Design, Synthesis, and Biological Activity of Dual Monoamine Oxidase A and Heat Shock Protein 90 Inhibitors, N-Methylpropargylamine-Conjugated 4-Isopropylresorcinol for Glioblastoma

Abstract

Monoamine oxidase A (MAO A) and heat shock protein 90 (HSP90) inhibitors have been shown to decrease the progression of glioblastoma (GBM) and other cancers. In this study, a series of MAO A/HSP90 dual inhibitors were designed and synthesized in the hope to develop more effective treatment of GBM. Compounds 4-b and 4-c are conjugates of isopropylresorcinol (pharmacophore of HSP90 inhibitor) with the phenyl group of clorgyline (MAO A inhibitor) by a tertiary amide bond substituted with methyl (4-b) or ethyl (4-c) group, respectively. They inhibited MAO A activity, HSP90 binding, and the growth of both TMZ-sensitive and -resistant GBM cells. Western blots showed that they increased HSP70 expression indicating reduced function of HSP90, reduced HER2 and phospho-Akt expression similar to MAO A or HSP90 inhibitor itself. Both compounds decreased IFN-γ induced PD-L1 expression in GL26 cells, suggesting they can act as immune checkpoint inhibitor. Further, they reduced tumor growth in GL26 mouse model. NCI-60 analysis showed they also inhibited the growth of colon cancer, leukemia, non-small cell lung and other cancers. Taken together, this study demonstrates MAO A/HSP90 dual inhibitors 4-b and 4-c reduced the growth of GBM and other cancers, and they have potential to inhibit tumor immune escape.

Graphical Abstract

1. Introduction

Glioblastoma (GBM) is the most aggressive primary brain tumor, which is responsible for majority deaths caused by primary brain tumor [1, 2]. Standard therapy, consisting with surgery, radiation therapy, and six maintenance cycles of temozolomide (TMZ) chemotherapy, medium survival time of patients with GBM are less than 2 years and five-year survival for GBM is 7% [2-4]. The poor survival of GBM indicates the urgent need in developing novel therapies for GBM [2, 5].

Monoamine oxidase A (MAO A) is a mitochondria isoenzyme which catalyzes the oxidative deamination of monoamine neurotransmitters and produces reactive oxygen species (ROS) [6, 7]. MAO A inhibitors have been used as antidepressants for decades [8]. The overexpression of MAO A has been shown in GBM, prostate cancer, and non-small cell lung cancer [9-12]. MAO A inhibitors reduce the tumor growth and increase the survival of mice bearing GBM, prostate cancer, and non-small cell lung cancer [10-14]. Thus MAO A is a valuable target for developing GBM treatment.

Heat shock protein 90 (HSP90) is a target for the GBM treatment. It is a chaperon protein which assists protein folding, transport, and degradation. HSP90 is essential for cell signal transduction and cell responses to stress [15]. HSP90 plays an important role for cancer cell to response to cellular stress and maintain cellular homeostasis [15, 16]. The expression of HSP90 is elevated in GBM cell lines and tissue compared to normal brain tissue [17-19]. HSP90 inhibitors can block the migration and invasion. It has been shown that HSP90 inhibitors, STA-9090 and AUY-922 inhibit the proliferation and induce pro-apoptotic effects in GBM in vitro and in vivo [20-22]. Overall, HSP90 can be recognized as a promising target for the treatment of GBM.

MAO A and HSP90 are both potential GBM targets. In this study, we designed a series of compounds with dual MAO A and HSP90 inhibition. Existing MAO A and HSP90 inhibitors were used as reference. Clorgyline is a selective MAO A inhibitor with high affinity. HSP90 inhibitors, AUY922 and STA-9090, both have 4-isopropylresorcinol group connected with 5-membered heterocycles, suggesting this group might be pharmacophore. The heterocyclic-based compounds have been an attractive structure and shown potential in various diseases [23, 24]. The heterocycle of HSP90 inhibitors, isoxazole of AUY922 and triazole of STA-9090, have been proven to be crucial in HSP90 ATP binding site and contributed to the pharmacokinetics properties [25, 26]. Both isoxazole and triazole are the bioisosteres of amide bond [27]. The design strategy here is to combine the pharmacophore of HSP90 inhibitors, 4-isopropylresorcinol with MAO A inhibitor, clorgyline by an amide linker mimicking the configuration of 5-membered heterocycles of HSP90 inhibitor with a ring-opening concept (Figure 1A). The 4-isopropylresorcinol was connected to 3-, 4-, or 5-position on phenyl ring of clorgyline through a secondary or tertiary amide linker with methyl, ethyl, propyl substituents (Figure 1B). A total of 12 compounds were synthesized and evaluated for their anti-GBM effect. The effective MAO A/HSP90 inhibitors were further studied for their mechanism and in vivo function.

Figure 1. Design rationale and structures of MAO A/HSP90 inhibitors.

(A) Structures of HSP90 inhibitors (AUY-922 and STA-9090, with 4-isopropylresorcinol group), MAO A inhibitor (clorgyline), and the structural-design rationale of MAO A and HSP90 dual inhibitors, (B) Structures of MAO A/HSP90 inhibitors in this study.

2. Result and discussion

2.1. Chemistry

MAO A/HSP90 inhibitors were synthesized by linking MAO A inhibitor, clorgyline, with pharmacophore of HSP90 inhibitor, 4-isopropylresorcinol through an amide linker. Isopropylresorcinol was linked with phenyl ring of clorgyline at 3-, 4-, or 5-position. Different alkyl groups were added on the amide linker and yielded compounds a (-H), b (-CH₃), c (-C₂H₅), and d (-C₃H₇) (Figure 1B).

MAO A/HSP90 inhibitors were synthesized as described in Scheme 1. Nucleophilic substitution of 2-cloro-nitrophenol with 1,3-dibromopropane provided compounds 3-2, 4-2, and 5-2. Nucleophilic substitution with N-methyl propargylamine yielded compounds 3-3, 4-3, and 5-3. Reduction of aromatic nitro group generated compounds 3-4, 4-4, and 5-4. Nucleophilic substitution of compounds 3-4 and 5-4 with produced acyl chloride provided compounds 3-5 and 5-5. Amide coupling of compound 4-4 with 2,4-bis(benzyloxy)-5-isopropyl benzoic acid generated compound 4-5. Cleavage of benzyl group yielded compounds 3-a, 4-a, and 5-a. Alkylation with various alkyl iodides under basic condition generated compounds 3-6-8; 4-6-8; and 5-6-8, respectively. Deprotection yielded compounds 3-b-d; 4-b-d; and 5-b-d, respectively. Details on the synthesis, isolation, and characterization of compounds can be found in experimental section and supporting information. The estimated purity of compounds 3-a-d, 4-a-d and 5-a-d is at least 95% as determined by HPLC analysis (supporting information).

Scheme 1. Synthesis of MAO A/HSP90 inhibitors 3-a-d, 4-a-d and 5-a-d^a.

^a Reagents and conditions: (1) 1,3-dibromopropane, Cs₂CO₃, ACN, reflux 80°C 11h; (2) N-methyl propargylamine, K₂CO₃, ACN, rt 2 days; (3) Fe, NH₄Cl, 90% C₂H₅OH in H₂O (9: 1), reflux 120°C 3h; (4) (i) 2,4-bis(benzyloxy)-5-isopropyl benzoic acid, SOCl₂, reflux 80°C 3-4h (ii) TEA, THF, 0°C to rt 8h, for 3-4 and 5-4; (5) 2,4-bis(benzyloxy)-5-isopropyl benzoic acid, EDC-HCl, HOBt, DIPEA, DMF, rt 6h, for 4-4; (6) various alkyl iodides, t-BuOK, THF, 0°C to rt overnight, for 3-6, 3-7, and 3-8; (7) various alkyl iodides, NaH, DMF, 0°C to rt overnight, for 4-6, 4-7, 4-8, 5-6, 5-7 and 5-8; (8) BCl₃ in hexane, 0°C to rt 10 min.

2.2. Biology

2.2.1. Inhibition of MAO A catalytic activity by MAO A/HSP90 inhibitors

The inhibition of MAO A catalytic activity of 12 MAO A/HSP90 inhibitors were determined in GBM GL26 cells, the results were shown in Table 1.

Table 1.

Effect of MAO A/HSP90 inhibitors on MAO A inhibition in mouse GL26 cells.

Compound	Position of 4- isopropylresorcinol (IPR) at the phenyl ring of clorgyline	Alkyl group	MAO A inhibition IC50 (μM)
3-a	3	R = H	2.36 ± 0.40
3-b	3	R = CH3	> 10
3-c	3	R = C2H5	2.27 ± 0.73
3-d	3	R = C3H7	0.88 ± 0.09
4-a	4	R = H	0.11 ± 0.01
4-b	4	R = CH3	1.77 ± 0.26
4-c	4	R = C2H5	4.58 ± 0.43
4-d	4	R = C3H7	3.86 ± 0.41
5-a	5	R = H	2.36 ± 0.19
5-b	5	R = CH3	> 10
5-c	5	R = C2H5	> 10
5-d Clorgyline STA-9090	5	R = C3H7	0.86 ± 0.01 0.0015 ± 0.0001 > 10

Among compounds with 4-isopropylresorcinol (IPR)at 3-position, compound 3-d with propyl substituent (-C₃H₇) showed more potent (IC₅₀: 0.88 μM) than other substitution. Compound 3-b with methyl substituent (-CH₃) was less potent (> 10 μM). Compound 3-a with hydrogen (-H, IC₅₀: 2.36 μM) and 3-c with ethyl (-C₂H₅, IC₅₀: 2.27 μM) showed moderate potency in between 3-d and 3-b. This suggests hydrogen (3-a) and ethyl (3-c) substituents has no difference. Effect of MAO A inhibition is not according to the size of alkyl group.

Among compounds with IPR at 4-position, compound 4-a with hydrogen (-H) showed more potent (IC₅₀: 0.11 μM) than other substitutions. Compound 4-b with methyl (-CH3, IC₅₀: 1.77 μM), 4-c with ethyl (-C₂H₅, IC₅₀: 4.58 μM) and 4-d with propyl (-C₃H₇, IC₅₀: 3.86 μM) showed less effective than compound 4-a. This result indicated that the replacement of hydrogen (4-a) with methyl (4-b), ethyl (4-c), or propyl (4-d) group of compounds with IPR at 4-position decrease MAO A inhibition.

Among compounds with IPR at 5-position, compound 5-d with a propyl substituent (-C₃H₇, IC₅₀: 0.86 μM) exhibited more potent. Compound 5-a with hydrogen (-H, IC₅₀: 2.36 μM) showed less effective. Compounds 5-b with methyl (-CH₃) and 5-c with ethyl substituents (-C₂H₅) showed less potent.

