Effective natural inhibitors targeting LSD1 by computational study

doi:10.21203/rs.3.rs-2618547/v1

Objective: This study aimed to screen lead compounds and drug candidates with an inhibitory effect on the function of LSD1 from the ZINC database.

Methods: We used computer-aided virtual technology to screen some agents that inhibit the function of LSD1. Initially, LibDock screened out some optimal compounds for docking with LSD1. These candidate compounds were subjected to ADME analysis (adsorption, distribution, metabolism, and excretion) and toxicity metrics. Molecular docking can determine the binding affinity between LSD1 and the ligand, and lastly, we applied molecular dynamics simulations to calculate the docking of the ligand-receptor complex.

Results: Two natural compounds, ZINC000001651126 and ZINC000000001083, found in the ZINC database, are potent inhibitors of LSD1. When ZINC000001651126 and ZINC000000001083 bind to LSD1, they show high binding affinity. They are not hepatotoxic and have a high tolerance to cytochrome P4502D6. In addition, ZINC000001651126 and ZINC000000001083 have less developmental toxicity potential, rodent carcinogenicity, and Ames mutagenicity.

Conclusions: ZINC000001651126 and ZINC000000001083 can be considered safe and ideal drug candidates for LSD1 inhibitors. This study can provide new ideas for future research and the application of LSD1 inhibitors.

Biological sciences/Cancer/Bone cancer

Biological sciences/Drug discovery/Drug screening

LSD1

Ewing sarcoma

virtual screening

drug candidate

computer-aided technology

Ewing sarcoma is an aggressive primary bone tumor frequently identified in children and adolescents [1]. It has the second-highest incidence of all primary bone tumors, accounting for 6 to 8 percent [2]. Ewing sarcoma metastasizes quickly and has a high degree of malignancy [3]. Traditional therapy methods for Ewing sarcoma, such as surgery, radiotherapy, and chemotherapy, are hard to reach notably satisfactory results [4]. The five-year survival rate for patients is less than 10 percent, and most patients are prone to dying within two years [5]. We should focus on finding interesting therapeutic strategies for Ewing sarcoma.

In Ewing sarcoma, there are pathological reciprocal translocations and gene fusions between chromosomes 11 and 22, which subsequently encode EWS-FLI1, the most common fusion protein of the oncoprotein EWS-ETS family [1]. The fusion protein can play a role similar to a transcription factor to regulate the downstream target gene circuits of Ewing sarcoma and drive tumorigenesis [6]. LSD1 (lysine-specific demethylase 1), an epigenetic regulator of gene transcription [7], is highly expressed in Ewing sarcoma cell lines and tumors, and elevated expression levels correlate with decreased overall survival [4]. LSD1 is a critical functional partner of the fusion gene EWS-FLI1 [8]. LSD1 and its downstream targets are involved in various biological processes such as tumor cell growth and, metastasis, embryonic development and play a crucial part in the occurrence and development of Ewing sarcoma [9]. It has been reported that the overexpression of LSD1 in Ewing sarcoma can decrease or inactivate the function of LSD1 to inhibit the proliferation and development of tumor cells. LSD1 inhibitor can alter the transcriptional activity of the fusion gene EWS-FLI and ruin the function of the encoded protein EWS-ETS [1, 8]. This indicates that the interruption of the EWS-FLI function by LSD1 inhibitors is an attractive therapeutic strategy for Ewing sarcoma. Numerous LSD1 inhibitors, such as SP-2509 and HCI2509, are undergoing clinical evaluation for tumor therapy [10]. Medication to stop mitochondrial dysfunction is called SP-2509. Its usage in conjunction with LSD1 could serve as a model for managing mitochondria-related cancers. Nonetheless, it is not specific and potent enough in its ability to inhibit LSD1 [11]. HCI2509 can markedly alter the up-and down-regulated transcriptional profiles of EWS/FLI and EWS/ERG with single-agent efficacy in a transplanted model of Ewing sarcoma. Nonetheless, its drug safety and tolerability still require continuous optimization of dosage in the future [12]. LSD1 inhibitors have both efficacy and limitations as potential therapeutic candidates and need to be screened before they can be generally used in the clinical setting of Ewing sarcoma.

This experiment conforms to the current trend of researching LSD1 inhibitors, explores the mechanism of action of LSD1 inhibitors in treating Ewing sarcoma, and discovers effective targeted therapy drugs for this purpose. Natural products can function as lead compounds, and then the product is converted and transformed into a new drug, also becoming a new drug source for the pharmaceutical industry [13]. This research will utilize various chemical methods and structural biology (such as molecular docking and virtual screening.) in the screening and identification of compounds capable of inhibiting LSD1 function. At the same time, we will further study the metabolism, absorption level, distribution characteristics, and side effects of these compounds, list the candidate compounds and their pharmacological properties and provide directional guidance for developing LSD1 inhibitors.

