**1. Introduction**

Genomic instability is one of the major driving forces of carcinogenesis. Studies of DNA repair mechanisms and their regulation are directly related to the search for the optimal ways of treating oncological and other human diseases. Chemotherapy is one of the main methods of treating various malignancy types. Anticancer chemotherapeutic

Filimonov, A.S.; Luzina, O.A.; Orlova, K.A.; Chernyshova, I.A.; Kornienko, T.E.; Malakhova, A.A.; Medvedev, S.P.; Zakharenko, A.L.; Ilina, E.S.; et al. New Hybrid Compounds Combining Fragments of Usnic Acid and Thioether Are Inhibitors of Human Enzymes TDP1, TDP2 and PARP1. *Int. J. Mol. Sci.* **2021**, *22*, 11336. https://doi.org/10.3390/

**Citation:** Dyrkheeva, N.S.;

Academic Editor: Maria Luisa Balestrieri

ijms222111336

Received: 30 September 2021 Accepted: 17 October 2021 Published: 20 October 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

drugs, by their principle of action, damage DNA in a targeted manner. They are powerful cell poisons that have a detrimental effect on the rapidly dividing cells of malignant tumors with a comparatively less negative damaging effect on the healthy, normally dividing cells and tissues of the organism. Although DNA repair is essential for a healthy cell, during anticancer therapy repair enzymes of the cancer cells counteract the efficacy of anticancer agents. Thus, DNA repair leads to a decrease in the effectiveness of therapy and contributes to the resistance of malignancies to chemotherapy. In recent years, much focus has been put on the DNA repair enzymes as targets for drug development. Researchers are actively searching for new compounds that suppress the DNA repair enzymes activity to increase the efficiency of anticancer therapy. Currently, tyrosyl-DNA phosphodiesterase 1 (TDP1) and poly(ADP-ribose) polymerase 1 (PARP1) are considered as promising target DNA repair enzymes for creating drugs [1,2].

TDP1 is involved in repairing stalled topoisomerase 1–DNA complexes by catalyzing the hydrolysis of the phosphodiester bond between the tyrosine residue of topoisomerase 1 (TOP1) and the 3' phosphate of DNA in the single-strand break generated by TOP1. TDP1 also catalyzes the cleavage of phosphodiester bonds in other DNA–protein adducts and a number of different lesions at the 3' end of DNA [3]. TDP1 plays a key role in removing the damage from DNA caused by the anticancer drugs used in clinical practice, such as topotecan (Tpc) and irinotecan, which are derivatives of the natural compound camptothecin [4,5]. Consequently, TDP1 activity may be a possible cause of tumor resistance to TOP1 inhibitors.

Tyrosyl-DNA phosphodiesterase 2 (TDP2) is a DNA repair enzyme that catalyzes the hydrolysis of dead-end complexes between DNA and the topoisomerase 2 (TOP2) active site tyrosine residue. TDP2 can remove a variety of covalent adducts from DNA through hydrolysis of a 50 phosphodiester bond, giving rise to DNA with a free 50 phosphate. TOP2 inhibitors stabilize the TOP2–DNA covalent complex and induce cell death [6]. TOP2 inhibitors (etoposide, doxorubicin) are widely used in clinical practice as antineoplastic drugs. TDP2 activity reduces the effectiveness of these drugs and, vice versa, TDP2 deficiency leads to a significant increase in sensitivity to TOP2 inhibitors [7,8]. The same as TDP1 inhibitors, TDP2 inhibitors can significantly increase the effectiveness of chemotherapy by synergizing with TOP2 inhibitors. The most effective TDP2 inhibitors today are deazaflavins [9,10]. Deazaflavin has been shown to synergize in vitro with etoposide at non-toxic concentrations [9]. It should be noted that deazaflavins have unsatisfactory pharmacokinetic characteristics, which makes it necessary to search for inhibitors of new structural types. Several TDP2 inhibitors of different chemical groups have already been proposed, but most of them had moderate efficacy [11]. Recently, though, some more efficient TDP2 inhibitors have been found [12–14]

TDP1 and TDP2 have little overlapping activity because TDP1 has a weak activity for 5 0 -phosphotyrosyl bonds, and TDP2 has a weak activity for 30 -phosphotyrosyl bonds [5]. Nevertheless, the ability of TDP1 and TDP2 to take on the functions of each other makes it highly promising to use the selective inhibitors of these two enzymes together, or to create agents capable of simultaneously inhibiting both TDP1 and TDP2. Simultaneous suppression of the activity of these two enzymes can be used to increase the effectiveness of a large set of clinically important anticancer drugs, TOP1 and TOP2 inhibitors. Recently, the first triple inhibitors of TOP1/TDP1/TDP2 have been discovered by Pommier's group [15], which exhibit only moderate activity against TDP1 and weak activity against TDP2.

PARP1 catalyzes the synthesis of poly(ADP-ribose) (PAR), which is responsible for the post-translational modification of proteins, as an immediate DNA damage response of the cell. PARP1 is a member of various DNA repair pathways [16,17]. Inhibition of PARP1 activity makes it possible to sensitize tumor cells to the action of chemotherapeutic drugs. Olaparib was the first example of therapeutic synthetic lethality in oncology approved by the FDA for the treatment of advanced ovarian cancers associated with defective BRCA1/2 [18]. To date, the FDA has approved four different PARP inhibitors, olaparib (2014), rucaparib (2016), niraparib (2017), and talazoparib (2018), for the treatment of

ovarian, fallopian tube, breast, and peritoneal cancers [19]. More inhibitors are in various stages of development and preclinical testing. In addition to cancer, PARP inhibitors are also promising in the treatment of cardiovascular diseases [20]. The anticancer effect of TOP1 inhibitors can be significantly enhanced by the simultaneous inhibition of PARP1 and TDP1. PARP1–TDP1 interplay was shown in a series of publications. PARylation of TDP1 enhances its recruitment to sites of DNA damage without interfering with the catalytic activity of TDP1 [21]. Briefly, it was shown [21] that the N-terminal domain of TDP1 directly binds the C-terminal domain of PARP1, and TDP1 is PARylated but not inactivated by PARP1. PARP1 is known to interact with TDP1 directly with K<sup>D</sup> 120 nM [22]. PARP1 stimulates the enzymatic activity of TDP1 on AP sites [23]. TDP1 can repair the covalent DNA–PARP1 crosslink at the apurinic/apyrimidinic (AP) site in double-stranded DNA [24,25]. Other authors also noticed that TDP1 together with PARP1 inhibition could be a successful cancer treatment strategy [26,27]. TDP1 expression is correlated with other DNA repair genes, including PARP1, BRCA2, and BRCA1 [28]. The combination of 70% to 90% Tdp1 knockdown and 10 mmol/L of the PARP1 inhibitor rucaparib was found to reduce the proliferation of A204, Birch, RH30, and CW9019 cells more than either of these treatments alone [29]. These facts make urgent the search for dual inhibitors of TDP1 and PARP1. In the other work [30], TDP1, together with PARP1, were shown to be essential cellular proteins in cancerogenic [31] human papillomavirus HPV18 replication, thus making those proteins good targets for developing HPV inhibitors. The authors also noticed that TDP1 and PARP1 inhibitors might also be effective against HPV-induced cancer. ment of ovarian, fallopian tube, breast, and peritoneal cancers [19]. More inhibitors are in various stages of development and preclinical testing. In addition to cancer, PARP inhibitors are also promising in the treatment of cardiovascular diseases [20]. The anticancer effect of TOP1 inhibitors can be significantly enhanced by the simultaneous inhibition of PARP1 and TDP1. PARP1–TDP1 interplay was shown in a series of publications. PARylation of TDP1 enhances its recruitment to sites of DNA damage without interfering with the catalytic activity of TDP1 [21]. Briefly, it was shown [21] that the N-terminal domain of TDP1 directly binds the C-terminal domain of PARP1, and TDP1 is PARylated but not inactivated by PARP1. PARP1 is known to interact with TDP1 directly with KD 120 nM [22]. PARP1 stimulates the enzymatic activity of TDP1 on AP sites [23]. TDP1 can repair the covalent DNA–PARP1 crosslink at the apurinic/apyrimidinic (AP) site in doublestranded DNA [24,25]. Other authors also noticed that TDP1 together with PARP1 inhibition could be a successful cancer treatment strategy [26,27]. TDP1 expression is correlated with other DNA repair genes, including PARP1, BRCA2, and BRCA1 [28]. The combination of 70% to 90% Tdp1 knockdown and 10 mmol/L of the PARP1 inhibitor rucaparib was found to reduce the proliferation of A204, Birch, RH30, and CW9019 cells more than either of these treatments alone [29]. These facts make urgent the search for dual inhibitors of TDP1 and PARP1. In the other work [30], TDP1, together with PARP1, were shown to be essential cellular proteins in cancerogenic [31] human papillomavirus HPV18 replication, thus making those proteins good targets for developing HPV inhibitors. The authors also noticed that TDP1 and PARP1 inhibitors might also be effective against HPV-induced cancer.

