N⁴-(3-bromophenyl)-7-(substituted benzyl) pyrrolo[2,3-d]pyrimidines as Potent Multiple Receptor Tyrosine Kinase Inhibitors: Design, Synthesis, and In vivo Evaluation

Abstract

With the goal of developing multitargeted receptor tyrosine kinase inhibitors that display potent inhibition against PDGFRβ and VEGFR-2 we designed and synthesized eleven N⁴-(3-bromophenyl)-7-(substitutedbenzyl) pyrrolo[2,3-d]pyrimidines 9a–19a. These compounds were obtained from the key intermediate N⁴-(3-bromophenyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine 29. Various arylmethyl groups were regiospecifically attached at the N7 of 29 via sodium hydride induced alkylation with substituted arylmethyl halides. Compounds 11a and 19a were potent dual inhibitors of PDGFRβ and VEGFR-2. In a COLO-205, in vivo tumor mouse model 11a demonstrated inhibition of tumor growth, metastasis, and tumor angiogenesis that was better than or comparable to the standard compound TSU-68 (SU6668, 8).

Introduction

Protein kinases are enzymes that transfer a phosphate group from ATP to the hydroxyl group of serine, threonine, or tyrosine of specific proteins inside cells.¹ The phosphorylation by these enzymes achieves an important function of signal transduction in eukaryotic cells and controls the processes of cell proliferation, metabolism, survival, and apoptosis. A mis-regulation of these tightly controlled processes results in the overexpression of kinases, which is associated with a variety of disease states including cancer.²

Angiogenesis is the process of formation of new blood vessels from existing vascular bed. Angiogenesis is crucial for a tumor to grow beyond 1 – 2 mm and for tumor invasion and metastasis. ^3,4 As a tumor grows in size, it becomes increasingly hypoxic and triggers the release of growth factors, particularly, vascular endothelial growth factor (VEGF), epidermal growth factor (EGF), and platelet derived growth factor (PDGF) among others. The growth factors activate adjacent blood vessels leading to angiogenesis.^5,6 Following angiogenesis, the tumor can grow at an increased rate.⁷

For angiogenesis to occur, the pro-angiogenic growth factors have to bind to members of the protein kinase family identified as receptor tyrosine kinases (RTKs). Following binding, the RTKs dimerize and undergo autophosphorylation, initiating a series of downstream events leading to proliferation migration and cell survival.⁸ The growth factor VEGF triggers angiogenesis by binding to a family of VEGF receptors, including VEGFR-2 (KDR), VEGFR-1 (Flt-1), and VEGFR-3 (Flt-4). Similarly PDGF binds to the family of PDGF receptors, Flt-3 (FMS-like tyrosine kinase-3), PDGFRβ, c-Kit (a stem cell factor receptor), and CSF1R (colony-stimulating factor 1 receptor).⁹

Abrogation of angiogenesis by inhibition of RTK signaling pathways is an important mechanism for the development of novel anticancer agents.^8,10,11 Early work in the area of RTK inhibitor discovery afforded small molecule inhibitors gefitinib 1 (Figure 1) (specific EGFR inhibitor, approved for limited use for the treatment of non-small cell lung cancer)¹² and erlotinib 2 (specific EGFR inhibitor, approved for the treatment of non-small cell lung cancer).¹² Recent data, however, shows that tumors treated with specific RTK inhibitors usually develop resistance through the upregulation of alternate kinase mediated pathways.¹³ In addition, cross-talk has been implicated between EGFR and other growth factor receptors involved in tumor development.¹⁴

Figure 1.

RTK Inhibitors

There are a number of RTK mediated processes that can promote angiogenesis, thus inhibitors targeting a broader range of RTKs may lead to a more robust antitumor response and prevent resistance by targeting two or more angiogenic pathways.^15,16 The FDA approval of several compounds targeting a broader-spectrum of RTKs lends credence to the development of single agents that target multiple RTKs. These include imatinib 3, sunitinib 4, sorafenib 5 and lapatinib 6 (Figure 1). Compound 3 inhibits Abelson Tyrosine Kinase (Abl), c-kit protein (CD117), and PDGFR, and has been approved for chronic myelogenous leukemia (CML),¹⁷ gastrointestinal stromal tumors (GISTs), and other malignancies. Compound 4 inhibits VEGFR-2, PDGFRβ, c-kit, and FMS-like tyrosine kinase 3 (FLT3)¹⁸ and has been approved for the treatment of renal cell carcinoma and imatinib-resistant GIST. Compound 5 inhibits Raf kinase, VEGFR-2, PDGFRβ, and c-kit¹⁹ and has been approved for advanced renal cell carcinoma. Compound 6 inhibits EGFR and erythroblastic leukemia viral oncogene homolog-2 (ErbB-2)²⁰ and has been approved for advanced metastatic breast cancer in conjunction with chemotherapy.

In angiogenesis, VEGFs cause a significant increase in the formation of blood vessels, but these vessels are immature and leaky. The formation of thicker, more stable vessels requires an encapsulation by pericytes. This process of encapsulation is driven by PDGFRβ signaling. Recent reports indicate that the inhibition of VEGFR-2 and PDGFRβ with two separate inhibitors produces a synergistic effect in early stage as well as late stage pancreatic islet cancer in mouse models.²¹ Timke et al.²² have investigated the therapeutic potential of a VEGFR inhibitor SU5416 (7) (Figure 1) and PDGFR inhibitor 8^23,24 in combination with radiotherapy in vitro and in vivo. It was observed that dual inhibition of VEGFR and PDGFR signaling was a more efficient antitumor mechanism than single pathway inhibition. In addition radiation markedly enhanced the therapeutic efficacy in vitro and in human glioblastoma and human prostrate carcinoma in mice.

We²⁵ previously reported a series of eleven N⁴-(3-bromophenyl)-6-(substituted benzyl) pyrrolo[2,3-d] pyrimidines 9–19 (Figure 1). These compounds were evaluated in human tumor cells known to over express high levels of specific RTKs. Compound 10 with a 2′-Me benzyl substitution exhibited toxicity against A431 cells (cell lines overtly dependent on EGFR for survival)²⁶ at values ten-fold better than the standard agent PD153035 (20) used in this assay (Table 1). In addition, 10 also demonstrated VEGFR-2 inhibition forty-fold better than the standard agent 7. EGFR and A431 cell lines were inhibited by 14 (with a 2′,4′-diCl benzyl substitution) at concentrations that were comparable to and five-fold better than the standard 20 respectively. Compound 13 with a 2′,5′-diOMe benzyl substitution inhibited VEGFR-2 and PDGFRβ at values seventeen-fold better than, and comparable to, standard agents 23 and AG1295 (22) respectively. Thus the N⁴-(3-bromophenyl)-6-(substituted benzyl) pyrrolo[2,3-d]pyrimidine scaffold with variations in the phenyl ring affords multi-kinase inhibition with potencies equivalent to or better than standard agents. In an attempt to develop compounds with multikinase inhibition potential it was of interest to explore the effect of moving the 6-position substituent in the parent compounds to the 7-position. Hence a series of N⁴-(3-bromophenyl)-7-(substituted benzyl) pyrrolo[2,3-d]pyrimidines–9a–19a (Figure 2) was designed. The selection of the substituents on the N7-position is the same as that at N6 in the parent series (9–19), and is rationalized below.