Virtual filtering of ZINC15 database for LSD1

Virtual screening of LSD1 must be applied to named ligands in natural databases with biological properties and functions for sale. First, we selected 17,931 ligands from the ZINC15 database and then created a reference ligand based on the chemical structure of LSD1. The created ligands were docked with 17931 ligands in LibDock, and it was discovered that 2280 ligands could be successfully linked to LSD1. Polar hotspots and apolar hotspots were formed after 2280 ligands were libdocked with LSD1. We choose the best pose of the hot spot for scoring and get the LibDock score. Tranylcypromine is an effective LSD1 inhibitor that can bind to LSD1 [14]. It can inhibit LSD1 from altering the transcriptional activity of the EWS/FLI signaling pathway and indirectly achieve the effect of treating Ewing sarcoma. Ultimately, we selected the top 30 compounds in the LibDock score for the next steps (Table 1).

ADME properties of ligands

According to the definition of water solubility, 13 compounds have good solubility in water. ZINC000004654957, ZINC000014658375, ZINC000008689957, ZINC000001721178, ZINC000002585906, ZINC000002563374 all have high blood-brain barrier permeability. Tranylcypromine and 26 compounds were identified as non-inhibitors of CYP2D6 with significant effects on drug metabolism. For liver toxicity, 20 compounds and Tranylcypromine were found to be nontoxic. Lastly, we also discovered that six compounds showed strong absorption when bound to plasma proteins, and 20 compounds had terrific absorption levels in the human gut (Table 2).

After that, we must consider the safety of the compounds. The TOPKAT module can detect the toxicity of the top 30 compounds and Tranylcypromine, such as potential features of developmental toxicity, carcinogenicity in rodents, and mutagenicity of Ames (Table 3). After the operation of TOPKAT, it can be seen that 20 compounds are non-mutant, and ten compounds have no developmental toxicity. The high rodent carcinogenicity of Tranylcypromine is only expressed in female rats but not in mice and male rats. Combined with the above results, the hepatotoxicity and Ames mutagenicity of ZINC000001651126 and ZINC000000001083 are at low levels, and they can be considered safe drugs. ZINC000001651126, ZINC000000001083, and Tranylcypromine can be put into a more in-depth study based on their similar chemical structures (Figure 2).

Ligand-binding analysis

We found that LSD1 could be used to explore the ligand-binding mechanism of candidate compounds. In the CDOCKER module, we attach ZINC000001651126, ZINC000000001083, and Tranylcypromine to the molecular structure of LSD1 one by one and start the computation of CDOCKER potential energy (Figure 3). The CDOCKER potential of Tranylcypromine was higher than that of ZINC000001651126 and ZINC000000001083 (Table 4), and compared with Tranylcypromine, LSD1 could bind to ZINC000001651126 and ZINC000000001083 with higher affinity. Hydrogen bonding and π-related interactions can be calculated and can be found in the structures of the compounds (Figures 3 and 4), ZINC000001651126 and ZINC000000001083, both showing several bond acceptor and donor atoms. They form several pairs of π-related interactions and hydrogen bonds with LSD1 in the complex (Tables 5 and 6).

Molecular dynamics simulations

We explore molecular dynamics simulations when considering how to evaluate the stability of ligand-LSD1 in the natural environment. We can use the CDOCKER module in molecular docking experiments to retrieve the original conformation. The potential energy and RMSD images of each complex are also listed in Figure 5. The traces of ZINC000001651126 and ZINC000000001083 reach equilibrium after 90 ps. The potential energies and RMSDs of these two complexes stabilized over time. Briefly, both ZINC000001651126 and ZINC000000001083 can interact with LSD1 to form complexes that can occur stably in natural environments that alter LSD1.

ZINC000001651126 and ZINC000000001083 inhibited the proliferation of Ewing sarcoma cells

In vitro, we examined the impact of ZINC000001651126 and ZINC000000001083 on the growth of Ewing sarcoma cells (Figure 6A). The 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay findings revealed that the viability of RD-ES cells declined with increasing drug concentration (P<0.05).

ZINC000001651126 and ZINC000000001083 lower LSD1 expression in Ewing's sarcoma cells

We employed Western blotting to assess LSD1 expression levels in Ewing sarcoma cells in order to confirm the effects of ZINC000001651126 AND ZINC000000001083 on LSD1. Western blotting results revealed that LSD1 expression decreased as drug concentration increased. LSD1 is involved in various biological processes such as Ewing sarcoma cell growth and metastasis. ZINC000001651126 and ZINC000000001083 reduce the expression level of LSD1, and the growth and metabolism of Ewing sarcoma cells are greatly inhibited (Figure 6B).

ZINC000001651126 and ZINC000000001083 inhibited the migration of Ewing sarcoma cells

Using a scratch test, it was shown that ZINC000001651126 and ZINC000000001083 had an impact on the invasion and migration of Ewing sarcoma cells. When compared to the ZINC000001651126 and ZINC000000001083 groups, the scratch area in the control group was dramatically reduced 24 hours later (Figure 6C). The control group saw a greater degree of reduction even though the scratch area of the ZINC000001651126 group, the ZINC000000001083 group, and the control group all decreased with time. These findings suggest that ZINC000001651126 and ZINC000000001083 possibly inhibit Ewing sarcoma cell invasion and migration.