olaparib (2014), rucaparib (2016), niraparib (2017), and talazoparib (2018), for the treat-

*Int. J. Mol. Sci.* **2021**, *22*, x FOR PEER REVIEW 3 of 19

We have previously discovered highly effective inhibitors of TDP1 based on the secondary metabolite of lichens usnic acid **1** (UA)—compounds **2**–**5** (Figure 1) [32–36]—the synergistic action of which, when combined with the TOP1 inhibitor Tpc, was confirmed in experiments on cell cultures and in animal models [36–38]. We have previously discovered highly effective inhibitors of TDP1 based on the secondary metabolite of lichens usnic acid **1** (UA)—compounds **2**–**5** (Figure 1) [32–36]—the synergistic action of which, when combined with the TOP1 inhibitor Tpc, was confirmed in experiments on cell cultures and in animal models [36–38].

**Figure 1.** Structures of usnic acid (**1**) and its derivatives (**2**–**5**) that inhibit TDP1 [32–36]. **Figure 1.** Structures of usnic acid (**1**) and its derivatives (**2**–**5**) that inhibit TDP1 [32–36].

We also previously studied the PARP1-inhibiting activity of various UA derivatives [39]. Most of the UA derivatives modified by the substituents in the C ring and the introduction of a fused ring or heterocycles into the A ring had low PARP1 affinity. The PARP1 inhibitory activity was facilitated by the introduction of aromatic substituents into the acyl fragment of the A ring of UA. Analysis of the literature showed that the pharmacophore fragments with respect to PARP1 are heterocycles containing one or several heteroatoms, primarily nitrogen (Figure 2) [40]. We also previously studied the PARP1-inhibiting activity of various UA derivatives [39]. Most of the UA derivatives modified by the substituents in the C ring and the introduction of a fused ring or heterocycles into the A ring had low PARP1 affinity. The PARP1 inhibitory activity was facilitated by the introduction of aromatic substituents into the acyl fragment of the A ring of UA. Analysis of the literature showed that the pharmacophore fragments with respect to PARP1 are heterocycles containing one or several heteroatoms, primarily nitrogen (Figure 2) [40].

*Int. J. Mol. Sci.* **2021**, *22*, x FOR PEER REVIEW 4 of 19

**Figure 2.** Some known inhibitors of PARP1. **Figure 2.** Some known inhibitors of PARP1. **Figure 2.** Some known inhibitors of PARP1.

We chose UA derivative **7** for the design of the joint TDP1 and PARP1 action inhibitors, which allow easy introduction of heteroatom-rich fragments into the acyl fragment of the ring A of UA (Figure 3). The activity of this type of UA derivative against DNA repair enzymes, including TDP1 and PARP1, has not been studied previously. We chose UA derivative **7** for the design of the joint TDP1 and PARP1 action inhibitors, which allow easy introduction of heteroatom-rich fragments into the acyl fragment of the ring A of UA (Figure 3). The activity of this type of UA derivative against DNA repair enzymes, including TDP1 and PARP1, has not been studied previously. We chose UA derivative **7** for the design of the joint TDP1 and PARP1 action inhibitors, which allow easy introduction of heteroatom-rich fragments into the acyl fragment of the ring A of UA (Figure 3). The activity of this type of UA derivative against DNA repair enzymes, including TDP1 and PARP1, has not been studied previously.

**Figure 3.** Retrosynthetic scheme for the synthesis of compound **7**. **Figure 3.** Retrosynthetic scheme for the synthesis of compound **7**. **Figure 3.** Retrosynthetic scheme for the synthesis of compound **7**.

In this work, we propose to use an approach for the synthesis of potential TDP1 inhibitors or dual inhibitors based on the construction of a derivative from a UA backbone, a flexible linker, and a heterocycle to synthesize a set of compounds derivatized through cycle A. The aim of this study was to assess the inhibitory ability of the UA derivatives against three DNA repair enzymes, search for selective and multitarget agents, study the inhibition mechanism assessment of toxicity, and study the prospects of using these compounds as potential pharmacological agents of in anticancer therapy. In this work, we propose to use an approach for the synthesis of potential TDP1 inhibitors or dual inhibitors based on the construction of a derivative from a UA backbone, a flexible linker, and a heterocycle to synthesize a set of compounds derivatized through cycle A. The aim of this study was to assess the inhibitory ability of the UA derivatives against three DNA repair enzymes, search for selective and multitarget agents, study the inhibition mechanism assessment of toxicity, and study the prospects of using these compounds as potential pharmacological agents of in anticancer therapy. In this work, we propose to use an approach for the synthesis of potential TDP1 inhibitors or dual inhibitors based on the construction of a derivative from a UA backbone, a flexible linker, and a heterocycle to synthesize a set of compounds derivatized through cycle A. The aim of this study was to assess the inhibitory ability of the UA derivatives against three DNA repair enzymes, search for selective and multitarget agents, study the inhibition mechanism assessment of toxicity, and study the prospects of using these compounds as potential pharmacological agents of in anticancer therapy.

#### **2. Results and Discussion 2. Results and Discussion 2. Results and Discussion**

#### *2.1. Chemistry 2.1. Chemistry 2.1. Chemistry*

In order to find out the structure–activity relationship (SAR), we synthesized the set of usnic acid (UA) derivatives of **7**, with five- and hexa-membered mono- and bicyclic heterocyclic along with non-heterocyclic and acyclic substituents. In order to find out the structure–activity relationship (SAR), we synthesized the set of usnic acid (UA) derivatives of **7**, with five- and hexa-membered mono- and bicyclic heterocyclic along with non-heterocyclic and acyclic substituents. In order to find out the structure–activity relationship (SAR), we synthesized the set of usnic acid (UA) derivatives of **7**, with five- and hexa-membered mono- and bicyclic heterocyclic along with non-heterocyclic and acyclic substituents.

The desired novel and known UA thioethers **7a**–**k** were synthesized using the procedure described early [41], by reaction of bromousnic acid **6** with the corresponding thiols in presence of NaOH. The target thioethers were obtained in 49–94% yield (Scheme 1). The desired novel and known UA thioethers **7a**–**k** were synthesized using the procedure described early [41], by reaction of bromousnic acid **6** with the corresponding thiols in presence of NaOH. The target thioethers were obtained in 49–94% yield (Scheme 1). The desired novel and known UA thioethers **7a**–**k** were synthesized using the procedure described early [41], by reaction of bromousnic acid **6** with the corresponding thiols in presence of NaOH. The target thioethers were obtained in 49–94% yield (Scheme 1).

**Scheme 1.** Synthesis of the thioether usnic acid derivatives. **Scheme 1.** Synthesis of the thioether usnic acid derivatives. **Scheme 1.** Synthesis of the thioether usnic acid derivatives.

UA derivative **9** was synthesized in two steps (Scheme 2). At first, the reaction of the bromousnic acid with sodium thiocyanate in acetone resulted in obtaining compound **8**, with a 95% yield. Then, UA thiocyanate derivative **8** was hydrolyzed. The hydrolysis was performed with (33%) sulfuric acid in glacial acetic acid to obtain the desired compound **9**, with a 38% yield. UA derivative **9** was synthesized in two steps (Scheme 2). At first, the reaction of the bromousnic acid with sodium thiocyanate in acetone resulted in obtaining compound **8**, with a 95% yield. Then, UA thiocyanate derivative **8** was hydrolyzed. The hydrolysis was performed with (33%) sulfuric acid in glacial acetic acid to obtain the desired compound **9**, with a 38% yield. UA derivative **9** was synthesized in two steps (Scheme 2). At first, the reaction of the bromousnic acid with sodium thiocyanate in acetone resulted in obtaining compound **8**, with a 95% yield. Then, UA thiocyanate derivative **8** was hydrolyzed. The hydrolysis was performed with (33%) sulfuric acid in glacial acetic acid to obtain the desired compound **9**, with a 38% yield.