In brief, compounds with IPR at 3-position and 4-position showed similar MAO A inhibition better than 5-position, with the exception of 3-b. Compound 4-a, 3-d, and 5-d showed most potent MAO A inhibition. The addition of IPR on phenyl group of clorgyline decreased MAO A inhibition compared to clorgyline (IC₅₀: 0.0015 μM) alone.

2.2.2. Effect of compounds 4-a-d on HSP90 inhibition

Compounds 4-a, 4-b, 4-c, and 4-d with IPR at 4-position showed more effective MAO A inhibition than other position (Table 1). Thus, compounds 4-a-d were further evaluated for their HSP90 inhibition using assay based on binding competition of fluorescently labeled geldanamycin (FITC-GM) with human HSP90α (Table 2). Geldanamycin is a known HSP90 inhibitor. Although compound 4-a with hydrogen (-H) showed highest MAO A inhibition among all compounds, 4-a was less potent (> 10 μM). Compound 4-b with methyl (-CH₃, IC₅₀: 0.019 μM) and 4-c with ethyl substituent (-C₂H₅, IC₅₀: 0.016 μM) showed equal potency. Compound 4-d (IC₅₀: 0.17 μM) showed about 10-fold less effective than 4-b and 10c. This result indicated that the secondary amide (4-a) may be unfavorable for HSP90 inhibition compared to tertiary amide (4-b, 4-c, 4-d). The propyl group (4-d) might contribute to the steric hindrance compared to methyl (4-b) and ethyl (4-c) substituents. Although compounds 4-b and 4-c showed equal potency compared to geldanamycin, none of these compounds were better than geldanamycin or STA-9090.

Table 2.

IC₅₀ of HSP90α inhibition by MAO A/HSP90 inhibitors 4-a, 4-b, 4-c and 4-d.

Compound	Position of IPR at the phenyl ring of clorgyline	Alkyl group	HSP90α inhibition IC50 (μM)
4-a	4	R = H	> 10
4-b	4	R = CH3	0.019
4-c	4	R = C2H5	0.016
4-d	4	R = C3H7	0.170
Geldanamycin			0.013
STA-9090			0.0065 [28]

2.2.3. Inhibition of cell growth by MAO A/HSP90 inhibitors in GBM cell lines

The inhibition of cell growth by MAO A/HSP90 inhibitors was determined in three GBM cell lines, mouse GL26, human U251S (TMZ-sensitive), and U251R (TMZ-resistant) GBM cells using MTT assay. The results were shown in table 3.

Table 3.

The effects of MAO A/HSP90 inhibitors on cell growth inhibition (IC₅₀) in three GBM cell lines.

Compound	Position of IPR at the phenyl ring of clorgyline	Alkyl group	Cell growth inhibition IC50 (μM)
			GL26	U251S	U251R
4-a	4	R = H	> 10	> 10	> 10
4-b	4	R = CH3	0.73 ± 0.06	1.68 ± 0.40	0.84 ± 0.25
4-c	4	R = C2H5	0.49 ± 0.28	1.09 ± 0.02	0.26 ± 0.10
4-d	4	R = C3H7	8.42 ± 0.85	> 10	> 10
Clorgyline			> 100	175 [11]	136 [11]
STA-9090			0.003 ± 0.0006	6.84 ± 1.07	> 10

Among compounds with IPR at 3-position (table not shown), compounds 3-b with methyl (-CH₃) and 3-c with ethyl substituent (-C₂H₅) showed similar growth inhibition in GL26 cells with IC₅₀ 4.48 ± 0.83 and 4.58 ± 0.28 μM, respectively. But they have less effect in U251S and U251R (> 10 μM). Compounds 3-a with hydrogen (-H) and 3-d with propyl substituent (-C₃H₇) were less potent (> 10 μM).

Among compounds with IPR at 4-position, compound 4-b with methyl (-CH₃) and 4-c with ethyl substituent (-C₂H₅), respectively, exhibited potent cell growth inhibition than other substitutions (Table 1). Compound 4-b showed IC₅₀ 0.73 μM in GL26, 1.68 μM in U251S, and 0.84 μM in U251R, 2-fold more potent in U251R than U251S. Similarly, compound 4-c showed potent growth inhibition with IC₅₀ 0.49 μM in GL26, 1.09 μM in U251S, 0.26 μM in U251R, 5-fold more potent in U251R than U251S. Both 4-b and 4-c have similar effect. These results suggest methyl (4-b) and ethyl (4-c) substituents have no difference in growth inhibition. Compounds 4-b and 4-c showed more effective growth inhibition than 4-a with hydrogen (-H, >10 μM) and 4-d with propyl (-C₃H₇, 8.4 μM in GL26), and clorgyline or STA-9090 alone, and more potent in TMZ-resistant U251R cells than TMZ-sensitive U251S cells.

Compounds with IPR at 5-position were less potent in all three cell lines (> 10 μM, table not shown).

These results show compounds with IPR at 4-position (4-b and 4-c) are more effective than 3- and 5-position in GBM cells. No difference in methyl (4-b) and ethyl (4-c) substituents was found. Compounds 4-b and 4-c showed more potent than clorgyline or STA-9090 alone in human GBM cells. They are more potent in TMZ-resistant U251R cells than in TMZ-sensitive U251S cells.

2.2.4. Inhibition of cell growth by compounds 4-b and 4-c in GBM cell lines

GBM is the most aggressive brain tumor and its heterogeneity increases the difficulty in the treatment. Herein, MAO A/HSP90 dual inhibitors were further tested on eight human GBM cell lines (Table 4.). These cell lines have different mutation and characteristics, which are important to be considered during drug development. U87MG cells have wild-type EGFR and p53, mutated PTEN, and decreased neurofibromin. U373MG cells have mutated IDH and p53 [29]. LN229 cells showed reduced expression of tumor suppressor and DNA repair-related genes p53, ATM, BRCA1 and BRCA2 [30]. A172 cells have large molecular weight EGFR and high expression of mesenchymal markers, CD90 and CD105, fibroblast activation protein, and tenascin C. While T98G cells lack of these antigen compared to A172 cells [29, 31]. P1S cells are patient-derived, TMZ-resistant and concurrent chemoradiotherapy (CCRT)-resistant GBM cells with positive O⁶-Methylguanine-DNA methyltransferase (MGMT) [32]. P3 cells are patient-derived GBM cells. P3R cells are patient-derived TMZ-resistant GBM cells with MGMT negative [32]. TMZ resistant characteristics of P3R and P1S cells was confirmed in the previous study [33].

Table 4.

The effects of MAO A/HSP90 inhibitors 4-b and 4-c on cell growth inhibition (IC₅₀) in eight human GBM cell lines.

GBM cell lines	IC50 (μM)		IC50 ratio (4-b/4-c)
	4-b	4-c
U87MG	0.35	0.36	0.97
U373MG	1.46	0.89	1.64
LN229	0.40	0.31	1.29
A172	3.31	0.89	3.72
T98G	0.46	0.46	1
P1S	0.86	0.68	1.26
P3	2.73	2.93	0.93
P3R	2.53	1.45	1.74

Compounds 4-b and 4-c showed moderate MAO A inhibition and effective HSP90 inhibition and they are most effective in cell growth inhibition in three GBM cell lines. Thus the growth inhibition of compounds 4-b and 4-c in these eight human GBM cell lines was further determined using WST-8 (CCK8) assay (Table 4.). Compounds 4-b and 4-c showed similar effectiveness in cell growth inhibition in these GBM cell lines. The IC₅₀ ratio of compound 4-b over 4-c were around 1 in most of cell lines, except A172. In A172 cells, compound 4-c exhibited 3.72-fold better cytotoxicity than compound 4-b.

2.2.5. Molecular docking and analysis of binding site interaction of compounds 4-b and 4-c at MAO A or HSP90

The structures of compounds 4-b and 4-c are designed based on HSP90 (4-isopropylresorcinol) and MAO A (clorgyline) inhibitors. These compounds were docked into the HSP90 and MAOA binding sites to investigate interactions they are able to establish. Both compounds form three hydrogen bonds to the same residues in the HSP90 binding site, including D54, D93, and T184 (Figure 2A). A hydrogen bond is present with the tertiary nitrogen and residue D54. The 4-isopropylresorcinol end of compounds 4-b and 4-c contains a hydroxyl group at the 1’ position of the aromatic ring that forms a hydrogen bond to residue D93. Previous reports have shown residue D93 contributes to HSP90 inhibitor binding [34]. A carbonyl present in both compounds facilitates an additional hydrogen bond to residue T184. Together, these hydrogen bonds may function as an anchor for the compounds in the HSP90 binding site. Additionally, several hydrophobic contacts were observed in the HSP90 binding site. This includes residues A55, M98, L107, and F138. The hydrophobic pocket generated by these residues makes contact with the aromatic rings of compounds 4-b and 4-c.

Figure 2. Docking poses of compounds 4-b and 4-c in HSP90 and MAO A.

Compounds 4-b (top) and 4-c (bottom) are docked into (A) HSP90 (PDB ID: 2VCI) and (B) MAO A (PDB ID: 2BXR) binding site. Both compounds generated similar interactions in the binding site. Docked compounds are shown in blue. HSP90 (gray) and MAO A (yellow) are shown as cartoons with binding site residues depicted as sticks. Hydrogen bonds are shown by green dotted lines.

In the MAO A binding site, compounds 4-b and 4-c also share similar interactions. A flavin-adenine dinucleotide (FAD) coenzyme and adjacent tyrosine residues Y407 and Y444 generate an “aromatic cage” within the MAO A binding site [35-37]. The inhibitors form a direct covalent bond to the FAD coenzyme through the covalent warhead located on the clorgyline moiety (Figure 2B). The MAO A binding site is a hydrophobic cavity [36, 37]. As a result, a series of hydrophobic contacts are generated by compounds 4-b and 4-c. This includes residues L97, F173, F208, V210, I325, and L337. Residue I335 may aid in MAO A selectivity, while residue Y407 forms part of the “aromatic cage” [37]. Both residues I335 and Y407 also contribute to hydrophobic contacts to compounds 4-b and 4-c. Hydrogen bonds are also observed with the 4-isopropylresorcinol terminal end of compounds 4-b and 4-c. Two hydrogen bonds are formed between the two hydroxyl moieties and residues G110 and V210. Overall, the structural similarity between compounds 4-b and 4-c generates similar interactions in the respective protein binding sites.

2.2.6. Effect of compounds 4-b and 4-c on protein expression in GL26 and U251R cells.

Activity of protein kinase B (PKB, Akt) or phosphor-Akt is increased in human GBM and help GBM cells proliferation and enhance tumor invasion [38, 39]. Our previous research showed that MAO A deletion reduce Akt phosphorylation and decrease tumor proliferation in mouse model of prostate cancer [40]. Akt is a client of HSP90. HSP90 inhibitors decrease Akt activity and induce apoptosis in GBM [18-20]. The Human epidermal growth factor receptor 2 (HER2) is overexpressed by up to 80% of GBM cases which is related to poor prognosis [41, 42]. HER1 and HER2 are both clients of HSP90, and inactivation of HSP90 leads to depletion and degradation of HER1/2 [43]. Thus, targeting Akt and HER2 via MAO A/HSP90 inhibition might be a strategy to decrease GBM progression.

