Synthesis and Biofilm Inhibition Studies of 2-(2-amino-6-arylpyrimidin-4-yl)quinazolin-4(3H)-ones

Abstract

Synthesis of novel 4(3H)-quinazolinonyl aminopyrimidine derivatives has been achieved via quinazolin only enones which in turn were obtained from 2-acyl-4(3H)-quinazolinone. They have been assayed for biofilm inhibition against Gram-positive (methicillin-resistant Staphylococcus aureus (MRSA) and Gram-negative bacteria (Acinetobacter baumannii). The analogues with 2,4,6-trimethoxy phenyl, 4-methylthio phenyl, and 3-bromo phenyl substituents (5h, 5j & 5k) have been shown to inhibit biofilm formation efficiently in MRSA with IC₅₀ values of 20.7-22.4 μM). The analogues 5h and 5j have demonstrated low toxicity in human cells in vitro and can be investigated further as leads.

Graphical Abstract

To create your abstract, type over the instructions in the template box below. Fonts or abstract dimensions should not be changed or altered.

The incidence of multidrug-resistant (MDR) bacteria is on the rise.¹ This situation has been further complicated by the antibacterial drug drought for the past 30 years, thus it is important to focus on identifying ways to address the health crisis caused by the same.² According to the Centers for Disease Control and Prevention (CDC), it is estimated that 2.8 million people are infected by antibiotic-resistant bacteria annually, and of those infected, about 35,000 die. The CDC predicts these numbers will only increase if pertinent attention is not brought to these persistent bacterial strains.³ Bacterial species such as Staphylococcus aureus, and Acinetobacter baumannii, have evolved to evade many current antibiotics. A common defence mechanism observed in MDR bacteria, and others, is the formation of biofilms.⁴ In general, biofilms are composed of surface attached community of bacteria enclosed in an extracellular matrix of biomolecules. It is known that bacteria in a biofilm, bacteria are atleast 1000-fold more resistant to antibiotics. Therefore, inhibition of biofilm formation is an attractive approach to combat the threat of MDR bacteria.⁵

The 4(3H)-quinazolinone, an important pharmacophore with many useful bioactivities⁶ found in numerous alkaloids and drugs currently in the market, has not been studied for antibacterial potential but not for the ability to inhibit biofilm development.⁷ The close congener, aminoquinazoline, has been shown to possess antibiofilm activity by several laboratories.

Notably, Shaw, et al., showed that N2, N4-disubstituted quinazoline analogues exhibited potent antibacterial activity against A. baumannii, displaying single-digit micromolar MICs, as well as eradicating 90% of cells within a biofilm at or near the minimum inhibitory concentration (MIC).⁸ These compounds also exhibited potent antibacterial activity against methicillin-resistant S. aureus (MRSA), with MIC values around 0.5 μM (Figure 1).⁹ Previous studies have also established that 2-aminoquinazoline derivatives have significant anti-biofilm activity against Mycobacterium smegmatis (IC₅₀ values around 15 μM).¹⁰ Additionally, multiple studies have shown that 2-aminopyrimidine derivatives possess anti-biofilm properties.¹¹Analogues of meridianin D, a 2-aminopyrimidine containing natural product show potent biofilm inhibitory activity against S. aureus (MRSA) (IC₅₀ values as low as 9 μM) (Figure 1).¹² Inspired by these findings, we have designed analogues of 2-(2-amino-6-arylpyrimidin-4-yl)quinazolin-4(3H)-ones (Figure 1) by combining the structural fragments of the leads, and tested therole of these analogues in the inhibition of biofilms. We have developed a facile synthetic route to access our designed compounds that centred on the condensation of quinazolinonyl enone intermediates with guanidine. Herein, we report the synthesis of the pyrimidinyl quinazolinone derivatives, their anti-biofilm activity against MRSA and A. baumanniias well as cytotoxicity in human colon cells in vitro.

Figure 1.

Anti-biofilm activity of analogues containing quinazoline or 2-aminopyrimidine

Our synthetic plan was to access the 2-(2-amino-6-arylpyrimidin-4-yl)quinazolin-4(3H)-ones through conjugate addition and subsequent condensation between guanidine and various quinazolinonyl enones (4a-v). We proposed to access these enones through the aldol condensation of 2-acetyl-4(3H)quinazolinone (3) with various substituted aromatic aldehydes.