Ewing sarcoma is the second most common bone and soft tissue sarcoma in children, accounting for 6 to 8 percent of primary malignant bone tumors15. The prognosis of Ewing sarcoma is related to various factors such as its location, size, lesion extent and tumor stage [16]. After radiotherapy and chemotherapy combined with surgery, most patients' survival status has improved. However, Ewing sarcoma has a poor prognosis if it grows in the pelvis, spine, or if it has metastasized [17]. The most critical features of Ewing sarcoma are specific chromosomal translocations, including the most common EWS-FLI1 gene fusion and EWSR1-ERG fusion [18, 19]. LSD1 is related to the transcriptional regulatory function of EWS-FLI1 on target genes. LSD1 and its downstream targets can promote the growth, metabolism and metastasis of Ewing sarcoma cells, contributing to the occurrence and development of Ewing sarcoma [20].

Based on the general expression of LSD1 in tumor cells such as osteosarcoma, Ewing sarcoma, and chondrosarcoma, down-regulation or elimination of LSD1 function will lead to inhibition of proliferation, invasion, and migration in cancer cells [14]. LSD1 inhibitors were developed based on this idea, such as SP-2509 and HCI2509, undergoing clinical evaluation for cancer treatment [11]. They can reverse the transcriptional activity of the fusion gene EWS-FLI and ultimately destroy the function of the encoded protein EWS-ETS [20]. SP-2509 in combination with LSD1 blocks DNA replication and transcription in mitochondrial-related tumors. However, the specificity and effectiveness of SP-2509 in inhibiting LSD1 are slightly insufficient [21]. HCI2509 can significantly reverse the up-and down-regulated transcriptional profiles of EWS/FLI and EWS/ERG with single-agent efficacy in a transplanted model of Ewing sarcoma. However, its drug safety and tolerability still require continuous optimization of dosage in the future [22]. LSD1 inhibitors have both efficacy and limitations as potential therapeutic candidates and need to be screened before they can be widely used in the Ewing sarcoma clinic.

LibDock technology, CDOCKER module, and TOPKAT module are used for virtual screening and analysis of ligands. We first executed a virtual screening of 17,931 biologically named ligands in the ZINC15 database. Through analysis of the results, we discovered that the molecular conformation of compounds with high LibDock scores was more stable than those with low LibDock scores. Compounds in the top 30 on the LibDock score were selected for the subsequent study.

ADMET contains adsorption, distribution, metabolism excretion, and toxicity indicators that can be used to evaluate the pharmacological profile of candidate compounds. ZINC000001651126 and ZINC000000001083 have good solubility in water and no hepatotoxicity, which is more convenient for follow-up research. Compared to the remaining compounds, they have the advantage of lower levels of rodent carcinogenicity, Ames mutagenicity, and potential developmental toxicity. Tranylcypromine has no hepatotoxicity and is less rodent carcinogenic and Ames mutagenic. It is a non-CYP2D6 inhibitor with poor absorption properties in plasma protein binding. Therefore, ZINC000001651126 and ZINC000000001083 are considered reliable candidates. On the other hand, although the remaining drugs have specific toxicity or side effects, they can still play a role in drug development. Taking all the results together, ZINC000001651126 and ZINC000000001083 are optimal compounds for further study.

The CDOCKER module can well reflect candidate compounds' bonding mechanisms and chemical bonds. The CDOCKER potential energy of ZINC000001651126 and ZINC000000001083 was discovered to be lower than that of Tranylcypromine by CDOCKER module computation. The lower the CDOCKER potential, the higher the binding affinity. It can be inferred that their binding affinity to LSD1 is higher than that of Tranylcypromine.

The stability of ligand-LSD1 in the natural environment must be verified with molecular dynamics simulations. Potential and RMSD images of the complexes formed by each ligand-LSD1 show that the trajectories of ZINC000001651126 and ZINC000000001083 reach equilibrium after 90 ps. These two complexes' potential energies and RMSDs are also gradually stabilized over time. It can be concluded from this that ZINC000001651126 and ZINC000000001083 can interact with LSD1 to form complexes that are stable in the natural environment that affects LSD1. Further analysis of the findings then continues with drug development and purification.

We can effectively identify ideal LSD1 inhibitors through the computer-aided virtual operation. Then continue to use LibDock technology and monitor the metabolism, distribution, absorption, excretion, and toxicity of LSD1 inhibitors. Molecular docking and molecular dynamics simulations have found that ZINC000001651126 and ZINC000000001083 have inhibitory effects on Ewing sarcoma when examining and evaluating the binding affinity and stability of the ligand-LSD1. Thousands of refinements and improvements to a drug are conducive to faster market standards, so we will still improve and perfect the various properties of the drug.

