Synthesis of some sulfonamide derivatives coupled with salicylamide or anisamide scaffold as potent PD-L1 inhibitors and their anti-proliferation assay

The present study explores the potential of the new sulfonamide derivatives coupled with salicylamide or anisamide scaffold as immune checkpoint PD-L1 inhibitors. Based on in silico virtual screening 32 derivatives were synthesized and tested in vitro as PD-L1 inhibitors using screening ELISA assay. Five compounds gave promised results with more than 50% inhibition. The most active, with activity 57.152%, was 5-Chloro-N-(4-(N-(3-uorophenyl)sulfamoyl)phenethyl)salicylamide (30). The other compounds were 5-Chloro-2-methoxy-N-(4-(N-(4-uorophenyl)sulfamoyl)benzyl)benzamide (4, 53.327%), 5-Chloro-2-methoxy-N-(4-(N-(4-methylphenyl)sulfamoyl)phenethyl)benzamide (17, 51.253%), 5-Chloro-N-(4-(N-(4-(tri�uoromethyl)phenyl)sulfamoyl)phenethyl)salicylamide (31, 51.058%) and 5-Chloro-2-methoxy-N-(4-(N-(2,4-diuorophenyl)sulfamoyl)benzyl)benzamide (7, 50.993). All compounds were found to be safe and have no cytotoxic effect against the �broblast cell lines. Moreover, compound 4, 7, 17, and 30 should no or little anti-proliferative activity against the cell lines use in this study except compound 4, which have anti-proliferative activity against PC-3 (66.640%). On the other hand, compound 31 should remark activities against the cell lines MCF7, DU-145, and PC-3. In general, human prostate cancer cell line PC-3 is highly sensitive towards some of the 32 compounds tested at 10 μ mol. The molecular docking study and ADMET (absorption, distribution, metabolism, excretion, and toxicity) analysis of the bioactive compounds were


