 (0.91 g, 90% overall yield of 2 steps) as a colorless oil. ¹H NMR (400 MHz, DMSO-d₆) δ 8.15 (d, J = 4.0 Hz, 1H), 8.09 (d, J = 8.0 Hz, 1H), 7.84 (d, J = 8.0 Hz, 1H), 7.35–7.23 (m, 6H), 7.00 (d, J = 8.0 Hz, 1H), 6.74 (s, 1H), 5.00–4.82 (m, 1H), 4.64–4.47 (m, 3H), 4.38–4.35 (m, 1H), 4.26–4.20 (m, 1H), 4.00–3.93 (m, 1H), 3.87–3.83 (m, 1H), 3.68–3.54 (m, 3H), 3.60 (s, 3H), 2.13–2.07 (m, 2H), 1.99–1.97 (m, 1H), 1.85–1.79 (m, 3H), 1.70–1.62 (m, 2H), 1.56–1.45 (m, 2H), 1.37 (s, 9H), 1.09–0.98 (m, 3H), 0.89–0.82 (m, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 173.91, 172.84, 171.83, 171.67, 169.27, 168.51, 155.32, 138.17, 128.19, 127.50, 127.39, 78.19, 72.13, 70.00, 67.02, 59.04, 56.16, 53.98, 52.35, 51.79, 50.33, 47.30, 30.98, 29.08, 28.20, 27.79, 24.17, 22.80, 21.29, 18.95; LRMS (ESI) m/z 749 [M + H]⁺.

Methyl N-(O-benzyl-N-((tert-butoxycarbonyl)-L-glutaminyl)-L-seryl)-O-(bis(benzyloxy)phosphoryl)-L-threonyl-L-prolyl-L-leucinate (19).

To a solution of 1H-tetrazole (1.03 g, 14.69 mmol) in MeCN (32.64 mL), dibenzyl N,N-diisopropylphosphoramidite (90% tech. Grade, 5.64 g, 14.69 mmol) was added at 0 °C. The mixture was stirred at room temperature for 1 h. To the mixture was added a solution of 18 (1.1 g, 11.47 mmol) in MeCN (10 mL) at 0 °C. The mixture was stirred at room temperature for 1 h. To the mixture was added 3-chloroperbenzoic acid (2.53 g, 14.69 mmol) at 0 °C. The mixture was stirred at 0 °C for 1 h, diluted with EtOAc, and quenched with saturated sodium bicarbonate. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (4 to 5% MeOH/CH₂Cl₂) to afford 19 (1.07 g, 72% overall yield of 2 steps) as a white solid. ¹H NMR (400 MHz, DMSO-d₆) δ 8.55 (d, J = 8.0 Hz, 1H), 8.22 (d, J = 8.0 Hz, 1H), 7.90 (d, J = 8.0 Hz, 1H), 7.38–7.22 (m, 16H), 7.01 (d, J = 8.0 Hz, 1H), 6.75 (s, 1H), 5.10–4.97 (m, 4H), 4.83 (t, J = 8.0 Hz, 1H), 4.65–4.60 (m, 2H), 4.47–4.40 (m, 2H), 4.34–4.31 (m, 1H), 4.26–4.20 (m, 1H), 4.00–3.93 (m, 1H), 3.67–3.54 (m, 3H), 3.60 (s, 3H), 2.13–2.06 (m, 2H), 1.99–1.94 (m, 1H), 1.85–1.78 (m, 2H), 1.77–1.64 (m, 4H), 1.58–1.43 (m, 2H), 1.36 (s, 9H), 1.31–1.19 (m, 3H), 0.84 (dd, J = 28.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 173.83, 172.92, 171.74, 171.36, 169.32, 166.73, 155.33, 138.09, 136.15, 128.46, 128.38, 128.22, 128.17, 127.79, 127.34, 78.13, 74.84, 72.19, 70.06, 67.02, 68.56, 59.15, 54.55, 53.97, 52.43, 51.75, 50.24, 47.30, 31.60, 29.15, 28.15, 27.67, 24.07, 22.75, 21.24, 17.66; LRMS (ESI) m/z 1009 [M + H]⁺.

Methyl N-(N-(L-glutaminyl)-O-benzyl-L-seryl)-O-(bis(benzyloxy)-phosphoryl)-L-threonyl-L-prolyl-L-leucinate (20).