**Scheme 2.** Synthesis of the thiocarbamate usnic acid derivative **9**. **Scheme 2.** Synthesis of the thiocarbamate usnic acid derivative **9**. **Scheme 2.** Synthesis of the thiocarbamate usnic acid derivative **9**.

The target sulfoxide derivatives of **10** were obtained using the procedure described in [42], which is by a peroxidation reaction of the corresponding thioethers **7g**,**h** with mCPBA in CH2Cl2 at 0 °C (Scheme 3). The desired compounds **10a**,**b** were obtained, with 70–75% yields. The target sulfoxide derivatives of **10** were obtained using the procedure described in [42], which is by a peroxidation reaction of the corresponding thioethers **7g**,**h** with mCPBA in CH2Cl2 at 0 °C (Scheme 3). The desired compounds **10a**,**b** were obtained, with 70–75% yields. The target sulfoxide derivatives of **<sup>10</sup>** were obtained using the procedure describedin [42], which is by a peroxidation reaction of the corresponding thioethers **7g**,**<sup>h</sup>** with mCPBA in CH2Cl<sup>2</sup> at 0 ◦C (Scheme 3). The desired compounds **10a**,**b** were obtained, with 70–75% yields.

*Int. J. Mol. Sci.* **2021**, *22*, x FOR PEER REVIEW 6 of 19

**Scheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**. **Scheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**. **Scheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**. **Scheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**. **Scheme 10aScheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**. **Scheme 3.** Synthesis of the sulfoxide usnic acid derivatives **10a**,**b**.

#### *2.2. Biology 2.2. Biology 2.2. Biology 2.2. Biology Biology 2.2. Biology 2.2. Biology 2.2. Biology*

#### 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity

We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory properties by measuring their IC50 values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentration range (IC50 1.7–25.2 μM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, except for **7d** (IC50 25.2 μM), whereas the compounds containing a six-membered heterocycle (**7i**-**k**) inhibit TDP1 at higher concentrations (IC50 > 11 μM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC50 2.2 μM and 1.4 μM, respectively) enhances the inhibitory properties of the compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**h** and **10a**,**b** (IC50 of 1.7 μM, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**. We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory properties by measuring their IC<sup>50</sup> values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentration range (IC<sup>50</sup> 1.7–25.2 µM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a fivemembered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**–**h** and **10a**,**b**), with an IC<sup>50</sup> of 1.4–4.4 µM, except for **7d** (IC<sup>50</sup> 25.2 µM), whereas the compounds containing a six-membered heterocycle (**7i**–**k**) inhibit TDP1 at higher concentrations (IC<sup>50</sup> > 11 µM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC<sup>50</sup> 2.2 <sup>µ</sup>M and 1.4 <sup>µ</sup>M, respectively) enhances the inhibitory properties ofthe compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**<sup>h</sup>** and**10a**,**<sup>b</sup>** (IC<sup>50</sup> of 1.7 <sup>µ</sup>M, 2.2 <sup>µ</sup>M, 2.1 <sup>µ</sup>M, and 1.4 <sup>µ</sup>M, respectively). Compounds **10a**,**<sup>b</sup>** aresulfoxide analogs of **7g**,**h**. We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory properties by measuring their IC50 values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentration range (IC50 1.7–25.2 μM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, except for **7d** (IC50 25.2 μM), whereas the compounds containing a six-membered heterocycle (**7i**-**k**) inhibit TDP1 at higher concentrations (IC50 > 11 μM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC50 2.2 μM and 1.4 μM, respectively) enhances the inhibitory properties of the compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**h** and **10a**,**b** (IC50 of 1.7 μM, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**. We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory properties by measuring their IC50 values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentration range (IC50 1.7–25.2 μM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, except for **7d** (IC50 25.2 μM), whereas the compounds containing a six-membered heterocycle (**7i**-**k**) inhibit TDP1 at higher concentrations (IC50 > 11 μM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC50 2.2 μM and 1.4 μM, respectively) enhances the inhibitory properties of the compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**h** and **10a**,**b** (IC50 of 1.7 μM, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**. tested all newly synthesized by measuring their All 14 obtained compounds (Table were shown to inhibit TDP1 in the range (IC50 μnot TDP1 We to affects the TDP1 inhibitory of There is a five-membered fragment in (**7b**,**c**,**10a**with M, for containing six-membered hetero-**7i**-TDP1 (IC50 > 11 μpresence of the six-membered (**7h** and **10b**, respectively) properties the compounds. the most of TDP1 compounds **h 10a**,of 1.7 μM, 2.2 μM, respectively). Compounds are analogs . 2.2.1. Real-Time Fluorescence Assay of TDP1 Activity We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory M). assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, μinhibit 2.2 μM and 1.4 μM, respectively) enhances the inhibitory properties of the compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**h** and **10a**,**b** (IC50 of 1.7 We tested all newly synthesized for TDP1 inhibitory properties by measuring their IC50 values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentra-1.7–25.2 UA not can to affects the TDP1 inhibitory activity of There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, except for **7d** (IC50 25.2 μM), whereas the compounds containing a six-membered heteroat the respectively) properties the compounds. Four the most of TDP1 compounds **h** and **10a**,of 1.7 <sup>μ</sup>M, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**. We tested UA and all 14 newly synthesized UA derivatives for their TDP1 inhibitory properties by measuring their IC50 values using a real-time fluorescent assay [43]. All 14 obtained compounds (Table were shown to inhibit TDP1 in the tion range (IC50 1.7–25.2 μM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the TDP1 inhibitory activity of the thioether derivatives. There is a five-membered heterocyclic fragment in the effective compounds (**7b**,**c**,**e**-**h** and **10a**,**b**), with an IC50 of 1.4–4.4 μM, for **7d**(IC50 μM), containing a six-membered heterocycle (**7i**-**k**) inhibit TDP1 at higher concentrations (IC50 > 11 μM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC50 2.2 μM and 1.4 μM, respectively) enhances the inhibitory properties of the compounds. Four of the most effective inhibitors of TDP1 were compounds **7g**,**h** and **10a**,**b** (IC50 of 1.7 μM, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**. We tested 14 synthesized assay [43]. obtained compounds (Table 1) were shown to inhibit TDP1 in the low-enough concentration range (IC50 1.7–25.2 μM). UA does not inhibit TDP1 in these concentrations. We can assume that the structure of the heterocyclic fragment bound to the sulfur atom affects the of thioether derivatives. five-membered clic fragment and with an except for **7d** (IC50 25.2 μM), whereas the compounds containing a six-membered heterocycle (**7i**-**k**) inhibit TDP1 at higher concentrations (IC50 > 11 μM). It is possible that the presence of a 4-halophenyl substituent in the six-membered heterocycle (**7h** and **10b**, IC50 μM respectively) of were compounds **7gh** μM, 2.2 μM, 2.1 μM, and 1.4 μM, respectively). Compounds **10a,b** are sulfoxide analogs of **7g,h**.

