Synthesis and biological activity of aminophthalazines and aminopyridazines as novel inhibitors of PGE₂ production in cells

Abstract

This article reports the synthesis and biological evaluation of a collection of aminophthalazines as a novel class of compounds capable of reducing production of PGE₂ in HCA-7 human adenocarcinoma cells. A total of 28 analogs were synthesized, assayed for PGE₂ reduction, and selected active compounds were evaluated for inhibitory activity against COX-2 in a cell free assay. Compound 2xxiv (R¹ = H, R² = p-CH₃O) exhibited the most potent activity in cells (EC₅₀ = 0.02 µM) and minimal inhibition of COX-2 activity (3% at 5 µM). Furthermore, the anti-tumor activity of analog 2vii was analyzed in xenograft mouse models exhibiting good anti-cancer activity.

Prostaglandin E₂ (PGE₂) is well known to play a pivotal role in processes associated with inflammation, pain and pyresis and is over expressed in various tumors where chronic inflammation has been linked to the growth of various cancerous tissues. Indeed, PGE₂ has been identified as the major prostaglandin associated with the progression of various tumor malignancies including that of the colon, lung, and breast.^1–5

Nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used as analgesics that function by inhibiting the activity of cyclooxygenase (COX) enzymes which are involved in the biosynthesis of a variety of prostaglandins, including PGE₂. At high doses NSAIDs have been shown to reverse the growth of colorectal tumors and patients using NSAIDs over long periods of time for chronic disorders (i.e rheumatoid arthritis) exhibit a lower risk of developing colon cancer.^1–4 It is postulated that the anti-cancer properties elicited by NSAIDs are primarily due to a reduction in PGE₂ levels supported by the observation that exogenous treatment with PGE₂ is shown to impede NSAID induced tumor regression. Historically, side effects with non-selective COX-1/2 inhibitors (Aspirin, Tylenol, Ibuprofen) driven by COX-1 mediated gastrointestinal intolerance, promoted efforts to develop specific COX-2 inhibitors (Celebrex^®, Bextra^® and Vioxx^®). However, the latter two drugs were subsequently withdrawn from the market due to a drug induced imbalance of prostacyclin (PGI₂) and thromboxane A₂ (TXA₂) leading to cases of myocardial infarction and thrombosis.^1–4 Thus the search for colon cancer modifying therapeutics is now geared towards the development of PGE₂ reducing drugs that do not alter COX activity.^7–11

In search of novel anti-tumor small molecules with the capacity to reduce cellular levels of PGE₂ without affecting in vitro COX-2 activity, our group has recently reported a series of 2-aminothiazoles with interesting biological profiles (Figure 1).¹¹ In particular, 1a exhibited the most potent cellular PGE₂ reducing activity of the entire series (EC₅₀ 90 nM), with only a nominal COX-2 inhibition [IC₅₀ >5 µM]. Furthermore, analog 1b exhibited promising anti-cancer activity in mouse xenograft models.¹¹ In continuation of our studies, we herein report the biological evaluation of a novel series of aminophthalazine analogs 2 with PGE₂ reducing character in HCA-7 colon cancer cells, Figure 1.

Figure 1.

Structures of 2-aminothiazoles 1a–b and general structure of aminophthalazines 2.

The aminophthalazines Table 1 (2i-xxiv), were assembled using methodology depicted in Scheme 1.^{12, 13} Briefly, 1,4-dichlorophthalazine 3 was treated at reflux with an equivalent amount of aniline 4 affording intermediates 5 in good yield. 5 smoothly underwent Suzuki-coupling with select boronic acids enabling formation of final aminophthalazine analogs 2i-xxiv in moderate yields. Analogs 6i-ii were obtained via standard N-methylation methodology from 2vi and 2vii respectively. Pyridazines 9i-ii were prepared from commercially available 1,4-dichloropyridazine 7 and p-toluidine 4 in a similar fashion to the aminophthalazines Scheme 2. Final compounds exhibited purities > 95%, as judged by LC/MS (254 and 214 nm) and evaporative light scattering (ELS).

Table 1.

Biological activity of aminophthalazines analogs 2i-xxiv, 6i-ii and pyridazines 9i-ii. ^a–^f