To a solution of 19 (0.83 g, 0.82 mmol) in CH₂Cl₂ (10 mL), trifluoroacetic acid (0.63 mL, 8.23 mmol) was added at 0 °C. The mixture was stirred at room temperature for 3 h and concentrated under reduced pressure. The residue was dissolved in CH₂Cl₂, solidified by adding Et₂O, and filtered to afford 20 (0.59 g, 79%) as a white solid. ¹H NMR (400 MHz, CD₃OD) δ 7.37–7.25 (m, 15H), 5.06–5.00 (m, 4H), 4.90 (d, J = 8.0 Hz, 1H), 4.79–4.73 (m, 1H), 4.70 (t, J = 4.0 Hz, 1H), 4.56–4.47 (m, 2H), 4.42–4.35 (m, 2H), 4.00 (t, J = 8.0 Hz, 1H), 3.84–3.67 (m, 3H), 3.67 (s, 3H), 2.48–2.43 (m, 2H), 2.16–2.07 (m, 3H), 2.00–1.93 (m, 2H), 1.87–1.80 (m, 1H), 1.75–1.69 (m, 1H), 1.65–1.56 (m, 2H), 1.42–1.29 (m, 3H), 0.95–0.85 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 177.10, 174.59, 174.04, 171.46, 171.38, 169.96, 168.94, 138.96, 137.18, 129.71, 129.67, 129.44, 129.31, 129.25, 128.82, 76.75, 74.35, 71.02, 70.64, 61.49, 56.71, 55.09, 53.74, 52.59, 52.38, 41.29, 31.53, 30.56, 28.07, 25.82, 25.73, 23.27, 21.90, 18.52; LRMS (ESI) m/z 909 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₅H₆₂N₆O₁₂P⁺ [M + H]⁺: 909.4158. Found: 909.4163.

General procedure for the synthesis of 1, 16(a-h), 21 and PMQSpTPL.

To a solution of 15 or 20 (1.0 equiv) in DMF (0.1 M), ((benzyloxy)carbonyl)-l-prolyl-l-methionine or several acyl acids (1.2 equiv), HATU (3.0 equiv), and DIPEA (5.0 equiv) were added at room temperature. The mixture was stirred at room temperature for 1 h, diluted with EtOAc, and quenched with H₂O. The organic layer was washed with brine, dried over MgSO₄, filtered, and concentrated under reduced pressure. The resulting residue was subjected to flash column chromatography on silica gel (5 to 10% MeOH/CH₂Cl₂) to afford corresponding amide compounds as a white solid. To a solution of the resulting amide (1.0 equiv) in EtOH/CH₂Cl₂/H₂O (0.02 M, 2/1/1), palladium hydroxide (0.5 equiv) was added at room temperature. The mixture was stirred at room temperature for 3 h under hydrogen atmosphere, filtered through celite pad, and concentrated under reduced pressure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford pure product (54 to 75% overall yield of 2 steps) as a solid.

Methyl ((3S,6S,9S,17R,19S)-6-(hydroxymethyl)-9-((S)-4-(methyl-thio)-2-((S)-pyrrolidine-2-carboxamido)butanamido)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (1).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 1 (62 mg, 65%). ¹H NMR (400 MHz, CD₃OD) δ 4.59–4.49 (m, 2H), 4.44–4.35 (m, 4H), 4.17–4.13 (m, 1H), 4.05–3.92 (m, 1H), 3.85–3.83 (m, 1H), 3.69 (s, 3H), 3.76–3.63 (m, 1H), 3.60–3.49 (m, 2H), 3.18–3.14 (m, 1H), 2.71–2.61 (m, 1H), 2.59–2.49 (m, 1H), 2.43–2.37 (m, 1H), 2.31–2.15 (m, 3H), 2.12–1.97 (m, 4H), 1.90 (s, 3H), 1.85–1.74 (m, 2H), 1.68–1.55 (m, 3H), 1.47–1.40 (m, 2H), 1.37–1.22 (m, 2H), 1.30–1.28 (m, 3H), 0.98–0.87 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 179.91, 175.79, 175.51, 174.63, 174.23, 173.06, 171.94, 171.66, 79.63, 74.26, 72.28, 69.31, 63.14, 62.76, 60.54, 60.43, 56.79, 55.72, 55.53, 52.65, 52.28, 49.85, 41.32, 40.64, 37.10, 33.03, 32.59, 30.74, 30.60, 30.42, 30.29, 30.22, 25.78, 23.87, 23.69, 23.32, 21.88, 18.98, 14.42, 10.44; LRMS (ESI) m/z 909 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₆H₆₂N₈O₁₅PS⁺ [M + H]⁺: 909.3787. Found: 909.3793.