Table 1.

IC₅₀ values (μM) of kinase inhibition and the A431 cytotoxicity assay

Compd #	PDGFR-β inhibition	VEGFR-2 inhibition	VEGFR-1 inhibition	EGFR inhibition	A431 cytotoxicity
9a	>50	>50	>50	1.67±0.3	31.8±6.3
10a	>50	0.25±0.04	>50	9.19±1.8	1.21±0.42
11a	>50	5.58±0.69	26.8±4.1	4.31±1.75	>50
12a	>50	8.28±0.69	42.7±6.1	17.42±3.9	28.6±5.1
13a	8.92±1.6	0.62±0.21	31.1±5.8	12.62±3.3	>50
14a	17.0±5.6	28.11±9.9	>50	0.23±0.06	2.8±1.1
15a	>50	>50	>50	19.77±5.6	33.5±6.2
16a	>50	5.08±0.83	19.2±4.3	>50	>50
17a	>50	>50	15.2±2.9	1.24±0.21	33.2±5.9
18a	>50	5.97±0.78	>50	6.16±1.2	23.5±5.2
19a	14.7±3.4	9.42±1.9	>50	>50	42.1±18.5
9a	>500	23.8±3.0	99.3±10.3	166.4±20.6	50.4±5.9
10a	159.6±26.3	113.4±17	129.3±20.4	113.3±18.9	15.7±2.8
11a	1.5±0.21	17.9±2.4	126.3±19.1	>200	88.4±10.2
12a	>500	65.3±7.9	79.9±8.4	99.9±18.6	16.1±2.2
13a		>200	>200	69.2±6	40.6
14a	34.2±4.4	>200	138.1±24.2	>200	15.3±1.9
15a	229.6±37.1	64.5±7.8	118.6±11.4	>200	19.2±3.0
16a	129.3±21.1	14.9±2.1	150±22.1	>200	36.3±4.9
17a	212.4±16.2	22.9±2.9	50.8±6.2	>200	20.4±3.5
18a	129.1±20.5	22.9±10.7	>200	>200	13.3±20.5
19a	1.8±0.29	25.7±4.6	156.5±25	>200	39.0±6.8
20				0.2±0.04	12.6±2.9
21			17.7±5.5
7		2.43±0.32			19.2±1.1
22	6.2±1.3
23		10.6±3.5

^aIC₅₀ values taken from ref. 25.

Figure 2.

Target compounds

Compound 9a (Figure 2) with a benzyl at N7 served as a comparison with compounds 10a to 19a (Figure 2) that possess varied substituents at the N7 position. The basis for including the 2′-Me benzyl group in 10a, 2′,5′-diOMe benzyl group in 13a, and 2′,4′-diCl benzyl group in 14a, was the potent RTK inhibition produced by the lead compounds 10, 13, and 14 (Figure 1) which have a similar substitution pattern at the 6-position.

The 2′-Cl phenyl group in 11a, 4′-Cl benzyl group in 12a, and 3′, 4′-diCl benzyl in group in 15a would afford information on activity obtained on having an electron withdrawing group (or groups) at the respective positions. The 3′,4′,5′-triOMe benzyl group in 19a would afford information on activity obtained on having electron donating groups at these positions.

The 1-naphthyl moiety in 16a, 2-naphthyl in 17a, and 4′-phenyl benzyl group in 18a were included to obtain information about bulk tolerance in this position in the RTK active site.

Chemistry

The synthesis of compounds 9a–19a is shown in Scheme 1. Compound 26 was obtained by cyclocondensation of chloroacetaldehyde 25 with 24 at reflux for 4 h as previously reported.²⁷ Chlorination of 26 with phosphorus oxychloride, in the presence of N,N′-dimethylaniline as a catalyst afforded 27. Nucleophilic displacement of the 4-chloro group in 27 with 3-bromoaniline 28 furnished the common intermediate 29. The target compounds 9a–19a were obtained in yields ranging from 5–47% by a sodium hydride induced regiospecific alkylation of 29 at the 7-position with the corresponding substituted arylmethyl halides 30a–k. The purification of the target compounds was tedious since the by-products had R_f values (TLC) close to the desired compounds 9a–19a. This necessitated the exploration of alternate synthetic procedures, particularly for large scale synthesis of the compounds for in vivo evaluation.

Scheme 1^a.

^aReagents and conditions: (a) sodium acetate, water, reflux, 4h; (b) phosphorus oxychloride, N,N′-dimethylaniline, reflux, 4h; (c) isopropanol, 3 drops concd HCl, reflux, 4h; (d) sodium hydride, dimethylformamide, 0 °C – 1 h, R.T. – 10 min–4h.

A gram scale synthesis of 11a was accomplished by an improved method²⁸ as shown in Scheme 2. A sodium hydride induced alkylation of the N7-position of 27 furnished 31. This was followed by a nucleophilic displacement of the 4-chloro group in 31 with 28 to afford 11a. This modified protocol provided a remarkable improvement in the overall reaction yield of 27 to 11a from a mere 5% (Scheme 1) to 39% (Scheme 2).

Scheme 2^a.

^aReagents and conditions: (a) sodium hydride, dimethylformamide, 0 °C – 30 min, R.T. – 2 h. (b) isopropanol, 3 drops concd HCl, reflux, 4h.

Biological Evaluation and Discussion

The RTK inhibitory activities of compounds 9a–19a (Table 1) were evaluated using a phosphotyrosine ELISA in human tumor cells known to express high levels of EGFR, VEGFR-1, VEGFR-2, and PDGFRβ.^29,30 Compounds known to inhibit a specific RTK were used as a positive control for these assays. Whole cell assays were used for measuring RTK inhibition as these assays afford more meaningful results for translation to in vivo studies.^25,31–34 To study the effect of compounds on cell proliferation, A431 cancer cells known to over express EGFR were used. EGFR has been shown to be a factor in the overall survival of A431 cells.²⁶ Cell-proliferation was measured using CYQUANT^®, a DNA intercalating dye shown to provide a linear approximation of cell number.³⁵ In this assay, purified angiogenic growth factors are positioned locally on a vascularized membrane of a developing chicken embryo along with possible inhibitors. Digitized images of the vasculature are taken 48 h after growth factor administration and the number of vessels per unit area is quantified to measure vascular density.

The IC₅₀ values of RTK inhibition vary under different assay conditions. Hence, we used a standard (control) compound in each of the evaluations. For EGFR, the standard was 20 (Figure 1); for VEGFR-1 the standard was 21; for VEGFR-2 the standard was 7; for PDGFRβ the standard was 22; for the cytotoxicity study against the growth of A431 cells in culture the standards were 20, 7 and 23.