Introduction
In recent years, cancer immunotherapy has achieved great clinical success, which makes it a current topic of intensive research and brings new hope to cancer patients.Among these, the immune checkpoint inhibitor is the most mature immunotherapy and has the highest share in the market now.Programmed cell death protein-1 or 2 and their Ligands 1 or 2 (PD-1, PD-2, PDL-1, or PDL-2) expressed on stimulated, regulatory-T lymphocytes, B, natural killers, monocytes, and dendritic cells [1,2].PD-1 and its ligand are maintaining immune tolerance through immune inhibitory signals to Tcells [3].PD-1 mediates tumors through interaction between the T cell-PD-1 expression and its ligand-PD-L1 on the tumor cell [4].Monoclonal antibodies (MoAbs) against PD-1 and PD-L1 have been reported as general immune therapeutics, especially in kidney cancer.Acting as inhibitor-monotherapy or in combination with others blocked the PD-1/PD-L1 interaction [5][6][7][8].Pembrolizumab and nivolumab are anti-PD-1 and gained approval from the Food and Drug Administration (FDA) in 2017, also; atezolizumab is an anti-PD-L1 MoAbs.The anti-PD-1 and anti-PD-L1 inhibitors have impressive antitumor effects in several malignancies.Blocking the PD-1/PD-L1 interaction stimulated cytotoxic T cells to eradicate cancer cells [9].Toxicity with PD-1 blocking agents is less than the toxicity with previous immune therapies such as interleukin 2 and cytotoxic Tlymphocyte antigen (CTLA)-4 blockade [10].However, immune therapeutic targeting of anti-PD-1: PD-L1/L2 also stimulated T cells against pathogens and tumors [11].
Egyptian Journal of Chemistry ________________________________________________ Egypt.J. Chem.67, No. 5 (2024) Compared with chemotherapy, FDA-approved PD-1/PD-L1 inhibitors have been found to cause fewer adverse events (AEs) in cancer patients who experienced no to fewer symptoms related to their treatment, less hematologic toxicity, and fewer immune-related AEs.Also, patients were less likely to discontinue treatment due to toxicities or experience treatment-related deaths.PD1/PD-L1 inhibitors are overall better tolerated than chemotherapy [12].Expression of PD-L1 on the tumor and immune cells membrane is associated with enhanced objective response rates to PD-1/PD-L1 inhibition [13].
In addition to the MoAbs, Aurigene Discovery Technologies, and Laboratoires Pierre Fabre, in 2014, announced a collaboration license for the first peptide inhibitor of the PD-1/PD-L1 pathway.Later, it was reported that more peptides and peptidomimetics serve as inhibitors of the PD-1/PD-L1 pathway [14], and their activity relationships were also discussed [15].Scientists from Bristol-Myers Squibb have discovered a novel series of small molecule inhibitors targeting PD-L1 [16].These small molecule inhibitors contain a mono-ortho substituted biphenyl moiety which links to another phenyl ring via a methylene amine group.Moreover, academic research groups worldwide have made enormous efforts to develop small molecules with chemical scaffolds such as biphenyl, biaryl, urea-linked, and indole/imidazole-based as potent immune checkpoint inhibitors [15,17].Sulfonamides are a group of drugs that have been used for a long time to treat many different diseases [18][19][20][21][22][23][24][25][26][27] and some of them work by affecting different cellular mechanisms [28][29][30][31][32][33][34][35].Sulfonamides can also help with male erectile dysfunction [36,37].Additionally, sulfonamides can block several important oncogenic signaling pathways [38,39].In 2011, Sharpe et al. from Harvard University first screened small molecules containing sulfamonomethoxine and sulfamethizole scaffold that could inhibit the PD-1/PD-L1 pathway.The results showed the sulfonamide derivatives had inhibitory activity at a concentration of 10 mmol/L [40].As the first discovered small molecule inhibitors, sulfonamides are promising as lead compounds for further modification [40,41].The notification from Harvard et al and our experience with sulfonamides as antitumor agents have motivated us to explore a new series of sulfonamides as potential inhibitors of the PD-1/PD-L1 pathway.Further development may lead to the design of an anticancer therapy based on orally delivered immune checkpoint inhibition.Our previous reports discussed the importance of sulfonamide new series coupled with salicylamide and/or anisamide scaffold as potent inhibitors for tubulin polymerization, dihydropteroate synthase enzyme (DHPS), and Inosine-5'-monophosphate dehydrogenase (IMPDH) [42][43][44].The present study reports the virtual screening of self-database, synthesis, and biological evaluation of a series of sulfonamide derivatives coupled with salicylamide or anisamide scaffold as potential PD-L1 inhibitors.The details of compound synthesis and enzyme inhibition were discussed.Evaluation of anti-tumor of the selected compounds was also evaluated.The molecular docking study and ADMET (absorption, distribution, metabolism, excretion, and toxicity) analysis of the bioactive compounds were used to elucidate the mode-of-action mechanism.

Chemistry
Two series of sulfonamide derivatives (series 1 and 2) were synthesized according to the published methods [42][43][44][45][46][47][48].All materials, reagents, and solvents were purchased from Sigma-Aldrich, Merck, Fisher chemicals, and Lab-Scan analytical sciences and were used without further purification. 1H and 13 C NMR spectra were recorded at 298 K on a JEOL ECA500 spectrometer ( 1 H at 500.16MHz and 13 C NMR at 125.76 MHz) and were processed using the Bruker Topspin 3.2 software. 1H and 13 C NMR spectra were referenced to 1 H signals of residual non-deuterated solvents and 13 C signals to the deuterated solvents. 1H NMR signals were reported with chemical shift values δ (ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet and br = broad), relative integral, coupling constants J (Hz) and assignments.Mass spectra were recorded on a JEOL DART+ HI RESOLUTION mass spectrometer, and ionization of all samples was carried out using ESI.Melting points were measured on an Electrothermal IA9100 digital melting point apparatus and were uncorrected.Analytical TLC was performed on Merck silica gel 60 F254 pre-coated aluminum plates (0.2 mm) and visualized under UV light (254 nm).

Series 2 (n=2)
The same method illustrated above was used for the synthesis of series 2 but the benzylamine was replaced with beta-phenethyl amine and finally, the benzenesulfonyl chloride derivative coupled with the appropriate amine for the synthesis of benzamides 10-19.