Compounds 4-b and 4-c were further evaluated for their pharmacodynamics, and western blot analysis was performed (Figure 3). GL26 and U251R cells were treated with 4-b, 4-c, clorgyline or STA-9090 at indicated concentration for 24 hours. The result showed that 4-b and 4-c concentration-dependently decreased HER2 and phospho-Akt expression similar to STA-9090 effect. Clorgyline also decreased phospho-Akt. Induction of HSP70 in response to HSP90 inhibition has been expected and used to indicate the resulting heat shock protein response when loss of HSP90 function [44]. Increases of HSP70 expression were observed in 4-b and 4-c treated cells in a dose-dependent manner. These results demonstrated that the mechanism of action of 4-b and 4-c was via inhibiting MAO A and HSP90 and thus reduce the proliferation of GBM cells.

Figure 3. The effect of MAO A/HSP90 inhibitors 4-b and 4-c on protein expression of GL26 and U251R cell lines using Western blot.

GL26 and U251R cells were incubated with vehicle, STA-9090, clorgyline, MAO A/HSP90 inhibitors 4-b and 4-c at indicated concentration (3-fold dilution from 3 μM to 0.1 μM) for 24 h. Cells were lysed and protein expression was analyzed using Western blot. HER2, HSP70, and phospho-Akt (Ser 473) expression in treated GL26 (A) and U251R (B) cells. Protein expression of each samples was first normalized to the amount of GAPDH and then normalized to the vehicle (no-treatment control).

2.2.7. MAO A/HSP90 inhibitors 4-b and 4-c decrease PD-L1 expression in mouse GBM cells

Programmed cell death protein 1 ligand 1 (PD-L1) expressed by tumor cells, can be identified by activated T cell with cell surface receptor, programmed cell death protein 1 (PD1). This PD1/PD-L1 pathway inhibit the activation of T cell as immune checkpoint and lead to tumor immune escape. The expression of PD-L1 can be increased by multiple proinflammatory molecules in tumor microenvironment (TME), including interferon-γ (IFN-γ), tumor necrosis factor (TNF-α), vascular endothelial growth factor (VEGF), IL-10, and IL-4. Although IFN-γ produced by T cell was considered as an anti-inflammatory cytokine, IFN-γ is the most potent inducer of PD-L1, which promote tumor immune escape. Also, IFN-γ mainly regulates PD-L1 expression in the GBM [45-47]. HSP90 inhibitors can decrease the expression of PD-L1. Thus, HSP90 inhibitors can increase the amount of activated T cells [48]. Further, MAO A deletion reduced immune suppression and hence stimulate immune, especially cytotoxic T cell in prostate cancer [49].

To evaluate the effect of 4-b and 4-c on PD-L1 expression, GL26 mouse GBM cells were treated with 4-b, 4-c, or clorgyline at indicated concentration for 24 h with or without 10 ng/mL of IFN-γ. After 24 h, PD-L1 expression was determined. Results showed that both 4-b and 4-c decreased IFN-γ induced PD-L1 expression concentration-dependently, with similar potency (Figure 4). MAO A inhibitor, clorgyline, the parent compound was less effective, showed decreased PD-L1 expression at high concentration, 20 μM. Compounds 4-b and 4-c decreased PD-L1 expression in HCT116 human colon cancer cells as well (Figure S49). These results indicated compounds 4-b and 4-c may have potential to be immune checkpoint inhibitors.

Figure 4. MAO A/HSP90 inhibitors 4-b and 4-c decreased IFN-γ induced PD-L1 expression in GL26 cells concentration-dependently.

GL26 were treated with 4-b, 4-c, or clorgyline at various concentration with or without 10 ng/mL IFN-γ for 24 h. (A) PD-L1 surface expression was then determined using flow cytometry analysis and represented as fluorescence intensity (FL1-H). (B)(C)(D) The mean fluorescence intensity (MFI) of each samples (PD-L1 antibody) after normalized to isotype control (secondary antibody alone) were represented as MFI ratio using bar figures, (B) compound 4-b, (C) compound 4-c, and (D) clorgyline.

2.2.8. Effect of MAO A/HSP90 inhibitors 4-b and 4-c on tumor size in GBM mouse model.

GL26 cells were subcutaneously implanted in the lower back of mice. On day 9 when tumor was palpable, the mice were treated with vehicle, clorgyline (10 mg/kg), compounds 4-b (25 mg/kg), and 4-c (25 mg/kg) intraperitoneally (IP) for 14 days. Tumor size was measured every days using caliper. At day 9 of treatment, tumor growth inhibition (TGI%) of compounds 4-b or 4-c treated group were 63.2% or 59.8% decreased compared to vehicle group. Clorgyline had no effect (Figure 5A). However, mice with 4-b treatment showed decreased body weight started at day 3 of treatment (Figure 5B). Mice were euthanized once their tumor size reach endpoint (diameter > 1.5 cm or volume >1.5 cm³) in accordance to the policy of Institutional Animal Care and Use Committee (IACUC). The result indicated that mice treated with compounds 4-b and 4-c showed significantly decreased tumor growth.

Figure 5. MAO A/HSP90 inhibitors 4-b and 4-c reduced tumor growth in GL26 mouse model.

C57BL/6J mice were implanted subcutaneously with 2 × 10⁵ GL26 cells. Nine days post tumor implantation, mice were treated with vehicle, clorgyline (10 mg/kg) daily, compounds 4-b (25 mg/kg) and 4-c (25 mg/kg) intraperitoneally daily for 3 days followed with every 3 days for a 14-day period. Mice were euthanized once tumor size reached endpoint (diameter > 1.5 cm or volume > 1.5 cm³) or based on health status. (A) Tumor size was determined by calipers. Percentage of tumor growth inhibition (TGI%) of compounds 4-b and 4-c was calculated compared to the average tumor size of vehicle treated group. All data presented as mean ± SD, *** P value < 0.01. (B) Body weight of mice were normalized to their weight of the day tumor implanted.

2.2.9. Cell growth inhibition of 4-b and 4-c in NCI-60 cell lines

Compounds 4-b and 4-c were submitted to the Developmental Therapeutics Program (DTP) of the National Cancer Institute (NCI) to evaluate their growth inhibition in NCI-60 human cancer cell lines. The concentration of 50% growth inhibition (GI₅₀), total growth inhibition (TGI), and lethal concentration (LC₅₀) of compounds 4-b and 4-c were shown (Figure S50 and S51). Both compounds 4-b and 4-c exhibited similar potency in all 60 cancer cell lines. Similar result was found in other human GBM cell lines and many other cancer, including leukemia, non-small cell lung cancer, and colon cancer (Table 5).

Table 5.

Growth inhibition (GI₅₀) of MAO A/HSP90 inhibitors 4-b and 4-c in human cancer cell lines determined by NCI-60 analysis.

Cancer cell lines	GI50 (μM)		GI50 ratio (4-b/4-c)
	4-b	4-c
Glioblastoma
SF-268	0.52	1.43	0.36
SF-295	0.29	0.32	0.91
SF-539	0.32	0.31	1.03
SNB-19	0.44	0.44	1
SNB-75	0.42	0.86	0.49
U251	0.32	0.37	0.86
Leukemia
HL-60	0.04	0.06	0.67
K-562	0.07	0.07	1
RPMI-8226	0.38	0.33	1.15
Non-small cell lung cancer
A549	0.19	0.14	1.36
NCI-H460	0.04	0.05	0.8
NCI-H522	0.08	0.07	1.14
Colon cancer
HCT-116	0.08	0.08	1
KM12	0.05	0.06	0.83
SW-620	0.05	0.24	0.21

3. Conclusion

In this study, a series of MAO A/HSP90 inhibitors (3-a-d, 4-a-d and 5-a-d) were synthesized. These inhibitors were composed by 4-isopropylresorcinol conjugated with phenyl group of clorgyline at 3-, 4-, 5-position through a secondary or tertiary amide linkage. The tertiary amide linker was substituted with methyl, ethyl, or propyl groups. Among these derivatives, compounds with IPR at 4-position, 4-b with methyl and 4-c with ethyl substituent, showed moderate MAO A inhibition and effective HSP90 inhibition. MAO A/HSP90 inhibitors 4-b and 4-c exhibited most effective cell growth inhibition than other compounds, clorgyline, or STA-9090 alone. 4-b and 4-c showed growth inhibition in TMZ-sensitive and TMZ-resistant GBM cell lines. In western blot analysis, 4-b and 4-c decreased HER2, phospho-Akt and increased HSP70, which demonstrated the mechanism of action via inhibiting MAO A and HSP90. Further, compound 4-b and 4-c decreased PD-L1 expression in GL26 GBM cells, which indicated they could be developed as immune checkpoint inhibitors. In GBM GL26 mouse models, 4-b and 4-c inhibited tumor growth by 63.2% and 59.8% size compared to vehicle group. In NCI-60 analysis, 4-b and 4-c exhibited effective cell growth inhibition in other human GBM and other cancer cell lines. In conclusion, this study showed MAO A/HSP90 inhibitors 4-b and 4-c are potential antineoplastic agent for the treatment of GBM and other cancers.

4. Experimental section

4.1. Chemistry

The progress of the reaction was monitored by Thin Layer Chromatography (TLC) using pre-coated sheets ALUGRAM Xtra SIL G/UV₂₅₄ Macherey Nagel. The samples were concentrated by Buchi rotary evaporator R-215, connected with heating bath B-491. NMR spectrum (¹H and ¹³C NMR) were obtained using the Bruker Fourier 300 and Agilent 600 DD2 NMR Spectrometer. The reported chemical shifts were presented in part per million (ppm, δ) with TMS as internal standard. The high resolution mass spectrum (HR-MS) were obtained using JEOL (JMS 700) electron impact (EI) mass spectrometer. Unless otherwise mentioned, compounds were purified using chromatography on a silica gel column (Merck Kieselgel 60, 230-400 mesh ASTM). The estimated purity of the compounds was determined using Hitachi 2000 series HPLC system with a C-18 column (Agilent ZORBAX Eclipse XDB-C18 5mm, 4.6mm, 150 mm). The purity of the final compounds was at least 95%.