Methyl ((3S,6S,9S,17R,19S)-6-(hydroxymethyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-9-(3-(trifluoromethyl)benzamido)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (16a).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16a (56 mg, 62%). ¹H NMR (400 MHz, CD₃OD) δ 8.17 (s, 1H), 8.11 (d, J = 8.0 Hz, 1H), 7.85 (d, J = 8.0 Hz, 1H), 7.67 (dd, J = 8.0, 8.0 Hz, 1H), 4.59–4.58 (m, 1H), 4.55–4.51 (m, 1H), 4.45–4.38 (m, 3H), 4.16 (s, 1H), 3.96–3.89 (m, 2H), 3.74–3.62 (m, 3H), 3.69 (s, 3H), 3.56–3.52 (m, 1H), 3.24–3.14 (m, 1H), 2.81–2.75 (m, 1H), 2.44–2.38 (m, 1H), 2.35–2.29 (m, 1H), 2.25–2.15 (m, 1H), 2.07–1.99 (m, 2H), 1.78 (s, 1H), 1.64–1.61 (m, 2H), 1.45–1.29 (m, 3H), 0.95 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.42, 174.54, 174.14, 173.67, 172.03, 171.15, 168.07, 136.07, 132.24, 131.62, 130.56, 129.26, 126.62, 125.51, 79.59, 73.13, 69.20, 60.51, 59.80, 56.78, 55.45, 54.69, 52.66, 52.25, 41.26, 40.63, 37.02, 32.73, 30.06, 25.75, 23.30, 21.85, 18.73; LRMS (ESI) m/z 853 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₄H₄₉F₃N₆O₁₄P⁺ [M + H]⁺: 853.2991. Found: 853.2996.

Methyl ((3S,6S,9S,17R,19S)-9-(3-(2-(dimethylamino)ethoxy)-benzamido)-6-(hydroxymethyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]-icosane-19-carbonyl)-L-leucinate (16b).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16b (62 mg, 68%). ¹H NMR (400 MHz, CD₃OD) δ 7.49–7.47 (m, 2H), 7.41 (t, J = 8.0 Hz, 1H), 7.21 (d, J = 8.0 Hz, 1H), 4.59–4.58 (m, 1H), 4.54–4.50 (m, 1H), 4.45–4.34 (m, 5H), 4.15 (s, 1H), 3.95–3.87 (m, 2H), 3.74–3.59 (m, 5H), 3.69 (s, 3H), 3.54–3.49 (m, 1H), 3.18–3.13 (m, 1H), 2.97 (s, 6H), 2.74–2.71 (m, 1H), 2.43–2.38 (m, 1H), 2.35–2.28 (m, 1H), 2.25–2.15 (m, 1H), 2.05–1.96 (m, 2H), 1.78–1.76 (m, 1H), 1.64–1.55 (m, 2H), 1.43–1.29 (m, 3H), 0.94 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.49, 174.57, 174.14, 173.65, 171.92, 171.43, 169.08, 159.13, 136.64, 130.96, 121.82, 119.45, 114.74, 79.62, 72.55, 69.28, 63.62, 63.00, 60.53, 60.32, 57.61, 55.54, 54.53, 52.67, 52.29, 44.06, 41.27, 40.62, 37.06, 32.76, 30.31, 25.78, 23.32, 21.88, 18.97; LRMS (ESI) m/z 872 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₇H₅₉N₇O₁₅P⁺ [M + H]⁺: 872.3801. Found: 872.3807.

Methyl ((3S,6S,9S,17R,19S)-6-(hydroxymethyl)-9-(3-isobutoxy-benzamido)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (16c).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16c (54 mg, 65%). ¹H NMR (400 MHz, CD₃OD) δ 7.38–7.33 (m, 3H), 7.08 (d, J = 8.0 Hz, 1H), 4.63–4.59 (m, 1H), 4.54–4.50 (m, 1H), 4.42–4.36 (m, 3H), 4.15 (s, 1H), 3.98–3.89 (m, 2H), 3.80–3.78 (m, 2H), 3.77–3.61 (m, 3H), 3.69 (s, 3H), 3.55–3.50 (m, 1H), 3.19–3.14 (m, 1H), 2.75–2.70 (m, 1H), 2.44–2.39 (m, 1H), 2.35–2.28 (m, 1H), 2.25–2.15 (m, 1H), 2.06–2.02 (m, 3H), 1.78 (s, 1H), 1.64–1.59 (m, 2H), 1.43–1.35 (m, 3H), 1.05–0.91 (m, 12H); ¹³C NMR (100 MHz, CD₃OD) δ 175.47, 174.59, 174.16, 173.61, 171.84, 171.63, 169.41, 160.70, 136.30, 130.65, 120.43, 119.11, 114.42, 79.57, 75.57, 72.11, 69.19, 63.00, 60.48, 56.72, 55.50, 54.45, 52.87, 52.24, 41.21, 40.61, 37.04, 32.68, 30.27, 29.43, 25.73, 23.28, 21.85, 19.48, 18.99; LRMS (ESI) m/z 857 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₇H₅₈N₆O₁₅P⁺ [M + H]⁺: 857.3692. Found: 857.3698.