IC50


**Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. **Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. **Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. **Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. **Table** PARP1 inhibition ability and cytotoxicity the thioether UA derivatives. **Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. **Table** and PARP1 inhibition ability and cytotoxicity the thioether UA derivatives. **Table 1.** TDP1, TDP2, and PARP1 inhibition ability and cytotoxicity of the thioether UA derivatives. TDP1, of


**Table 1.** *Cont.*

#### 2.2.2. Type of Inhibition of TDP1 Enzyme Reaction for the Most Effective Compounds 2.2.2. Type of Inhibition of TDP1 Enzyme Reaction for the Most Effective Compounds

The reversible enzyme inhibitors were classified as competitive, uncompetitive, noncompetitive, or mixed, depending on the binding of the enzyme, to the enzyme–substrate complex (ES) or to the enzyme–substrate–inhibitor triple complex (ESI). The enzyme could bind the inhibitor at the different steps of catalysis (before or after the substrate) and at the substrate-binding site or another site. The effect of the inhibitor on enzymatic activity can be analyzed using graphical representations of the Michaelis–Menten equation. We defined the type of inhibition for the four most effective inhibitors of TDP1: **7g**,**h** and **10a**,**b** (IC50~1.4–2.2 μM, Table 1) by obtaining the dependence of the reaction rate on the substrate concentration and dependence of Vmax and KM on the concentration of the inhibitor under steady-state reaction conditions. For all four TDP1 inhibitors, we observed a fall in both the Vmax and KM values with an increasing concentration of the inhibitor, which is typical for the uncompetitive type of inhibition (Table S1, Figure S25) when the inhibitor binds the enzyme–substrate complex but does not bind free enzyme. Thus, it was determined that no studied inhibitors demonstrate the competitive mechanism of inhibition. TDP1 has a catalytic pocket with a narrow positively charged cleft where it binds DNA and with a relatively large cleft that contains a mixed charge distribution for the TOP1 fragment binding [44–46]. It is known that TDP1 hydrolyzes the DNA–TOP1 adducts via two coordinated nucleophilic attacks of two His residues (His263 and His493) with the covalent intermediate complex formation [45,46]. The enzyme reaction catalyzed by TDP1 could be inhibited in two steps: the first one is the inhibition of the TDP1 binding to the DNA substrate (the nucleophilic attack of His263). The second step is releasing TDP1 from the transition complex (nucleophilic attack by a water molecule activated by His493). Inhibition of the second step prevents the release of the DNA 3′-phosphate product from the DNA–TDP1 complex. Spinocerebellar ataxia with axonal neuropathy (SCAN1) is a neurological disease that is caused by a TDP1 His493Arg mutation. This mutation prevents the second step of hydrolysis of the intermediate complex and the enzyme remains covalently bound to DNA [47]. The uncompetitive inhibitors also were at the after and at the substrate-binding site or another site. The effect of the inhibitor on enzymatic defined the enzyme–substrate TOP1 attacks to the DNA substrate (the nucleophilic attack of His263). The second step is releasing the ataxia enzyme remains covalent competitive, or mixed, depending on the binding of the enzyme, to the enzyme–substrate complex (ES) or to the enzyme–substrate–inhibitor triple complex (ESI). The enzyme bind substrate) site. of activity can be analyzed using graphical representations of the Michaelis–Menten equation. We defined the type of inhibition for the four most effective inhibitors of TDP1: **7g**,**h** and **10a**,**b** (IC50~1.4–2.2 μM, Table 1) by obtaining the dependence of the reaction rate on the substrate concentration and dependence of Vmax and KM on the concentration of the steady-state reaction a fall in both the Vmax and KM values with an increasing concentration of the inhibitor, which is typical for the uncompetitive type of inhibition (Table S1, Figure S25) when the inhibitor binds the enzyme–substrate complex but does not bind free enzyme. Thus, it was determined that no studied inhibitors demonstrate the competitive mechanism of binds DNA and with a relatively large cleft that contains a mixed charge distribution for the TOP1 fragment binding [44–46]. It is known that TDP1 hydrolyzes the DNA–TOP1 adducts via two coordinated nucleophilic attacks of two His residues (His263 and His493) with the covalent intermediate complex formation [45,46]. The enzyme reaction catalyzed by TDP1 could be inhibited in two steps: the first one is the inhibition of the TDP1 binding to DNA TDP1 from the transition complex (nucleophilic attack by a water molecule activated by His493). Inhibition of the second step prevents the release of the DNA 3′-phosphate product from the DNA–TDP1 complex. Spinocerebellar ataxia with axonal neuropathy (SCAN1) is a neurological disease that is caused by a TDP1 His493Arg mutation. This second enzyme remains covalently bound to DNA [47]. The uncompetitive inhibitors also prevent the second step of the reaction, stabilizing the enzyme–DNA covalent complex, which can lead to the accumulation of the single-strand breaks in the cell. in concentration of and relatively mixed the TOP1 fragment binding [44–46]. It is known that TDP1 hydrolyzes the DNA–TOP1 by a enzyme also prevent the second step of the reaction, stabilizing the enzyme–DNA covalent complex, 1.4 + 0.2 >50 20 ± 3 27 ± 2 - + The reversible enzyme inhibitors graphical defined the four for inhibitor binds the enzyme–substrate complex but does not bind free enzyme. Thus, it TOP1 the coordinated of second step prevents covalent accumulation of 11.9 ± 0.4 ND ND ND - + a fall in both the Vmax and KM values with an increasing concentration of the inhibitor, TDP1 from the transition complex (nucleophilic attack by a water molecule activated by 16.9 <sup>±</sup> 2.4 ND ND ND + + complex (ES) or to the enzyme–substrate–inhibitor triple complex (ESI). The enzyme could bind the inhibitor at the different steps of catalysis (before or after the substrate) the the the of **7g**,**h** and **10a**,**b** (IC50~1.4–2.2 μM, Table 1) by obtaining the dependence of the reaction rate on the substrate concentration and dependence of Vmax and KM on the concentration of the inhibitor under steady-state reaction conditions. For all four TDP1 inhibitors, we observed type was determined that no studied inhibitors demonstrate the competitive mechanism of inhibition. TDP1 has a catalytic pocket with a narrow positively charged cleft where it DNA large via His with the covalent intermediate complex formation [45,46]. The enzyme reaction catalyzed by TDP1 could be inhibited in two steps: the first one is the inhibition of the TDP1 binding attack His263). molecule activated step product from the DNA–TDP1 complex. Spinocerebellar ataxia with axonal neuropathy (SCAN1) is a neurological disease that is caused by a TDP1 His493Arg mutation. This mutation prevents the second step of hydrolysis of the intermediate complex and the covalently The uncompetitive inhibitors also step the cell. 19.6 <sup>±</sup> 2.8 ND ND ND + + The reversible enzyme inhibitors were classified as competitive, uncompetitive, noncompetitive, or mixed, depending on the binding of the enzyme, to the enzyme–substrate complex (ES) or to the enzyme–substrate–inhibitor triple complex (ESI). The enzyme could bind the inhibitor at the different steps of catalysis (before or after the substrate) and at the substrate-binding site or another site. The effect of the inhibitor on enzymatic activity can be analyzed using graphical representations of the Michaelis–Menten equation. We defined the type of inhibition for the four most effective inhibitors of TDP1: **7g**,**h** and **10a**,**b** (IC50~1.4–2.2 µM, Table 1) by obtaining the dependence of the reaction rate on the substrate concentration and dependence of Vmax and K<sup>M</sup> on the concentration of the inhibitor under steady-state reaction conditions. For all four TDP1 inhibitors, we observed a fall in both the Vmax and K<sup>M</sup> values with an increasing concentration of the inhibitor, which is typical for the uncompetitive type of inhibition (Table S1, Figure S25) when the inhibitor binds the enzyme–substrate complex but does not bind free enzyme. Thus, it was determined that no studied inhibitors demonstrate the competitive mechanism of inhibition. TDP1 has a catalytic pocket with a narrow positively charged cleft where it binds DNA and with a relatively large cleft that contains a mixed charge distribution for the TOP1 fragment binding [44–46]. It is known that TDP1 hydrolyzes the DNA–TOP1 adducts via two coordinated nucleophilic attacks of two His residues (His263 and His493) with the covalent intermediate complex formation [45,46]. The enzyme reaction catalyzed by TDP1 could be inhibited in two steps: the first one is the inhibition of the TDP1 binding to the DNA substrate (the nucleophilic attack of His263). The second step is releasing TDP1 from the transition complex (nucleophilic attack by a water molecule activated by His493). Inhibition of the second step prevents the release of the DNA 30 -phosphate product from the DNA–TDP1 complex. Spinocerebellar ataxia with axonal neuropathy (SCAN1) is a neurological disease that is caused by a TDP1 His493Arg mutation. This mutation prevents the second step of hydrolysis of the intermediate complex and the enzyme remains covalently bound to DNA [47]. The uncompetitive inhibitors also prevent

prevent the second step of the reaction, stabilizing the enzyme–DNA covalent complex,

which can lead to the accumulation of the single-strand breaks in the cell.