4.1.1. 1-(3-Bromopropoxy)-2-chloro-3-nitrobenzene (3-2)

To a solution of 2-chloro-3-nitrophenol (5 g, 28.8 mmol) in 20 mL acetonitrile (ACN) was added into a mixture of 1,3-dibromopropane (23 mL, 231.2 mmol) and CS₂CO₃ (18.7 g, 57.4 mmol) in 30 mL ACN. The resultant solution was heated to reflux under N₂ atmosphere for 11 h. The reaction mixture was suspended in H₂O (300 mL) and extracted with EtOAc (3 × 300 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 20% EtOAc in n-hexane to give compound 3-2 in 92% yield. ¹HNMR (300 MHz, DMSO-d₆): δ_H 7.63-7.59 (m, 1 H), 7.57-7.51 (m, 1H), 4.29 (t, J = 6.0 Hz, 2H), 3.71 (t, J = 6.3 Hz, 2H), 2.33 (p, J = 6.3 Hz, 2H).

4.1.2. 1-(3-Bromopropoxy)-2-chloro-4-nitrobenzene (4-2)

Following the procedure as described for the preparation of 3-2, reaction of 2-chloro-4-nitrophenol (2 g, 11.5 mmol), 1,3-dibromopropane (9.2 mL, 92 mmol), and CS₂CO₃ (7.5 g, 23 mmol) in ACN (50 mL) gave compound 4-2 in 97% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 8.31 (d, J = 2.7 Hz, 1H), 8.23 (dd, J = 9.3, 2.7 Hz, 1H), 7.63 (d, J = 9.0 Hz, 1H), 4.33 (t, J = 6.0 Hz, 2H), 3.68 (t, J = 6.3 Hz, 2H), 2.33 (p, J = 6.3 Hz, 2H).

4.1.3. 2-(3-Bromopropoxy)-l-chloro-4-nitrobenzene (5-2)

Following the procedure as described for the preparation of 3-2, reaction of 2-chloro-4-nitrophenol (3 g, 17.29 mmol), 1,3-dibromopropane (13 mL, 138.32 mmol), and CS₂CO₃ (11 g, 34.60 mmol) in ACN (50 mL) gave compound 5-2 in 95% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.83-7.78 (m, 2H), 7.53-7.50 (m, 1H), 4.28 (t, J = 5.7 Hz, 2H), 3.66 (t, J = 6.3 Hz, 2H), 2.42 (p, J = 6.0 Hz, 2H).

4.1.4. N-(3-(2-Chloro-3-nitro phenoxy) propyl)-N-methyl prop-2-yn-l-amine (3-3)

To a solution of compound 3-2 (4.5 g, 15.3 mmol) in 20 mL ACN was added K₂CO₃ (2.37 g, 16.8 mmol) in 7 mL H₂O, followed by N-methyl propargylamine (2.6 mL, 30.6 mmol). The resultant solution stirred under N₂ atmosphere at room temperature for 2 days. The reaction mixture was suspended in H₂O (300 mL) and extracted with EtOAc (3 × 300 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 50% EtOAc in n-hexane to give compound 3-3 in 76% yield. ¹HNMR (300 MHz, DMSO-d₆): δ_H 7.56-7.42 (m, 3H), 4.16 (t, J = 6.0 Hz, 2H), 3.29 (d, J = 2.4 Hz, 2H), 3.07 (t, J = 2.4 Hz, 1H), 2.53-2.49 (m, 2H), 2.19 (s, 3H), 1.87 (p, J = 6.6 Hz, 2H).

4.1.5. N-(3-(2-Chloro-4-nitro phenoxy) propyl)-N-methyl prop-2-yn-l-amim (4-3)

Following the procedure as described for the preparation of 3-3, reaction of compound 4-2 (2.7 g, 9.17 mmol), in 20 mL ACN was added K₂CO₃ (1.39 g, 10.1 mmol) in 7 mL H₂O, followed by N-methyl propargylamine (1.5 mL, 18.3 mmol) to give compound 4-3 in 66% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 8.26 (d, J = 3.0 Hz, 1H), 8.18 (dd, J = 9.3, 2.7 Hz, 1H), 7.33 (d, J = 9.3 Hz, 1H), 4.22 (t, J = 6.0 Hz, 2H), 3.15 (d, J = 2.4 Hz, 2H) 3.06 (t, J = 2.4 Hz, 1H), 2.53 (t, J = 6.9 Hz, 2H), 2.20 (s, 3H), 1.89 (p, J = 6.3 Hz, 2H).

4.1.6. N-(3-(2-Chloro-5-nitro phenoxy) propyl)-N-methyl prop-2-yn-l-amine (5-3)

Following the procedure as described for the preparation of 3-3, reaction of compound 5-2 (2.7 g, 9.17 mmol), in 20 mL ACN was added K₂CO₃ (1.39 g, 10.1 mmol) in 7 mL H₂O, followed by N-methyl propargylamine (1.5 mL, 18.3 mmol) to give compound 5-3 in 66% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.79-7.75 (m, 2H), 7.51-7.47 (m, 1H), 4.19 (t, J = 6.3 Hz, 2H), 3.37 (d, J = 2.4 Hz, 2H), 2.66 (t, J = 6.9 Hz, 2H), 2.33 (s, 3H), 2.22 (t, J = 2.4 Hz, 1H), 2.03 (p, J = 6.6 Hz, 2H).

4.1.7. 2-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)aniline (3-4)

To a solution of compound 3-3 (3.5 g, 12.38 mmol), iron (3.40 g, 61.9 mmol), and ammonium chloride (1.32 g, 24.76 mmol) in 50 mL 10% EtOH in H₂O. The resultant mixture was heated to reflux at 120°C under N₂ atmosphere for 4 h. The reaction mixture was filtered through Celite pad and the filtrate was collected and concentrated to half in vacuo. The residue was suspended in H₂O (300 mL) and extracted with EtOAc (3 × 300 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo to give compound 3-4 in 65% yield. ¹HNMR (300 MHz, DMSO-d₆): δ_H 6.91 (dd, J = 8.1, 8.1 Hz, 1H), 6.38 (dd, J = 8.4, 1.2 Hz, 1H), 6.26 (dd, J = 8.1, 1.5 Hz, 1H), 5.26 (s, 2H), 3.96 (t, J = 6.3 Hz, 2H), 3.30 (d, J = 2.4 Hz, 2H), 3.09 (t, J = 2.4 Hz, 1H), 2.53-2.50 (m, 2H), 2.20 (s, 3H), 1.82 (p, J = 6.6 Hz, 2H).

4.1.8. 3-Chloro-4-(3-(methyl (prop-2-yn-l-yl) amino) propoxy) aniline (4-4)

Following the procedure as described for the preparation of 3-4, reaction of compound 4-3 (1.8 g, 6.4 mmol), iron (1.75 g, 3.2 mmol) and ammonium chloride (0.68 g, 12.8 mmol) in 20 mL 10% EtOH in H₂O gave compound 4-4 in 66% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 6.82 (d, J = 8.7 Hz, 1H), 6.62 (d, J = 2.4 Hz, 1H), 6.46 (dd, J = 8.7, 2.7 Hz, 1H), 4.88 (s, 2H), 3.88 (t, J = 6.3 Hz, 2H), 3.28 (d, J = 2.4 Hz, 2H), 3.08 (t, J = 2.1 Hz, 1H), 2.48 (t, J = 7.2 Hz, 2H), 2.19 (s, 3H), 1.77 (p, J = 6.6 Hz, 2H).

4.1.9. 4-Chloro-3-(3-(methyl (prop-2-yn-l-yl) amino)propoxy) aniline (5-4)

Following the procedure as described for the preparation of 3-4, reaction of compound 5-3 (2.77 g, 9.79 mmol), iron (2.70 g, 49 mmol) and ammonium chloride (1.04 g, 19.6 mmol) in 30 mL 10% EtOH in H₂O gave compound 5-4 in 84% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.07 (d, J = 8.4 Hz, 1H), 6.29 (d, J = 2.4 Hz, 1H), 6.22 (d, J = 8.4 Hz, 2.4 Hz, 1H), 4.05 (t, J = 6.0 Hz, 2H), 3.84 (br.s, 2H), 3.57 (d, J = 2.4 Hz, 2H), 2.91 (t, J = 7.5 Hz, 2H), 2.54 (s, 3H), 2.39 (t, J = 2.7 Hz, 1H), 2.13 (p, J = 6.3 Hz, 2H).

4.1.10. 2,4-Bis(benzyloxy)-N-(2-chloro-3-(3-(methylyl)amino)propoxy)phenyl)-5-isopropyl benzamide (3-5)

To a solution 2,4-bis(benzyloxy)-5-isopropyl benzoic acid (0.75g, 1.98 mmol) in 3 mL SOCl₂ was heated to reflux for 3 h. The resultant mixture was concentrated in vacuo to give crude mixture of 2,4-bis(benzyloxy)-5-isopropylbenzoyl chloride. To the solution of 2,4-bis(benzyloxy)-5-isopropylbenzoyl chloride in 5 mL THF was added the solution of compound 3-4 (0.5g, 1.98 mmol) in 5 mL THF, and TEA (0.67 mL, 4.95 mmol) dropwise in an ice-bath. After 15 min, the ice-bath was removed and the resultant solution was stirred at room temperature for 8 h. The reaction mixture was suspended in H₂O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 50% EtOAc in n-hexane to give compound 3-5 in 60% yield. ¹HNMR (300 MHz, DMSO-d₆): δ_H 10.50 (s, 1H), 8.32 (dd, J = 8.7, 1.2 Hz, 1H), 8.17 (s, 1H), 7.40-7.34 (m, 10H), 7.21 (dd, J = 8.4, 8.4 Hz, 1H), 6.68 (dd, J = 8.4, 1.2 Hz, 1H), 6.51 (s, 1H), 5.31 (s, 2H), 5.00 (s, 2H), 4.10 (t, J = 6.0 Hz, 2H), 3.48 (br.s, 2H), 3.30 (p, J = 6.6 Hz, 1 H), 2.79 (t, J = 7.2 Hz, 2H), 2.45 (s, 3H), 2.30 (br.s, 1H), 2.09-2.00 (m, 2H), 1.24 (d, J = 6.9 Hz, 6H).

4.1.11. 2,4-Bis(benzyloxy)-N-(3-chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropylbenzamide (4-5)

To a solution of compound 4-4 (1.2 g, 4.75 mmol), EDC-HCl (1.47 g, 9.5 mmol), HOBt (1.08 g, 7.13 mmol), DIPEA (2.1 mL, 9.5 mmol) in 7.5 mL DMF was stirred in room temperature for 30 min. After 30 min, the solution was added 2,4-bis(benzyloxy)-5-isopropylbenzoic acid (1.78 g, 4.75 mmol). The resultant solution was stirred under N₂ atmosphere for 6 h. The reaction mixture was suspended in H₂O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 80% EtOAc in n-hexane to give compound 4-5 in 80% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 9.92 (s, 1H), 7.67 (s, 1H), 7.59-7.56 (m, 2H), 7.53-7.50 (m, 2H), 7.48-7.35 (m, 8H), 7.30 (dd, J = 8.7, 2.7 Hz, 1H), 7.04 (d, J = 9.0 Hz, 1H), 7.01 (s, 1H), 4.02 (t, J = 6.3 Hz, 2H), 3.30 (d, J = 2.4 Hz, 2H), 3.28-3.19 (m, 3H), 3.094 (t, J = 2.1 Hz, 1H), 2.51 (t, J = 1.8 Hz, 3H) coinciding with DMSO-d₆, 2.20 (s, 3H), 1.83 (p, J = 6.9 Hz, 2H), 1.17 (d, J = 6.6 Hz, 6H).