Methyl ((3S,6S,9S,17R,19S)-9-(3-(2-(dimethylamino)ethoxy)-5-isobutoxybenzamido)-6-(hydroxymethyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]-icosane-19-carbonyl)-L-leucinate (16d).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16d (67 mg, 67%). ¹H NMR (400 MHz, DMSO-d₆) δ 7.00 (s, 1H), 6.99 (s, 1H), 6.54 (s, 1H), 4.50–4.33 (m, 3H), 4.18–4.15 (m, 4H), 4.06 (s, 3H), 3.78–3.76 (m, 3H), 3.59 (s, 3H), 3.60–3.55 (m, 4H), 3.05–2.90 (m, 3H), 2.86–2.67 (m, 3H), 2.36 (s, 6H), 2.18–2.08 (m, 3H), 2.05–1.98 (m, 3H), 1.84–1.82 (m, 3H), 1.71–1.69 (m, 2H), 1.56–1.45 (m, 4H), 1.23 (s, 3H), 1.18–1.16 (m, 2H), 0.98 (d, J = 8.0 Hz, 6H), 0.87 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 172.97, 172.40, 171.69, 171.28, 169.82, 168.62, 165.40, 159.77, 159.21, 136.20, 106.19, 105.98, 104.13, 77.83, 74.04, 67.81, 65.24, 61.62, 58.94, 58.21, 56.98, 55.64, 53.52, 52.51, 51.76, 50.50, 44.82, 40.42, 35.83, 31.55, 29.58, 29.04, 27.75, 24.15, 22.72, 21.53, 19.08, 18.37; LRMS (ESI) m/z 944 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₁H₆₇N₇O₁₆P⁺ [M + H]⁺: 944.4376. Found: 944.4382.

Methyl ((3S,6S,9S,17R,19S)-9-(3,5-diisobutoxybenzamido)-6-(hydroxymethyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (16e).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16e (62 mg, 63%). ¹H NMR (400 MHz, CD₃OD) δ 6.97 (s, 1H), 6.61 (s, 1H), 4.54–4.52 (m, 2H), 4.45–4.38 (m, 3H), 4.15 (s, 1H), 3.96–3.89 (m, 2H), 3.77–3.73 (m, 4H), 3.77–3.69 (m, 3H), 3.69 (s, 3H), 3.55–3.52 (m, 1H), 3.20–3.14 (m, 1H), 2.81–2.73 (m, 1H), 2.44–2.38 (m, 1H), 2.35–2.28 (m, 1H), 2.25–2.15 (m, 1H), 2.06–2.02 (m, 4H), 1.78 (s, 1H), 1.62–1.61 (m, 2H), 1.45–1.29 (m, 3H), 1.04–0.91 (m, 18H); ¹³C NMR (100 MHz, CD₃OD) δ 175.50, 174.57, 174.12, 173.82, 171.92, 171.41, 169.44, 160.83, 136.92, 106.89, 105.65, 79.61, 75.69, 72.81, 69.19, 60.53, 60.16, 56.65, 55.45, 54.57, 52.65, 52.24, 41.30, 40.64, 37.04, 32.73, 30.14, 29.47, 25.76, 23.33, 21.86, 19.52, 18.85; LRMS (ESI) m/z 929 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₁H₆₆N₆O₁₆P⁺ [M + H]⁺: 929.4267. Found: 929.4273.

4-(3-(2-(Dimethylamino)ethoxy)-5-(((3S,6S,9S,17R,19S)-6-(hydroxymethyl)-19-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)-carbamoyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosan-9-yl)carbamoyl)-phenoxy)butanoic acid (16f).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16f (60 mg, 59%). ¹H NMR (400 MHz, CD₃OD) δ 7.04 (s, 1H), 7.03 (s, 1H), 6.76 (s, 1H), 4.57–4.51 (m, 3H), 4.44–4.40 (m, 2H), 4.36–4.34 (m, 4H), 4.17 (s, 1H), 4.05 (s, 2H), 3.95–3.90 (m, 3H), 3.70–3.62 (m, 4H), 3.70 (s, 3H), 3.54–3.50 (m, 3H), 3.19–3.13 (m, 1H), 2.92 (s, 6H), 2.81–2.66 (m, 1H), 2.43–2.38 (m, 2H), 2.32–2.31 (m, 1H), 2.25–2.15 (m, 1H), 2.07–2.01 (m, 4H), 1.77–1.76 (m, 1H), 1.65–1.59 (m, 2H), 1.43 (s, 3H), 1.34–1.28 (m, 1H), 0.93 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 174.29, 172.96, 172.40, 171.69, 171.28, 169.76, 168.58, 165.35, 159.49, 158.99, 136.25, 106.18, 104.27, 77.84, 69.98, 67.86, 67.01, 64.54, 61.56, 58.86, 58.18, 56.41, 55.63, 53.55, 52.47, 51.75, 50.49, 44.21, 35.83, 31.51, 30.25, 29.60, 29.00, 24.24, 24.14, 22.72, 21.51, 18.35; LRMS (ESI) m/z 974 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₁H₆₅N₇O₁₈P⁺ [M + H]⁺: 974.4118. Found: 974.4124.