No study design and measurement are perfect, and this study inevitably has several shortcomings. For example, it was not clear what signaling pathway the inhibitor worked through. The specificity and sensitivity of drugs targeting LSD1 require more biological and functional work. We will detect half-maximal inhibitory concentration, half-maximum inhibitory concentration, and half-maximum inhibitory concentration. Indicators such as effective concentration, LD50, and ED50 can comprehensively evaluate the properties of inhibitors.

This study employs a range of computer-aided chemical and structural techniques to select and evaluate ZINC compounds from a large pool of natural medicines with functions to inhibit LSD1 possibly. After careful design and precise calculations, both ZINC000001651126 and ZINC000000001083 can be considered safe and reliable drug candidates with less potential developmental toxicity. They have a high binding affinity for LSD1 and can competitively inhibit LSD1 and reduce the expression of LSD1 in Ewing sarcoma cells. ZINC000001651126 and ZINC000000001083 possibly inhibit the proliferation and migration of Ewing sarcoma cells, and the inhibitory effect was positively correlated with the drug concentration. In addition, we provide many drug candidates with pharmacological properties that facilitate the continuous development and improvement of LSD1 inhibitors.

ZINC database and Discovery studio software

Discovery Studio software is a system capable of simulating tiny and macromolecules, which can determine and refine the pathways of lead compounds and drug candidates, and then utilize structural biology and structural chemistry calculations to screen, design, and modify potential drugs [23].

LibDock technology and ADME module in DS4.5 (Discovery Studio 4.5 software) were used for virtual screening, and CDOCKER was used for docking studies. We screened LSD1 inhibitors from natural products in the ZINC15 database. The Irwin and Shouchet labs at UCSF (University of California, San Francisco) provide the ZINC15 database for free commercial use.

LibDock for virtual filtering

High-throughput screening (HTS) is initiated in the search for LSD1 inhibitors, which can help discover compounds with affinity for LSD1. Tranylcypromine is essentially a complete drug library, so for hundreds of analogs, the establishment of SAR is also applicable [24]. When LibDock operates on the virtual filtering step, it uses probes and grid points to compute protein hotspots, then arranges ligands through protein hotspots and performs an intelligent minimization algorithm [25]. In the charm force field, LibDock will minimize the ligands so that good interaction can be formed between the ligands [26]. When ligands are minimized, all ligand positions are determined according to the ordering of the ligand score. The chemical structure of LSD1 is illustrated in Fig. 1. LSD1 is identified as 2Z3Y in the Protein Data Bank (PDB), and the chemical structure of 2Z3Y was downloaded and imported into LibDock's environment. In order to obtain the chemical structure of LSD1, as shown in Fig. 2, heteroatoms and crystal water in the structure need to be removed, and polar hydrogen added. The structure with radius 13 centered on the inhibitor is selected. We removed the inhibitor from the model and balanced the crystal structure for subsequent structural analysis. After the chemical structure of Tranylcypromine is retrieved from the ZINC database and the binding site of Tranylcypromine is selected as the active docking site, the prepared ligand can be attached to the active site, and then virtual screening can be carried out. The ligands were finally sorted and grouped according to their corresponding LibDock scores.

Toxicity indicators and ADME characteristics of ligands

The ADME and TOPKAT modules in DS4.5 can be applied to the distribution, metabolism, excretion, and absorption levels of candidate compounds and can also detect toxicity and other properties. Before starting the selection of LSD1 inhibitors, we must consider the pharmacological properties of the compounds, including BBB, PPB levels, water-solubility, CYP2D6, hepatotoxicity, Ames mutagenicity, rodent carcinogenicity, developmental toxicity potential (DTP). After using the ADME module to predict the pharmacological properties of 17,931 ligands and Tranylcypromine, we obtained the water solubility, hepatotoxicity, and plasma protein binding (PPB) properties, CYP2D6 binding properties, blood-brain barrier (BBB) levels, and intestinal absorption levels. The results are also listed in Tables 2 and 3. It can be seen from the results that ZINC000001651126 and ZINC000000001083 have no hepatotoxicity, are not easy to pass the blood-brain barrier, have a high tolerance to CYP2D6, and have good water solubility. In addition, ZINC000001651126 and ZINC000000001083 have low Ames mutagenicity and can also be considered safe and desirable compounds.

Molecule docking and pharmacophore prediction

The CDOCKER module of DS4.5 can be applied to molecular docking studies. Through the query, it can be seen that CDOCKER can analyze the posture of the tested molecule and perform high-precision docking under the guidance of the CHARMm force field. The interaction generated by the CHARMm force field can act on receptors and ligands and is reflected in the binding affinity of receptors and ligands. During the docking process between the ligand and the receptor, the water molecule changes from the crystalline state to the fixed state, constantly changing the state of the ligand-receptor complex [27, 28]. So we remove water molecules and add hydrogen atoms to the protein. When the ligand-receptor binding reaches the optimal docking state, we should adapt the parameter TOP HITS to 10 and pose cluster radius to 0.5. The ligand-receptor complex will dock with LSD1 to initiate the CDOCKER process [29, 30].