4.1.12. 2,4-Bis(benzyloxy)-N-(4-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl benzamide (5-5)

Following the procedure as described for the preparation of 3-5, a solution 2,4-bis(benzyloxy)-5-isopropyl benzoic acid (2.78 g, 7.4 mmol) in 4 mL SOCl₂ was heated to reflux for 3 h. The crude mixture of 2,4-bis(benzyloxy)-5-isopropylbenzoyl chloride in 5 mL THF was added the solution of compound 5-4 (2.33 g, 9.25 mmol) in 5 mL THF, and TEA (3.2 mL, 23.13 mmol) dropwise in an ice-bath to give compound 5-5 in 74% yield. ¹HNMR (300 MHz, CD₃OD): δ_H 7.97 (s, 1H), 7.48-7.28 (m, 10H), 7.02 (d, J = 8.7 Hz, 1H), (d, J = 2.1 Hz, 1H), 6.72 (s, 1H), 6.63 (dd, J = 8.7, 2.1 Hz, 1H), 5.10 (s, 1H), 5.09 (s, 1H), 3.75 (t, J = 6.0 Hz, 2H), 3.33 (d, J = 2.4 Hz, 2H), 3.29-3.18 (m, 1H), 2.65-2.60 (m, 3H), 2.31 (s, 3H), 1.88 (p, J = 6.3 Hz, 2H), 1.19 (d, J = 6.9 Hz, 6H).

4.1.13. 2,4-Bis(benzyloxy)-N-(2-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl-N-methylbenzamide (3-6)

To a solution of compound 3-5 (0.35 g, 0.57 mmol) in dry 5 mL THF was added potassium tert-butoxide (0.10g, 0.86 mmol) in an ice-bath, after 15 min, followed by methyl iodide (72 μL, 1.14 mmol). After another 15 min, the ice-bath was removed and the resultant mixture was stirred at room temperature under N₂ atmosphere for overnight. The reaction mixture was suspended in H₂O (50 mL) and extracted with EtOAc (3 × 50 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 80% EtOAc in n-hexane to give compound 3-6 in 54% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.37-7.27 (m, 10H), 7.04 (s, 1H), 6.90 (dd, J = 8.1, 8.1 Hz, 1H), 6.72-6.68 (m, 2H), 6.27 (s, 1H), 5.11-4.97 (m, 2H), 4.85 (s, 2H), 4.05-3.91 (m, 2H), 3.37 (s, 3H), 3.34 (d, J = 2.4 Hz, 2H), 3.16-3.07 (m, 1H), 2.64-2.58 (m, 2H), 2.32 (s, 3H), 2.20 (t, J = 2.4 Hz, 1H), 1.94 (p, J = 6.9 Hz, 2H), 1.05 (d, J = 6.9 Hz, 3H), 0.94 (d, J = 6.9 Hz, 3H).

4.1.14. 2,4-Bis(benzyloxy)-N-(3-chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl-N-methylbenzamide (4-6)

To a solution of compound 4-5 (0.40 g, 0.65 mmol) in 5 mL DMSO was added anhydrous NaH (31 mg, 1.3 mmol) in an ice-bath, after 15 min, followed by methyl iodide (57 μL, 0.91 mmol). After another 15 min, the ice-bath was removed and the resultant mixture was stirred at room temperature under N₂ atmosphere for overnight. The reaction mixture was suspended in H₂O (50 mL) and extracted with EtOAc (3 × 50 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 80% EtOAc in n-hexane to give compound 4-6 in 87% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 7.42-7.33 (m, 9H), 7.19 (br.s, 1H), 6.94 (br.s, 1H), 6.61 (br.s, 1H), 3.97 (br.s, 1H), 3.26-3.23 (m, 6H), 3.08 (s, 2H), 2.46-2.43 (m, 2H), 1.79 (p, J = 4.5 Hz, 2H), 1.01 (d, J = 6.3 Hz, 6H).

4.1.15. 2,4-Bis(benzyloxy)-N-(4-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl-N-methylbenzamide (5-6)

Following the procedure as described for the preparation of 4-6, reaction of compound 5-5 (0.69 g, 1.13 mmol), anhydrous NaH (54 mg, 2.26 mmol), methyl iodide (100 μL, 1.58 mmol) in 5 mL DMSO gave compound 5-6 in 34% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.39-7.26 (m, 10 H), 7.13 (br.s, 1H), 7.08 (d, J = 8.1 Hz, 1H), 6.50 (br.s, 2H), 6.26 (br.s, 1H), 4.92 (s, 2H), 4.83 (s, 2H), 3.61 (br.s, 2H), 3.40 (s, 3H), 3.26-3.19 (m, 3H), 2.52 (t, J = 6.9 Hz, 2H), 2.28 (s, 3H), 2.19 (t, J = 2.4 Hz, 1H), 1.80 (p, J = 6.6 Hz, 2H), 1.13 (d, J = 6.9 Hz, 6H).

4.1.16. 2,4-Bis(benzyloxy)-N-(2-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-N-ethyl-5-isopropylbenzamide (3-7)

Following the procedure as described for the preparation of 3-6, reaction of compound 3-5 (0.60 g, 0.98 mmol), potassium tert-butoxide (165 mg, 1.47 mmol), ethyl iodide (158 μL, mmol) in 5 mL THF gave compound 3-7 in 64% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.48-7.26 (m, 10 H), 7.10 (s, 1H), 6.90 (dd, J = 8.1, 8.1 Hz, 1H), 6.70 (dd, J = 8.4, 7.8 Hz, 2H), 6.34 (s, 1H), 5.07-4.97 (m, 2H), 4.88 (s, 2H), 4.44-4.32 (m, 1H), 4.03-3.89 (m, 2H), 3.48-3.41 (m, 1H), 3.32 (d, J = 2.4 Hz, 2H), 3.21-3.12 (m, 1H), 2.66-2.56 (m, 2H), 2.31 (s, 3H), 2.22 (t, J = 2.4 Hz, 1H), 1.97-1.88 (m, 2H), 1.20 (t, J = 7.2 Hz, 3H), 1.10 (d, J = 6.9 Hz, 2H), 1.00 (d, J = 6.9 Hz, 2H).

4.1.17. 2,4-Bis(benzyloxy)-N-(3-chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-N-ethyl-5-isopropylbenzamide (4-7)

Following the procedure as described for the preparation of 4-6, reaction of compound 4-5 (0.69 g, 1.13 mmol), anhydrous NaH (54 mg, 2.26 mmol), ethyl iodide (55 μL, 0.69 mmol) in 5 mL DMSO gave compound 4-7 in 87% yield. ¹H NMR (300 MHz, DMSO-d₆): δ_H 7.44-7.36 (m, 11H), 7.14 (br.s, 1H), 6.92-6.89 (3H), 6.61 (br.s, 1H), 5.01 (br.s, 4H), 3.96 (br.s, 2H), 3.72 (br.s, 2H), 3.39-3.36 (m, 4H), 3.26 (s, 2H), 3.07 (s, 2H), 2.16 (s, 3H), 1.78 (br.s, 2H), 1.01 (d, J = 4.8 Hz, 6H).

4.1.18. 2,4-Bis(benzyloxy)-N-(4-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-N-ethyl-5-isopropylbenzamide (5-7)

Following the procedure as described for the preparation of 4-6, reaction of compound 5-5 (0.77 g, 1.26 mmol), anhydrous NaH (60 mg, 2.52 mmol), ethyl iodide (142 μL, 1.76 mmol) in 5 mL DMSO gave compound 5-7 in 66% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.38-7.25 (m, 10 H), 7.10-7.05 (m, 2H), 6.49 (br.s, 2H), 6.26 (br.s, 1H), 4.93 (s, 2H), 4.84 (s, 2H), 3.90 (br.s, 2H), 3.61 (br.s, 2H), 3.29-3.19 (m, 3H), 2.51 (t, J = 6.6 Hz, 2H), 2.27 (s, 3H), 2.19 (t, J = 2.1 Hz, 1H), 1.81 (br.s, 2H), 1.17 (br.s, 3H), 1.13 (d, J = 6.9 Hz, 6H).

4.1.19. 2,4-Bis(benzyloxy)-N-(2-chloro-3-(3-(methyl(prop-2-yn-l-yl) amino) propoxy) phenyl)-5-isopropyl-N-propylbenzamide (3-8)

Following the procedure as described for the preparation of 3-6 reaction of compound 3-5 (0.40 g, 0.65 mmol), potassium tert-butoxide (110 mg, 0.98 mmol), 1-iodopropane (130 μL, 1.30 mmol) in 5 mL THF gave compound 3-8 in 87% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.47-7.45 (m, 2H), 7.40-7.27 (m, 8H), 7.07 (s, 1H), 6.89 (dd, J = 8.1, 8.1 Hz, 1H), 6.72-6.68 (m, 2H), 6.33 (s, 1H), 5.10-4.96 (m, 2H), 4.88 (s, 2H), 4.37-4.27 (m, 1H), 4.07-3.91 (m, 2H), 3.34-3.32 (m, 3H), 3.16 (p, J = 6.9 Hz, 1H), 2.66-2.57 (m, 3H), 2.30 (br.s, 3H), 1.96-1.90 (m, 2H), 1.71-1.51 (m, 2H), 1.09 (d, J = 6.9 Hz, 3H), 0.99 (d, J = 6.9 Hz, 3 H), 0.92 (t, J = 7.2 Hz, 3H).

4.1.20. 2,4-Bis(benzyloxy)-N-(3-chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl-N-propylbenzamide (4-8)

Following the procedure as described for the preparation of 4-6, reaction of compound 4-5 (0.30 g, 0.49 mmol), anhydrous NaH (23 mg, 0.98 mmol), 1-iodopropane (66 μL, 0.68 mmol) in 5 mL DMSO gave compound 4-8 in 90% yield. ¹HNMR (300 MHZ, DMSO-d₆): δ_H 7.44-7.31 (m, 11H), 7.15 (br.s, 1H), 6.89 (br.s, 3H), 6.64 (br.s, 1H), 5.02 (s, 4H), 3.96 (br.s, 2H), 3.68 (br.s, 2H), 3.26 (s, 2H), 3.06 (s, 2H), 2.46 (t, J = 6.3 Hz, 2H), 2.16 (s, 3H), 1.78 (t, J = 6 Hz, 2H), 1.43 (br.s, 2H), 1.01 (d, J = 6Hz, 6H), 0.81 (br.s, 3H).

