Inhibition of Tubulin Polymerization by Select Alkenyldiarylmethanes

Abstract

During studies on the alkenyldiarylmethane (ADAM) class of non-nucleoside reverse transcriptase inhibitors (NNRTIs), analogues were discovered that exhibit low micromolar and sub-micromolar cytotoxicities. Since the ADAMs are structurally related to the tubulin polymerization inhibitor CC-5079, a set of fourteen ADAMs were tested for inhibition of tubulin polymerization in an attempt to identify the biological target responsible for their cytotoxicity. The results indicate that, overall, the ADAMs are poor inhibitors of tubulin polymerization. However, the two most cytotoxic compounds, 15 and 16, are in fact active as inhibitors of tubulin assembly with IC₅₀ values of 3.7 ± 0.3 and 2.8 ± 0.2 µM, respectively, and they both inhibit the binding of colchicine to tubulin. Both compounds were investigated for anticancer activity in the National Cancer Institute's panel of 60 human cancer cell lines, and both compounds consistently displayed submicromolar cytotoxicities with mean-graph midpoint (MGM) values of 0.31 ± 0.08 and 0.47 ± 0.09 µM, respectively.

Over the past twenty-five years, infection by the human immunodeficiency virus (HIV) has reached pandemic proportions, and an estimated 41 million people are believed to be carriers of the acquired immune deficiency syndrome’s (AIDS) etiological agent.¹ HIV has caused the deaths of more than 25 million people since its first major appearance in 1981¹, and developing a cure for HIV-infection is one of the major challenges currently facing medical science. Several FDA-approved drugs are available to combat HIV infections and AIDS progression. Unfortunately, the rapid mutation rate of HIV allows the virus to develop resistance to many antiviral agents as early as two months after initial anti-HIV treatment. Thus, until a cure is discovered, development of antiviral therapeutics that are active against both the wild-type and drug-resistant forms of HIV is a primary goal for AIDS researchers.^2–4

The alkenyldiarylmethane (ADAM) class of non-nucleoside reverse transcriptase inhibitors (NNRTIs) was discovered over ten years ago. The lead compounds 1 and 2 retain antiviral activity against the common HIV-1 reverse transcriptase (RT) drug-resistance mutations K103N and Y188C.^5–8 For this reason, the development of the ADAMs as potential antiviral therapeutics has been pursued. It has been established that the ADAMs exert their antiviral properties through the allosteric inhibition of HIV-1 RT. However, the observation that some ADAM analogues do not inhibit HIV-1 RT and yet still exhibit anti-HIV activity indicates that, at least, certain ADAMs interact with another viral or cellular entity, and this has led us to investigate other molecular targets.⁹

Another series of ADAMs has recently been developed by scientists at Celgene Corp. as potent inhibitors of inflammation, phosphodiesterase type 4 activity, and tubulin polymerization, where tubulin inhibition involves binding of the inhibitor to the same site as the natural antimitotic agent, colchicine (3, Chart 1).^10–13 The structural similarity between Celgene’s inhibitor CC-5079 (Chart 1) and our own antiviral agents led us to consider that the ADAMs may also exhibit one or more of the properties displayed by the Celgene inhibitors. It is well known that disrupting microtubule homeostasis causes cells to undergo apoptosis¹¹, and the low micromolar toxicity generally displayed by the ADAM class of NNRTIs led to the hypothesis that the ADAMs’ cytotoxic properties may originate from the inhibition of tubulin polymerization by binding to the colchicine binding site. Additionally, inhibition of tubulin polymerization may also account for the RT-independent antiviral activity displayed by many ADAM analogues. The vital HIV protein Tat is known to, among other things, stabilize the microtubule framework of HIV-infected cells by binding to microtubule-associated protein (MAP) binding sites.¹⁴ It has been proposed that the interactions between Tat and microtubules help facilitate the replication of HIV and may also contribute to the mechanism of HIV-related cell death.¹⁴ In light of this information, the inhibition of tubulin polymerization by a select group of ADAMs was investigated. Herein we report the syntheses, antiviral activity, and tubulin inhibitory effects of ADAMs 4–17.

Chart 1.