Biological evaluation 2.2.1. In vitro PD-L1 inhibition assay
The inhibitory activity of the compounds as PD-L1 inhibitors was determined using PD-1 [Biotinylated]: PD-L1 Inhibitor Screening ELISA Assay Pair (AcroBiosystems, USA) following the manufacturer's protocol.Briefly, the 96 wells in the plate were coated with human PD-L1 protein followed by applying 50 µM of the potential compounds to each well.Unblocked proteins were then bound to human PD-1-Biotin.Streptavidin-HRP followed by TMB was then added to each well and the ability of compounds to inhibit PD-1: PD-L1 binding was determined by comparing OD readings.The absorbance was obtained at 450 nm using a Synergy™ HTX Multi-Mode Microplate Reader (BioTek, Winooski, VT, USA).

Virtual screening and Molecular docking
The X-ray diffraction pdb files 5j8o corresponding to the structure of human programmed cell death (hPD-L1) were downloaded from the protein data bank (www.rcsb.org/pdb/welcome.do)with resolution 2.30 and 2.30 A°, respectively.Their energy was minimized using the YASARA Energy Minimization Server (www.rcsb.org/pdb/welcome.do)and all water, bound ligands, and cofactors were removed.The conformers corresponding to each compound under investigation were created, energy minimized and saved as pdb file format using OSIRIS Datawarrior v5.5.0 (www.openmolecules.org).The pdbqt file format of the proteins and the compounds were created using MGL Tools 1.5.6 (www.mgltools.scripps.edu).Docking calculation was performed using Auto Dock Vena [49] and PyRx 0.8 (www.mgltools.scripps.edu).The results were analyzed based on the binding affinity of the ligand at the inhibitory pocket and hydrogen bonding formed using BIOVA Studio Visualizer v21.1.0.20298 (www.3dsbiova.com).

Virtual screening 3.1.1. Database
One of the most important computational techniques recently used in drug discovery is virtual screening.It is used to identify the structures, which are most likely to bind with drug targets such as protein receptors or enzymes.Recently, several highresolution crystal structures establishing the structural foundations of human PD-1 (hPD-1) and PD-L1 (hPD-L1) were published and available in the protein data bank [15].The obtained crystals provide important information about the hPD-1/hPD-L1 interaction and the details of the binding modes.Unlike the traditional approaches to drug discovery, which rely on a step-wise synthesis and screening of large numbers of compounds to optimize activity profiles, scientists have used computer models of new chemical entities to help define the activity profiles, geometries, and reactivity of the target compounds.Furthermore, in medicinal chemistry, biological activity is dependent on the three-dimensional positioning of the pharmacophore.So, the improvement of new mathematical models that describe chemical phenomena and the development of more program interfaces coupled with the availability of high-technology computer hardware has provided scientists with a new set of computational tools.To rationalize the prospected biological activity and to describe the binding mode of the active compounds with their predicted intracellular target, docking analysis should be carried out.
In the present study, virtual screening of the selfdatabase of some sulfonamide derivatives as inhibitors for the PD-1/PD-L1 pathway was created first, followed by synthesis and determination of its bioactivity.The self-database of the sulfonamide derivatives was designed with the assistance of different molecular modeling and drug discovery computer programs.This creation was made with the understanding that protein inhibitors should have small molecular sizes that fit in the inhibitory pocket containing hydroxyl, amino, or oxo groups to form hydrogen bonds and an aryl group to allow for hydrophobic interaction.Additionally, the flexibility of the inhibitors was improved by gradually increasing the number of internal single bonds (Fig. 1).The 2D structure corresponding to each compound was drawn and saved in sck file format using BIOVA Draw v21.1.NET.It was converted to a 3D structure and saved as mol file format using BIOVA Studio Visualizer v21.1.0.20298 (www.3dsbiova.com).The structure in mol file format was energy minimized and saved as pdb file format using the VEGA ZZ program (Drug Design Laboratory website).The conformers corresponding to each compound were created, energy minimized and saved as pdb file format using OSIRIS Datawarrior v5.5.0 (www.openmolecules.org).Fig. 1.The structure of the sulfonamides database used in the virtual screening.