Accordingly, our synthesis process started with identifying a synthetic route to 2-acetylquinazolin-4(3H)-one (3) amenable to large scale synthesis. There were only a few reports available in literature for the synthesis of 3 that involved either the use of selenium dioxide¹³ or triphenyl phosphine,¹⁴ reagents problematic for our intended large-scale synthesis. The benzodiazepine ring contraction route¹⁵ also failed to deliver 3 in satisfactory yields in our hands. We then resorted to pyruvic acid route¹⁶ utilized by Hart et al., which yielded the desired compound 3 albeit in poor yield in our hands due to an unknown dimeric side product. After a great deal of experimentation, we successfully modified the preparation of the diamide (2) that resulted in a 61% yield on a >100 g reaction scale. The dehydrative cyclization of 2 using mild bases provided compound 3 in 65% yield on a reaction scale of 10 g. In addition to suppressing the unwanted dimer formation, the modification of utilizing aq. basic conditions allowed us to run the subsequent aldol condensation in the same pot to obtain quinazolinonyl enones (4a-4v) in a single operation at an improved overall yield (75–89%), thus rendering the method practical and straightforward to implement (Scheme 2).

Scheme 2.

Synthesis of 2-(aryl) aminopyrimidinyl quinazolin-4(3H)-ones.

Under basic conditions, the enones (4a-4v) were then refluxed in ethanol in the presence of guanidine to yield the target pyrimidinyl quinazolinones (5a-5v, Scheme 2). The amassed collection was assayed for their anti-biofilm activity against Gram-positive (MRSA) and Gram-negative (A. baumannii) bacteria.

Compounds were initially screened for the ability to inhibit biofilm formation by A. baumannii ATC 19606 and MRSA ATCC 43300 at a concentration of 100 μM using a crystal violet reporter assay as previously described.¹⁷ All compounds that inhibited biofilm formation by greater than 50% as compared to an untreated control at this concentration were subjected to a dose-response assay to determine the IC₅₀ value (which we define as the concentration at which the compound effects a 50% reduction in the amount of biofilm). This class of compounds did not prove to exhibit antibiofilm activity against A. baumannii, with none of the compounds tested inhibiting biofilm formation by more than 25% at 100 μM (Table S1 Supporting Information). Against MRSA however, 14 of the 22 compounds inhibited biofilm formation by more than 50% at 100 μM, and IC₅₀ values were therefore determined (Table 1).

Table 1.

IC₅₀ values for inhibition of MRSA ATCC 43300 biofilm formation.

Compound	IC50 (μM)
5a	34.4 ± 0.54
5d	84.8 ± 6.63
5f	50.5 ± 5.55
5g	38.4 ± 1.70
5h	21.2 ± 3.33
5j	22.4 ± 3.44
5k	20.7 ± 2.71
5l	47.3 ± 4.50
5m	56.2 ± 2.32
5o	42.7 ± 6.77
5q	86.5 ± 7.08
5s	37.3 ± 2.83
5u	71.0 ± 11.32
5v	56.9 ± 13.60

The unsubstituted phenyl derivative 5a exhibited moderate activity, returning an IC₅₀ of 34.41 μM, while incorporation of a methyl substituent at either the 2- or 4-position of the phenyl ring (5b and c) resulted in reduced activity. The monomethoxy derivative 5d, and the 3,4- and 3,5-di-methoxy derivatives 5e and 5f displayed reduced activity compared to the parent, however the 2,4-di-methoxy analogue 5g exhibited comparable activity to the parent. The addition of a second ortho-methoxy group to this scaffold in compound 5h resulted in increased activity, with this compound exhibiting an IC₅₀ of 21.2 μM, while the 3,4,5-tri-methoxy isomer 5i displayed considerably reduced activity. The thioether derivative 5j displayed comparable activity to 5h (IC₅₀ 22.4 μM). The most active compound from this series was the 3-bromo derivative 5k which exhibited an IC₅₀ value of 20.7 μM, while the 4-bromo derivative 5l displayed reduced activity. Other halogen substituents including 2-chloro (5m), and 2,4-dichloro (5o) were moderately active while the 4-chloro (5n) and 4-fluoro (5p) derivatives did not inhibit biofilm formation by more than 50% at 100 μM. Placing a trifluoromethyl substituent at the 4-position (5q) resulted in reduced activity compared to the parent. Benzyl derivatives followed the same trend as the methoxy derivatives in that placement at the 3-position (5r) reduced activity, while placement at the 4-position (4s) conferred moderate activity comparable to the parent. Finally, the effect of incorporation of aromatic substituents was investigated, with the naphthyl and pyridinyl derivatives 5t and 5u effecting less than 50% inhibition at 100 μM, and the thiophenyl and furan derivatives 5u and 5v exhibiting reduced activity compared to the parent.