Cmpd	R1	R2	X	PGE2 (%)d	COX-2 (%)e	PGE2 EC50 /µMf
2i	p-CH3O	p-C(O)NH(CH2)2OH	C	39.2 ± 13	-	-
2ii	p-OH	H	C	26.7 ± 5	-	-
2iii	p-SO2NH2	H	C	NR	-	-
2iv	p-COOEt	H	C	18.4 ± 3	-	-
2v	p-OCH2C(O)NH2	p-CH3O	C	NR	-	-
2vi	p-CH3	H	C	98.9 ± 1	23.0 ± 1	0.031 ± 0.07
2vii	p-CH3	p-CH3O	C	98.3 ± 3	NR	0.032 ± 0.02
2viii	p-CH3	m-CH3O	C	62.0 ± 8	-	-
2ix	p-CH3	o-CH3O	C	77.2 ± 0	NR	0.59 ± 0.04
2x	p-CH3	H	N	15.1 ± 4	-	-
2xi	p-CH3	p-N(CH3)2	C	94.4 ± 1	14.2 ± 1	0.42 ± 0.002
2xii	p-CH3	p-OEt	C	75.3 ± 9	NR	0.57 ± 0.06
2xiii	p-CH3	p-OCH(CH3)2	C	76.7 ± 3	NR	0.70 ± 0.25
2xiv	p-CH3	3,4-OCH2O-	C	59.0 ± 8	-	-
2xv	p-CH3	m-Cl	C	75.2 ± 8	NR	0.86 ± 0.10
2xvi	p-CH3	o-Cl	C	97.2 ± 1	NR	0.38 ± 0.01
2xvii	p-CH3	m-Ph	C	84.2 ± 1	NR	0.54 ± 0.05
2xviii	p-Et	p-CH3O	C	85.0 ± 11	NR	0.93 ± 0.01
2xix	m-CH3	H	C	74.1 ± 6	NR	0.76 ± 0.30
2xx	m-CH3	p-CH3O	C	87.5 ± 5	NR	0.64 ± 0.34
2xxi	m-Cl	H	C	79.4 ± 9	10.1 ± 7	0.50 ± 0.4
2xxii	p-PhO	H	C	71.2 ± 15	NR	-
2xxiii	p-PhO	p-CH3O	C	92.6 ± 5	NR	0.09 ± 0.05
2xxiv	H	p-CH3O	C	97.3 ± 1	3.0 ± 10	0.02 ± 0.01
6i	CH3	CH3O	-	87.0 ± 2	NR	0.05 ± 0.02
6ii	CH3O	H	-	92.4 ± 1	21.4 ± 27	0.220 ± 0.006
9i	CH3	H	-	62.5 ± 1	-	-
9ii	CH3	CH3O	-	49.0 ± 4	-	-

^aPreviously described inhibitor of cellular PGE₂ levels, 1b, was used as a positive control (PGE₂ 10.9 ± 4.5 and EC₅₀ = 0.33 ± 0.09 uM). ¹¹

^bNR, no observed reduction.

^c- not determined.

^d% of inhibition of PGE₂ levels in HCA-7 cells at 1 µM concentration ± SD (n = 3).

^e% of inhibition of COX-2 levels in vitro at 5 µM concentration ± SD (n = 3).

^fEC₅₀ for PGE₂ level reduction in HCA-7 cells ± SD (n = 3).

Scheme 1.

Synthesis of aminophthalazines 2i-xxiv and 6i-ii (R¹, R² and X vary, see Table 1). ¹³

^a Reagents and conditions: (i) (a) EtOH, reflux, 0.5 h; (b) NaOH (aq). (ii) Boronic acid (2 equiv.), K₂CO₃ (2 equiv.), Bis(triphenylphosphine)palladium(II) dichloride (0.05 equiv.), dioxane-H₂O (4:1), microwave irradiation, 100 °C, 1 h. (iii) NaH (1.5 equiv.), MeI (1.5 equiv.), DMF, 0 °C to rt.

Scheme 2.

Synthesis of aminopyridazines 9i-ii.

^a Reagents and conditions: (i) (a) EtOH, reflux, 0.5 h; (b) NaOH (aq). (ii) Boronic acid (2 equiv.), K₂CO₃ (2 equiv.), Bis(triphenylphosphine)palladium(II) dichloride (0.05 equiv.), dioxane-H₂O (4:1), microwave irradiation, 100 °C, 1 h.

All compounds were screened for their ability to reduce PGE₂ production in HCA-7 colon cancer cells at 1 µM concentration and activities are summarized as percentage reduction of PGE₂ levels Table 1.¹⁴ Compounds that exhibited reduction of PGE₂ levels higher than 70% were tested for COX-2 inhibition at 5 µM in an in vitro cell free assay, with Celecoxib incorporated as a positive control in both PGE₂ and COX-2 assays.¹⁵ IC₅₀ values for COX-2 inhibition were not determined, as no inhibitory activity against COX-2 > 50% was observed. Compounds exhibiting > 70% reduction of PGE₂ levels, and < 50% COX-2 inhibition were pushed forward for EC₅₀ determinations of PGE₂ reducing level ability.