the second step of the reaction, stabilizing the enzyme–DNA covalent complex, which can lead to the accumulation of the single-strand breaks in the cell. 2.2.3. The Effect of TDP1 Inhibitors on Cells Viability

#### 2.2.3. The Effect of TDP1 Inhibitors on Cells Viability The low intrinsic cytotoxicity of the additional drug component in the clinically used

*Int. J. Mol. Sci.* **2021**, *22*, x FOR PEER REVIEW 8 of 19

The low intrinsic cytotoxicity of the additional drug component in the clinically used anticancer cocktails with traditional drugs is very important for treatment. This approach in chemotherapy can reduce the potential resulting toxic load on the body in the case of a combined action of these inhibitors on the activity of several targets. The aim of this part of the study was to demonstrate the ability of the found TDP1 inhibitors to enhance the cytotoxicity of the clinically used camptothecin derivative Tpc on the cancer cells. First, we analyzed the intrinsic cytotoxicity for 10 of the most effective TDP1 inhibitors (IC<sup>50</sup> 1.4–6.6 µM, Table 1) on the HEK293A, HEK293FT, and HeLa cell lines by colorimetric tests. The effects of the tested compounds on cell survival are shown on Figure 4 and the CC<sup>50</sup> values for three cell lines are shown in Table 1. The intrinsic cytotoxicity of a number of the studied compounds is quite high, namely, they have close IC<sup>50</sup> values against the TDP1 and CC<sup>50</sup> values on one or several studied cell lines. Most of the compounds generally were more toxic for the HEK293FT and HeLa cell lines than for the HEK293A cell line (Table 1 and Figure 4). The cytotoxicity of the three leader compounds, **7g**, **10a**, and **7h**, was quite high. Interestingly, the sulfoxide analogs **10a**,**b** were less cytotoxic than their thioester analogs **7g**,**h**. Compound **10b** was one of the least toxic among the ten tested compounds. anticancer cocktails with traditional drugs is very important for treatment. This approach in chemotherapy can reduce the potential resulting toxic load on the body in the case of a combined action of these inhibitors on the activity of several targets. The aim of this part of the study was to demonstrate the ability of the found TDP1 inhibitors to enhance the cytotoxicity of the clinically used camptothecin derivative Tpc on the cancer cells. First, we analyzed the intrinsic cytotoxicity for 10 of the most effective TDP1 inhibitors (IC50 1.4– 6.6 μM, Table 1) on the HEK293A, HEK293FT, and HeLa cell lines by colorimetric tests. The effects of the tested compounds on cell survival are shown on Figure 4 and the CC50 values for three cell lines are shown in Table 1. The intrinsic cytotoxicity of a number of the studied compounds is quite high, namely, they have close IC50 values against the TDP1 and CC50 values on one or several studied cell lines. Most of the compounds generally were more toxic for the HEK293FT and HeLa cell lines than for the HEK293A cell line (Table 1 and Figure 4). The cytotoxicity of the three leader compounds, **7g**, **10a**, and **7h**, was quite high. Interestingly, the sulfoxide analogs **10a,b** were less cytotoxic than their thioester analogs **7g,h**. Compound **10b** was one of the least toxic among the ten tested compounds.

**Figure 4.** TDP1 inhibitors' intrinsic cytotoxicity on HEK293A, HEK293FT, and HeLa cells, and the dose-dependent action of the most effective compounds. The values were normalized on the control wells with the cells in 1% DMSO. **Figure 4.** TDP1 inhibitors' intrinsic cytotoxicity on HEK293A, HEK293FT, and HeLa cells, and the dose-dependent action of the most effective compounds. The values were normalized on the control wells with the cells in 1% DMSO.

#### 2.2.4. PARP1 and PARP2 Activity in the Presence of TDP1 Inhibitors 2.2.4. PARP1 and PARP2 Activity in the Presence of TDP1 Inhibitors

The next step of our study was to test the inhibitors for their ability to inhibit PARP1 and PARP2. UA and the 14 synthesized compounds were tested for the suppression of the ability of PARP1 and PARP2 to include a radioactive label into the acid-insoluble reaction product, poly(ADP-ribose) (PAR). None of the tested compounds inhibited PARP2 in the applied reaction conditions. All the results obtained for PARP1 inhibition are shown in Table 1. Eight compounds exhibited a weak inhibitory activity against PARP1 (40–80% of residual activity) at a 1 mM concentration, working at significantly higher concentration ranges than against TDP1. The best effect we observed was for **9**; the residual activity was 40 ± 5% under the reaction conditions (0.5 mM concentration of the compound), indicated in Materials and Methods. For PARP1, compounds with smaller substituents, an aliphatic or benzene moiety, had a higher inhibitory effect. The next step of our study was to test the inhibitors for their ability to inhibit PARP1 and PARP2. UA and the 14 synthesized compounds were tested for the suppression of the ability of PARP1 and PARP2 to include a radioactive label into the acid-insoluble reaction product, poly(ADP-ribose) (PAR). None of the tested compounds inhibited PARP2 in the applied reaction conditions. All the results obtained for PARP1 inhibition are shown in Table 1. Eight compounds exhibited a weak inhibitory activity against PARP1 (40–80% of residual activity) at a 1 mM concentration, working at significantly higher concentration ranges than against TDP1. The best effect we observed was for **9**; the residual activity was 40 ± 5% under the reaction conditions (0.5 mM concentration of the compound), indicated in Materials and Methods. For PARP1, compounds with smaller substituents, an aliphatic or benzene moiety, had a higher inhibitory effect.

#### 2.2.5. TDP2 Activity in the Presence of TDP1 Inhibitors 2.2.5. TDP2 Activity in the Presence of TDP1 Inhibitors

To study the ability of the investigated compounds to inhibit TDP2 activity, we separated the TDP2 reaction products under denaturing conditions in polyacrylamide gel (PAGE). We tested the ability of TDP2 to eliminate tyrosine residues from the 5' end of the oligonucleotide substrate in the absence and presence of TDP1 inhibitors. UA did not inhibit TDP2 (Table 1, Figure S26). All 14 newly synthesized UA derivatives inhibited TDP2 at a 1 μM concentration, but almost did not inhibit its activity at a 0.200 mM To study the ability of the investigated compounds to inhibit TDP2 activity, we separated the TDP2 reaction products under denaturing conditions in polyacrylamide gel (PAGE). We tested the ability of TDP2 to eliminate tyrosine residues from the 5' end of the oligonucleotide substrate in the absence and presence of TDP1 inhibitors. UA did not inhibit TDP2 (Table 1, Figure S26). All 14 newly synthesized UA derivatives inhibited TDP2 at a 1 µM concentration, but almost did not inhibit its activity at a 0.200 mM concentration.

concentration. The residual TDP2 activity for the best four TDP1 inhibitors, **7g**,**h** and

The residual TDP2 activity for the best four TDP1 inhibitors, **7g**,**h** and **10a**,**b**, was from 30 to 60%, depending on the tested compound, at a 1 mM concentration of the inhibitor. We tested the ability of compounds **7g**,**h** and **10a**,**b** to inhibit TDP2 activity at 0.2 mM and 0.5 mM concentrations (Figure S26). All four tested compounds inhibited TDP2 insignificantly, at higher concentration ranges than TDP1. Interestingly, the best TDP1 inhibitors, **7g**,**h** and **10a**, exhibited the highest effect on TDP2. **10a**,**b**, was from 30 to 60%, depending on the tested compound, at a 1 mM concentration of the inhibitor. We tested the ability of compounds **7g**,**h** and **10a**,**b** to inhibit TDP2 activity at 0.2 mM and 0.5 mM concentrations (Figure S26). All four tested compounds inhibited TDP2 insignificantly, at higher concentration ranges than TDP1. Interestingly, the best TDP1 inhibitors, **7g**,**h** and **10a**, exhibited the highest effect on TDP2. 2.2.6. The Effect of TDP1 Inhibitors on Cell Viability in Combination with Topotecan

#### 2.2.6. The Effect of TDP1 Inhibitors on Cell Viability in Combination with Topotecan TDP1 is able to cleave the TOP1–DNA complex, thus preventing the action of Tpc

