Microwave-Assisted Synthesis of Novel Pyrazolo[3,4-g][1,8]naphthyridin-5-amine with Potential Antifungal and Antitumor Activity

The microwave assisted reaction between heterocyclic o-aminonitriles 1 and cyclic ketones 2 catalyzed by zinc chloride led to new series of pyrazolo[3,4-b][1,8]naphthyridin-5-amines 3 in good yields. This procedure provides several advantages such as being environmentally friendly, high yields, simple work-up procedure, broad scope of applicability and the protocol provides an alternative for the synthesis of pyrazolonaphthyridines. The whole series showed antifungal activities against Candida albicans and Cryptococcus neoformans standardized strains, being compounds with a 4-p-tolyl substituent of the naphthyridin scheleton (3a, 3d and 3g), the most active ones mainly against C. albicans, which appear to be related to their comparative hydrophobicity. Among them, 3d, containing a cyclohexyl fused ring, showed the best activity. The anti-Candida activity was corroborated by testing the three most active compounds against clinical isolates of albicans and non-albicans Candida strains. These compounds were also screened by the US National Cancer Institute (NCI) for their ability to inhibit 60 different human tumor cell lines. Compounds 3a and 3e showed remarkable antitumor activity against cancer cell lines, with the most important GI50 values ranging from 0.62 to 2.18 μM.


Introduction
Naphthyridines are fused nitrogen heterocycles present in many natural and synthetic compounds of particular interest in Medicinal Chemistry due to their diverse biological activities.They have showed a broad range of interesting pharmacological activities, such as anti-inflammatory [1][2][3][4], analgesic, antiaggressive [5], anticancer [6], antibacterial [7], antitumor [8], antihypertensive [9] and antiallergic [10] ones.They also showed to be useful starting materials for the synthesis of various policyclic heterocycles of biological interest.Due to their biological and synthetic importance, the development of effective routes to synthesize naphthyridines continues to be an active area of research for synthetic organic chemists [11].A survey of the literature shows that the major synthetic approaches used to prepare the naphthyridine system involved condensation of 2-aminopyridine derivatives with carbonyl compounds containing an activated methylene group [12][13][14][15][16][17][18][19] or with β-ketoesters [20].
Microwave irradiation (MWI) is a technique that has been employed in a number of applications in synthetic chemistry and has been observed to have several advantages compared to traditional methods of synthesis.Due to selective heating in the microwave, the occurrence of side reactions is avoided.Microwave assisted organic synthesis (MAOS) has also emerged as a powerful tool for high-throughput procedures.This can improve the yield and purity of the final compounds in short reaction times through the precise control of parameters such as power irradiation, pressure and temperature [21][22][23][24][25][26][27].
On the other hand, the Friedländer annulation is one of the most simple and straightforward approaches to the synthesis of poly-substituted pyridines and related aza-heterocycles or aza-aromatic compounds [28][29][30].For this kind of Friedländer reaction, two points are of concern.One is the catalyst, which belongs to one of two categories: the proton acid [31][32][33][34] or the Lewis acid [35][36][37][38][39].The other point is that the structures of a normal Friedländer condensation product are either a pyridine or a quinoline skeleton.
Cancer is the major health problem that threatens people worldwide.Since many of the current pharmacotherapeutic drugs have problems with toxicity and drug resistance, there is a strong need for the discovery and development of effective new anticancer drugs [40].Among the wide range of compounds tested as potential anticancer agents, compounds containing the six-membered heterocyclic pyridine and/or pyrimidine, have attracted significantly attention.
In the last years, fungi have emerged as major cause of human infections especially among immunocompromised hosts having an enormous impact on morbidity and mortality [1][2][3][4].Because most patients with invasive fungal infections are immunocompromised, the success of treatments is more dependent on the efficacy of the antifungal agent than on the immune system [41,42].Unfortunately, the available antifungal agents are limited and it took 30 years for the newest class of antifungal drugs, the echinocandins [43,44], to appear on the market.Furthermore, the usual therapy for cryptococcal meningitis, a disease produced by Cryptococcus neoformans that kills most AIDs patients worldwide, is based on amphotericin B and flucytosine, which were discovered nearly 50 years ago [45].There is, therefore, an urgent need for new antifungal chemical structures for treating infections produced by these fungi, alternatives to the existing ones [46].
Because of this great need for new antifungal and/or anticancer structures, and considering that other naphthyridines-bearing structures have shown antimicrobial and antitumor activities [4,5], herein we report the synthesis of a novel series of pyrazolo [3,4-b][1,8]naphthyridin-5-amines by the Friedländer condensation of o-aminonitrile 1 with cyclic ketones 2. The whole series was tested for antifungal activity against standardized as well clinical strains of the clinically important fungi C. neoformans and species of Candida genus.All compounds were also evaluated to determine antitumor activity.