3′-(2-(Dimethylamino)ethoxy)-5′-(((3S,6S,9S,17R,19S)-6-(hydroxymethyl)-19-(((S)-1-methoxy-4-methyl-1-oxopentan-2-yl)-carbamoyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosan-9-yl)carbamoyl)-[1,1′-biphenyl]-4-carboxylic acid (16g).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16g (64 mg, 61%). ¹H NMR (400 MHz, CD₃OD) δ 8.08 (d, J = 8.0 Hz, 2H), 7.79 (d, J = 8.0 Hz, 2H), 7.49–7.24 (m, 3H), 4.60 (s, 1H), 4.54–4.44 (m, 5H), 4.36–4.31 (m, 1H), 4.13 (s, 1H), 3.93–3.86 (m, 2H), 3.74–3.59 (m, 5H), 3.69 (s, 3H), 3.53–3.48 (m, 1H), 3.18–3.13 (m, 1H), 3.00 (s, 6H), 2.81–2.66 (m, 1H), 2.43–2.38 (m, 1H), 2.35–2.28 (m, 1H), 2.25–2.15 (m, 1H), 2.05–1.96 (m, 2H), 1.79–1.78 (m, 1H), 1.62–1.60 (m, 2H), 1.45–1.28 (m, 3H), 0.94 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.53, 174.56, 174.14, 173.83, 171.09, 170.52, 169.46, 168.85, 159.53, 145.27, 142.90, 137.20, 131.40, 129.43, 129.12, 128.69, 128.18, 120.77, 118.06, 114.31, 79.65, 73.35, 69.37, 63.66, 63.00, 60.51, 59.94, 57.55, 55.47, 54.69, 52.65, 52.27, 44.05, 41.35, 40.65, 37.07, 32.83, 30.27, 25.77, 23.34, 21.87, 18.80; LRMS (ESI) m/z 992 [M + H]⁺. HRMS (ESI) m/z calculated for C₄₄H₆₃N₇O₁₇P⁺ [M + H]⁺: 992.4013. Found: 992.4018.

Methyl ((3S,6S,9S,17R,19S)-9-(3,5-bis((2,5,8,11-tetraoxatridecan-13-yl)oxy)benzamido)-6-(hydroxymethyl)-2,5,8,12-tetraoxo-3-((R)-1-(phosphonooxy)ethyl)-16-oxa-1,4,7,13-tetraazabicyclo[15.2.1]icosane-19-carbonyl)-L-leucinate (16h).

Compound 15 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 16h (68 mg, 54%). ¹H NMR (400 MHz, CD₃OD) δ 7.03 (s, 2H), 6.72 (s, 1H), 4.54–4.50 (m, 3H), 4.43–4.40 (m, 4H), 4.17–4.16 (m, 5H), 3.99–3.97 (m, 1H), 3.86–3.84 (m, 5H), 3.78–3.69 (m, 11H), 3.67–3.64 (m, 5H), 3.63–3.60 (m, 10H), 3.53–3.51 (m, 4H), 3.17–3.13 (m, 1H), 2.76–2.73 (m, 1H), 2.43–2.38 (m, 1H), 2.35–2.27 (m, 1H), 2.21–2.17 (m, 1H), 2.05–2.01 (m, 2H), 1.78 (s, 1H), 1.67–1.61 (m, 2H), 1.45–1.43 (m, 2H), 1.39–1.36 (m, 13H), 1.33–1.29 (m, 1H), 0.95 (dd, J = 16.0, 8.0 Hz, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 175.44, 174.51, 174.12, 173.66, 171.34, 171.13, 169.01, 161.37, 137.19, 107.43, 106.23, 79.58, 72.73, 71.52, 71.29, 71.26, 71.18, 71.01, 70.63, 69.18, 69.02, 60.50, 60.13, 59.13, 56.59, 55.77, 55.45, 54.53, 52.65, 52.23, 41.28, 40.62, 37.04, 32.73, 30.25, 25.75, 23.33, 21.88, 18.76; LRMS (ESI) m/z 1197 [M + H]⁺. HRMS (ESI) m/z calculated for C₅₁H₈₆N₆O₂₄P⁺ [M + H]⁺: 1197.5426. Found: 1197.5431.