4.1.21. 2,4-Bis(benzyloxy)-N-(4-chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-5-isopropyl-N-propylbenzamide (5-8)

Following the procedure as described for the preparation of 4-6, reaction of compound 5-5 (0.50 g, 0.82 mmol), anhydrous NaH (40 mg, 1.64 mmol), 1-iodopropane (111 μL, 1.15 mmol) in 5 mL DMSO gave compound 5-8 in 70% yield. ¹HNMR (300 MHz, CDCl₃): δ_H 7.37-7.26 (m, 10H), 7.07 (br.s, 2H), 6.50 (br.s, 2H), 6.26 (br.s, 2H), 4.94 (s, 2H), 4.86 (s, 2H), 3.82 (br.s, 2H), 3.63 (br.s, 2H), 3.32-3.18 (m, 3H), 2.53 (t, J = 6.9 Hz, 2H), 2.30 (s, 3H), 2.20 (t, J = 2.1 Hz, 1H), 1.82 (br.s, 2H), 1.60-1.58 (m, 2H), 1.12 (d, J = 6.6 Hz, 6H), 0.93-0.89 (m, 3H).

4.1.22. N-(2-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropylbenzamide (3-a)

To a solution of compound 3-5 (0.3 g, 0.49 mmol) in 5 mL dry CH₂Cl₂ was added BCl₃ in hexane (2 mL, 1.96 mmol) dropwise in an ice-bath. The resultant solution was stirred at room temperature in N₂ atmosphere for 10 min. The reaction mixture was suspended in H₂O (50 mL) and extracted with CH₂Cl₂ (3 × 50 mL). The combined organic layer was dried over anhydrous MgSO₄, filtered and concentrated in vacuo. The residue was chromatography over a silica gel column eluted by 2% MeOH in CH₂Cl₂ to give compound 3-a in 58% yield. HPLC purity: 98.89%; m.p: 88-89 °C; ¹HNMR (300 MHz, DMSO-d₆): δ_H 11.63 (s, 1H), 10.90 (br.s, 1H), 10.72 (s, 1H), 10.16 (s, 1H), 8.04 (dd, J = 9.3, 1.2 Hz, 1H), 7.78 (s, 1H), 7.30 (t, J = 8.4 Hz, 1H), 6.93 (dd, J = 8.4, 1.2 Hz, 1H), 6.53 (s, 1H), 4.19-4.12 (m, 4H), 3.82 (br.s, 1H), 3.28 (br.s, 1H), 3.17-3.08 (m, 2H), 2.80 (s, 3H), 2.19 (p, J = 6.9 Hz, 2H), 1.15 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 167.75, 161.45, 158.21, 155.62, 138.22, 129.31, 129.14, 128.56, 117.38, 115.00, 110.36, 110.15, 103.43, 81.23, 73.04, 67.41, 54.53, 46.41, 40.91, 27.75, 25.60, 23.06; HR-MS (ESI) for C₂₃H₂₈ClN₂O₄ [M+H]⁺: calcd, 431.1738; found 431.1737.

4.1.23. N-(2-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropyl-N-methylbenzamide (3-b)

Following the procedure as described for the preparation of 3-a, reaction of compound 3-7 (0.13 g, 0.21 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (0.80 mL, 0.84 mmol) gave compound 3-b in 30% yield. HPLC purity: 97.56%; m.p: 65-66 °C; ¹HNMR (300 MHz, CD₃OD): δ_H 7.25 (dd, J = 8.1, 8.1 Hz, 1H), 7.03 (dd, J = 8.4, 1.2 Hz, 1H), 6.91 (dd, J = 1.2 Hz, 1H), 6.64 (br.s, 1H), 6.20 (s, 1H), 4.14-4.07 (m, 2H), 3.58 (d, J = 2.4 Hz, 2H), 3.31 (s, 3H) (coincide with CD₃OD solvent peak), 2.94-2.89 (m, 3H), 2.83 (t, J = 2.4 Hz, 1H), 2.52 (s, 3H), 2.07 (p, J = 6.3 Hz, 2H), 0.81 (d, J = 10.8 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 173.65, 159.77, 157.29, 130.85, 129.35, 127.92, 122.56, 122.20, 113.65, 103.31, 76.23, 68.44, 53.63, 46.24, 41.75, 28.10, 27.57, 26.91, 26.78, 23.03; HR-MS (ESI) for C₂₄H₃₀ClN₂O₄ [M+H]⁺: calcd, 445.1894; found, 445.1895.

4.1.24. N-(2-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amitio)propoxy)phenyl)-N-ethyl-2,4-dihydroxy-5-isopropylbenzamide (3-c)

Following the procedure as described for the preparation of 3-a, reaction of compound 3-8 (0.39 g, 0.61 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (2.45 mL, 2.44 mmol) gave compound 3-c in 45% yield. ¹HNMR (300 MHz, CD₃OD): δ_H 7.31 (dd, J = 8.4, 8.1 Hz, 1H), 7.07 (dd, J = 8.4, 1.2 Hz, 1H), 7.00 (dd, J = 7.8, 1.5 Hz, 1H), 6.63 (br.s, 1H), 6.19 (s, 1H), 4.22-4.15 (m, 4H), 4.03-3.96 (m, 1H), 3.76-3.64 (m, 1H), 3.46 (dd, J = 8.4, 6.6 Hz, 2H), 3.36-3.35 (m, 2H), 2.99 (s, 3H), 2.94-2.87 (m, 1H), 2.32-2.22 (m, 2H), 1.19 (t, J = 7.2 Hz, 3H), 0.81 (d, J = 6.0 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 173.18, 159.65, 156.77, 130.84, 129.19, 127.73, 124.10, 122.81, 113.69, 103.32, 81.22, 73.04, 67.62, 54.29, 46.39, 40.85, 26.80, 25.61, 23.08, 22.97, 12.78; HR-MS (ESI) for C₂₅H₃₂ClN₂O₄ [M+H]⁺: calcd, 459.2051; found, 459.2052.

4.1.25. N-(2-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropyl-N-propylbenzamide (3-d)

Following the procedure as described for the preparation of 3-a, reaction of compound 3-9 (0.35 g, 0.53 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (2.13 mL, 2.24 mmol) gave compound 3-d in 41% yield. HPLC purity: 95.69%; m.p: semisolid at rt; ¹HNMR (300 MHz, CD₃OD): δ_H 7.30 (dd, J = 8.1, 8.1 Hz, 1H), 7.07-7.00 (m, 2H), 6.63 (br.s, 1H), 6.19 (s, 1H), 4.20-4.09 (m, 4H), 3.59-3.49 (m, 1H), 3.41-3.36 (m, 2H), 3.00-2.89 (m, 4H), 2.29-2.17 (m, 2H), 1.68-1.58 (m, 2H), 1.30 (br.s, 2 H), 0.924 (t, J = 12 Hz, 3 H), 0.81 (d, J = 6.0 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 173.23, 159.54, 157.10, 130.83, 129.01, 127.60, 126.80, 123.64, 122.67, 113.56, 103.28, 77.76, 76.19, 6814, 53.83, 5277, 46.27, 41.44, 33.02, 28.07, 26.92, 23.05, 22.96, 21.57, 11.68; HR-MS (ESI) for C₂₆H₃₄ClN₂O₄ [M+H]⁺: calcd, 473.2207; found, 473.2210.

4.1.26. N-(3-Chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropylbenzamide (4-a)

Following the procedure as described for the preparation of 3-a, reaction of compound 4-5 (0.2 g, 0.33 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (1.5 mL, 1.31) gave compound 4-a in 20% yield. HPLC purity: 97.59 %; m.p: 168-169 °C; ¹H NMR (300 MHz, DMSO-d₆): δ_H 7.78 (d, J = 2.7 Hz, 1H), 7.71 (s, 1H), 7.49 (dd, J = 9.0, 2.4 Hz, 1H), 7.13 (d, J = 9.0 Hz, 1H), 6.34 (s, 1H), 4.06 (t, J = 6.3 Hz, 2H), 3.31 (d, J = 2.1 Hz, 4H), 3.09 (t, J = 2.4 Hz, 1H), 2.21 (s, 3H), 1.85 (p, J = 6.9 Hz, 2H), 1.17 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 169.66, 161.53, 160.94, 152.70, 133.19, 128.68, 127.35, 124.92, 123.69, 122.52, 114.92, 108.73, 103.69, 78.66, 75.14, 68.51, 53.52, 46.17,41.95, 28.09, 27.98, 23.04; HR-MS (ESI) for C₂₃H₂₈ClN₂O₄ [M+H]⁺ : calcd, 431.1738; found, 431.1739.

4.1.27. N-(3-Chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-3-isopropyl-N-methylbenzamide (4-b)

Following the procedure as described for the preparation of 3-a, reaction of compound 4-7 (0.35 g, 0.56 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (1.5 mL, 1.31 mmol) gave compound 4-b in 29% yield. HPLC purity: 99.12 %; m.p: 89-90 °C; ¹H NMR (300 MHz, DMSO-d₆): δ_H 10.32 (s, 1H), 7.33 (d, J = 2.1 Hz, 1H), 7.09-7.02 (m, 2H), 6.58 (s, 1H), 6.18 (s, 1H), 4.02 (t, J = 6.3 Hz, 2H), 3.28-3.26 (m, 6H), 3.07 (t, J = 2.1 Hz, 1H), 2.93-2.84 (m, 1H), 2.48 (t, J = 12 Hz, 2H), 2.18 (s, 3H), 1.81 (p, J = 7.2 Hz, 2H), 0.84 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 173.27, 159.27, 158.74, 154.07, 140.69, 129.940, 128.88, 127.79, 127.15, 124.10, 115.12, 103.32, 81.40, 72.88, 67.57, 54.39, 46.42, 40.84, 38.98, 26.91, 25.52, 24.21, 23.00; HR-MS (ESI) for C₂₄H₃₀ClN₂O₄ [M+H]⁺ : calcd, 445.1894; found 445.1892.