Chemistry
In our study, several conditions were tested for the first time, including diverse solvents, temperatures, heating source and catalysts in order to find the best reaction conditions for the synthesis of 3. The reactions were carried out from o-aminonitrile 1a (R = CH3) and cyclohexanone as a model reaction (Scheme 1).When ethanol was used as the solvent and the mixture was subjected to reflux, the desired product, 3d, was obtained in low yields (30%, 35% and 40%, entries 2, 3 and 4) after 10 h.In a modified protocol, the reaction was performed under MWI, obtaining significant improvements evidenced by better yields of the target product (40%, 45% and 75%, entries 6, 7 and 8) and shorter reaction times.As it is clearly observed in entries 1 and 5, the reaction did not take place without a catalyst.So, we performed the reaction with a Lewis acid catalyst.Results showed that both the aluminum chloride and the p-toluenesulfonic acid, in either conventional refluxing or MWI, favored the product of Friedländer condensation 3 (Table 1), but the yields were not very high, whereas anhydrous zinc chloride favored the formation of the product in higher yields in both conditions.The best catalyst, anhydrous zinc chloride, was therefore chosen as the catalyst for this new transformation.To determine the extent of application of the cyclization reaction, the same conditions were used for some selected cyclic ketones (of five, six and seven members) to get the pyrazolo [3,4-  All the structures 3a-i were characterized by IR (Infrared), 1 H-NMR, 13 C-NMR (Proton-and carbon-Nuclear Magnetic Resonance), MS (mass-spectrometry) spectra and elemental analyses.Formation of the pyrazolo [3,4-b][1,8]naphthyridin-5-amines 3 was unequivocally established by NMR data of the products.The chemical shifts and multiplicities of the protons were in accordance with the expected values.For example, signals for the protons of the phenyl of compounds 3 were found between 7.20 and 8.30 ppm.The signal for NH2 appears as a broad singlet between 6.10 and 6.80 ppm, the signals of methyl protons of CH3 appears as singlets between 1.74 and 1.84 ppm and all aliphatic protons, corresponding to the five-, six-and seven-membered, appear between 1.46 and 3.87.
A possible mechanism of the cyclization reaction between 1 and 2 is depicted in Scheme 2. The reaction seemed to proceed via the initial formation of the imine intermediate 5 by the normal Friedländer reaction [47,48], which subsequently affords the final compound 3 by an intramolecular nucleophilic cyclization on the nitrile.Scheme 2. Proposed mechanism.