Methyl O-phosphono-N-(3-(trifluoromethyl)benzoyl)-L-glutaminyl-L-seryl-L-threonyl-L-prolyl-L-leucinate (21).

Compound 20 (100 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford 21 (67 mg, 75%). ¹H NMR (400 MHz, CD₃OD) δ 8.25–8.14 (m, 2H), 7.86 (d, J = 8.0 Hz, 1H), 7.69 (dd, J = 8.0, 8.0 Hz, 1H), 4.70–4.64 (m, 1H), 4.62–4.57 (m, 1H), 4.52–4.49 (m, 1H), 4.46–4.39 (m, 2H), 3.87–3.69 (m, 4H), 3.70 (s, 3H), 2.51–2.43 (m, 2H), 2.26–2.23 (m, 1H), 2.19–2.16 (m, 1H), 2.07–2.04 (m, 1H), 1.97–1.93 (m, 2H), 1.78–1.73 (m, 1H), 1.64–1.59 (m, 2H), 1.43–1.34 (m, 3H), 0.97–0.88 (m, 6H); ¹³C NMR (100 MHz, DMSO-d₆) δ 174.32, 172.87, 171.91, 171.35, 169.72, 167.51, 164.92, 134.92, 131.72, 129.60, 127.64, 125.40, 124.23, 122.70, 71.64, 61.54, 59.30, 55.25, 53.24, 21.83, 50.41, 47.47, 41.54, 31.67, 29.24, 27.32, 24.15, 22.79, 21.35, 17.97, 16.71; LRMS (ESI) m/z 811 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₂H₄₇F₃N₆O₁₃P⁺ [M + H]⁺: 811.2885. Found: 811.2891.

Methyl N-L-prolyl-L-methionyl-L-glutaminyl-L-seryl-O-phosphono-L-threonyl-L-prolyl-L-leucinate (PMQSpTPL).

Compound 20 (10 mg, 0.11 mmol) was transformed into the target compound using the following procedure. The residue was dissolved in MeOH, solidified by adding Et₂O, and filtered to afford PMQSpTPL (68 mg, 71%). ¹H NMR (400 MHz, CD₃OD) δ 4.59–4.52 (m, 3H), 4.47–4.36 (m, 4H), 3.97–3.89 (m, 1H), 3.80–3.73 (m, 2H), 3.70 (s, 3H), 3.52–3.46 (m, 1H), 3.42–3.33 (m, 3H), 2.58–2.54 (m, 1H), 2.51–2.46 (m, 1H), 2.36–2.33 (m, 2H), 2.23–2.16 (m, 3H), 2.12–2.09 (m, 4H), 2.06–2.01 (m, 3H), 1.99–1.91 (m, 1H), 1.89–1.86 (m, 1H), 1.76–1.71 (m, 1H), 1.67–1.57 (m, 2H), 1.41–1.37 (m, 2H), 1.33–1.32 (m, 3H), 1.29–1.26 (m, 2H), 0.96–0.90 (m, 6H); ¹³C NMR (100 MHz, CD₃OD) δ 177.90, 174.85, 174.74, 174.54, 174.45, 173.61, 172.38, 170.74, 74.21, 71.81, 68.82, 62.80, 61.66, 61.05, 59.25, 57.17, 55.77, 54.82, 54.67, 52.70, 52.34, 49.85, 43.78, 41.13, 32.49, 31.09, 30.63, 28.37, 26.07, 25.94, 25.86, 25.55, 25.40, 23.31, 21.89, 19.19, 18.03, 15.31, 13.17, 10.85; LRMS (ESI) m/z 867 [M + H]⁺. HRMS (ESI) m/z calculated for C₃₄H₆₀N₈O₁₄PS⁺ [M + H]⁺: 867.3682. Found: 867.3687.

ELISA-Based Plk1-PBD Binding Inhibition Assay.

