Carboxyxanthones: Bioactive Agents and Molecular Scaffold for Synthesis of Analogues and Derivatives

Xanthones represent a structurally diverse group of compounds with a broad range of biological and pharmacological activities, depending on the nature and position of various substituents in the dibenzo-γ-pyrone scaffold. Among the large number of natural and synthetic xanthone derivatives, carboxyxanthones are very interesting bioactive compounds as well as important chemical substrates for molecular modifications to obtain new derivatives. A remarkable example is 5,6-dimethylxanthone-4-acetic acid (DMXAA), a simple carboxyxanthone derivative, originally developed as an anti-tumor agent and the first of its class to enter phase III clinical trials. From DMXAA new bioactive analogues and derivatives were also described. In this review, a literature survey covering the report on carboxyxanthone derivatives is presented, emphasizing their biological activities as well as their application as suitable building blocks to obtain new bioactive derivatives. The data assembled in this review intends to highlight the therapeutic potential of carboxyxanthone derivatives and guide the design for new bioactive xanthone derivatives.


Introduction
Xanthones (9H-xanthen-9-ones) are an important class of oxygenated three-membered heterocyclic compounds with a dibenzo-γ-pyrone scaffold (1, Figure 1) [1]. Over the years, considerable interest has been attracted in xanthone derivatives mainly because of their diverse range of biological/pharmacological activities [2][3][4][5]. The xanthone scaffold is considered a privileged structure [6,7], which can belong to the pharmacophoric moiety for the activity exhibited or as a substituent group associated with other chemical cores to modulate diverse biological responses [3].
Naturally-occurring xanthones can be found as secondary metabolites in diverse terrestrial sources including higher plants, fungi, lichens [8,9] as well as isolated from marine invertebrates, such as sponges, tunicates, mollusks and bryozoans, in addition to algae and marine microorganisms (cyanobacteria and fungi) [10,11]. They comprise a variety of different types of substituents in certain positions of the xanthone scaffold, leading to a vast diversity of biological/pharmacological [3] as well as different physicochemical and pharmacokinetic properties [12,13], being a remarkable basis for the discovery of new potential drug candidates.
Currently, there are many drugs on the market and in clinical trials, which were isolated or based on natural products [14][15][16], highlighting that natural compounds, such as xanthone derivatives, have always been a source of inspiration for the discovery of new therapeutic agents [14]. Some commercially available extracts with human health promotion properties present xanthone derivatives in composition [9]. Nevertheless, biosynthetic pathways only allow the presence of certain groups in specific positions on the xanthone scaffold. Therefore, the total synthesis strategy allows access to structures that otherwise could not be reached within the natural product as a launching platform for molecular modification [17]. In fact, with proper synthetic pathways, many other substituents can be introduced into the xanthone scaffold affording the development of more diverse compounds for biological activity and structure-activity relationship (SAR) studies [18], as well as other applications such as preparation of fluorescence probes [19,20] or stationary phases for liquid chromatography [21][22][23]. For the last several years, the isolation and synthesis of new bioactive xanthone derivatives using different synthetic methodologies has remained in the area of great interest of our group, as exemplified in [24][25][26][27][28][29][30][31][32][33][34][35].
Among the large number of natural and synthetic xanthone derivatives, those containing a carboxylic group have shown great significance in medicinal chemistry. A remarkable example is 5,6dimethylxanthone-4-acetic acid (DMXAA, Vadimezan, ASA404, 2, Figure 1), a simple carboxyxanthone derivative, which reached phase III clinical trials towards antitumor activity [36].
This review aims to describe the research findings on biological and pharmacological activities of natural and synthetic carboxyxanthone derivatives. Their applications as suitable chemical substrates to obtain new analogues and derivatives are also presented.

Natural Carboxyxanthone Derivatives
Typically, natural xanthones are classified in six main groups, depending on the nature of the substituents in the xanthone scaffold: simple xanthones, glycosylated xanthones, prenylated xanthones, bis-xanthones, xanthonolignoids and miscellaneous [3,9]. More recently, Masters and Bräse [8] subdivided the natural xanthones in monomers and dimers/heterodimers. Regarding the structural characteristics of natural carboxyxanthone derivatives, in this review they are classified into simple carboxyxanthone derivatives, prenylated carboxyxanthone derivatives, caged carboxyxanthone derivatives, and carboxyxanthone derivatives bound or fused to polysubstituted oxygenated heterocycles.

Natural Carboxyxanthone Derivatives
Typically, natural xanthones are classified in six main groups, depending on the nature of the substituents in the xanthone scaffold: simple xanthones, glycosylated xanthones, prenylated xanthones, bis-xanthones, xanthonolignoids and miscellaneous [3,9]. More recently, Masters and Bräse [8] subdivided the natural xanthones in monomers and dimers/heterodimers. Regarding the structural characteristics of natural carboxyxanthone derivatives, in this review they are classified into simple carboxyxanthone derivatives, prenylated carboxyxanthone derivatives, caged carboxyxanthone derivatives, and carboxyxanthone derivatives bound or fused to polysubstituted oxygenated heterocycles.