Dynamic simulation experiment

The optimal binding conformation of each compound is subjected to molecular dynamics simulation experiments one by one. We first construct an orthogonal box to hold the ligand-receptor complex and then utilize the water model to dissolve it. To simulate the physiological environment of the CHARMm force field, we inject solid chloride ions into the system and minimize the CHARMm force field and energy. We also need to adjust the Target Temperature and Reference Pressure to 300K and 1.0 and the duration and step size to 100 ps and one fs. The linear constraint solving algorithm is incorporated in the computation of fixed hydrogen bonds. The calculation of the long-distance electrostatic field involves using the particle grid Ewald algorithm. Finally, we determine the complicated initial settings for reference. The analysis of ballistic structure, energy, and RMSD requires using the DS4.5 ballistics protocol.

Cell lines

Ewing sarcoma cells (RD-ES) from the American Type Culture Collection were employed in this investigation. The fetal bovine serum supplemented with Dulbecco's modified Eagle medium (DMEM, Hyclone, USA) was used to cultivate the cells (Gibco, MD, USA). The Petri plates were kept in a 37°C, 5% CO2, and 95% air environment.

MTT assay

A 96-well culture plate with 500 Ewing sarcoma cells per well was plated, and different dosages of ZINC000001651126 and ZINC000000001083 were made. MTT (5 mg/mL; Sigma, St. Louis, MO, USA) was dissolved in phosphate-buffered saline to check for cell viability. Fresh DMEM and diluted MTT (1:10, 10 percent MTT) were added to the medium at the start of the test, and it was then incubated at 37°C for 3.5 hours. After removing the growth media, the azo crystal was dissolved in a 200 L solution of dimethyl sulfoxide. An ELx800 absorbance microplate (BioTek Instruments, VT, USA) may be used to monitor the absorbance at 570 nm when MTT expression levels are decreased.

Western blotting

ZINC000001651126 and ZINC000000001083 were administered at various dosages to Ewing sarcoma cells (RD-ES) that were in the logarithmic growth phase and injected on 6-well plates at a density of 2×10⁵ cells/well. Total proteins were recovered from lysed cells after a 48-hour culture and sorted by electrophoresis. The cell membrane then received the protein. Primary antibodies against LSD1 and -actin were applied after the protein bands of interest were cleaved. After cleaning, secondary antibodies were incubated on the cell membrane. An improved chemiluminescence detection technique was used to see the membranes at the end (Pierce; Thermo Fisher Scientific, Inc.).

In vitro scratch test

On a 24-well PermanoxTM plate, Ewing sarcoma cells (RD-ES) were first grown. Apply the tip of a 10-µL pipette needle uniformly along the ruler to a consistent cell-free region in each well once the entire dish has been covered. Then carefully run DMEM over the loose cells. Then, various concentrations of ZINC000001651126 and ZINC000000001083 were administered to the cells. The monopoly area photos were taken with a phase-contrast microscope at three-time intervals, 0, 6, and 24 hours. The surviving damaged area and scratch breadth were assessed at six locations on each photograph.

Statistical analysis

GraphPad software was used for all data processing and analysis (GraphPad Prism version 9.0.0 for Windows, GraphPad Software, San Diego, California USA, www.graphpad.com”). To compare the two continuous variable groups and assess the statistical significance of normally distributed variables, the t-test was employed. The Mann-Whitney U test was used to assess the differences between non-normally distributed variables and the independent variables (i.e., the Wilcoxon rank sum test). P < 0.05 considered statistically significant.

Lysine specific demethylase 1, LSD1;

Adsorption, distribution, metabolism and excretion, ADME;

Developmental toxicity potential, DTP;

Cytochrome P4502D6, CYP2D6;

Ewing sarcoma breakpoint region 1-friend leukemia virus integration 1, EWS-FLI1;

Plasma protein binding, PPB;

Blood brain barrier, BBB;

Toxicity prediction by Computer assistive technology, TOPKAT;

Root Mean Square Deviation, RMSD;

Discovery Studio 4.5 software,DS4.5;

University of California, San Francisco, UCSF;

Protein Data Bank, PDB;

High-throughput screening, HTS;

Chemistry at harvard Macromolecular Mechanics, charmm;

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide, MTT.

CONFLICT OF INTEREST

The authors declare no conflict of interest.

Data availability statement

The crystal structure of Lysine-specific demethylase1 can be deposited in the Worldwide Protein Data Bank (PDB DOI: 10.2210/pdb2Z3Y/pdb).

AUTHOR CONTRIBUTIONS

Dong Zhu designed experiments; Liangxin Lin wrote the manuscript; Qifan Yang, Ye Jiang, Liangxin Lin, and Yilin Yang carried out experiments; Pengcheng Zhou analyzed the results of the experiments. Xinyu Wang supervised the study and contributed to data analysis.