TDP1 is able to cleave the TOP1–DNA complex, thus preventing the action of Tpc and decreasing the efficiency of this clinically used drug. The inhibition of TDP1 activity can increase the efficiency of Tpc. Next, we checked the cytotoxic effect of the combination of Tpc and examined the TDP1 inhibitors compared to Tpc alone on HeLa cells and HEK293 cells. and decreasing the efficiency of this clinically used drug. The inhibition of TDP1 activity can increase the efficiency of Tpc. Next, we checked the cytotoxic effect of the combination of Tpc and examined the TDP1 inhibitors compared to Tpc alone on HeLa cells and HEK293 cells. Only one compound, **10a**, of the most effective TDP1 inhibitors (IC50 2.1 μM, Table 1)

Only one compound, **10a**, of the most effective TDP1 inhibitors (IC<sup>50</sup> 2.1 µM, Table 1) showed promising synergy on HeLa cells in conjunction with Tpc. We observed the suppressed cell growth in the joint presence of the TDP1 inhibitor and Tpc on HeLa cells compared to Tpc alone (Figure 5). We observed the synergistic effect both when we titrated **10a** with addition of fixed Tpc (1 µM or 2 µM) and vice versa when we titrated Tpc with addition of fixed **10a** (5 µM). The CC<sup>50</sup> value of **10a** decreased from 15 µM (only **10a**) to 3 µM (**10a** with 2 µM Tpc). The CC<sup>50</sup> value of Tpc decreased from 2 µM (only Tpc) to 0.7 µM (Tpc with 5 µM **10a**). showed promising synergy on HeLa cells in conjunction with Tpc. We observed the suppressed cell growth in the joint presence of the TDP1 inhibitor and Tpc on HeLa cells compared to Tpc alone (Figure 5). We observed the synergistic effect both when we titrated **10a** with addition of fixed Tpc (1 μM or 2 μM) and vice versa when we titrated Tpc with addition of fixed **10a** (5 μM). The CC50 value of **10a** decreased from 15 μM (only **10a**) to 3 μM (**10a** with 2 μM Tpc). The CC50 value of Tpc decreased from 2 μM (only Tpc) to 0.7 μM (Tpc with 5 μM **10a**).

**Figure 5.** Left panel: The influence of Tpc at 1 μM and 2 μM concentrations on **10a** cytotoxicity. B: The influence of the **10a** derivative at a 5 μM concentration on Table 1. DMSO, for the red curve on the cells treated with 1 μM topotecan and 1% DMSO, and for the blue curve on the cells treated with 2 μM topotecan and 1% DMSO, in order to exclude the effect of topotecan itself and see the effect of the drug combination. Right panel: cells treated with 1% DMSO (black line) or 5 μM **10a** (red line) were taken as 100% in order to evaluate the effect of the combination rather than the individual substance **10a**. **Figure 5.** Left panel: The influence of Tpc at 1 µM and 2 µM concentrations on **10a** cytotoxicity. B: The influence of the **10a** derivative at a 5 µM concentration on Table 1. DMSO, for the red curve on the cells treated with 1 µM topotecan and 1% DMSO, and for the blue curve on the cells treated with 2 µM topotecan and 1% DMSO, in order to exclude the effect of topotecan itself and see the effect of the drug combination. Right panel: cells treated with 1% DMSO (black line) or 5 µM **10a** (red line) were taken as 100% in order to evaluate the effect of the combination rather than the individual substance **10a**.

In our previous works, we checked the cytotoxic effect of Tpc and the TDP1 inhibitors—monoterpene 3-carene-derived compounds [48,49] and UA combined with monoterpenoids [49], measured separately and jointly with Tpc—using a panel of HEK293FT and HEK293A *Tdp1* knockout isogenic clones. We showed that *Tdp1* knockout cells were more sensitive to Tpc compared to WT cells. The data on the HEK293FT mutants were of low reproducibility [48], which is why we decided to change the basic cell line to HEK293A [49]. In this work, we created new *PARP1*-/- HEK293A cells using the CRISPR–Cas9 approach (Figure S27). The sensitivity of the *PARP1*-/- cell line to Tpc was lower than WT HEK293A cells (CC50 50 ± 5 nM and 27 ± 4 nM respectively, Figure S28). We checked the intrinsic cytotoxicity of the leader compound (**10a**) on the HEK293 wild type and *TDP1*-/- and *PARP1*-/- cell lines using a colorimetric test (Figure 6). The cytotoxicity of **10a** was nearly the same for all the cell lines, with a CC50 of 15–20 μM In our previous works, we checked the cytotoxic effect of Tpc and the TDP1 inhibitors —monoterpene 3-carene-derived compounds [48,49] and UA combined with monoterpenoids [49], measured separately and jointly with Tpc—using a panel of HEK293FT and HEK293A *Tdp1* knockout isogenic clones. We showed that *Tdp1* knockout cells were more sensitive to Tpc compared to WT cells. The data on the HEK293FT mutants were of low reproducibility [48], which is why we decided to change the basic cell line to HEK293A [49]. In this work, we created new *PARP1*-/- HEK293A cells using the CRISPR–Cas9 approach (Figure S27). The sensitivity of the *PARP1*-/- cell line to Tpc was lower than WT HEK293A cells (CC<sup>50</sup> 50 ± 5 nM and 27 ± 4 nM respectively, Figure S28). We checked the intrinsic cytotoxicity of the leader compound (**10a**) on the HEK293 wild type and *TDP1*-/- and *PARP1*-/- cell lines using a colorimetric test (Figure 6). The cytotoxicity of **10a** was nearly the same for all the cell lines, with a CC<sup>50</sup> of 15–20 µM (Figure 6). We did not observe any effect of the joint presence of **10a** and Tpc on any of these cells (data not shown). There was no difference between the WT and mutant cells.

**Figure 6.** Intrinsic cytotoxicity of **10a** on HEK293FT and HEK293A cells, as well as on the Tdp1 and *PARP1* knockout HEK293A cell lines (HEK293A *Tdp1*-/- and HEK293A *PARP1*-/-); the dosedependent action. The values were normalized to the control wells with the cells in 1% DMSO. **Figure 6.** Intrinsic cytotoxicity of **10a** on HEK293FT and HEK293A cells, as well as on the Tdp1 and *PARP1* knockout HEK293A cell lines (HEK293A *Tdp1*-/- and HEK293A *PARP1*-/-); the dosedependent action. The values were normalized to the control wells with the cells in 1% DMSO.

(Figure 6). We did not observe any effect of the joint presence of **10a** and Tpc on any of these cells (data not shown). There was no difference between the WT and mutant cells.

#### **3. Materials and Methods 3. Materials and Methods**

#### *3.1. Chemistry 3.1. Chemistry*

The analytical and spectral studies were conducted in the Chemical Service Center for the collective use of the Siberian Branch of the Russian Academy of Sciences. PMR and 13C NMR spectra were recorded in CDCl3 or DMSO-d6 using solvent resonances (1H 7.24 ppm, 13C 76.90 ppm, and 1H 2.50 ppm, 13C 39.50, respectively) as the standards on a Bruker AV-400 spectrometer (Bruker Corporation, Germany; operating frequencies 400.13 MHz for 1H and 100.61, for 13C). Mass spectra (ionizing-electron energy 70 eV) were measured with a DFS Thermo Scientific high-resolution mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Macherey–Nagel silica gel (63–200 μ) was used for the column chromatography. Thin-layer chromatography was performed on TLC Silica gel The analytical and spectral studies were conducted in the Chemical Service Center for the collective use of the Siberian Branch of the Russian Academy of Sciences. PMR and <sup>13</sup>C NMR spectra were recorded in CDCl<sup>3</sup> or DMSO-d<sup>6</sup> using solvent resonances ( <sup>1</sup>H 7.24 ppm, <sup>13</sup>C 76.90 ppm, and <sup>1</sup>H 2.50 ppm, <sup>13</sup>C 39.50, respectively) as the standards on a Bruker AV-400 spectrometer (Bruker Corporation, Germany; operating frequencies 400.13 MHz for <sup>1</sup>H and 100.61, for <sup>13</sup>C). Mass spectra (ionizing-electron energy 70 eV) were measured with a DFS Thermo Scientific high-resolution mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). Macherey–Nagel silica gel (63–200 µ) was used for the column chromatography. Thin-layer chromatography was performed on TLC Silica gel 60 plates (UV-254, Merck, Darmstadt, Germany) (Figures S1–S24).