Antifungal Activity
The antifungal properties of compounds 3a-3i were tested first against two clinically important fungal species, Candida albicans and C. neoformans, which were selected due to the following facts: C. albicans is among the most common cause of opportunistic fungal infections in immunocompromised hosts, although recently, non-albicans Candida species have been increasingly identified among Candida-infected patients [49].
In turn, C. neoformans is the most frequent cause of meningitis and is one of the most important HIV-related fatal opportunistic mycosis, which has killed more than 650,000 immunocompromised patients worldwide up to date [46].Although the incidence of disease tends to decline in countries with highly active anti-retroviral therapy, the outcome of infection is influenced by a variety of factors including the antifungal resistance and new strategies including new structural types with anti-cryptococcal activity are highly welcome [50].
For a more comprehensive analysis of the antifungal results, we grouped the compounds in two series, (i) and (ii).Series (i) includes compounds with different rings (cyclopentyl, cyclohexyl or cycloheptyl) fused to the naphthyridines scheleton and the same R (sub-series i.1 with CH3, i.2 with Cl or i.3 with OCH3), which allowed having a look at the influence of the size of the fused ring on the antifungal activity.Series (ii) includes compounds with different R (CH3, Cl or OCH3), but the same fused ring moiety (subseries ii.1 with cyclopentyl; ii.2 with cyclohexyl and ii.3 with cycloheptyl), which allowed analyzing the role played by the different R substituents in the antifungal activity.
Compounds were evaluated by using the standardized microbroth dilution method M-27A3 for yeasts of Clinical and Laboratory Standards Institute [51], which assures confident and reproducible results.
Results of the whole series are expressed as the percentages of inhibition of each fungus in the range 250-3.9µg•mL −1 and are presented as Supplementary Table 1 (Table S1).
For the sake of clarity, Table S1 was summarized in Table 3 by using the MICs at different endpoints, such as MIC100 MIC80 and MIC50 (minimum concentration that inhibits 100%, 80% and 50% of growth) that have showed to consistently represent the in vitro activity of compounds [52].
From Table 3, it is clear that all compounds displayed some degree of activity against C. albicans and C. neoformans.However, C. albicans showed to be more sensitive for the whole series than C. neoformans, since eight of the nine compounds showed MIC100 against C. albicans below 250 µg•mL −1 (range = 31.2-250µg•mL −1 ), while only four of the nine compounds showed MICs100 below 250 µg•mL −1 against C. neoformans (range = 125-250 µg•mL −1 ).The same analysis can be performed with MIC80 and MIC50.
Regarding the activity against C. albicans, the comparison of the activity of compounds with same ring and different R (compare 3a/3b/3c, 3d/3e/3f or 3g/3h/3i) showed that the type of ring does not play a crucial role in the activity, since the most active compounds, 3a, 3d and 3g, possess different rings, cyclopentyl, cyclohexyl and cycloheptyl, respectively.Instead, the three compounds share the feature of having a 4-p-tolyl moiety.
The comparison of the activity of the three compounds possessing a methyl groups as R can be clearly observed in Figure 1.
From Figure 1, it is clear that 3d, possessing a cyclohexyl ring, is the most active compound, while 3a and 3g (with cyclopentyl and cycloheptyl, respectively) possessed lower activities than 3d, but similar to each other.Table 3. Minimum inhibitory concentrations (MIC100, MIC80 and MIC50) and minimum fungicidal concentrations (MFC) of 3a-i grouped by their structural features against standardized strains of Candida albicans and C. neoformans.It is known that logP (the logarithm of the partition coefficient in a biphasic system, e.g., n-octanol/water) describes the macroscopic hydrophobicity of a molecule, which is a factor that determines its ability to penetrate fungal cell membranes and to reach the interacting sites, thus influencing the antifungal activity of compounds [53,54].In order to establish a correlation between logP and the activity of 3a-i, if any, logP of each compound was calculated and correlated to the percentage of inhibition of each compound at a selected concentration (125 µg•mL −1 ).Table 4 shows the values of logP and Figure 2 plots logP vs. activity of all compounds of the series.For the calculation of logP, we used quantum mechanical at semi-empirical level using Mopac, with the parametric method 3 (PM3).The molecular modeling were prepared using CS Chem-Office Software version 9.0 (Cambridge software) [55].The models were minimizedation until the root mean square (RMS) gradient value reached a value smaller than 0.0001 kcal•mol −1 .The lowest energy structure was used for each molecule to calculate logP values.Interesting enough, the most active compounds, 3a, 3d, 3g and 3h, possess logP values between 5.62 and 6.61.The rest of these compounds possess values of logP lower than 4.8 (mean value).The results showed above suggest that the antifungal activity of compounds 3 would be related to their hydrophobicity.