FUNDING

This study was supported by the National Natural Science Foundation of China (Grant No12072129).

Theisen, E.R., et al., Therapeutic opportunities in Ewing sarcoma: EWS-FLI inhibition via LSD1 targeting. Oncotarget, 2016. 7(14): p. 17616-17630.
Jiang, J., et al., CCT6A, a novel prognostic biomarker for Ewing sarcoma. Medicine, 2021. 100(4).
Gartrell, J. and C. Rodriguez-Galindo, Ewing sarcoma: investigational mono- and combination therapies in clinical trials. Expert Opinion on Investigational Drugs, 2021. 30(6): p. 653-663.
Pishas, K.I., et al., Therapeutic Targeting of KDM1A/LSD1 in Ewing Sarcoma with SP-2509 Engages the Endoplasmic Reticulum Stress Response. Molecular Cancer Therapeutics, 2018. 17(9): p. 1902-1916.
Bradford, K., et al., Primary Renal Ewing Sarcoma in Children and Young Adults. Journal of Pediatric Hematology Oncology, 2020. 42(8): p. 474-481.
Sheffield, N.C., et al., DNA methylation heterogeneity defines a disease spectrum in Ewing sarcoma. Nature Medicine, 2017. 23(3): p. 386-395.
Treesaranuwattana, T., et al., Lysine-Specific Demethylase-1 Deficiency Increases Agonist Signaling Via the Mineralocorticoid Receptor. Hypertension, 2020. 75(4): p. 1045-1053.
Theisen, E.R., et al., Chromatin profiling reveals relocalization of lysine-specific demethylase 1 by an oncogenic fusion protein. Epigenetics, 2021. 16(4): p. 405-424.
Zheng, Y.-C., et al., A Systematic Review of Histone Lysine-Specific Demethylase 1 and Its Inhibitors. Medicinal Research Reviews, 2015. 35(5): p. 1032-1071.
Fang, Y., G. Liao, and B. Yu, LSD1/KDM1A inhibitors in clinical trials: advances and prospects. Journal of Hematology & Oncology, 2019. 12(1).
Tokarsky, E.J., et al., Mitochondrial Dysfunction is a Driver of SP-2509 Drug Resistance in Ewing Sarcoma. Molecular cancer research : MCR, 2022.
Sankar, S., et al., Reversible LSD1 Inhibition Interferes with Global EWS/ETS Transcriptional Activity and Impedes Ewing Sarcoma Tumor Growth. Clinical Cancer Research, 2014. 20(17): p. 4584-4597.
Fang, Y., et al., Natural products as LSD1 inhibitors for cancer therapy. Acta Pharmaceutica Sinica B, 2021. 11(3): p. 621-631.
Zheng, Y.C., et al., Irreversible LSD1 Inhibitors: Application of Tranylcypromine and Its Derivatives in Cancer Treatment. Current Topics in Medicinal Chemistry, 2016. 16(19): p. 2179-2188.
Beduk-Esen, C.S., et al., Ewing sarcoma in an infant and review of the literature. The Turkish journal of pediatrics, 2019. 61(5): p. 760-764.
Chen, W., et al., Using machine learning techniques predicts prognosis of patients with Ewing sarcoma. Journal of Orthopaedic Research, 2021. 39(11): p. 2519-2527.
Potratz, J., et al., Ewing Sarcoma: Clinical State-of-the-Art. Pediatric Hematology and Oncology, 2012. 29(1): p. 1-11.
Haybaeck, J. and A. Roessner, Malignant round cell tumors The Ewing sarcoma and beyond. Pathologe, 2020. 41(2): p. 116-122.
Teicher, B.A., et al., Characteristics of human Ewing/PNET sarcoma models. Annals of Saudi Medicine, 2011. 31(2): p. 174-182.
Romo-Morales, A., et al., Catalytic inhibition of KDM1A in Ewing sarcoma is insufficient as a therapeutic strategy. Pediatric Blood & Cancer, 2019. 66(9).
Harancher, M.R., et al., Antiviral Properties of the LSD1 Inhibitor SP-2509. Journal of Virology, 2020. 94(19).
Theisen, E.R., et al., Reversible inhibition of lysine specific demethylase 1 is a novel anti-tumor strategy for poorly differentiated endometrial carcinoma. Bmc Cancer, 2014. 14.
Zhong, S., et al., Selected by gene co-expression network and molecular docking analyses, ENMD-2076 is highly effective in glioblastoma-bearing rats. Aging-Us, 2019. 11(21): p. 9738-9766.
Lesuisse, D., et al., Discovery of the first non-ATP competitive IGF-1R kinase inhibitors: Advantages in comparison with competitive inhibitors. Bioorganic & Medicinal Chemistry Letters, 2011. 21(8): p. 2224-2228.
Rao, S.N., et al., Validation studies of the site-directed docking program LibDock. Journal of Chemical Information and Modeling, 2007. 47(6): p. 2159-2171.
Brooks, B.R., et al., CHARMM: The Biomolecular Simulation Program. Journal of Computational Chemistry, 2009. 30(10): p. 1545-1614.
Kalani, K., et al., In silico and in vivo anti-malarial studies of 18β glycyrrhetinic acid from Glycyrrhiza glabra. PLoS One, 2013. 8(9): p. e74761.
Sarvagalla, S., et al., Identification of ligand efficient, fragment-like hits from an HTS library: structure-based virtual screening and docking investigations of 2H- and 3H-pyrazolo tautomers for Aurora kinase A selectivity. Journal of Computer-Aided Molecular Design, 2015. 29(1): p. 89-100.
Haider, M.K., H.O. Bertrand, and R.E. Hubbard, Predicting fragment binding poses using a combined MCSS MM-GBSA approach. J Chem Inf Model, 2011. 51(5): p. 1092-105.
Hou, T., et al., Assessing the performance of the MM/PBSA and MM/GBSA methods. 1. The accuracy of binding free energy calculations based on molecular dynamics simulations. J Chem Inf Model, 2011. 51(1): p. 69-82.