Essentially following the procedure described previously,¹⁰ a Plk1-PBD binding biotinylated PBIP1 p-T78 peptide was first diluted to a final concentration of 0.3 μM with coating solution (KPL, Inc., Gaithersburg, MD) and then 100 μL of the resulting solution was immobilized on a 96-well streptavidin-coated plate (Nalgene Nunc, Rochester, NY). The immobilized materials were washed with phosphate-buffered saline + 0.05% Tween 20 (PBST) once and incubated with 200 μL of 1% BSA in PBS (blocking buffer) for 1 h to block the unoccupied surface. HEK293A cell lysates expressing HA-EGFP-Plk1 were prepared in TBSN (20 mM Tris-Cl (pH 8.0), 150 mM NaCl, 0.5% NP-40, 5 mM EGTA, 1.5 mM EDTA, 20 mM p-nitropheylphosphate and protease inhibitor cocktail (Roche)) supplemented with 40 mM p-nitrophenyl phosphate and a protease inhibitor cocktail (Roche). Immediately after mixing the lysates with the indicated concentration of macrocyclic phosphopeptides, the samples were applied to the biotinylated p-T78 peptide-coated ELISA wells, incubated with constant rocking for 1 h at 25 °C, and then washed five times with PBST (pH 7.4) + 0.5% Tween-20). To detect bound HA-EGFP-Plk1, samples were incubated with 100 μL/well of anti-HA antibody (0.5 μg/mL in the blocking buffer) for 2 h and washed five times. Afterward, the plates were incubated with 100 μL/well of secondary antibody at the 1:1000 dilution in the blocking buffer, washed five times with PBST, and then incubated with 100 μL/well of 3,3′,5,5′-tetramethylbenzidine substrate solution (Sigma, St. Louis, MO) until a desired absorbance was achieved. Reactions were terminated by the addition of 100 μL/well of stop solution (Cell Signaling Technology, MA) and the optical densities were measured at 450 nm using a Perkin-Elmer EnSpire Multimode Plate reader (Perkin-Elmer, MA). Data acquired from three independent experiments were plotted using GraphPad Prism 7 and IC₅₀ values were determined.

Binding Specificities for PBDs of Plk1 Using a Fluorescence Polarization Assay.

5-Carboxyfluorescein-labeled peptides, 5-carboxyfluorescein-Ahx-DPPLHS-pT-AI-NH₂,²³ 5-carboxyfluorescein-Ahx-GPMQTS-pT-PKNG, and 5-carboxyfluorescein-Ahx-GPLATS-pT-PKNG⁴⁵ (final concentrations: 10 nM) were incubated with corresponding Plk1, Plk2 and Plk3-PBDs (PBD concentration was used at EC₈₅) purified from E. coli in a binding buffer containing 10 mM Tris-Cl (pH 8.0), 1 mM EDTA, 50 mM NaCl, and 0.01% Nonidet P-40. Binding reactions were carried out in the presence of testing peptides at the indicated concentrations. Fluorescence polarization was measured 30 min after mixing all components in the 384-well format using Molecular Devices Spectra Max Paradigm with a Multi-Mode Microplate Detection platform. Data acquired from three independent experiments were plotted using GraphPad Prism 7.

X-ray Crystallography.

Plk1-PBD protein (residues 371–603) was purified using the procedure previously described.⁸ Crystals were grown using the hanging drop vapor diffusion method. Plk1-PBD protein at 12 mg/mL in 10 mM Tris pH 8.0, 0.5 M NaCl, 10 mM DTT, 2% DMSO and 2 mM 16a were mixed with an equal volume of reservoir solution consisting of 0.1 M Bis-Tris pH 5.5, 2.75 M NaCl. Crystals appeared overnight and reached maximum size over several days. Crystals were cryo-protected in 0.1 M Bis-Tris pH 5.5, 3 M NaCl, 28% (v/v) glycerol, 10 mM DTT, 2% DMSO, 2 mM 16a, and data were collected at 100 K on a Mar345 image plate detector with a Rigaku RU-300 home X-ray source. The data were processed with the HKL⁴⁶ and CCP4⁴⁷ software suites. The structure was solved by molecular replacement using EPMR⁴⁸ and chain A of structure 3FVH⁸ (RCSB accession code) as a search model, and refined using PHENIX⁴⁹ with manual fitting in XtalView.⁵⁰ The figure was created using Molscript⁵¹ and PyMOL (http://www.pymol.org/). The coordinates and structure factors have been deposited with the RCSB accession code 7MSO.

Molecular Docking of Macrocyclic Phosphopeptide Bound to Plk1-PBD.