A number of methods have been developed for the synthesis of the ADAM scaffold, and the syntheses of ADAMs 4–10,^†,¹⁵ 12,¹⁵ 13,¹⁵ and 17¹⁶ have been published. ADAMs 11, 14, 15, and 16 were constructed via the general cross-coupling route depicted in Scheme 1. Sonogashira coupling of aryl halide 18^†,^†,¹⁷ and terminal alkyne 19,¹⁵ followed by hydrostannation, affords stannane intermediate 20. The stannane and aryl halide 21 are coupled via the Stille reaction to obtain the desired analogue.

Scheme 1.

a) 5 mol% PdCl₂(PPh₃), 10 mol% CuI, Et₃N, THF; b) 2 mol% Pd(PPh₃)₄, Bu₃SnH, THF, 0 °C; c) 10 mol% Pd(PPh₃)₄, 20–100 mol% CuI, CsF, DMF, 60 °C.

The tubulin polymerization inhibitory data^18–20 for ADAMs 4–17 are presented in Table 1, together with the antiviral data^*,^6,21–23 associated with the compounds. Nevirapine is included for antiviral activity comparisons, while colchicine and combretastatin A-4 are well-known inhibitors of tubulin polymerization, with the latter compound an exceptionally potent inhibitor of the binding of radiolabeled colchicine to tubulin. The majority of the compounds studied for inhibition of tubulin polymerization were chosen on the basis of their acute cytotoxicity, which one would expect to correlate with tubulin destabilizing activity. To our surprise, only two of the fourteen analogues investigated were capable of inhibiting tubulin polymerization at concentrations lower than 40 µM, despite the structural similarities observed between the compounds and Celgene’s inhibitors of tubulin polymerization. These data refute the hypothesis that inhibition of tubulin polymerization is the general source of ADAMs' activity. However, the tubulin IC₅₀’s observed for ADAMs 15 and 16 are not substantially greater than that of the potent antimitotic agent combretastatin A-4, indicating the extreme cytotoxicity observed with these two analogues most likely results from disruption of the microtubule network. Inhibitors of tubulin polymerization often induce cell death at concentrations much lower than their in vitro tubulin IC₅₀ values in cell-free systems, which accounts for the 1,000-fold difference in tubulin inhibitory activity and cytotoxicity observed for ADAMs 15 and 16. The ADAMs 12 and 13, which are structurally related to 15 and 16, were completely inactive as inhibitors of tubulin polymerization. Evidently, the relatively small structural change involving the movement of the nitrile from the para position in 16 to the meta position in 15 only causes a small drop in activity, but the more drastic alterations embodied in 12 and 13 are not tolerated.

Table 1.

Antiviral and tubulin polymerization inhibitory activities of ADAMs 4–17.

Compound	HIV-1 RT IC50(µM)a	EC50 (µM)b			CC50 (µM)c		Tubulin IC50 (µM)d
		1RF	1IIIB	2ROD	CEM-SS	MT-4
4	71	2.3	3.2	N.A.e	40	36	>40
5	0.3	0.013	0.60	2.5	32	160	>40
6	>100	13	2.6	29	>200	>198	>40
7	N.T.f	N.A.e	N.T.f	N.T.f	29	N.T.f	>40
8	99	8.2	3.0	N.A.e	>100	62	>40
9	100	53	N.T.f	N.T.f	20	N.T.f	>40
10	67	2.9	N.A.e	N.A.e	33	7.0	>40
11	40	N.A.e	N.A.e	N.A.e	1.1	0.97	>40
12	33	N.A.e	N.A.e	N.A.e	1.2	1.6	>40
13	5.2	N.A.e	N.A.e	N.A.e	0.78	1.1	>40
14	73	N.A.e	N.A.e	N.A.e	2.1	4.2	>40
15	93	N.A.e	N.A.e	N.A.e	0.004	0.20	3.7 ± 0.3
16	>100	N.A.e	N.A.e	N.A.e	0.004	0.34	2.8 ± 0.2
17	0.60	N.A.e	1.2	N.A.e	2.9	12	>40
Nevirapine	0.08424	0.0015	0.053	N.A.e	N.T.f	15	N.T.f
Combretastatin A-4	N.T.f	N.T.f	N.T.f	N.T.f	N.T.f	N.T.f	1.2 ± 0.1
Colchicine	N.T.f	N.T.f	N.T.f	N.T.f	N.T.f	N.T.f	3.8 ± 0.1

^aInhibitory activity versus HIV-1 RT with poly(rC).oligo(dG) as the template primer.