Target Protein
PD-L1 is a type I transmembrane protein and the PD-1/PD-L1 interaction releases a co-inhibitory signal to activated T cells, which induces T cell apoptosis, energy, and functional exhaustion.As a result, T cells' activities to antigenic stimuli including proliferation, cytokine secretion, and cytolytic activity are reduced.Normally, PD-1 and widely expressed PD-L1 act as an intrinsic negative feedback system to down-regulate the immune system and promote self-tolerance by suppressing T cell inflammatory activity.However, the overexpressed PD-L1 on the surface of tumor cells can induce T-cell dysfunction and suppress antigenspecific T-cell responses [15].The crystal structure of human programmed cell death (hPD-L1, 5g8o) that has been solved by X-ray crystallography was downloaded from the Protein Data Bank website (www.rcsb.org)as a pdb file format.Its structure was corrected and energy was minimized using the YASARA Energy Minimization Server (www.rcsb.org/pdb/welcome.do).All ligands and cofactors were removed and prepared for virtual screening in pdbqt file format using the MGL Tools Program (www.mgltools.scripps.edu).The inhibitory pocket of the target protein was analyzed and the amino acid chains that surround the native ligand were defined.It composed of Chain A (ILE54, VAL55, TYR56, ASN63, GLN66, VAL68, VAL76, MET115, ILE116, SER117, ALA121, ASP122, TYR123) and chain B (PHE19, ILE54, VAL55, TYR56, MET115, ILE116, SER117, ALA121, ASP122, TYR123, LYS124).The native ligand and the surrounding amino acids are illustrated in Fig. 2.

Docking calculation and Results analysis
The sulfonamide database created was docked into the inhibitory pocket of the hPD-L1 binding site to deduce the binding affinity (kcal/mol) using AutoDock Vena and PyRx programs (www.mgltools.scripps.edu).The binding affinity is used to describe the affinity between a ligand (such as a drug) and a protein, i.e. how tightly a ligand binds to a particular protein.Ligand-protein affinities (kcal/mol) are influenced by non-covalent intermolecular interactions such as hydrogen bonding, electrostatic interactions, hydrophobic and Van der Waals forces.The result analysis was performed using BIOVA Studio Visualizer v21.1.0.20298 (www.3dsbiova.com).By taking into consideration the binding affinity of the native ligand (6GZ), compounds with binding affinity as low as -9.5 kcal/mol are considered promising compounds, and subjected to synthesis and testing directly against the target protein.Table 1 illustrates the promising compounds, their binding affinity, and the hydrogen bonds formed.
Fig. 2. A. The inhibitory pocket of the hPD-L1 (PDB ID: 5j8o).The outermost surface of the enzyme was illustrated as an electrostatic potential map and the native inhibitor as a gray ball and stick; B. The amino acids surrounding the inhibitory pocket are illustrated as lines and the native ligand is a pink ball and stick.
In series 1, the amidation of 5-chloro-o-anisic acid with benzylamine was performed using a mixed anhydride method to afford 5-chloro-2-methoxy-Nbenzylbenzamide [47].In series 2, benzyl amine was replaced with beta-phenethylamine to increase the flexibility of the produced compound by increasing the possible rotation around the single bonds.The benzamides formed were treated with chlorosulfonic acid in an ice bath for 24 hours to yield the corresponding benzenesulfonyl chloride derivatives.Using the reported method [42], benzenesulfonyl chloride derivatives were coupled with appropriate aromatic amines in the presence of sodium carbonate in a mixture of dichloromethane (DCM) and water (2:1 in ratio) to yield benzamide derivatives 4-19.