60 plates (UV-254, Merck, Darmstadt, Germany) (Figures S1-S24). (+)-UA {**1**, [α]D +478° (*c* 0.1, CHCl3)} was isolated from a mixture of *Usnea* lichen species using the literature method [50]. Bromousnic acid **2** was synthesized according to the literature method [51]. Usnic acid derivative **9** was synthesized according to the (+)-UA {**1**, [α]<sup>D</sup> +478◦ (*c* 0.1, CHCl3)} was isolated from a mixture of *Usnea* lichen species using the literature method [50]. Bromousnic acid **2** was synthesized according to the literature method [51]. Usnic acid derivative **9** was synthesized according to the literature [35].

literature [35]. Synthetic starting materials, reagents, and solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA), Acros Organics (Geel, Belgium), and AlfaAesar (Heysham, UK) (95–99% pure). All chemicals were used as described unless otherwise noted. Reagent-grade solvents were redistilled prior to use. Synthetic starting materials, reagents, and solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA), Acros Organics (Geel, Belgium), and AlfaAesar (Heysham, UK) (95–99% pure). All chemicals were used as described unless otherwise noted. Reagentgrade solvents were redistilled prior to use.

### 3.1.1. Reaction of 2 with Thiols (General Method)

3.1.1. Reaction of 2 with thiols (general method). A weighed portion of KOH (1.1 mmol), MeOH (6 mL) and the appropriate thiol (1.1 mmol) were placed into a flask, stirred at room temperature for 10–15 min, treated with a solution of **2** (1 mmol) in CH2Cl2 (2 mL), stirred at room temperature for 2–3 h until the reaction was finished (TLC monitoring), washed with distilled H2O (two times the volume), dried over MgSO4, and concentrated. If necessary, the solid was A weighed portion of KOH (1.1 mmol), MeOH (6 mL) and the appropriate thiol (1.1 mmol) were placed into a flask, stirred at room temperature for 10–15 min, treated with a solution of **2** (1 mmol) in CH2Cl<sup>2</sup> (2 mL), stirred at room temperature for 2–3 h until the reaction was finished (TLC monitoring), washed with distilled H2O (two times the volume), dried over MgSO4, and concentrated. If necessary, the solid was chromatographed over silica gel using a CH2Cl<sup>2</sup> eluent.

chromatographed over silica gel using a CH2Cl2 eluent. *(2R)-4-acetyl-10-{2-[(dimethylcarbamothioyl)sulfanyl]acetyl}-5,11,13-trihydroxy-2,12- (2R)-4-acetyl-10-{2-[(dimethylcarbamothioyl)sulfanyl]acetyl}-5,11,13-trihydroxy-2,12-dimethyl-8-oxatricyclo [7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 7a*

*dimethyl-8-oxatricyclo [7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 7a*  Yellow amorphous powder, yield 94%. 1H NMR (CDCl3, δ) 1.77 (3H, s), 2.09 (3H, s), 2.65 (3H, s), 3.46 (3H, s), 3.54 (3H, s), 4.92 (2H, m), 6.02 (1H, s), 11.11 (OH, s), 12.81 (OH, Yellow amorphous powder, yield 94%. <sup>1</sup>H NMR (CDCl3, δ) 1.77 (3H, s), 2.09 (3H, s), 2.65 (3H, s), 3.46 (3H, s), 3.54 (3H, s), 4.92 (2H, m), 6.02 (1H, s), 11.11 (OH, s), 12.81 (OH, s), 18.84 (OH, s). <sup>13</sup>C NMR (CDCl3, δ): 7.4, 27.7, 31.9, 41.5, 45.6, 47.9, 58.8, 98.5, 100.9, 104.1, 105.0, 109.3, 154.7, 157.7, 163.5, 178.8, 191.4, 194.5, 195.2, 197.8, 201.6. HRMS m/z 463.0750 (calcd for C21H21O7N<sup>2</sup> <sup>32</sup>S2, 463.0754).

*(2R)-4-acetyl-10-[2-(4,5-dihydro-1,3-thiazol-2-ylsulfanyl)acetyl]-5,11,13-trihydroxy-2,12-dimethyl-8-oxatricyclo[7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 7b*

Yellow amorphous powder, yield 49%. <sup>1</sup>H NMR (CDCl3, δ) 1.73 (3H, s), 2.06 (3H, s), 2.63 (3H, s), 3.41 (2H, bt), 4.15 (2H, bt), 4.43 (2H, m), 5.99 (1H, s), 11.09 (OH, s), 12.71 (OH, s), 18.81 (OH, s). <sup>13</sup>C NMR (CDCl3, δ): 7.5, 27.8, 31.9, 43.0, 58.8, 63.7, 75.7, 98.7, 100.2, 104.1, 105.1, 109.5, 154.6, 157.9, 163.7, 167.7, 178.7, 191.5, 194.2, 197.8, 201.7. HRMS m/z 461.0592 (calcd for C21H19O7N32S2, 461.0598).

*(2R)-4-acetyl-10-{2-[(5-amino-1,3,4-thiadiazol-2-yl)sulfanyl]acetyl}-5,11,13-trihydroxy-2,12 dimethyl-8-oxatricyclo[7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 7c*

Yellow amorphous powder, yield 81% <sup>1</sup>H NMR (DMSO-d6, δ) 1.71 (3H, s), 1.99 (3H, s), 2.56 (3H, s), 4.62 (2H, s), 6.89 (1H, s), 7.32 (NH2, s), 11.56 (OH, bs), 12.75 (OH, s). <sup>13</sup>C NMR (DMSO-d6, δ): 7.6, 27.9, 31.5, 44.2, 58.3, 98.7, 100.0, 105.1, 105.6, 107.5, 107.6, 148.7, 154.8, 157.2, 157.4, 162.3, 162.6, 170.2, 178.3, 190.9, 195.4, 197.4, 201.0. HRMS m/z 475.0507 (calcd for C20H17O7N<sup>3</sup> <sup>32</sup>S2, 475.0502).

*(2R)-4-acetyl-10-[2-(1H-1,3-benzodiazol-2-ylsulfanyl)acetyl]-5,11,13-trihydroxy-2,12-dimethyl-8-oxatricyclo[7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 7f*

Yellow amorphous powder, yield 54%. <sup>1</sup>H NMR (CDCl3, δ) 1.70 (3H, s), 2.07 (3H, s), 2.64 (3H, s), 4.67 (2H, m), 5.94 (1H, s), 7.16 (2H, m), 7.47 (2H, m), 11.18 (OH, s), 12.64 (OH, bs), 18.78 (OH, bs). <sup>13</sup>C NMR (CDCl3, δ): 7.3, 27.7, 31.7, 42.9, 58.5, 98.6, 100.0, 104.1, 104.9, 109.1, 113.8, 122.1, 138.4, 148.5, 154.5, 157.9, 163.3, 178.3, 191.3, 194.5, 197.6, 201.5. HRMS m/z 492.0981 (calcd for C25H20O7N<sup>2</sup> <sup>32</sup>S, 492.0986).

*(2R)-4-acetyl-5,11,13-trihydroxy-2,12-dimethyl-10-[2-(pyridin-2-ylsulfanyl)acetyl]-8-oxatricyclo[7.4.0*.*0 2,7]trideca-1(13),4,6,9,11-pentaen-3-one 7i*

Yellow amorphous powder, yield 90% <sup>1</sup>H NMR (CDCl3, δ) 1.76 (3H, s), 2.09 (3H, s), 2.65 (3H, s), 4.70 (2H, m), 5.92 (1H, s), 6.98 (1H, dd, J<sup>1</sup> = 7.3 Hz, J<sup>2</sup> = 5.0 Hz), 7.26 (1H, d, J = 7.3 Hz), 7.49 (1H, dt, J<sup>1</sup> = 7.3 Hz, J<sup>2</sup> = 1.5 Hz), 8.33 (1H, d, J = 5.0 Hz), 11.09 (OH, s), 12.86 (OH, s), 18.83 (OH, s). <sup>13</sup>C NMR (CDCl3, δ): 7.5, 27.8, 32.0, 40.3, 58.9, 98.5, 100.5, 104.1, 105.1, 109.5, 119.8, 122.0, 136.1, 149.3, 154.7, 156.8, 157.7, 163.8, 179.0, 191.6, 195.9, 197.9, 201.7. HRMS m/z 453.0874 (calcd for C23H19O7N32S, 453.0877).