Second-Order Studies with Clinical Isolates
In order to gain insight into the actual inhibitory capacity of 3a, 3d and 3g against C. albicans, the three compounds were tested not only against the ATCC standardized strain but also on six clinical strains of C. albicans (isolated from patients suffering from mycoses) and on four non-albicans Candida strains, such as C. glabrata, C. parapsilopsis, C. krusei and C. tropicalis, all of them provided by CEREMIC (see Experimental).The selection of these non-Candida spp.was due to that fact that these four spp, along with C. albicans, are responsible for more than 90% of all Candida infection all over the world and also in Latin American countries [49].
The minimum inhibitory concentration (MIC) values of 3a, 3d and 3g were determined against this new panel by determining MIC100, MIC80 and MIC50.These results are shown in Table 5.As it can be seen in Table 5, clinical isolates of Candida genus showed similar sensitivity to 3a, 3d and 3g than the standardized strain C. albicans ATCC 10231, corroborating the anti-Candida activity of the three compounds.

Anticancer Activity
As a preliminary screening, structures of all new compounds (i.e., 3a-j) were submitted to the Developmental Therapeutics Program (DTP) at National Cancer Institute (NCI) for evaluation of their anticancer activity against different human cell lines.All the submitted structures (3a-j) were selected and subjected to the preliminary evaluation against the 60 tumor cell lines at a single dose of 10 mM after 48 h of incubation.The output from the single dose screening was reported as a mean graph available for analysis by the COMPARE program (data not shown).The results of this first assay showed that compounds 3a and 3e were active.
Then, the second screening was made in order to determine cytostatic activity of active compound against the 60 tumor cell lines represented in leukemia, melanoma, lung, colon, brain, breast, ovary, kidney and prostate panels; where the testing results were expressed according to the following three parameters: GI50, which is the molar concentration of the compounds required to inhibit the growing of the cell lines to 50% (relative to untreated cells).TGI as the molar concentration that causes total growth inhibition, and LC50, which is a parameter of cytotoxicity and reflects the molar concentration needed to kill 50% of the cells [56].The active compounds were evaluated at five concentration levels (100, 10, 1.0, 0.1, and 0.01 mM) and the test consisted of a 48 h continuous drug exposure protocol using sulforhodamine B (SRB) protein assay to estimate cell growth.Details of this evaluation method, and the complementary information related with the activity pattern over all cell lines, have been published [57][58][59][60][61].The compound 3a show a remarkable activity against 39 human tumor cell lines (Table 6), with low values of GI50 ≈ 10 −6 μM, being SR (Leukemia; GI50 = 0.62 μM, LC50 > 100 μM), HCT-15 (Colon Cancer; GI50 = 1.98 μM, LC50 > 100 μM) and MCF7 (Breast Cancer; GI50 = 1.62 μM, LC50 > 100 μM) the most sensitive strains.In a similar way, compound 3e also showed an interesting activity against 41 human tumor cell lines, remarking that for SR and CCRF-CEM (Leukemia) with GI50 Values of 1.84, 2.18 μM and LC50 values of 79.5 μM and >100 μM, respectively.The cytotoxic effects associated with compounds 3a and 3e were measured as LC50 goes from 6.79 to >100 μM, indicating a low toxicity of these compounds for normal human cell lines, as required for development of potential antitumor agents.

General Information
All the reagents were commercially available and used without any further purification.The solvents used were of analytical grade.Microwave experiments were carried out on a focused microwave reactor (300W CEM Discover).TLC analyses were performed on Merck TLC-plates aluminum silica gel 60 F254.Melting points were determined in a Büchi Melting Point Apparatus and are uncorrected.The IR analysis was performed on a Shimadzu FTIR 8400 spectrophotometer in KBr disks. 1 H-and 13 C-NMR spectra were run on a Bruker AVANCE 400 spectrometer operating at 400 MHz and 100 MHz, respectively, using dimethylsulfoxide-d6 as solvents and tetramethylsilane as internal standard.The mass spectra were scanned on a Shimadzu GCMS-QP 2010 and Hewlett-Packard HP Engine-5989 spectrometer (equipped with a direct inlet probe) and operating at 70 eV.The elemental analyses were obtained using a LECO CHNS-900 elemental analyzer and the values are within ± 0.4% of the theoretical values.