Biological evaluation 3.3.1. In vitro PD-L1 inhibition assay
Immune checkpoint signaling pathways are the focus of contemporary cancer research.PD-1 is one of the best-characterized checkpoint proteins.The binding between PD-1 and its ligand PD-L1 inhibits T-cell activation and allows cancer cells to evade the body's immune surveillance.Thus, pharmacological inhibition of PD-1 or its ligands has been considered a promising strategy by many oncologists [50].
The PD-1 [Biotinylated]: PD-L1 inhibitor screening enzyme-linked immunosorbent assay (ELISA) Kit is designed to facilitate the identification and characterization of novel inhibitors of the PD-1 pathway.The assay utilizes the binding of biotinylated human PD-1 to immobilized human PD-L1 in a functional ELISA assay using a simple colorimetric sandwich ELISA platform.The experiment consists of 4 easy steps: a) overlaid the plate with human PD-L1.b) Added the potential compounds corresponding to 50 μM to the test.c) Added human PD-1 biotin to bind coated human PD-L1.d) Added Streptavidin-HRP followed by 3,3',5,5'-Tetramethylbenzidine (TMB) or other Horseradish Peroxidase (HRP) colorimetric substrate.
Finally, the ability of potential compounds to inhibit PD-1: PD-L1 binding is determined by comparing optical density (OD) readings between different experimental groups.Table 2 illustrates the results of the inhibitory screening in order from the most active to the least active.The pink and red dashed lines showed pi-sigma and pi-pi stacked interaction, respectively.The pall red dashed lines illustrated the hydrophobic interaction with A: HIS69, ALA121, and B: ALA18, ILE54.The yellow dashed lines showed pi-sulfur interaction with MET115.Fig. 6. 3D (a) and 2D (b) binding mode of compound 30 (colored according to the functional groups) docked into the hotspot of human programmed cell death 1 (PDB ID: 5j8o).The rest of the enzyme structure was deleted from the view to clarify the docked conformer.The green dashed lines indicate hydrogen bonds with A: GLN66 and B: ALA121.The blue dashed lines showed the polar contact of the fluorine atom with B: MET115 and B: ILE116.The pink and red dashed lines showed pi-sigma and pi-pi stacked interaction, respectively.The pall red dashed lines illustrated the hydrophobic interaction with A: HIS69, ALA112, MET115, and B: ALA18.

In vitro anti-proliferative activity and cytotoxicity effect
Within the framework of our efforts exerted in the evaluation of the sulfonamide derivatives coupled with salicylamide or anisamide scaffold [42][43][44][45][46][47][48], in vitro anti-proliferative activity against four types of human cell lines was tested at 10 μM.The objective of this study is to evaluate the safety of PD-L1 active compounds and compare their anti-proliferative activity with that of other compounds under investigation.The compounds were tested against human breast cancer cell lines MCF-7 which have estrogen, progesterone, and glucocorticoid receptors [51].Also, two human prostate cancer cell lines DU-145 and PC-3 were considered to be the standard prostate cancer cell lines used in therapeutic research and drug development [51][52][53].The second cell line did not respond to androgens, glucocorticoid, or fibroblast growth factors, but it was influenced by epidermal growth factors.Fibroblast was used in the present study as a normal cell line and it is a type of biological cell that synthesizes the extracellular matrix and collages, producing the structural framework role in wound healing [54].The results of the anti-proliferative activity and their cytotoxic effect on the fibroblast cell line are illustrated in Table 3.The compounds that have inhibitory activity against PD-L1 (4, 7, 17, 30, and 31) were found to be safe and did not have any cytotoxic effects on the fibroblast cell lines.Moreover, compounds 4, 7, 17, and 30 showed little to no anti-proliferative activity against the cell lines used in this study except compound 4, which displayed anti-proliferative activity against PC-3 (66.640%).On the other hand, compound 31 showed remarkable activities against the cell lines MCF7, DU-145, and PC-3 with 67.756%, 44.403%, and 64.737% respectively.Overall, the human prostate cancer cell line PC-3 is highly sensitive toward the tested compounds as compound 23 is the most active with 83.760%, and compounds 4, 19, 22, 24, 29, 31, and 33 display antiproliferative activity with more than 50%.Table 3*.Percentage of the antiproliferative activity of the compounds under investigation against the tested human cancer cell lines and their cytotoxic effect on fibroblast cell lines at 10 µM.

In silico ADMET prediction
In silico physicochemical properties, lipophilicity, solubility, absorption, distribution, metabolism, and toxicity parameters of compounds 4, 7, 17, 30, and 31 are predicted using the admetSAR website [55] and are summarized in Table 4.The promising compound selected is found to be favorable in an acceptable prediction.