4.1.28. N-(3-Chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-N-ethyl-2,4-dihydroxy-5-isopropyl benzamide (4-c)

Following the procedure as described for the preparation of 3-a, reaction of compound 4-8 (0.26 g, 0.41 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (1.7 mL, 1.2 mmol) gave compound 4-c in 28% yield. HPLC purity 99.69 %; m.p: 68-69 °C; ¹HNMR (300 MHz, DMSO-d₆): δ_H 10.37 (s, 1H), 9.69 (s, 1H), 7.31 (t, J = 1.2 Hz, 1H), 7.05 (d, J = 1.2 Hz, 2H), 6.58 (s, 1H), 6.18 (s, 1H), 4.04 (t, J = 6.3 Hz, 2H), 3.78-3.71 (m, 3H), 3.64 (br.s, 1H), 3.16 (s, 1H), 2.92-2.81 (m, 3H), 2.45 (br.s, 3H), 1.97-1.93 (m, 2H), 1.05 (t, J = 6.9 Hz, 3H), 0.83 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 172.81, 159.17, 154.19, 138.78, 130.75, 128.82, 127.05, 124.09, 115.02, 103.30, 81.41, 72.85, 67.55, 54.39, 46.59, 46.41, 40.83, 26.88, 25.50, 23.01, 12.96; HR-MS (ESI) for C₂₅H₃₂ClN₂O₄ [M+H]⁺: calcd, 459.2051; found 459.2050.

4.1.29. N-(3-Chloro-4-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropyl-N-propyl benzamide (4-d)

Following the procedure as described for the preparation of 3-a, reaction of compound 4-9 (0.37 g, 0.56 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (2.30 mL, 2.24 mmol) gave compound 4-d in 32% yield. HPLC purity 99.47%; mp: semisolid at rt; ¹HNMR (300 MHz, DMSO-d₆): δ_H 10.29 (s, 1H), 9.63 (s, 1H), 7.30 (br.s, 1H), 7.04 (s, 2H), 6.56 (s, 1H), 6.17 (s, 1H), 4.03 (t, J = 6.3 Hz, 2H), 3.68 (t, J = 7.5 Hz, 3H), 3.20-3.13 (m, 2H), 2.93-2.84 (m, 1H), 2.61 (br.s, 1H), 2.29 (s, 3H), 1.87 (t, J = 6.0 Hz, 2H), 1.47 (q, J = 7.2 Hz, 3H), 0.86-0.83 (m, 10H); ¹³C NMR (150 MHz, CD₃OD): δ 172.95, 159.17, 154.63, 138.55, 130.49, 128.85, 128.50, 127.04, 124.13, 114.84, 103.31, 77.27, 76.62, 68.25, 53.66, 53.22, 46.25, 41.68, 27.38, 26.87, 23.00, 21.75, 11.59; HR-MS (ESI) for C₂₆H₃₄ClN₂O₄ [M+H]⁺ : calcd, 473.2207; found, 473.2204.

4.1.30. N-(4-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropyl benzamide (5-a)

Following the procedure as described for the preparation of 3-a reaction of compound 5-5 (0.33 g, 0.53 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (2.13 mL, 2.13 mmol) gave compound 5-a in 48% yield. HPLC purity: 98.97%; m.p: 139-140 °C; ¹HNMR (300 MHz, CD₃OD): δ_H 7.76 (s, 1H), 7.72 (d, J = 2.4 Hz, 1H), 7.34 (d, J = 8.7 Hz, 1H), 7.10 (dd, J = 8.7, 2.1 Hz, 1H), 6.36 (s, 1H), 4.27-4.23 (m, 4H), 3.54 (br.s, 2H), 3.42 (t, J = 2.7 Hz, 1H), 3.25-3.16 (m, 1H), 3.05 (s, 3H), 2.33 (p, J = 12 Hz, 2 H), 1.25 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 169.45, 161.71, 160.41, 155.06, 139.77, 130.93, 130.83, 128.89, 127.74, 118.67, 115.72, 108.98, 108.47, 103.62, 81.49, 72.79, 67.23, 54.62, 40.85, 28.07, 27.95, 25.42, 23.00; HR-MS (ESI) for C₂₃H₂₈ClN₂O₄ [M+H]⁺ : calcd, 431.1738; found 431.1737.

4.1.31. N-(4-Chloro-3-(3-(methyl(prop-2-yn-l-yl) amino) propoxy) phenyl)-2,4-dihydroxy-5-isopropyl-N-methyl benzamide (5-b)

Following the procedure as described for the preparation of 3-a, reaction of compound 5-7 (0.23 g, 0.37 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (1.5 mL, 1.48 mmol) gave compound 5-b in 60% yield. HPLC purity: 99.68%; m.p: 86-87 °C; ¹HNMR (300 MHz, CD₃OD): δ_H 7.37 (d, J = 8.1 Hz, 1H), 6.90-6.86 (m, 2H), 6.66 (s, 1H), 6.20 (s, 1H), 4.17 (d, J = 2.4 Hz, 2H), 4.02 (t, J = 5.7 Hz, 2H), 3.46-3.40 (m, 5H), 3.39 (t, J = 2.4 Hz, 1H), 2.99-2.93 (m, 4H), 2.22 (p, J = 7.5 Hz, 2H), 0.87 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 173.13, 159.49, 159.11, 155.53, 146.64, 131.56, 130.84, 129.05, 127.20, 122.10, 121.29, 114.15, 110.96, 103.33, 81.41, 72.82, 67.41, 54.29, 46.37, 40.79, 38.77, 26.91, 25.37, 22.93; HR-MS (ESI) for C₂₄H₃₀ClN₂O₄ [M+H]⁺ : calcd, 445.1894; found 445.1889.

4.1.32. N-(4-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-N-ethyl-2,4-dihydroxy-5-isopropylbenzamide (5-c)

Following the procedure as described for the preparation of 3-a, reaction of compound 5-8 (0.51 g, 0.80 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (3.20 mL, 3.2 mmol) gave compound 5-c in 32% yield. HPLC purity: 99.28%; m.p: 58-59 °C; ¹HNMR (300 MHz, CD₃OD): δ_H 7.37 (dd, J = 6.9, 1.8 Hz, 1H), 6.87-6.84 (m, 2H), 6.64 (s, 1H), 6.20 (s, 1H), 4.17 (d, J = 2.7 Hz, 2H), 4.03 (t, J = 5.7 Hz, 2H), 3.94 (q, J = 7.2 Hz, 2H), 3.43 (t, J = 7.8 Hz, 2H), 3.38 (t, J = 2.7 Hz, 1H), 3.01-2.92 (m, 4H), 2.26-2.17 (m, 2H), 1.20 (t, J = 7.2 Hz, 3H), 0.86 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 172.71, 159.45, 159.41, 155.60, 144.94, 131.56, 129.03, 127.12, 122.33, 122.21, 115.05, 111.00, 103.34, 81.42, 72.85, 67.49, 54.31, 46.61, 46.40, 40.82, 26.92, 25.41, 22.96, 13.12; HR-MS (ESI) for C₂₅H₃₂ClN₂O₄ [M+H]⁺ : calcd, 459.2051; found 459.2052.

4.1.33. N-(4-Chloro-3-(3-(methyl(prop-2-yn-l-yl)amino)propoxy)phenyl)-2,4-dihydroxy-5-isopropyl-N-propylbenzamide (5-d)

Following the procedure as described for the preparation of 3-a, reaction of compound 5-9 (0.35 g, 0.53 mmol) in 5mL dry CH₂Cl₂ and BCl₃ in hexane (2.14 mL, 2.12 mmol) gave compound 5-d in 28% yield. HPLC purity: 97.66%; m.p: semisolid at rt; ¹HNMR (300 MHz, CD₃OD): δ_H 7.36 (d, J = 8.7 Hz, 1H), 6.87-6.85 (m, 2H), 6.63 (s, 1H), 6.20 (s, 1H), (d, J = 2.1 Hz, 2H), 4.02 (t, J = 5.1 Hz, 2H), 3.85 (t, J = 7.8 Hz, 2H), 3.41 (br.s, 1H), 3.01-2.92 (m, 4H), 2.23 (br.s, 2H), 1.68-1.60 (m, 2H), 1.30 (br.s, 2H), 0.94 (t, J = 7.2 Hz, 3H), 0.86 (d, J = 6.9 Hz, 6H); ¹³C NMR (150 MHz, CD₃OD): δ 172.82, 159.26, 159.00, 145.06, 131.46, 130.79, 128.90, 127.07, 122.19, 122.05, 115.04, 81.61, 72.68, 67.65, 54.46, 53.18, 46.65, 41.15, 28.03, 26.85, 25.46, 24.16, 22.95, 21.93; HR-MS (ESI) for C₂₆H₃₄ClN₂O₄ [M+H]⁺ : calcd, 473.2207; found 473.2207.

4.2. Biology

4.2.1. Cell culture

GL26 mouse GBM cells, U251S and U251R human GBM cells were obtain as described previously.[11] U87, U373, LN229, A172, and T98G human GBM cell lines were obtained from American type culture collection (ATCC). P1S, P3, and P3R cells were derived from GBM patient surgical specimens.[50] All GBM cell lines were grown in Dulbecco’s Modified Eagle’s Medium (Life Technologies, Carlsbad, CA) with 10% fetal bovine serum and 1% penicillin/streptomycin at 37°C in a humidified atmosphere of 5% CO₂ and 95% air. P3R cells were grown with the same medium with 50 μM TMZ.

4.2.2. MTT assay

GL26, U251S and U251R cells were seeded at 5×10³ cells/well in a 96-well plate 24 hours before addition of compounds. Compound-containing medium was added into wells and 96-well plates were incubated for 48 h. After 48 h, medium was replaced with 0.5 mg/mL MTT-containing medium and incubated for 3 hours. DMSO was used to dissolve the purple formazan product in each well. Absorbance was recorded by a microplate reader Synergy HTX (Bio-Tek, Winooski, VT, USA) at 570 nm and data was plotted using GraphPad Prism based on a sigmoidal dose-response equation. The experimental groups were normalized to control (no treatment) which was set as 100%. Experiment were performed in triplicate. IC₅₀ values were calculated using GraphPad Prism based on a sigmoidal dose-response equation.

4.2.3. MAO A inhibition assay

MAO A activity was determined using radio assay as described previously.[51] For MAO A inhibition in GL26 cells, to determine the MAO A inhibition of compounds, the total protein of GL26 cells homogenate were treated with compounds at indicated concentration and pre-incubated for 20 min at 37°C. After pre-incubation, the samples were added 100 μL of 1 mM ¹⁴C-labeled 5-hydroxytryptamine (¹⁴C-5-HT) and incubated for 20 min in the assay buffer (50 mM sodium phosphate buffer, pH 7.4) at 37°C. Then the reaction was terminated with 100 μL of 6 N HCl_(aq). The radio activity of ¹⁴C-labeled 5-HIAA product was extracted with 50% EtOAc in Benzene (6 mL), and centrifuged at 2500 rpm for 7 min. The organic layer (4 mL) was combined with 5 mL Budget-Solve^™ (RPI, USA) and the radioactivity of reaction products was measured using Beckman/PerkinElmer Scintillation Counter. 100% MAO A activity is 39.05 (nmol/20 min/mg protein). IC₅₀ values were calculated using GraphPad Prism based on a sigmoidal dose-response equation.