Microorganisms and Media
For the antifungal evaluation, standardized strains from the American Type Culture Collection (ATCC), Rockville, MD, USA, Reference Center in Mycology (CEREMIC, CCC, Rosario, Argentina) were used.Standardized strains: C. albicans ATCC 10231 and C. neoformans ATCC 32264; clinical isolates of Candida genus were provided by CCC.Voucher specimens of the isolated are presented in Table 5. Strains were grown on Sabouraud-chloramphenicol agar slants for 48 h at 30 °C, were maintained on slopes of Sabouraud-dextrose agar (SDA, Oxoid) and sub-cultured every 15 days to prevent pleomorphic transformations.Inocula were obtained according to reported procedures, [51] and adjusted to 1 × 10 3 -5 × 10 3 cells with colony forming units (CFU)/mL.

Fungal Growth Inhibition Percentage Determination
Yeasts broth microdilution technique M27-A3 of CLSI [51] was performed in 96-well microplates.For the assay, compound test wells (CTWs) were prepared with stock solutions of each compound in DMSO (maximum concentration ≤ 1%), diluted with RPMI-1640, to final concentrations of 250-3.9μg•mL −1 .An inoculum suspension (100 μL) was added to each well (final volume in the well = 200 μL).A growth control well (GCW) (containing medium, inoculum, and the same amount of DMSO used in a CTW, but compound-free) and a sterility control well (SCW) (sample, medium, and sterile water instead of inoculum) were included for each fungus tested.Microtiter trays were incubated in a moist, dark chamber at 30 °C for 48 h for both yeasts.Microplates were read in a VERSA Max microplate reader (Molecular Devices, Sunnyvale, CA, USA).Amphotericin B was used as positive control.Tests were performed in triplicate.Reduction of growth for each compound concentration was calculated as follows: % of inhibition = 100 − (OD 405 CTW − OD 405 SCW)/(OD 405 GCW − OD 405 SCW).The means ± SD (standard deviations) were used for constructing the dose-response curves representing % inhibition vs. concentration of each compound.Dose-response curves were constructed with SigmaPlot 11.0 software.

MIC100, MIC80 and MIC50 Determinations
Three endpoints were defined from the dose-response curves.Minimum Inhibitory concentration (MIC) resulting in total fungal growth inhibition was named MIC100, while MIC80 and MIC50 were defined as the minimum concentration that inhibits 80% or 50% of the fungal growth, respectively.