Conclusions
Based on in silico virtual screening of sulfonamides self-database against immune checkpoint PD-L1, 32 sulfonamide derivatives coupled with salicylamide or anisamide scaffold were synthesized and tested in vitro as PD-L1 inhibitor using screening ELISA assay.Five compounds (30, 4, 17, 31, 7) gave promising results with more than 50% inhibition.All compounds were found to be safe and have no cytotoxic effect against the fibroblast cell lines.Moreover, compounds 4, 7, 17, and 30 showed little to no anti-proliferative activity against the cell lines used in this study except compound 4, which displayed anti-proliferative activity against PC-3.On the other hand, compound 31 showed remarkable activities against the cell lines MCF-7, DU-145, and PC-3.Overall, the human prostate cancer cell line PC-3 is highly sensitive toward some of the 32 compounds tested at 10 μM.The results of this study indicate a new class of safe sulfonamide derivatives as PD-L1 inhibitors, which can be developed to increase their activity against immune checkpoint PD-L1.

Conflicts of interest
The authors declare that they have no competing interests.

Formatting of funding sources
The study was supported by the Pharmaceutical and Drug Industries Research Institute, National Research Centre, Egypt under Grant (Number 12060114).

Fig. 3 .
Fig. 3. 3D (a) and 2D (b) binding mode of compound 4 (colored according to the functional groups) docked into the hotspot of human Programmed cell death 1 (PDB ID: 5j8o).The rest of the enzyme structure was deleted from the view to clarify the docked conformer.The green dashed lines indicate hydrogen bonds with B: ALA121.The blue dashed lines showed the polar contact of the fluorine atom with B: ILE54, VAL55, and MET115.The pink and red dashed lines showed pi-sigma and pi-pi stacked interaction, respectively.The pall red dashed lines illustrated the hydrophobic interaction with A: MET115, and B: ALA18.

Fig. 4 .
Fig. 4. 3D (a) and 2D (b) binding mode of compound 7 (colored according to the functional groups) docked into the hotspot of human programmed cell death 1 (PDB ID: 5j8o).The rest of the enzyme structure was deleted from the view to clarify the docked conformer.The green dashed lines indicate hydrogen bonds with B: ALA121.The blue dashed lines showed the polar contact of the fluorine atom with B: ILE54, VAL55, ILE116, and ASP122.The pink dashed lines showed pi-sigma interaction.The pall red dashed lines illustrated the hydrophobic interaction with A: MRT115 and B: ALA18.

Fig. 5 .
Fig. 5. 3D (a) and 2D (b) binding mode of compound 17 (colored according to the functional groups) docked into the hotspot of human programmed cell death 1 (PDB ID: 5j8o).The rest of the enzyme structure was deleted from the view to clarify the docked conformer.The green dashed lines indicate hydrogen bonds with B: PHE19.The pink and red dashed lines showed pi-sigma and pi-pi stacked interaction, respectively.The pall red dashed lines illustrated the hydrophobic interaction with A: HIS69, ALA121, and B: ALA18, ILE54.The yellow dashed lines showed pi-sulfur interaction with MET115.

Fig. 7 .
Fig. 7. 3D (a) and 2D (b) binding mode of compound 31 (colored according to the functional groups) docked into the hotspot of human Programmed death 1 (PDB ID: 5j8o).The rest of the enzyme structure was deleted from the view to clarify the docked conformer.The green dashed lines indicate hydrogen bonds with A: GLN66, MET115, and B: PHE19, ALA121.The blue dashed lines showed the polar contact of the fluorine atom with A: ALA121 and B: ILE54, MET115.The pink and red dashed lines showed pi-sigma and pi-pi stacked interaction, respectively.The pall red dashed lines illustrated the hydrophobic interaction with B: ALA18 and TYR56.

Table 1 .
Compounds were selected for synthesis and testing as hPD-L1 inhibitors according to the virtual screening and formed hydrogen bonds with the amino acid residues.

Table 2 .
The inhibitory activity percentage of the potential compounds (50 μM) in order according to the most active to the least active.

, 17, 30, and 31) have
the ability to inhibit PD-L1 with more than 50%.Four of them have fluorine atoms in ring C, while the 5 th (17) has a methyl group in the para

Table 4 .
In silico ADMET predictions of the selected active compounds as a PD-L1 inhibitor