Since the inhibitory activities are determined in cells, a definite structure-activity relationship cannot be determined for 9a–19a and RTK inhibition. Compounds 11a with a 2′-Cl benzyl substitution and 19a with a 3′,4′,5′-triOMe benzyl substitution inhibited PDGFRβ at single digit micromolar concentrations. Compounds 11a and 19a inhibited PDGFRβ 4-fold and 3-fold better respectively than the standard 22. The 2′,4′-diCl benzyl substituted 14a, inhibited PDGFRβ at an IC₅₀ 5.5- fold lower than 22. Compounds 16a (1-naphthyl substituted) and 18a (4-phenyl benzyl substituted) had similar potencies and were 20- fold less active than 22. The 2-naphthyl substituted 17a was 34-fold less potent than 22. Hence bulky 7- substituents were not tolerated (16a–18a). An electron withdrawing group at the 2′-position (2′-Cl benzyl in 11a) is favorable for activity, but an electron donating group in this position (2′-Me benzyl in 10a), or an unsubstituted phenyl (9a) is not favored. Moving the 2′-chloro group to the 4′-position (compare 11a to 12a), or disubstitution with chloro groups (14a, 15a) is detrimental to activity. The most active compound in the parent series in this assay is the 2′,5′-diOMe benzyl substituted 13. The most active compound in the N7 substituted series, the 2′-Cl benzyl substituted 11a is >33-fold more active than its parent regioisomer 11.

In the VEGFR-2 assay, compounds 9a with an unsubstituted phenyl was 2-fold less potent than the standard 23. The 2-naphthyl substituted 17a was also 2-fold less active than 23, while the 1-naphthyl substituted 16a was 1.4-fold less active than 23 and was the most active compound in this series, in the VEGFR-2 assay. The electron withdrawing 2′-Cl benzyl substituted 11a was the next most active compound and was 1.6- fold less active than 23; however the electron donating 2′-Me benzyl substituted 10a was 11-fold less potent than 23. Comparison of 11a and 10a, suggests that an electron withdrawing substituent at the 2′– position is beneficial for activity relative to an electron donating group. The 4′-Cl benzyl substituted 12a was 6-fold less active than 23, while the 2′,4′-diCl benzyl substituted 14a was inactive. The 3′,4′-diCl benzyl substituted 15a was also 6-fold less active than 23. The 4-phenyl benzyl substituted 18a was only 2-fold less active than 23 indicating that bulk is tolerated in this position. The 3′,4′,5′-triOMe benzyl substituted 19a was also 2-fold less active than 23. The 2′,5′-diOMe benzyl substituted 13a was inactive even at 200 μM. The most active compound in the parent series, the 2′-Me benzyl substituted 10, is 450-fold more active relative to its regioisomer 10a. The most active compound in the N7 substituted series, the 1-naphthyl substituted 16a is 3-fold less active than its parent regioisomer 16.

In the VEGFR-1 assay, compound 9a, with an unsubstituted benzyl group was 5.6-fold less active than the standard 21. Compound 10a has a methyl group in the 2′ position of the benzyl substituent (compared to 9a). This compound is 7-fold less active than standard 21. Compound 11a bears a 2′-chloro benzyl substituent and is also 7-fold less active than 21. The 2′-substituted compounds 10a and 11a were less active than the unsubstituted 9a. Compound 16a with a (naphthalene-1-yl)methyl substitution, and, bulk at the 2′ position, is 8.4-fold less active than 21. Conversely the (naphthalene-2-yl)methyl substitution in compound 17a is beneficial for activity, with this compound being only 2.8- fold less active than 21 and the most potent compound in this series against VEGFR-1. Compound 12a with a 4′-Cl benzyl is 4.5- fold less active than 21; while 11a with a 2′-Cl benzyl is 7- fold less active than 19. Excessive bulk in the 4′-position is not tolerated as exemplified by the (biphenyl-4-yl)methyl substituted 18a which is inactive at a concentration of 200 μM. Dichloro benzyl substituted compounds 14a and 15a were 8- and 7- fold less active than 21 respectively. The 2′,5′-dimethoxy benzyl substituted compound is inactive at 200 μM. The 3′,4′,5′-triOMe benzyl substituted compound 19a, which lacks the 2′-substitution is about 8.8- fold less active than 21. Compared to the parent 6- benzyl substituted series, 9a–19a showed a decrease in VEGFR-1 inhibition.

In the EGFR assay, the 2′,5′-dimethoxy benzyl substituted compound 13a exhibited 2-digit micromolar inhibition. Though this compound was the most potent compound of the series in this assay, its activity was 346- fold lower than the standard 20. Compounds 10a with a 2′-Me benzyl and 12a with a 4′-Cl benzyl substitution were about 500- fold less potent than 20. Compound 9a with an unsubstituted phenyl group was about 800- fold less active than 20. EGFR inhibition was abolished on variations with a bulky substitution (1-naphthyl methyl in 16a, 2-naphthyl methyl in 17a, and 4-phenyl benzyl in 18a). In addition, alternative di-substitution (2′,4′-diCl benzyl in 14a, 3′,4′-diCl benzyl in 15a) and tri-substitution (3′,4′,5′-triOMe benzyl in 19a) also lead to inactivity in the EGFR assay. Relative to the parent 6-benzyl substituted series (9–19), 9a–19a showed a decrease in activity in the EGFR assay.

The most potent compound, in the A431 cytotoxicity assay, the 4-phenyl benzyl substituted 18a was equipotent to standard 20, 1.4- fold better than 7 and 1.7- fold less active than 23. The 2′-Me benzyl substituted 10a, 4′-Cl benzyl substituted 12a, and 2′,4′-diCl benzyl substituted 14a were equipotent to the standard 20, and 1.2- fold more potent than 7. Every compound had an improved A431 cytotoxicity value relative to their corresponding EGFR kinase IC₅₀’s. This is presumably due to the varied transport and/or pharmacokinetic differences of analogs between the two cell systems.

Taken together, transposing the 6- benzyl substituent in the parent compounds to the 7- position produced significant differences in the inhibition of the various RTKs. Although the EGFR inhibition present in the parent series was lost in 9a–19a, some of these compounds (11a, 19a) were much more potent PDGFRβ inhibitors. The variation in activity of analogs 9a–19a may either be due to the stringent requirements for binding in individual RTKs and/or due to differences in cell permeability of these compounds.

From this study, compounds 11a and 19a were identified as inhibitors of multiple RTKs. The discovery of these compounds is significant because they are potent inhibitors of VEGFR-2. In addition, 11a and 19a inhibit PDGFRβ at concentrations better than the standard agents. As stated above VEGFR-2 and PDGFRβ are the two principal mediators of angiogenesis. Compared to the most potent multikinase inhibitor 11a, the most potent multikinase inhibitor in the parent series – compound 13 – had a 2′,5′-diOMe benzyl group at N6 and exhibited inhibition against PDGFRβ at values 1.5-fold less than the standard 22 and in addition, demonstrated VEGFR-2 inhibition seventeen-fold better than the standard 23 (Table 1).