Anticancer Activity
The human tumor cell lines of the cancer-screening panel were grown in RPMI 1640 medium containing 5% fetal bovine serum and 2 mM L-glutamine.For a typical screening experiment, cells are inoculated into 96 well microtiter plates.After cell inoculation, the microtiter plates were incubated at 37 °C, 5% CO2 and 95% air, and 100% relative humidity for 24 h prior to addition of tested compounds.After 24 h, two plates of each cell line were fixed in situ with TCA, to represent a measurement of the cell population for each cell line at the time of sample addition (Tz).The samples were solubilized in dimethyl sulfoxide (DMSO) at 400-fold the desired final maximum test concentration and stored frozen prior to use.At the time of compounds addition, an aliquot of frozen concentrate was thawed and diluted to twice the desired final maximum test concentration with complete medium containing 50 µg•mL −1 gentamicin.Additionally, four 10-fold or ½ log serial dilutions were made to provide a total of five drug concentrations plus control.Aliquots of 100 µL of these different sample dilutions were added to the appropriate microtiter wells already containing 100 µL of medium, resulting in the required final sample concentrations [58].After the tested compounds were added, the plates were incubated for an additional 48 h at 37 °C, 5% CO2 and 95% air, and 100% relative humidity.For adherent cells, the assay was terminated by the addition of cold TCA.Cells were fixed in situ by the gentle addition of 50 µL of cold 50% (w/v) TCA (final concentration, 10% TCA) and incubated for 60 min at 4 °C.The supernatant was discarded, and plates were washed five times with tap water and air-dried.Sulforhodamine B (SRB) solution (100 µL) at 0.4% (w/v) in 1% acetic acid was added to each well, and plates were incubated for 10 min at room temperature.After staining, unbound dye was removed by washing five times with 1% acetic acid and the plates were air-dried.Bound stain was subsequently solubilized with 10 mM trizma base, and the absorbance was read on an automated plate reader at a wavelength of 515 nm.Using the seven absorbance measurements (time zero (Tz), control growth in the absence of drug (C), and test growth in the presence of drug at the five concentration levels (Ti)), the percentage growth was calculated at each of the drug concentrations levels.Percentage growth inhibition was calculated as: [(Ti − Tz)/(C − Tz)] × 100 for concentrations for which Ti > Tz, and [(Ti − Tz)/Tz] × 100 for concentrations for which Ti < Tz.Three dose response parameters were calculated for each compound.Growth inhibition of 50% (GI50) was calculated from [(Ti − Tz)/(C − Tz)] × 100 = 50, which is the drug concentration resulting in a 50% lower net protein increase in the treated cells (measured by SRB staining), as compared to the net protein increase seen in the control cells.The drug concentration resulting in total growth inhibition (TGI) was calculated from Ti = 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), indicating a net loss of cells following treatment, was calculated from [(Ti − Tz)/Tz] × 100 = −50.Values were calculated for each of these three parameters if the level of activity is reached; however, if the effect was not reached or was exceeded, the value for that parameter was expressed as greater or less than the maximum or minimum concentration tested [58][59][60][61].

Conclusions
In this article we described the microwave-assisted synthesis of novel pyrazolo [3,4g][1,8]naphthyridin-5-amine 3 by the Friedländer condensation of o-aminonitrile 1 with cyclic ketones 2 using zinc chloride as a catalyst.The presented synthetic procedure is an environmentally friendly, simple, and high yielding method for the preparation of compounds 3. The whole series of compounds 3 were tested against standardized strains of the clinically important fungi, C. albicans and C. neoformans, showing C. albicans to be more sensitive to the whole synthetic series than C. neoformans.Compounds 3a, 3d and 3g, all possessing a p-tolyl substituent, but different fused rings, were the most active structures.They showed good antifungal activity against a second panel of clinical isolates of albicans and non-albicans Candida species and appear as good models for the development of new analogues with improved activity.The antitumor evaluation data revealed that compounds 3a and 3e exhibited remarkable activity with GI50 values in the range from 10−6 M against different cancer cell lines.

Table 1 .
Optimization of reaction conditions for the synthesis of compounds 3d.

Table 2 .
Scope of the reaction.

albicans ATCC 10231 C. neoformans ATCC 32264 MIC 100 MIC 80 MIC 50 MIC 100 MIC 80 MIC 50
Figure 1.Comparative curves of the growth inhibition of C. albicans ATCC 10231 produced by compounds 3a, 3d and 3g at different concentrations.Inhibition percentages are the means ± SD obtained from experiments in triplicate.

Table 5 .
The 100%, 80% and 50% inhibitory concentrations (MIC100, MIC80 and MIC50) of 3a, 3d, and 3g against clinical isolates of C. albicans and non-albicans Candida strains.For the sake of comparison, MIC100, MIC80 and MIC50 of all compounds against the ATCC 10231 standardized strain of C. albicans are included.

Table 6 .
In vitro testing expressed as growth inhibition of cancer cell lines for compounds 3a and 3e a .
cells during the drug incubation.Determined at five concentration levels (100, 10, 1.0, 0.1, and 0.01 μM); c LC 50 is a parameter of cytotoxicity and reflects the molar concentration needed to kill 50% of the cells.