4.2.4. HSP90 N-terminal domain assay

The assay was conducted by Reaction Biology Corporation (Malvern, USA) as described previously.[52] Compounds were tested at 10 doses, with a 3-fold serial dilution starting at 10 μM. Briefly, human HSP90α recombinant enzyme in HSP90 assay buffer (50 mM NaCl, 10 mM MgCl₂, 20mM Hepes, 0.02% Brij^™ 35, 2 mM DTT, 1% DMSO, 0.02 mg/mL BSA, pH 7.5) was added compounds at indicated concentration and pre-incubated in 37°C for 30 min. The samples were added with fluorescently labeled geldanamycin (FITC-GM) and incubated in 37°C for 3 hours. After incubation, the fluorescent polarization (FP) of samples were measure (excitation λ = 475-495 nm, emission λ = 518-538 nm), mP and % probe binding were calculated. The reduced fluorescence indicated the effect of inhibition.

4.2.5. Cell Viability CCK8 assay

U87MG, U373MG, LN229, A172, T98G, P1S, P3, and P3R human GBM cells were seeded at 2×10³ cells/well in a 96-well plate 24 hours before addition of compounds. Compound-containing medium was added into wells and 96-well plates were incubated for 72 h. After 72 h, cell viability was estimated using CCK8 assay[53] and data was plotted using GraphPad Prism based on a sigmoidal dose-response equation. The experimental groups were normalized to control (no treatment) which was set as 100%. Experiment were performed in triplicate. IC₅₀ values were calculated using GraphPad Prism based on a sigmoidal dose-response equation.

4.2.6. Molecular docking

The HSP90 (PDB ID: 2VCI) and MAO A (PDB ID: 2BXR) protein structures were downloaded from the Protein Data Bank.[54] The structures were prepared using the Schrodinger Protein Preparation Wizard module in the Maestro software suite.[55, 56] This included the removal of water atoms and the addition of hydrogen atoms and relevant charges. The co-crystal ligands (PDB Ligand ID: 2GJ and MLG) were selected as the centroid for the generation of the grids. Compounds were prepared for docking using Schrodinger’s LigPrep module with default settings. In short, low-energy 3D structures with relevant protonation states were generated for input compounds. Finally, molecular docking was performed using the Glide.[56]

4.2.7. Western blot analysis

GL26 and U251R cells were seeded at 3 × 10⁵ cells/well in 6-well plates 24 hours before addition of compounds. Cells were treated with compound-containing medium for 24 hours. After treatment, medium was removed and the cells were washed twice with ice-cold PBS. The cells were lysed with RIPA buffer (150 mM NaCl, 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS, 50 mM Tris pH 8.0) with protease inhibitor cocktail (Sigma-Aldrich). The lysate samples were centrifuged in centrifuge tubes at 4°C at 12,000 rpm for 20 min and the supernatant was collected. The protein concentration of samples was determined using Pierce^™ BCA protein assay kit (Thermo Fisher). Cell lysate samples was separated by 10% SDS-PAGE and transferred onto PVDF membranes (Bio-Rad). The membranes were blocked with 5% non-fat milk for 1 hour and probed with primary anti-HER2 (29D8) rabbit mAb (#2165, Cell Signaling Technology), anti-HSP70 rabbit mAb (#4872, Cell signaling Technology), anti-phospho-Akt (Ser473) rabbit mAb (#2118-1, Epitmics), anti-GAPDH (6C5) mouse mAb (Santa Cruz Biotechnology) in cold room overnight, respectively. Anti-mouse HRP-linked IgG (#7076S, Cell Signaling Technology) and anti-rabbit HRP-linked IgG (#7074S, Cell Signaling Technology) were used to detected the primary antibodies in room-temperature for 1 hour. The membranes were scanned using iBright FL1000 Gel/Cell Imaging System (Thermo Fisher) and the bands were analyzed using iBright Analysis Software (Thermo Fisher).

4.2.8. Analysis of PD-L1 expression on cell surface using flow cytometry

GL26 cells were seeded in 6 well-plates (3.6 × 10⁵ cells/well) and treated with compounds 4-b, 4-c, or clorgyline at various concentration with or without 10 ng/mL IFN-γ for 24 h. After 24 h, cells were suspended at 3.6 × 10⁵ cells/tube in PBS buffer containing 0.05% (w/v) BSA, and incubated with goat anti-PD-L1 antibody (Cat# AF1019-SP, R&D systems) for 1 h. After 1 h, cells were washed with PBS and incubated with FITC-conjugated donkey anti-goat antibody (Cat# 705-096-147, Jackson ImmunoResearch Laboratories) for 1 h. After washing the cells with PBS to remove the unbound antibodies, FITC fluorescence of the cells was measured using Sony SA3800 Spectral Analyzer flow cytometry (Sony Biotechnology, San Jose, CA, USA). The data and mean fluorescence intensity (MFI) of FITC signal were analyzed by FlowJo software (BD Biosciences, San Jose, CA, USA).

HCT116 cells were seeded in 6 well-plates (3 × 10⁵ cells/well) and treated with compounds 4-b, and 4-c at various concentration with or without 20 ng/mL IFN-γ for 48 h. After 48 h, cells were suspended at 3 × 10⁵ cells/tube in PBS buffer containing 0.05% (w/v) BSA, and incubated with mouse anti-PD-L1 antibody (Cat# 14-5983-82, Thermo Fisher Scientific) for 1 h. After 1 h, cells were washed with PBS and incubated with FITC-conjugated goat anti-mouse antibody (Cat# 115-096-071, Jackson ImmunoResearch Laboratories) for 1 h. After washing the cells with PBS to remove the unbound antibodies, FITC fluorescence of the cells was measured using FACScan flow cytometry (BD Biosciences, San Jose, CA, USA). The data and mean fluorescence intensity (MFI) of FITC signal were analyzed by CellQuest software (BD Biosciences, San Jose, CA, USA).

4.2.9. GL26 GBM mouse model

Animal protocol (protocol number: 20212) was approved by Institutional Animal Care and Use Committee (IACUC) of USC. 8-9 weeks old male mice (C57BL/6J) were purchased from the Jacksons Laboratory, USA. 2 × 10⁵ GL26 mouse GBM cells were subcutaneously implanted into the lower back of the mice. After tumor visible in mice back at 9^th day post tumor implantation, mice were treated with vehicle, 10 mg/kg of clogyline daily, 25 mg/kg of compound 4-b, 25 mg/kg of 4-c, and 5 mg/kg NMI for 14 days intraperitoneally. For treatment groups of 4-b and 4-c, i.p. injection was given to mice daily at first 3 days, and then every 3 days since 4^th day of treatment. For treatment group of NMI, i.p. injection was given to mice every other day. Tumor size was measured using calipers, and tumor volume was calculated by the formula: length × width² ÷ 2. Tumors were dissected and weight once the mouse reached endpoint (diameter > 1.5 cm or volume >1.5 cm³) or based on health status.

4.2.10. NCI-60 screening

The procedure of compound submission is shown on the website (https://dtp.cancer.gov/organization/dscb/compoundSubmission/submissionProcedures.htm) and NCI-60 screen methodology is shown on the website (https://dtp.cancer.gov/discovery_development/nci-60/methodology.htm) of Developmental Therapeutics Program (DTP). Briefly, 60 human cancer cell lines were seeded in 96-well plates, respectively, in a densities of 5000 to 40000 cells per well depending on the doubling time of each cell lines 24 h before addition of drugs at various concentrations. To measure the cell population before drug addition, each cell lines were fixed with TCA as the population at time zero $(\mathrm{Tz})$. After 48 h incubation with drugs, the cell viability was measured using sulforhodamine B (SRB) assay. The percentage of growth is calculated at each drug concentration based on seven absorbance measurements [time zero $(\mathrm{Tz})$, control growth (C), cell growth in the presence of drug at various concentrations $(\mathrm{Ti})$] using the formula:

$$\begin{array}{cc}\hfill & [(\mathrm{Ti}-\mathrm{Tz})∕(\mathrm{C}-\mathrm{Tz})]\times 100\text{for concentration with}\mathrm{Ti}\ge \mathrm{Tz}\text{or}\hfill \\ \hfill & [(\mathrm{Ti}-\mathrm{Tz})∕\mathrm{Tz}]\times 100\text{for concentration with}\mathrm{Ti}<\mathrm{Tz}\hfill \end{array}$$

The growth inhibition of 50% was calculated from $[(\mathrm{Ti}-\mathrm{Tz})∕(\mathrm{C}-\mathrm{Tz})]\times 100=50$. The Total growth inhibition (TGI) was calculated from $\mathrm{Ti}=\mathrm{Tz}$. The LC₅₀ (concentration of drug resulting in a 50% reduction in the measured protein at the end of the drug treatment as compared to that at the beginning) was calculated from $[(\mathrm{Ti}-\mathrm{Tz})∕(\mathrm{C}-\mathrm{Tz})]\times 100=-50$.

Figure 6. A proposed working model for how MAO A/HSP90 dual inhibitors regulate the glioma growth.

Dual MAO A/HSP90 inhibitors, 4-b and 4-c, reduced HER2, and phospho-Akt, which are related to tumor growth. They also reduced PD-L1 expression, which inhibits T cell activation. Thus, 4-b and 4-c contribute to decrease tumor growth.

Highlights.

• Conjugating clorgyline and 4-isopropylresorcinol as MAO A/HSP90 dual inhibitors.

• 4-b and 4-c, inhibit TMZ-sensitive and -resistant glioblastoma cell growth.

• 4-b and 4-c increased HSP70, reduced HER2, phospho-Akt and PD-L1 expression.

• 4-b and 4-c inhibit glioblastoma growth in GL26 mouse model.

Abbreviation used

MAO A

monoamine oxidase A

GBM

glioblastoma

HSP90

heat shock protein 90

TMZ

temozolomide

RTK

receptor tyrosine kinase

PI3K

phosphoinositide 3-kinase

Rb

retinoblastoma protein

EGFR

epidermal growth factor receptor

PKB, AKT

protein kinase B

HER2

human epidermal growth factor 2

ROS

reactive oxygen species

MAPK

mitogen-activated protein kinase

PD-L1

programmed cell death protein 1 ligand

PD-1

death protein 1

Cs₂CO₃

cesium carbonate

ACN

acetonitrile

K₂CO₃

potassium carbonate

NH₄Cl

ammonium chloride

SOCl₂

thionyl chloride

TEA

triethylamine

THF

tetrahydrofuran

HOBt

hydroxybenzotriazole

EDC-HCl

1-(3-Dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride

DIPEA

N,N-diisopropylethylamine

DMF

N,N-dimethylformamide

t-BuOK

potassium tert-butoxide

NaH

sodium hydride

rt

room temperature

BCl₃

boron trichloride

TLC

thin layer chromatography

NMR

nuclear magnetic resonance