Molecular Modeling Studies

Molecular modeling studies were carried out using Flexx 3.1.2³⁶ and MOE 2008.10³⁷ in VEGFR2 (pdb code 1YWN)³⁸ and a homology model of PDGFRβ³⁹ Figures 3–5 depict the docked poses of 11 and its regioisomer 11a as a representative study in VEGFR2 and the homology model of PDGFRβ Figure 3 shows the superimposition of the best scored docked poses of 11 and 11a in the ATP binding site of VEGFR-2. The binding site of ATP competitive inhibitors in RTKs consists of a Hinge region, two hydrophobic binding sites (Hydrophobic Region I and II) and a Sugar binding pocket (Figure 3).^25,40–42 Compounds 11 and 11a can be seen to adopt different docked conformations in the active site. The pyrrolo[2,3-d]pyrimidine ring of 11a occupies the adenine binding portion of the ATP binding site. The 2-NH₂ moiety is involved in a hydrogen bond with Glu915 in the Hinge region while the N3 and 4-anilino NH are involved in hydrogen bonds with the backbone of Cys917 in the Hinge region. Additional hydrophobic interactions of the pyrrolo[2,3-d]pyrimidine ring with Leu1033 (not labeled) can stabilize the docked pose. In this pose, the N7 benzylic substitution extends towards Hydrophobic region I and is involved in interactions with Val846, Ala864, Val897 and Val914. The N7 benzylic substitution also interacts with the side chain carbon atoms of Glu883 and Cys1043. The N⁴-(3-bromophenyl) is extended towards Hydrophobic region II and interacts with the side chains of Phe916, Leu838(not shown) and Leu1033 (not shown).

Figure 3.

Stereo view. Superimposition of docked poses of 11 (red) and 11a (white) in the ATP binding site of VEGFR2. (PDB: 1YWN)

Figure 5.

Stereo view. Superimposition of the best scored pose (white) and alternate docked pose (blue) of 11a in the putative binding site of PDGFRβ The putative binding site of PDGFRβ was generated by homology modeling.³⁹

In contrast, the pyrrolo[2,3-d]pyrimidine scaffold of 11 docks in a flipped conformation compared to 11a described above that enables the formation of three hydrogen bonds with the hinge region (Figure 3). The 2-NH₂, N1 and pyrrole NH of 11 form hydrogen bonds with the backbone of Glu915 and Cys917. Additionally, the pyrrolo[2,3-d]pyrimidine scaffold forms hydrophobic interactions with Leu838 (not shown), Val846, Ala864, Val897 and Leu1033 (not shown). This mode places the N⁴-(3-bromophenyl) moiety in Hydrophobic region I where it forms hydrophobic interactions with Val846, Ala864, Leu887, Val897, Val912 and Val914. Additional hydrophobic interactions with the side chain carbon atoms of Lys866 and Glu883 stabilize this docked pose. The 6-benzylic substitution extends towards Hydrophobic region II and forms interactions with Leu838 (not shown), side chain atoms of Phe916, Cys917, Lys918 and Gly920 (not shown). Thus, molecular modeling and docking studies suggests that the 7-benzylic substitution forces 11a to adopt a binding mode different from that docked for the 6-benzylic compound 11 in VEGFR2.

There is no reported crystal structure of PDGFRβ bound to a ligand. Hence a homology model of PDGFRβ was built using the structure of c-KIT kinase complex (pdb code 1PKG) as a template.³⁰ Docking studies were performed with 11 and 11a as described above for VEGFR-2. Compound 11a binds to the ATP binding site of PDGFRβ with the pyrrolo[2,3-d]pyrimidine portion occupying the adenine binding site (Figure 4). Critical hydrogen bonds with the Hinge region are maintained in this binding mode: the aniline NH hydrogen bonds with the backbone of Cys684 while the N3 and 2-NH₂ moieties form hydrogen bonds with Glu682. This binding mode places the N⁴-(3-bromophenyl) moiety in Hydrophobic region I and forms hydrophobic interactions with Val614, Ala632 (not shown), Val665 and Leu833 (not shown) and with the side chain carbon atoms of Lys634 and Thr681. The N7-benzyl substituent lies in the Sugar binding pocket where it can interact with Leu606 (not shown), Val614, Val615 (not shown) and Ala848. A similar binding mode was seen with the best scored pose of 11 in PDGFRβ. The 6-benzylic substituent of 11 accesses the same region accessed by the 7-benzylic substituent of 11a however in a different conformation as shown in Figure 4. It was interesting to note that alternate binding modes of 11a that scored 1–2 kcal/mol higher than the best docked pose indicate different bound conformations of the 7-benzylic substituent (Figure 5). In the alternate binding mode, the 7-benzylic substituent accesses Hydrophobic region II instead of the sugar binding pocket as is seen in the bound conformation in Figure 5 where it interacts with Phe916 and Leu938 (not shown). Compound 11, which has a 6-benzylic substituent cannot access Hydrophobic region II in the poses seen in Figures 3 and 4. Similar poses were observed from the docking studies on other compounds. These binding modes suggested that the presence of multiple docked poses could translate into differences in the activity and selectivity of these compounds against different kinases as compared to the parent series.

Figure 4.

Stereo view of the docked pose of 11 (red) and 11a (white) in the putative binding site of PDGFRβ The putative binding site of PDGFRβ was generated by homology modeling.³⁰

On the basis of its in vitro activity, compound 11a was selected for in vivo evaluation in a mouse model to determine its effect on tumor growth, metastasis, and angiogenesis.

To examine whether compound 11a was effective in vivo in reducing tumor volume and metastasis, a syngeneic mouse tumor model was used. Athymic mice implanted with COLO-205 metastatic colon cancer cells, shown to overexpress PDGFRβ were used.⁴³ In a six week toxicity determination study, it was found that 35 mg/kg of 11a and 10 mg/kg (MTD) 8⁴⁴ caused no overt toxicity in mice (data not shown). At day 10 after implantation, the multitargeted inhibitor 8 at the MTD of 10 mg/kg three times weekly and 11a at 35 mg/kg three times weekly were administered to mice (Fig. 6A). Tumor growth was measured using calipers. It was found that 8 and 11a resulted in significant decreases in primary tumor growth rate (Fig. 6B) compared to the control. At the end of the experiment, tumors were taken and stained for blood vessels as described in the methods section. It was found that both 8 and 11a resulted in a significant decrease in primary tumor vascularity as compared to control animals (Fig. 6C). Finally, livers were stained for the presence of metastases and it was found that 11a, but not 8, resulted in significantly reduced numbers of COLO-205 metastases to the liver (Fig. 6D). Thus 11a demonstrated significant tumor inhibitory activity, antiangiogenic activity and antimetastatic activity in a COLO-205 mouse tumor xenograft model and is being considered for further preclinical development.

Figure 6.

Tumor growth (A), primary tumor growth rate (B), primary tumor vascularity (C), and liver metastasis of COLO-205 tumor xenografts after various treatments (D). COLO-205 cells were implanted into male athymic mice as in the research design and methods section. Compounds were administered at the listed doses three times weekly and tumor growth measured using calipers. At the experiment end, tumors were stained for vascularity and livers stained for metastases as in research design and methods section. Data represent the average +/− SD of 6–9 animals. * P>0.05, ** P<0.01, *** P<0.001 by one way ANOVA and Repeated Measures post test. The average tumor growth rate for control treated animals (B) is 100.2 mm³/day, average number of tumor vessels in control-treated animals are 14.3/field (C) and average metastases in control treated animals are 1.2/lobe.