*(2R)-4-acetyl-5,11,13-trihydroxy-2,12-dimethyl-10-[2-(pyrimidin-2-ylsulfanyl)acetyl]-8-oxatricyclo[7.4.0*.*0 2,7]trideca-1(13),4,6,9,11-pentaen-3-one 7j*

Yellow amorphous powder, yield 92% <sup>1</sup>H NMR (CDCl3, δ) 1.77 (3H, s), 2.09 (3H, s), 2.65 (3H, s), 4.65 (2H, m), 5.98 (1H, s), 6.96 (1H, t, J = 6.4 Hz), 8.45 (2H, d, J = 6.4 Hz), 11.09 (OH, s), 12.82 (OH, s), 18.83 (OH, s). <sup>13</sup>C NMR (CDCl3, δ): 7.5, 27.8, 32.0, 41.4, 58.9, 98.5, 100.5, 104.1, 105.1, 109.5, 116.77, 154.7, 157.2, 157.4, 163.7, 170.7, 178.9, 191.5, 195.2, 197.8, 201.7. HRMS m/z 454.0823 (calcd for C22H18O7N<sup>2</sup> <sup>32</sup>S, 454.0829).

#### 3.1.2. Hydrolysis of Compound **8**

Compound **8** (1 mmol) was dissolved in glacial acetic acid (15 mL). The sulfuric acid water solution (30%, 0.5 mL) was added and the mixture was stirred for 4 h. After that, the resulted mixture was treated with water and the precipitate that formed was filtered off and dried in air. The solid was chromatographed over silica gel using a CH2Cl<sup>2</sup> eluent with an MeOH gradient.

*(2R)-4-acetyl-10-[2-(carbamoylsulfanyl)acetyl]-5,11,13-trihydroxy-2,12-dimethyl-8-oxatricyclo[7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 8*

Yellow amorphous powder, yield 38%. <sup>1</sup>H NMR (CDCl3, δ) 1.74 (3H, s), 2.07 (3H, s), 2.64 (3H, s), 4.43 (2H, m), 5.80 (NH2, bs) 6.00 (1H, s), 11.12 (OH, s), 12.76 (OH, s), 18.82 (OH, s). <sup>13</sup>C NMR (CDCl3, δ): 7.5, 27.8, 31.9, 40.8, 58.8, 98.7, 100.2, 104.2, 105.1, 109.5, 154.7, 158.0, 163.7, 167.7, 178.7, 191.5, 194.9, 197.8, 201.7. HRMS m/z 419.0671 (calcd for C19H17O8N32S, 419.0669).

## 3.1.3. Oxidation of Compounds **7g**,**h**

Thioether **7g** or **7h** (1 mmol) was dissolved in methylene chloride (5 mL) on an ice bath. Meta-chloroperoxybenzoic acid (3 mmol) was added and the resulted solution was stirred for 30 min at 0 ◦C. After that, the mixture was treated with a saturated sodium sulfite solution (3 mL) and the resulted mixture was stirred for 1 h at room temperature. Finally, the mixture was extracted with methylene chloride, dried over magnesium sulfate, and evaporated.

*(2R)-4-acetyl-10-[2-(1,3-benzothiazole-2-sulfinyl)acetyl]-5,11,13-trihydroxy-2,12-dimethyl-8 oxatricyclo[7.4.0.02,7]trideca-1(13),4,6,9,11-pentaen-3-one 10a*

Yellow amorphous powder, yield 75%. <sup>1</sup>H NMR (CDCl3, δ) 1.64 and 1.69 (3H, s), 2.06 and 2.09 (3H, s), 2.63 and 2.63 (3H, s), 4.72 and 5.01 (2H, m), 5.93 and 5.98 (1H, s), 7.46 (2H, m), 8.00 (2H, m), 11.18 and 11.21 (OH, s), 12.52 and 12.59 (OH, s), 18.78 (OH, bs). <sup>13</sup>C NMR (CDCl3, δ): 7.4 and 7.4, 27.7, 31.7 and 31.8, 58.5 and 58.6, 68.0 and 68.6, 99.0, 101.3 and 101.4, 104.4 and 104.5, 105.0 and 105.1, 109.5 and 109.6, 122.1 and 122.2, 123.9 and 124.0, 126.4 and 127.0, 136.0 and 136.1, 153.5, 154.5 and 154.5, 158.8 and 158.9, 163.8 and 164.0, 176.1 and 176.2, 178.1, 190.4 and 190.6, 191.4, 197.6, 201.6 and 201.7.

#### *3.2. Biology*

## 3.2.1. Preparation of Labeled Oligonucleotides

A single-stranded oligodeoxynucleotide 5'-FAM-AAC GTC AGG GTC TTC C-BHQ-1-30 containing a 6-carboxamido fluorescein (6-FAM) fluorophore at the 50 -end and a Black Hole Quencher 1 (BHQ-1) at the 30 -end was employed as an internally-quenched probe for real-time detection of TDP1 activity [43]. Another single-stranded oligonucleotide 5 0 -Tyr-AAC GTC AGG GTC TTC C-FAM-30 containing a 6-FAM label at the 30 -end and an Ltyrosine residue attached via the phenolic OH group to the 50 -terminal phosphate was used as a substrate for detection of TDP2 activity (Tyr = HOOC-CH(NH2)-CH2-C6H4-*p*-O~).

Oligonucleotides were synthesized essentially as described in a previous paper [33]. Briefly, both oligonucleotides were assembled on a Biosset ASM-800 automated DNA/RNA synthesizer (Novosibirsk, Russia) on 200 nmol scale by β-cyanoethyl phosphoramidite chemistry. The doubly labeled probe was prepared from standard 50 -DMTr-deoxynucleoside 3 0 -phosphoramidites, 6-FAM phosphoramidite for 50 -labeling, and a BHQ-1 CPG support for attachment of a Black Hole QuencherTM BHQ-1 (all from Glen Research, Sterling, VA, USA), deprotected and purified under standard conditions as described above [33].

Synthesis of a 50 -tyrosinyl oligodeoxynucleotide substrate 50 -Tyr-AAC GTC AGG GTC TTC C-FAM-3' was carried out on a specially prepared CPG 500 Å support containing *N*α-Fmoc-L-tyrosine with an unprotected phenol group esterified via the carboxy group to the 5'-OH group of a thymidine residue linked to the support through the 30 -OH by a succinate linker. Synthesis of the tyrosine support was accomplished as described previously [33]. In contradistinction to the previous paper, in this work the solid-phase DNA assembly was carried out with 'reversed' 30 -DMTr-deoxynucleoside 50 -phosphoramidites (ChemGenes, Wilmington, MA, USA) in the 50→3 0 direction using a free phenolic OH of the supportbound tyrosine as an anchoring point to introduce the tyrosine residue onto the 50 -end. 6-FAM phosphoramidite (Glen Research, Sterling, VA, USA) was coupled at the last cycle of the synthesis to place the label at the 30 -end. The oligonucleotide with a phosphotyrosine residue was cleaved from support by alkaline hydrolysis of the ester linkages (100 µL of 0.1 M NaOH, 1 h at 25 ◦C). Excess NaOH was quenched by ca. 50 µL of Amberlyst cation exchange resin beads in NH<sup>4</sup> + form. The solid support was discarded, the supernatant and aqueous washings (2 × 100 µL) were combined, evaporated to dryness, and treated with 200 µL of a conc. (ca. 28%) aqueous NH<sup>3</sup> solution at 55 ◦C for 16 h to remove all the remaining *N*-protecting groups from oligonucleotide. The DNA substrate was isolated, purified, and analyzed as described previously [33]: [M-H]− calc. 5666.83, found 5666.90.

Human recombinant tyrosyl-DNA phosphodiesterase 1 (TDP1) and human recombinant poly(ADP-ribose)-polymerase1 (PARP1) were expressed in the *E. coli* system and purified as described [52,53].