^bEC₅₀ is the concentration required to inhibit 50% of the cytopathic effect of HIV-1_RF in CEM-SS cells, HIV-1_IIIB in MT-4 cells, or HIV-2_ROD in MT-4 cells.

^cCC₅₀ is the cytotoxic concentration required to cause cell death for 50% of the mock-infected CEM-SS or MT-4 cells.

^dThe concentration required to inhibit 50% of tubulin polymerization in vitro. The reaction mixtures contained 10 µM tubulin, and the extent of assembly after 20 min at 30 °C in 0.8 M monosodium glutamate + 0.4 mM GTP was measured.

^eNot Active.

^fNot Tested.

In order to confirm that ADAMs 15 and 16 are in fact binding to the colchicine site of tubulin, the inhibition of colchicine binding to tubulin was determined at 37 °C in reaction mixtures incubated for 10 min and containing tubulin (1 µM), 15 or 16 (5 µM), and [³H]colchicine (5 µM). ADAMs 15 and 16 inhibited [³H]colchicine binding by 24 ± 1% and 42 ± 3%, respectively. Combretastatin A-4, which was used as a positive control, inhibited [³H]colchicine binding by 98 ± 0.3 %.

The ADAMs 15 and 16 were examined for antiproliferative activity against the human cancer cell lines in the National Cancer Institute screen, in which the activity of each compound was evaluated with approximately 55 different cancer cell lines of diverse tumor origins. The GI₅₀ values obtained with selected cell lines, along with the mean graph midpoint (MGM) values, are summarized in Table 2. The MGM is based on a calculation of the average GI₅₀ for all of the cell lines tested (approximately 55) in which GI₅₀ values below and above the test range (10⁻⁸ to 10⁻⁴ molar) are taken as the minimum (10⁻⁸ molar) and maximum (10⁻⁴ molar) drug concentrations used in the screening test. Both ADAMs 15 and 16 consistently produced submicromolar GI₅₀ values in these human cancer cell lines, resulting in MGM values of 0.31 ± 0.08 and 0.47 ± 0.09 µM, respectively.

Table 2.

Cytotoxicities of Alkenyldiarylmethanes 15 and 16.

	Compound GI50 (µM)a
Cell Line or MGMb	15	16
Lung (HOP-62)	0.76	0.55 ± 0.08
Colon (HCT-116)	0.49	0.46 ± 0.08
CNS (SF-539)	0.24 ± 0.06	0.30 ± 0.04
Melanoma (LOX IMVI)	0.59 ± 0.17	0.72 ± 0.19
Ovarian (OVCAR-3)	0.10 ± 0.06	0.20 ± 0.03
Renal (SN12C)	0.50 ± 0.07	0.78 ± 0.32
Prostate (DU-145)	0.32 ± 0.08	0.57 ± 0.14
Breast (MDA-MB-435)	0.089 ± 0.051	0.19 ± 0.02
MGM	0.31 ± 0.08	0.47 ± 0.09

^aConcentration for 50% growth inhibition

^bMean graph midpoint for growth inhibition of all human cancer cell lines found to be sensitive.

The results with analogues 4, 6, 8, and 10 are consistent with the conclusion that inhibition of tubulin polymerization is not required for the ADAMs' RT-independent antiviral mechanism. These compounds each exhibited RT-independent antiviral activity, as can be seen by comparing their in vitro RT inhibitory activities with their respective cytoprotective activities; yet, none of these four analogues had a significant effect on tubulin polymerization at 40 µM.

In summary, select ADAMs were evaluated for inhibition of tubulin polymerization, and the results indicate that most members of this structural class of compounds are poor inhibitors. These results do not support the hypothesis that disruption of microtubule stability is the origin of the ADAMs' cytotoxicity, nor do the results support an important role for tubulin in the ADAMs’ RT-independent antiviral mechanism. Additional studies are required to elucidate the so far unknown antiviral mechanism. Despite the poor tubulin inhibition generally exhibited by most members of the ADAM class described here, the highly cytotoxic analogues 15 and 16 were identified as potent tubulin destabilizing agents with activities not substantially greater than the activity of to the natural antimitotic agent combretastatin A-4. In light of this information, 15 and 16 could potentially be used as leads in the development of ADAM-based inhibitors of tubulin polymerization for the treatment of cancer.

Supporting Information

Physical constants, elemental analysis data, and experimental procedures for the synthesis of compounds 11, 14, 15, and 16 are included in the Supporting Information section of this report.

Chart 2.