In summary, eleven novel N⁴-(3-bromophenyl)-7-(substitutedbenzyl) pyrrolo[2,3-d]pyrimidines 9a–19a were designed and synthesized as RTK inhibitors. The biological evaluation showed that several analogs afforded potent inhibition against PDGFRβ, VEGFR-2 and the A431 cytotoxicity assay. Of the analogs evaluated, 11a and 19a were better than or comparable to the standard compounds used in these assays against two principal RTKs responsible for angiogenesis in tumors – PDGFRβ and VEGFR-2. Compound 11a was evaluated in a COLO-205 xenograft mouse model, where it exhibited tumor growth inhibition, antimetastatic and antiangiogenic effects better than or comparable to standard 8. Using the analogs described in this study as leads, structure based design and synthesis of novel multi-RTK inhibitors are currently in progress. In addition, selected compounds from this study are being evaluated in preclinical studies and in other tumor xenograft assays. One of the observed differences on transposing the 6- benzyl substituent in the parent compounds to the 7- position was that - although the EGFR inhibition present in the parent series was lost in 9a–19a, some of these compounds (11a, 19a) were much more potent PDGFRβ inhibitors. Molecular modeling studies indicate that access of Hydrophobic region II by the 7-benzyl substituent in this series could explain, in part, the observed differences in activities as compared to the 6-substituted lead series of compounds.

Experimental Section

Analytical samples were dried in vacuo (0.2 mm Hg) in a CHEM-DRY drying apparatus over P₂O₅ at 80 °C. Melting points were determined on a MEL-TEMP II melting point apparatus with FLUKE 51 K/J electronic thermometer and are uncorrected. Nuclear magnetic resonance spectra for proton (¹H NMR) were recorded on a Bruker WH-300 (300 MHz) or a Bruker 400 MHz/52 MM (400 MHz) spectrometer. The chemical shift values are expressed in ppm (parts per million) relative to tetramethylsilane as an internal standard: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad singlet. Thin-layer chromatography (TLC) was performed on Whatman Sil G/UV254 silica gel plates with a fluorescent indicator, and the spots were visualized under 254 and 366 nm illumination. Proportions of solvents used for TLC are by volume. Column chromatography was performed on a 230–400 mesh silica gel (Fisher, Somerville, NJ) column. Elemental analyses were performed by Atlantic Microlab, Inc., Norcross, GA. Element compositions are within (0.4% of the calculated values and indicate >95% purity. Fractional moles of water or organic solvents frequently found in some analytical samples could not be prevented in spite of 24–48 h of drying in vacuo and were confirmed where possible by their presence in the ¹H NMR spectra. All solvents and chemicals were purchased from Aldrich Chemical Co. or Fisher Scientific and were used as received. High resolution mass spectra (HRMS) were obtained for compounds 13a and 18a and TLC of these compounds produced a single spot in three different solvent systems.

4-chloro-7H-pyrrolo[2,3-d]pyrimidin-2-amine (27)

To a round-bottomed flask were added 26 (3 g, 19.9 mmol) of and 3 drops of N,N′ –Dimethyl aniline. The mixture was suspended on phosphorus oxychloride (250 mL). The reaction was continued at reflux for 4 h. After cooling to room temperature, the solvent was evaporated under reduced pressure and the residue was treated with ice water (20 mL) in an ice bath. The pH of the suspension was adjusted to 7 with concentrated ammonium hydroxide solution. The precipitate was filtered, air-dried and then dissolved in methanol. Silica gel (6 g) was added to the solution which was then evaporated to dryness to form a plug. The silica gel plug obtained was loaded onto a silica gel column and eluted with 2% methanol in chloroform. Fractions corresponding to the product (TLC) were pooled and evaporated to dryness under reduced pressure to afford 27 (10%) as a sparkling white solid: mp 215–216 °C; TLC R_f 0.56 (CHCl₃/CH₃OH, 5:1 with 2 drops of concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 6.24 (s, 1H, C5–CH), 6.49 (bs, 2H, NH₂, exch), 7.08 (s, 1H, C6–CH), 11.46 (s, 1H, NH, exch). Anal. (C₆H₅ClN₄•0.25H₂O) C, H, N, Cl.

N⁴-(3-bromophenyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (29)

To a round bottomed flask were added 27 (111 mg, 0.65 mmol)and 3-bromoaniline 28 (226 mg, 1.31 mmol) in isopropanol (10 mL). Following this 3 drops of concd HCl were added. The reaction was continued at reflux for 4 h at the end of which the solvent was evaporated under reduced pressure. The residue was dissolved in methanol (10 mL), and silica gel (250 mg) was added to the solution which was then evaporated to dryness to form a plug. The silica gel plug obtained was loaded onto a silica gel column and eluted with 2% methanol in chloroform. Fractions corresponding to the product (TLC) were pooled and evaporated to dryness under reduced pressure to afford 29 (70%) as a green solid: mp 210–212.5 °C; TLC R_f 0.56 (CHCl₃/CH₃OH, 5:1 with 2 drops of concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.81 (bs, 2H, NH₂, exch), 6.55 (s, 1H, C5–CH), 6.79 (s, 1H, C6–CH), 7.11 (s, 1H, Ar–H), 7.22 (d, J = 12 Hz, 1H, Ar–H), 8.05 (d, J = 12 Hz, 1H, Ar–H), 8.14 (s, 1H, Ar–H), 9.04 (s, 1H, NH, exch), 10.91 (s, 1H, NH, exch). Anal. (C₁₂H₁₀BrN₅• 0.1H₂O) C, H, N, Br.

General procedure for the synthesis of compounds 9a–19a

To a round bottomed flask was added 29 and dissolved in 5 ml of DMF. The solution was cooled to 0 °C and sodium hydride was added. After stirring at 0 °C for 1 hour, the appropriate (halomethyl)arene was added. The reaction was continued at R.T. till the appearance of a new spot (TLC) after which the reaction was quenched with water. The water phase was extracted with chloroform. The organic phase was dried over sodium sulfate and evaporated under reduced pressure. The residue was dissolved in methylene chloride, 250 mg silica gel was added to the solution which was then evaporated to dryness to form a plug. The silica gel plug obtained was loaded onto a silica gel column and eluted with 1% methanol in chloroform. Fractions corresponding to the product (TLC) were pooled and evaporated to dryness under reduced pressure to afford the product.