Aminophthalazines 2i-v, characterized by various polar substituents at both the C-1 phenyl (R²) and C-4 aniline (R¹) rings, generally exhibited poor reduction of cellular PGE₂ levels (0–39%), possibly due to a reduced ability to permeate the cellular membrane. Replacement with R¹ = p-CH₃ (2vi-xvii) resulted in good to high levels of inhibition of cellular PGE₂ levels (59–99%), the only exception being 2x (R² = H, X = N, 15%), characterized by a polar pyrimidine ring instead of the phenyl ring. In this group of analogs, the activity was influenced by the nature and position of the substituent R² on the C-1 phenyl ring. In detail, when this ring was unsubstituted (2vi), a strong reduction of PGE₂ levels was observed (99%; EC₅₀ 0.031 µM). Surprisingly, a para-methoxy substituent (R²) led to the same level of inhibitory activity as 2vi (2vii, 98%; EC₅₀ 0.032 µM). Movement of the methoxy group R² to the meta and ortho positions resulted in a reduction of activity (2viii and 2ix, 62 % and 79 %, respectively; 2ix EC₅₀ 0.59 µM). Similar to what was observed for analog 2vii, a dimethylamine substituent at the para position of the C-1 phenyl ring (R²) led to potent reduction of PGE₂ levels (2xi, 94 %; EC₅₀ 0.42 µM), albeit 10 fold less than 2vii. Enlargement of the para-methoxy (R²) to ethoxy (2xii) or isopropoxyl (2xiii) resulted in partial loss of activity (75 % and 98 %; EC₅₀ 0.57 and 0.70 µM, respectively). A consistent reduction in activity was also observed when a bulkier 3,4-methylenedioxy substituent was added to the phenyl ring R² (2xiv, 59%).

Introduction of a bulky hydrophobic substituent at the meta position of the C-1 phenyl ring (R²), including chloro (2xv, 75%; EC₅₀ 0.86 µM) or phenyl (2xvii, 85%; EC₅₀ 0.54 µM) led to moderated activity, whilst movement of a chlorine atom from the meta to the ortho position (2xvi) improved activity (97%; EC₅₀ 0.38 µM). This result may be explained by restricted rotation of the phenyl ring and associated lowering of entropic barriers to binding, resulting from bulky substituent in the ortho position. Elongation of the para methyl group on the C-4 of the phenyl ring (R¹) to ethyl (2xviii, 85%; EC₅₀ 0.93µM) resulted in slightly reduced levels of activity when compared to 2vii (R² p-CH₃O; EC₅₀ 0.032µM). Movement of the para methyl group on the C-4 aniline ring to the meta position led to a decrease in the observed activity (R¹, 2xix and 2xx, 74% and 87%; EC₅₀ 0.76 and 0.64 µM, respectively), compared with the activity previously observed for 2vi-vii (99% and 98%; EC₅₀ 0.031 and 0.032 µM). Replacement of the meta methyl R¹ with a bioisosteric chlorine (2xxi, 74%; EC₅₀ 0.76 µM) resulted in a similar level of activity to 2xix. Introduction of a para phenoxy substituent on the C-1 phenyl ring (R², 2xxii-xxiii) maintained strong reduction of PGE₂ cellular levels, surprisingly only when the C-4 anilino ring (R¹) was substituted with a para methoxy group (92%; EC₅₀ = 0.09 µM). Interestingly, removal of the methoxy group from the aniline ring (R¹ = H, R² = p-CH₃O) resulted in slight increase of cellular activity (2xxiv, 97%; EC₅₀ = 0.02 µM), delivering the most functionally potent compound observed in cells.

N-Methylation of 2vi-vii to afford 6i and 6ii respectively, interestingly only slightly impaired the capacity to reduce cellular PGE₂ levels (87 % and 92%; EC₅₀ 0.05 µM and 0.220 µM) when compared to structurally related analogs 2vi-vii. Finally, replacement of the phthalazine scaffold with a pyridazine ring resulted in reduced level of activity (62 % and 49 %, 9i and 9ii, respectively) and no further efforts were pursued with this heterocycle.

Remarkably for this class of compounds, the most functionally active compounds in cells exhibited only negligible inhibition of COX-2 activity in vitro at 5 µM, with % inhibition comprised between 0 and 30%. As such, no dose response curves for COX-2 were determined.

In conclusion, we have prepared and evaluated twenty-six aminophthalazine and two pyridazine analogs for their capacity to reduce cellular levels of PGE₂ in HCA-7 cells. The in vitro inhibitory activity against COX-2 was also determined, leading to the identification of potent inhibitors of PGE₂ levels with negligible activity against COX-2. Compounds 2vi-vii, xvi and xxiv exhibited the highest reduction of PGE₂ levels, with percentage of PGE₂ reduction between 97.2 and 98.9% and EC₅₀ values comprised between 0.038 and 0.02 µM respectively. Furthermore, 2vii was tested for its effect on tumor growth in mouse xenograft models expressing HCA-7 cells and was confirmed to have good anti-cancer activity with tumor versus control value (T/C) of 34% for HCA-7 human colonic adenocarcinoma cell lines, at a dosing schedule of 100 mg/kg i.p. over 10 days. Work is now on going to elucidate the mode of action of these molecules with tagged probes. Future synthetic efforts will focus on further development of new scaffolds with improved physicochemical properties to further develop established SAR and promote discovery of new candidates for in vivo evaluation in mouse xenograft models.