Automated docking simulation of 16e, 16h, and 21 on to the PBD was performed using Autodock4 and our crystal structure of the PBD in complex with 16a (PDB ID: 7MSO). The protein structure was initially kept rigid, while 16e, 16h, and 21 were set free to undergo flexible docking. Polar hydrogen was added using the hydrogen module in AutoDock tools (ADT) for the PBD. Then Gasteiger–Marsilli partial charges were assigned. Docking of 16e, 16h, and 21 to PBD was performed using the empirical free energy function and Lamarckian genetic algorithm, applying a standard protocol with a default GA parameter. For the local search, so-called Solis and Wets algorithm was applied, using a maximum 1000 iterations. The binding energy between the PBD and inhibitor was evaluated using atom affinity potentials that are precalculated on grid maps using AutoGrid. The grids were chosen to be sufficiently large to include not only the inhibitor-bound pocket but also a significant part of the surrounding surface. The grid maps had dimensions of 40 × 42 × 40 ų with 0.375 Å spacing between grid points. The docking conformation of the inhibitor was selected according to the criteria of the lowest binding free energy of AutoDock results. Molecular graphics of selected refined docking models with inhibitor were generated using PyMol package (http://www.pymol.org/).

Zebrafish Experiments.

16a and 16e were microinjected into one-cell stage zebrafish embryos using a PV820 Pneumatic PicoPump (WPI, Sarasota, FL, USA). For immunostaining, embryos were fixed in 4% paraformaldehyde. Embryos were rinsed three times in phosphate-buffered saline (PBS), in cold methanol for 5 min, and then rinsed in MeOH/PBS-Tween 20 (PBST) and PBST for each 5 min. Embryos were blocked for 3 h using Blocking Reagent (5% BSA/ 10% goat serum in PBS/1% DMSO/0.5% Triton-X100/0.1% Tween 20) with mild rotating. Embryos were incubated overnight at 4 °C with a polyclonal anti-phospho-H3 antibody (1:1000, Cell Signaling Technology, Danvers, MA, USA). Embryos were washed 4× in PBDTT (1 h each) before addition of biotin-labeled goat anti-rabbit antibody (1:2000) in 1% BSA/PBDTT after blocking for 4 h. Antibody staining was detected using the Vectastain Elite ABC reagent and Vector DAB substrate kit to the manufacturer’s specifications. Embryos used in this study were obtained from the Zebrafish Center for Disease Modeling (ZCDM; Daejeon, Republic of Korea). Animal experiments were conducted according to guidelines approved by the Animal Care and Use Committee of Chungnam National University.

Cell Culture, Analysis of Aberrant Mitotic Population and Indirect Immunofluorescence Microscopy.

Human HEK293A and HeLa cervical carcinoma cell line CCL2 cells were purchased from the American Type Culture Collection (ATCC) and cultured as recommended by the ATCC. For ELISA assays, cell lysates containing HA-EGFP-Plk1 were prepared by infecting HEK293A cells with adenovirus expressing HA-EGFP-Plk1. To determine the mitotic effect of the indicated macrocyclic phosphopeptide in cultured cells, asynchronously growing HeLa cells were treated with 200 μM of the indicated peptides for 4 h. The resulting cells were fixed with 4% paraformaldehyde for 15 min and then immunostained with the indicated antibodies and stained with 0.1 μg/mL of 4′,6′-diamidino-2-phenylindole (DAPI) as described previously.¹⁷ Confocal images were acquired using Zeiss LSM780 equipped with a plan-apochromat 40× (N.A. 1.3) and 63× (N.A. 1.4) oil-immersion objective lenses (Cal Zeiss Microscopy, LLC), and 12-bit, 0.5-μm z-steps. To quantify fluorescence signal intensities of Plk1, images acquired under the same settings were first processed to generate the maximum intensity projection of z-stacks. The resulting images were analyzed using the Zeiss ZEN v2.1 software (Carl Zeiss Microscopy, LLC.)

Cell Proliferation Assay.

To examine the cellular efficacy of Plk1-PBD inhibitors generated in this study, MTS [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt]-based assays were performed using Cell Titer 96 AQueous One Solution Cell Proliferation Assay kit (Promega, WI) according to the manufacturer’s instructions. In brief, asynchronously growing HeLa cells were plated at a density of 2000 cells/100 μL medium in a 96-well plate in triplicate. The cells were treated with compounds at the indicated concentration for 24 h followed by another 24 h treatment after replacing them with fresh medium (a total of continuous 48 h treatment). The resulting cells were treated with 20 μL of the MTS solution, incubated at 37 °C for 1–4 h in a humidified 5% CO₂ atmosphere, and subjected to absorbance measurement at 490 nm using the Perkin-Elmer EnSpire Multimode 96-well plate reader (Waltham, MA).

ABBREVIATIONS

ADT

autodock tools

ATCC

American type culture collection

DAPI

4′,6′-diamidino-2-phenylindole

FP

fluorescence polarization

TMB

3,3′,5,5′-tetramethylbenzidine