7-benzyl-N⁴-(3-bromophenyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (9a)

Reaction of 29 (210 mg, 0.69 mmol), 30a (236 mg, 1.38 mmol) and sodium hydride (16 mg, 0.65 mmol), using the general procedure described above, gave 11a: yield 36%, mp >200 °C; TLC R_f 0.87 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.30 (s, 2H, CH₂), 6.03 (bs, 2H, NH₂, exch), 6.60 (d, J = 9 Hz, 1H, Ar H), 6.71 (s, 1H, C5-CH), 6.91 (s, 1H, C6-CH), 7.12–7.15 (s, 1H, Ar–H), 7.22–7.31 (m, 4H, Ar–H), 7.49–7.51 (m, 2H, Ar–H), 8.07–8.13 (m, 2H, Ar–H), 9.17 (s, 1H, NH, exch). Anal. (C₁₉H₁₆N₅Br.5CHCl₃) C, H, N, Br.

N⁴-(3-bromophenyl)-7-(2-methylbenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (10a)

Reaction of 29 (500 mg, 1.64 mmol), 30b (607 mg, 3.28 mmol) and sodium hydride (39 mg, 1.64 mmol), using the general procedure described above, gave 10a: yield 34%, mp 177 °C; TLC R_f 0.85 (CHCl₃/CH₃OH, 5:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 2.30 (s, 3H, CH₃), 5.19 (s, 2H, CH₂), 5.97 (bs, 2H, NH₂, exch), 6.61 (m, 1H, Ar–H), 6.65 (s, 1H, C5-CH), 6.77 (s, 1H, C6-CH), 7.05–7.27 (m, 5H, Ar–H), 7.95–8.13 (m, 2H, Ar–H), 9.12 (s, 1H, NH, exch). Anal. (C₂₀H₁₈N₅Br) C, H, N, Br.

N⁴-(3-bromophenyl)-7-(2-chlorobenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (11a)

Method A: Reaction of 29 (200 mg, 0.65 mmol), 30c (270 mg, 1.31 mmol) and sodium hydride (15 mg, 0.62 mmol), using the general procedure described above, gave 11a: yield 7%.

Method B: To a round bottomed flask were added 31 (320 mg, 1.1 mmol)and 3-bromoaniline 28 (1.42 g, 8.25 mmol) in isopropanol (10 mL), followed by the addition of 3 drops of concd HCl. The reaction was continued at reflux for 4 h at the end of which the solvent was evaporated under reduced pressure. The residue was dissolved in methanol (10 mL), and silica gel gel (3 g) was added to the solution which was then evaporated to dryness to form a plug. The silica gel plug obtained was loaded onto a silica gel column and eluted with plain chloroform. Fractions corresponding to the product (TLC) were pooled and evaporated to dryness under reduced pressure to afford 11a (81%) as a white solid: mp 182.8 °C; TLC R_f 0.87 (CHCl₃/CH₃OH, 5:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.28 (s, 2H, CH₂), 6.02 (bs, 2H, NH₂, exch), 6.59 (d, J = 6 Hz, 1H, Ar–H), 6.68 (s, 1H, C5-CH), 6.89 (s, 1H, C6-CH), 7.12 (d, J = 6 Hz, 1H, Ar–H), 7.21–7.29 (m, 3H, Ar–H), 7.49 (d, J = 9 Hz, 1H, Ar–H), 8.05–8.11 (m, 2H, Ar–H), 9.17 (s, 1H, NH, exch). Anal. (C₁₉H₁₅N₅BrCl) C, H, N, Br, Cl.

N⁴-(3-bromophenyl)-7-(4-chlorobenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (12a)

Reaction of 29 (200 mg, 0.65 mmol), 30d (211 mg, 1.31 mmol) and sodium hydride (15 mg, 0.62 mmol), using the general procedure described above, gave 12a: yield 47%, mp 127–128 °C; TLC R_f 0.88 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.19 (s, 2H, CH₂), 6.04 (bs, 2H, NH₂, exch), 6.63 (d, J = 3 Hz, 1H, C5-CH), 6.89 (d, J = 3 Hz, 1H, C6-CH), 7.14–8.09 (m, 8H, Ar–H), 9.14 (s, 1H, NH, exch). Anal. (C₁₉H₁₅N₅BrCl) C, H, N, Br, Cl.

N⁴-(3-bromophenyl)-7-(2,5-dimethoxybenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (13a)

Reaction of 29 (300 mg, 0.98 mmol), 30e (368 mg, 1.97 mmol) and sodium hydride (22 mg, 0.93 mmol), using the general procedure described above, gave 13a: yield 10%, mp 150.9 °C; TLC R_f 0.71 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 3.56 (s, 3H, OCH₃), 3.79 (s, 3H, OCH₃), 5.11 (s, 2H, CH₂), 6.00 (bs, 2H, NH₂, exch), 6.11 (s, 1H, Ar–H), 6.64 (d, J = 3 Hz, 1H, C5-CH), 6.77–6.79 (m, 1H, Ar–H), 6.85 (d, J = 3 Hz, 1H, C6-CH), 6.92–8.11 (m, 5H, Ar–H), 9.13 (s, 1H, NH, exch). HRMS calcd for C₂₁H₂₀N₅O₂Br 453.0800, found 453.0788.

N⁴-(3-bromophenyl)-7-(2,4-dichlorobenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (14a)

Reaction of 29 (100 mg, 0.32 mmol), 30f (125 mg, 0.64 mmol) and sodium hydride (7 mg, 0.32 mmol), using the general procedure described above, gave 14a: yield 24%, mp 191.4 °C; TLC R_f 0.86 (CHCl₃/CH₃OH, 5:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆) δ 5.26 (s, 2H, CH₂), 6.05 (bs, 2H, NH₂, exch), 6.58 (d, J = 6 Hz, 1H, Ar–H), 6.69 (s, 1H, C5–CH), 6.89 (s, 1H, C6–CH), 7.13–7.35 (m, 3H, Ar–H), 7.67 (s, 1H, Ar–H), 8.07–8.11 (m, 2H, Ar–H), 9.18 (s, 1H, NH, exch). Anal. (C₁₉H₁₄N₅BrCl₂) C, H, N, Br, Cl.

N⁴-(3-bromophenyl)-7-(3,4-dichlorobenzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (15a)

Reaction of 29 (200 mg, 0.65 mmol), 30g (256 mg, 1.31 mmol) and sodium hydride (15 mg, 0.62 mmol), using the general procedure described above, gave 15a: yield 16%, mp 136.3 °C; TLC R_f 0.78 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆) δ 5.20 (s, 2H, CH₂), 6.02 (bs, 2H, NH₂, exch), 6.64 (d, J = 3 Hz, 1H, C5–CH), 6.92 (d, J = 3 Hz, 1H, C6–CH), 7.09–8.10 (m, 7H, Ar–H), 9.13 (s, 1H, NH, exch). Anal. (C₁₉H₁₄BrCl₂N₅•0.25CHCl₃) C, H, N, Br, Cl.

N⁴-(3-bromophenyl)-7-(1-naphthylmethyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (16a)

Reaction of 29 (500 mg, 1.64 mmol), 30h (579 mg, 3.28 mmol) and sodium hydride (39 mg, 1.64 mmol), using the general procedure described above, gave 16a: yield 18%, mp 237.5 °C; TLC R_f 0.92 (CHCl₃/CH₃OH, 5:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.67 (s, 2H, CH₂), 6.01 (bs, 2H, NH₂, exch), 6.64 (d, J = 3 Hz, 1H, C5–CH), 6.82 (d, J = 3 Hz, 1H, C6–CH), 6.93–8.21 (m, 11H, Ar–H), 9.12 (s, 1H, NH, exch). Anal. (C₂₃H₁₈N₅Br) C, H, N, Br.