In general, the structure activity relationship (SAR) data generated from this initial library indicates that electron donating groups placed at the 2- or 4- (or both) positions leads to higher activity, while placement of such groups at the 3-position leads to reduced activity (as seen for compounds 5h and 5j). Incorporation of bromine at the 3-position confers increased activity (compound 5k) but when a halogen is placed at the 2- or 4- position a reduction in activity is observed.

We next tested the ability of the library to disperse preformed MRSA biofilms as previously described¹⁶ however none of the compounds effected more than 25% dispersion (compared to biofilms treated with fresh media alone) at 100 μM. Finally MICs of all compounds against MRSA ATCC 433300 were recorded to ensure that the observed biofilm inhibition activity was not a result of planktonic toxicity, and all compounds retuned MICs of greater than 100 μM (highest concentration tested).

We have also assessed whether the most active compounds (5h, 5j, and 5k) exhibit any toxicity to human cells at the concentrations that were used to inhibit biofilm formation. Normal human colon cells (CCD-18 Co cells) were cultured in 96-well plates for three days followed by the addition of the three active compounds in triplicate (5h, 5j and 5k) at three concentrations of each compound, 50 μM, 25 μM and 12.5 μM. MTT cytotoxicity assay was performed on day 1 following the addition of the compounds as shown in Figure 2. The control samples in this study were DMSO (1 μL) treated cells since the trace amount of DMSO in the wells also had some cytotoxic effect on the cells. An additional positive control was of the CC-18 Co cells not exposed to DMSO or the compounds were used to compare the effect of just DMSO and DMSO in combination with compounds on the cells. This control was higher in viability indicating damage to the cells in the presence of DMSO (1 μL and 0.5 μL) though in a low volume. In the case of compound, 5h, the cellular toxicity was minimal even at the highest concentration used (50 μM). Whereas in the case of compound 5J, 50 μM proved to more cytotoxic whereas the lower concentrations were comparable to the DMSO controls. Among the three compounds used in this study, 5h demonstrated the lowest cytotoxicity, in comparison with the DMSO controls whereas compound 5K had the highest toxicity. Compound 5k had significant toxicity even at the lowest concentration used (12.5 μM) and comparable to the toxicity at the highest concentration used (50 μM).

Figure 2.

Normalized MTT absorbance indicating the cell viability of CCD-18 Co cells after 1-day treatment with the three bioactive compounds (5h, 5k, and 5j).

In conclusion, novel 2-(2-amino-6-arylpyrimidin-4-yl) quinazolin-4(3H)-ones were designed and accessed via guanidine condensation with the quinazolinonyl aryl enones that were in turn derived from 2-acetyl 4(3H) quinazolinone. The synthetic collection was assayed for the antibiofilm inhibition against a representative gram-positive (MRSA) and representative gram-negative bacteria (A. baumannii). The screening results revealed that some of the analogues (5h, 5j, 5k) inhibited biofilm formation efficiently in MRSA (IC₅₀ ~20 μM) while further optimization through additional diversification could potentially augment activity further. The cytotoxicity assay of the active compounds revealed that compounds 5h and 5j have low toxicity at lower concentrations whereas compound 5k was highly toxic to the cells making compounds 5h and 5k as possible leads for further modification. Efforts are underway in this direction and results will be reported in future.

Scheme 1.

Synthesis of 2-acetylquinazolin-4(3H)-one and aryl quinazolinonyl enones.