Table 1. Top 30 Ranked Compounds with Higher LibDock Scores
Number	Compounds	Libdock Score	Number	Compounds	Libdock Score
1	ZINC000004654957	171.831	16	ZINC000000340372	142.751
2	ZINC000014658375	162.486	17	ZINC000001651126	140.419
3	ZINC000004557101	159.544	18	ZINC000015270205	140.061
4	ZINC000003831332	159.455	19	ZINC000001632198	140.026
5	ZINC000013540278	157.689	20	ZINC000001721178	139.856
6	ZINC000008689957	153.379	21	ZINC000100191509	138.963
7	ZINC000013449412	153.25	22	ZINC000005115722	138.963
8	ZINC000003875576	148.932	23	ZINC000004025169	136.361
9	ZINC000004474575	148.908	24	ZINC000002585906	135.492
10	ZINC000008844372	148.704	25	ZINC000014642997	135.032
11	ZINC000003087481	147.292	26	ZINC000000001083	134.586
12	ZINC000004654839	146.958	27	ZINC000238730544	132.958
13	ZINC000002561271	144.894	28	ZINC000002563374	132.258
14	ZINC000005845723	144.862	29	ZINC000004098683	131.934
15	ZINC000095630290	144.034	30	ZINC000004655389	130.954

Table 2. Adsorption, Distribution, Metabolism, and Excretion Properties of Compounds
Number	Compounds	Solubility Level	BBB Level	CYP2D6	Hepatotoxicity	Absorption Level	PPB Level
1	ZINC000004654957	3	0	0	0	1	0
2	ZINC000014658375	2	0	0	0	0	1
3	ZINC000004557101	3	4	0	0	1	0
4	ZINC000003831332	0	4	0	0	3	1
5	ZINC000013540278	3	4	0	0	1	0
6	ZINC000008689957	3	0	0	0	0	0
7	ZINC000013449412	2	1	0	0	0	1
8	ZINC000003875576	2	1	1	0	0	1
9	ZINC000004474575	2	4	0	0	2	1
10	ZINC000008844372	3	4	0	1	1	0
11	ZINC000003087481	2	1	1	0	0	1
12	ZINC000004654839	0	4	0	0	3	1
13	ZINC000002561271	2	1	0	1	0	1
14	ZINC000005845723	1	1	0	1	0	1
15	ZINC000095630290	2	2	0	1	0	1
16	ZINC000000340372	2	1	1	0	0	1
17	ZINC000001651126	3	2	0	0	0	1
18	ZINC000015270205	3	1	0	0	0	1
19	ZINC000001632198	2	2	0	1	0	1
20	ZINC000001721178	1	0	0	1	1	1
21	ZINC000100191509	3	2	0	0	0	1
22	ZINC000005115722	3	2	0	0	0	1
23	ZINC000004025169	3	2	0	0	0	1
24	ZINC000002585906	1	0	1	1	2	1
25	ZINC000014642997	3	3	0	1	0	1
26	ZINC000000001083	3	2	0	0	0	1
27	ZINC000238730544	2	1	0	1	0	1
28	ZINC000002563374	2	0	0	0	0	1
29	ZINC000004098683	2	4	0	1	2	0
30	ZINC000004655389	3	2	0	0	0	1
31	Tranylcypromine	4	2	0	0	0	0

BBB, blood-brain barrier; CYP2D6, cytochrome P-450 2D6; PPB, plasma protein binding

Aqueous-solubility level: 0, extremely low; 1, very low, but possible; 2, low; 3, good.

BBB level: 0, very high penetrant; 1, high; 2, medium; 3, low; 4, undefined.

CYP2D6 level: 0, noninhibitor; 1, inhibitor.

Hepatotoxicity: 0, nontoxic; 1, toxic.