N⁴-(3-bromophenyl)-7-(2-naphthylmethyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (17a)

Reaction of 29 (200 mg, 0.65 mmol), 30i (290 mg, 1.31 mmol) and sodium hydride (15 mg, 0.62 mmol), using the general procedure described above, gave 17a: yield 19%, mp 130.2 °C; TLC R_f 0.73 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.37 (s, 2H, CH₂), 5.98 (bs, 2H, NH₂, exch), 6.65 (d, J = 3 Hz, 1H, C5–CH), 6.92 (d, J = 3 Hz, 1H, C6–CH), 7.12–8.12 (m, 11H, Ar–H), 9.11 (s, 1H, NH, exch). HRMS calcd for C₂₃H₁₈N₅Br 443.0745, found 443.0792.

7-Biphenyl-4-ylmethyl-N⁴-(3-bromo-phenyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (18a)

Reaction of 29 (500 mg, 1.64 mmol), 30j (666 mg, 3.28 mmol) and sodium hydride (37 mg, 1.55 mmol), using the general procedure described above, gave 18a: yield 10%, mp 86.2 °C; TLC R_f 0.82 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); ¹H NMR (DMSO-d₆): δ 5.24 (s, 2H, CH₂), 5.99 (bs, 2H, NH₂, exch), 6.64 (s, 1H, C5–CH), 6.91 (s, 1H, C6–CH), 7.07–8.11 (m, 13H, Ar–H), 9.12 (s, 1H, NH, exch). HRMS calcd for C₂₅H₂₀N₅Br 469.0902, found 469.1001.

N⁴-(3-Bromo-phenyl)-7-(3,4,5-trimethoxy-benzyl)-7H-pyrrolo[2,3-d]pyrimidine-2,4-diamine (19a)

Reaction of 29 (400 mg, 1.31 mmol), 30k (570 mg, 2.63 mmol) and sodium hydride (31 mg, 1.31 mmol), using the general procedure described above, gave 19a: yield 14%, mp 113.6 °C; ¹H NMR (DMSO-d₆): δ 3.61 (s, 3H, OCH₃), 3.69 (s, 6H, 2 x OCH₃), 5.10 (s, 2H, CH₂), 5.99 (bs, 2H, NH₂, exch), 6.60 (s, 1H, C5–CH), 6.60 (s, 2H, Ar–H), 6.90 (s, 1H, C6–CH), 7.12–8.12 (m, 4H, Ar–H), 9.09 (s, 1H, NH, exch). TLC R_f 0.85 (CHCl₃/CH₃OH, 10:1 with 2 drops concentrated NH₄OH); Anal. (C₂₂H₂₂N₅BrO₃) C, H, N, Br.

4-chloro-7-(2-chlorobenzyl)-7H-pyrrolo[2,3-d]pyrimidin-2-amine (31)

To a round bottomed flask was added 27 (60 mg, 0.35 mmol) and dissolved in 5 ml of DMF. The solution was cooled to 0 °C and sodium hydride (9 mg, 0.33 mmol) was added. After stirring at 0 °C for 1 hour 30c (81 mg, 0.39 mmol) was added. The reaction was continued at R.T. for 2 h, at the end of which the reaction was quenched with water. The water phase was extracted with ethylacetate. The organic phase was dried over sodium sulfate and evaporated under reduced pressure. The residue was dissolved in methylene chloride, 250 mg silica gel was added to the solution which was then evaporated to dryness to form a plug. The silica gel plug obtained was loaded onto a silica gel column and eluted with 4: 1 hexane: ethylacetate. Fractions corresponding to the product (TLC) were pooled and evaporated to dryness under reduced pressure to afford the product 31 as a white solid: yield 48%, mp 148 °C; TLC R_f 0.23 (Hexane/EtOAc, 4:1); ¹H NMR (DMSO-d₆): δ 5.32 (s, 2H, CH₂), 6.39 (d, J = 4 Hz, 1H, C5–CH), 6.65–6.73 (bs, 2H, NH₂, exch), 6.71 (d, J = 16 Hz, 1H, Ar-H), 7.17–7.18 (d, 1H, C6–CH), 7.22–7.50 (m, 4H, Ar–H). Anal. (C₁₃H₁₀Cl₂N₄) C, H, N, Cl.

Molecular Modeling and Computational Studies

The X-ray crystal structure of VEGFR2 at 1.71Å resolution was obtained from the protein database (PDB ID 1YWN)³⁸. This crystal structure contains VEGFR2 in complex with a novel 4-amino-furo[2,3-d]pyrimidine. Docking studies were performed using the docking suite of Molecular Operating Environment software (MOE 2008.10)³⁷ and Flexx 3.1.2.⁴⁵ After addition of hydrogen atoms, the protein was then “prepared” using the LigX function in MOE. LigX is a graphical interface and collection of procedures for conducting interactive ligand modification and energy minimization in the active site of a flexible receptor. In LigX calculations, the receptor atoms far from the ligand are constrained and not allowed to move while receptor atoms in the active site of the protein are allowed to move but are subject to tether restraints that discourage gross movement.³⁷ This procedure was performed with the default settings. Ligands were built using the molecule builder function in MOE and were energy minimized to its local minima using the MMF94X forcefield to a constant of 0.05 kcal/mol. Ligands were docked into the active site of the prepared protein using the docking suite as implemented in MOE. The docking was restricted to the active site pocket residues using Alpha triangle placement method. Refinement of the docked poses was carried out using the Forcefield refinement scheme and scored using Affinity dG scoring system. Around 30 poses were returned for each compound at the end of each docking run. The docked poses were manually examined in the binding pocket to ensure quality of docking and to confirm absence of steric clashes with the amino acid residues of the binding pocket.

To validate the utility of MOE 2008.10 for docking ligands into the active site, the native ligand in the crystal structure (PDB: 1YWN) was built using the molecule builder, energy minimized using MMFF94x forcefield to a gradient of 0.05 kcal/mol and docked into the active site using the above parameters. The best docked pose of the native ligand displayed an RMSD of 0.6031 Å compared to the crystal structure pose. MOE 2008.10 was thus validated for our docking process.

Docking with Flexx 3.1.2 was carried out in by defining the active site as a sphere of ~6.5 Å from the ligand. The protonation states utilized for the proteins and the ligands were calculated using the default settings. Water molecules in the active site were permitted to rotate freely. Ligands for docking were prepared using MOE and energy minimized using the MMF94X forcefield to a constant of 0.05 kcal/mol. Triangle matching was used as the placement method and the docked poses were scored using default settings. The docked poses were exported and visualized in MOE. The best docked pose had an RMSD of 0.7686 Å. Flexx 3.1.2 was thus validated for our docking process.

Docking studies were performed for 11 and 11a in MOE and Flexx using a similar procedure. Poses from the docking experiment performed in Flexx were exported and visualized in MOE.

Cells

All cells were maintained at 37 °C in a humidified environment containing 5% CO₂ using media from Mediatech (Hemden, NJ). A-431 cells were from the American Type Tissue Collection (Manassas, VA).

Chemicals

All growth factors (bFGF, VEGF, EGF, and PDGF-β) were purchased from Peprotech (Rocky Hill, NJ). PD153035, SU5416, AG1295, and CB676475 (4-[(4′-chloro-2′-fluoro)phenylamino]-6,7-dimethoxyquinazoline) were purchased from Calbiochem (San Diego, CA). The CYQUANT cell proliferation assay was from Molecular Probes (Eugene, OR). All other chemicals were from Sigma Chemical unless otherwise noted.

Antibodies

The PY-HRP antibody was from BD Transduction Laboratories (Franklin Lakes, NJ). Antibodies against EGFR, PDGFR-β, FGFR-1, Flk-1, and Flt-1 were purchased from Upstate Biotech (Framingham, MA).

Phosphotyrosine ELISA

Cells used were tumor cell lines naturally expressing high levels of EGFR (A431), Flk-1 (U251), Flt-1 (A498), PDGFR-β (SF-539), and FGFR-1 (NIH OVCAR-8). Expression levels at the RNA level were derived from the NCI Developmental Therapeutics Program (NCI-DTP) web site public molecular target information (http://www.dtp.nci.nih.gov/mtargets/mt_index.html). Briefly, cells at 60–75% confluence were placed in serum-free medium for 18 h to reduce the background of phosphorylation. Cells were always >98% viable by Trypan blue exclusion. Cells were then pretreated for 60 min with 10, 3.33, 1.11, 0.37, and 0.12 μM compounds followed by 100 ng/ml EGF, VEGF, PDGF-BB, or bFGF for 10 min. The reaction was stopped and cells permeabilized by quickly removing the media from the cells and adding ice-cold Tris-buffered saline (TBS) containing 0.05% Triton X-100, protease inhibitor cocktail, and tyrosine phosphatase inhibitor cocktail. The TBS solution was then removed and cells fixed to the plate for 30 min at 60 °C and further incubation in 70% ethanol for an additional 30 min. Cells were further exposed to block (TBS with 1% BSA) for 1 h, washed, and then a horseradish peroxidase (HRP)-conjugated phosphotyrosine (PY) antibody added overnight. The antibody was removed, cells were washed again in TBS, exposed to an enhanced luminal ELISA substrate (Pierce Chemical, Rockford, IL), and light emission measured using a UV product (Upland, CA) BioChemi digital darkroom. The known RTK-specific kinase inhibitor, PD153035, was used as a positive control compound for EGFR kinase inhibition; SU5416 for Flk1 kinase inhibition; AG1295 for PDGFR-β kinase inhibition; and CB676475 (4-[(4′-chloro-2′-fluoro)phenylamino]-6,7-dimethoxyquinazoline) was used as a positive control for both Flt1 and Flk1 kinase inhibition. Data were graphed as a percent of cells receiving growth factor alone and IC₅₀ values were estimated from two to three separate experiments (n = 8–24) using hand-drawn probit plots. In each case, the activity of a positive control inhibitor did not deviate more than 10% from the IC₅₀ values listed in the text.

CYQUANT cell proliferation assay

As a measure of cell proliferation, the CYQUANT cell counting/proliferation assay was used as previously described.³⁵ Briefly, cells are first treated with compounds for 12 h and then allowed to grow for an additional 36 h. The cells are then lysed and the CYQUANT dye, which intercalates into the DNA of cells, is added and after 5 min the fluorescence of each well measured using a UV product BioChemi digital darkroom. A positive control used for cytotoxicity in each experiment was cisplatin, with an apparent average IC₅₀ value of 8.2 ± 0.65 μM. Data are graphed as a percent of cells receiving growth factor alone and IC₅₀ values estimated from two to three separate experiments (n = 6–15) using probit plots.

Statistics

All analysis was done using Prism 4.0. (GraphPad Software, San Diego, CA).

Whole animal toxicity assay

To first determine the maximal tolerated dose (MTD) of a drug, groups of four male NCr nu/nu mice at 8 weeks of age without tumors were injected with 5, 10, 15, 25, 35, and 45 mg/kg 11a and 5, 10, 15 and 20 mg/kg 8 three times weekly on Monday AM, Wednesday noon and Friday PM. Weights were taken and animals observed for acute distress during the first 96 hr after injection. No apparent toxicity and no significant weight loss (P>0.05; one-way ANOVA with Repeated Measures Post-Test) was observed throughout a 6 week period after drug administration.

COLO-205 human metastatic colon mouse xenograft

One million COLO-205 (liver colonizing; American Type Culture Collection) human metastatic colon cancer cells were injected SQ into the lateral flank of athymic NIH-II male mice, 8 weeks in age. Animals are monitored every other day for the presence of tumors. At the time in which most tumors are measurable by calipers (days 8–11 after implantation), animals with tumors are randomly sorted into treatment groups. DMSO stocks (30 mM) of drugs are further dissolved into sterile water for injection and the optimal dose (from MTD studies) injected intraperitoneally (IP). The length (long side), width (short side) and depth of the tumors was measured using digital Vernier Calipers each Monday, Wednesday, and Friday. Tumor volume is calculated using the formula length x width x depth. Tumor growth rate is was calculated using a linear regression analysis algorithm with the software GraphPad Prism 4.0.c (GraphPad software, San Diego, CA). At the experiment’s end, animals were humanly euthanized using carbon dioxide, tumors and liver excised, fixed in 20% neutral buffered formalin for 8–10 hr, embedded into paraffin, and hematoxylin-eosin (H&E) stain of 3–4 separate tissue sections completed to span the tumor/liver. Together with the OUHSC Department of Pathology core, metastases per liver lobe were counted using the H&E stained sections. The metastases could be seen as purple clusters of disorganized cells on the highly organized largely pink liver. Together with the OUHSC Department of Pathology core, blood vessels per unit area on primary tumors are also counted in 5 fields at 100x magnification and averaged.

Abbreviations

ATP

adenosine 5′triphosphate

VEGF

vascular endothelial growth factor

EGF

epidermal growth factor

PDGF

platelet derived growth factor

RTK

receptor tyrosine kinase

KDR

kinase insert domain receptor

VEGFR-2

vascular endothelial growth factor receptor-2

Flt-1

fms-related tyrosine kinase 4

VEGFR-1

vascular endothelial growth factor receptor-1

Flt-4

fms-related tyrosine kinase 4

VEGFR-3

vascular endothelial growth factor receptor-3

Flt-3

FMS-like tyrosine kinase-3

PDGFRβ

platelet derived growth factor receptor-β

CSF1R

colony stimulating factor 1 receptor

CML

chronic myelogenous leukemia

GIST

gastrointestinal stromal tumor

Abl

abelson tyrosine kinase

ErbB-2

erythroblastic leukemia viral oncogene homolog-2

TLC

thin layer chromatography

CAM

chorioallantoic membrane