Human-intestinal absorption level: 0, good; 1, moderate; 2, poor; 3, very poor.

PPB: 0, absorbent weak; 1, absorbent strong.

Table 3. Toxicities of Compounds
Number	Compounds	Mouse NTP		Rat NTP		Ames	DTP
Number	Compounds	Female	Male	Female	Male	Ames	DTP
1	ZINC000004654957	0	0	0	0	0	0
2	ZINC000014658375	0.019	1	1	0	1	1
3	ZINC000004557101	0	0.006	0	0	0	0
4	ZINC000003831332	1	1	0	0.993	0	1
5	ZINC000013540278	0	0.001	0	0	0	0
6	ZINC000008689957	0	0	0	0	0	0
7	ZINC000013449412	1	0	0	0.989	0	0.996
8	ZINC000003875576	1	1	0	0.997	0	0.999
9	ZINC000004474575	0	0.004	0	0	0.001	0
10	ZINC000008844372	1	1	1	1	0	1
11	ZINC000003087481	0	0.999	1	0.001	0	1
12	ZINC000004654839	1	1	0	0.994	0.019	1
13	ZINC000002561271	0	0	0.996	1	1	0.002
14	ZINC000005845723	0	0	1	0	0.224	0
15	ZINC000095630290	0.049	1	0.001	1	0.03	1
16	ZINC000000340372	0	0.999	1	0.001	0	1
17	ZINC000001651126	1	0	0	0.037	0.005	1
18	ZINC000015270205	1	0.116	0	0.642	0	0
19	ZINC000001632198	0	1	1	0.019	1	0.988
20	ZINC000001721178	0	0	0	0.022	0	1
21	ZINC000100191509	1	0	0	0.832	0.207	1
22	ZINC000005115722	1	0	0	0.832	0.207	1
23	ZINC000004025169	0.207	1	0	0.003	0	0.999
24	ZINC000002585906	1	0.551	1	0	0	0.147
25	ZINC000014642997	0	0.999	0.016	0.995	0	0.834
26	ZINC000000001083	0.385	0.76	0.641	0.663	0.001	0.959
27	ZINC000238730544	0	0	0.815	1	1	1
28	ZINC000002563374	1	1	1	0.997	0	0
29	ZINC000004098683	0	1	1	1	0	1
30	ZINC000004655389	0.207	1	0	0.003	0	0.999
31	Tranylcypromine	0	0	0.992	0	0.011	0

NTP,U.S.NationalToxicologyProgram;DTP,developmentaltoxicitypotential.

NTP<0.3(noncarcinogen);>0.8(carcinogen).

Ames<0.3(nonmutagen);>0.8(mutagen).

DTP<0.3(nontoxic);>0.8(toxic).

Table 4. CDOCKER Potential Energy of Compounds with LSD1
Compounds	-CDOCKER Potential Energy (kcal/mol)
ZINC000001651126	54.7042
ZINC000000001083	59.9534
Tranylcypromine	29.14

Table 5. Hydrogen Bond Interaction Parameters for Each Compound with LSD1 Residues
Receptor	Compound	Donor Atom	Receptor Atom	Distances(Å)
2Z3Y	ZINC000001651126	SER289:HN	ZINC000001651126:O1	2.05487
		SER289:HG	ZINC000001651126:O1	2.01472
		ARG310:HE	ZINC000001651126:O9	2.00932
		ZINC000001651126:H28	GLU308:OE2	2.34169
		ALA331:HA	ZINC000001651126:O21	2.75456
	ZINC000000001083	SER289:HG	ZINC000000001083:O1	2.09737
		ARG310:HE	ZINC000000001083:O9	2.16062
		ZINC000000001083:H26	GLU308:OE2	2.16053
		ZINC000000001083:H27	TRP756:O	1.9218
	Tranylcypromine	Tranylcypromine:H11	THR624:O	2.34383
	Tranylcypromine	Tranylcypromine:H12	THR624:O	2.09162

Table 6. π-Related Interaction Parameters for Each Compound with LSD1
Receptor	Compound	Donor Atom	Receptor Atom	Distances(Å)
2Z3Y	ZINC000001651126	TYR761	ZINC000001651126	4.84957
		ZINC000001651126	PRO626	4.61317
		ZINC000001651126	ARG316	5.44445
		ZINC000001651126	ALA331	3.88625
	ZINC000000001083	ZINC000000001083	PRO626	4.64088
		ZINC000000001083	ARG316	4.58062
		ZINC000000001083	ALA331	4.93468
		ZINC000000001083	VAL811	5.0674
		ZINC000000001083	ALA814	5.40379
	Tranylcypromine	Tranylcypromine	VAL288	5.16145
		Tranylcypromine	ARG316	4.62655
		Tranylcypromine	ALA814	4.0202

No competing interests reported.

Effective natural inhibitors targeting LSD1 by computational study

Status:

Version 1

Abstract

Figures

Introduction

Result

Discussion

Conclusions

Materials And Methods