Monodictyxanthone (5)
In 2007, Krick et al. [39] isolated a new carboxyxanthone, monodictyxanthone (5) (Figure 2), from the fungus genus Monodictys putredinis and tested it in a series of bioassays for potential cancer chemopreventive activities. The results showed dose-dependent Cytochrome P450 1A activity inhibition and a slight inhibition of the enzyme aromatase [39].
2.1.4. Yicathin C (7) Sun et al. [41] reported, in 2013, the isolation of yicathin C (7) (Figure 2), from the inner tissue of the marine red alga Gymnogongrus flabelliformis. Yicathin C (7) was assayed for antibacterial and antifungal activities using a standard agar diffusion test. Inhibitory activity against E. coli, S. aureus and C. lagenarium was observed [41]. In addition, it was found that this marine carboxyxanthone exhibited weak brine shrimp (Artemia salina) toxicity [41].

Glomexanthones A-C (21-23)
The isolation of glomexanthones A-C (21-23) (Figure 2), from an ethanol extract of Polygala glomerata, was described by Li et al., in 2014 [54]. These compounds were subjected to neuroprotection bioassays in human neuroblastoma SK-N-SH cells and showed moderate neuroprotective effects on L-Glutamic acid-induced cellular damage [54].

Xanthofulvin (96)
In 1993, the pharmaceutical company Hoffmann-La Roche AG, in the person of Dr. Masubuchi, filed a patent on the isolation of a new carboxyxanthone, xanthofulvin (96) (Figure 7), from cultures of Eupenicillium sp. NR7125 [92]. This compound (96) was found to have good inhibitory activity against the enzyme chitin synthase [92]. A decade later, in 2003, Kumagai et al. [93] isolated compound 96 from cultures of Penicillium sp. SPF-3059, and demonstrated that it also exhibited semaphorin inhibitory activity. In the same year, Kikuchi et al. [94] and Kaneko et al. [95] reported that xanthofulvin (96) was the first described Sema3A inhibitor in both in vitro and in vivo studies promoting spinal cord regeneration. Recently, it was evaluated for inhibition of cysteine synthase enzyme by Mori et al. [96] showing inhibitory activity against both EhCS1 and EhCS3. Recently, the mechanism of action of xanthofulvin (96) and vinaxanthone (91) for inhibition of Sema3A have been described [97].

Synthetic Carboxyxanthone Derivatives
Michael and Kostanecki introduced one of the first methods for the synthesis of xanthones, which involved the distillation of a mixture of a phenol, O-hydroxybenzoic acid, and acetic anhydride [101,102]. Since then, several other routes affording higher yields and less drastic experimental

Synthetic Carboxyxanthone Derivatives
Michael and Kostanecki introduced one of the first methods for the synthesis of xanthones, which involved the distillation of a mixture of a phenol, O-hydroxybenzoic acid, and acetic anhydride [101,102]. Since then, several other routes affording higher yields and less drastic experimental conditions have been developed [103][104][105][106][107][108][109][110].
In 2002, an improved synthesis of DMXAA (2) was developed by optimization of the synthesis of the key intermediate 3,4-dimethylanthranilic acid via nitration of 3,4-dimethylbenzoic acid and separation by crystallization [145]. A higher overall yield was obtained from 3,4-dimethylbenzoic acid, specifically 22%. Seven years later, a new short and efficient synthesis of DMXAA (2) was reported using 3,4-dimethylbenzoic acid as starting material [146]. The synthetic pathway comprises of four steps, being the key steps the dibromination of 3,4-dimethylbenzoic acid, followed by the regioselective coupling with 2-hydroxyphenylacetic acid and further cyclodehydration, in an overall yield of 51%. From a biological activity perspective, it is evident that DMXAA (2) may be a useful scaffold for the development of other bioactive compounds and, over the years, several analogues and derivatives have been developed. In 2006, Gobbi et al. [147], synthetized several carboxylated DMXAA (2) analogues (119-134) with potential antitumoral activity ( Figure 10). The synthesis was performed through a multi-step pathway by derivatization of 4-allyl-3-hydroxy-9H-xanthen-9-one. All compounds were tested for antiproliferative activity towards human ovarian adenocarcinoma 2008 cell line, and cisplatin-resistant subline C13* [147]. It was found that compounds 119 and 128 presented good ability to inhibit 2008 cell line [148]. Most of the other compounds only presented cytotoxic activity at the highest tested concentration [147].
In the same study, Gobbi et al. [147] also described another 12 XAA derivatives (135-146) ( Figure  10), specifically the intermediates for synthesis of the analogues 119-134; however, they were not tested for cytotoxic activity. From a biological activity perspective, it is evident that DMXAA (2) may be a useful scaffold for the development of other bioactive compounds and, over the years, several analogues and derivatives have been developed. In 2006, Gobbi et al. [147], synthetized several carboxylated DMXAA (2) analogues (119-134) with potential antitumoral activity ( Figure 10). The synthesis was performed through a multi-step pathway by derivatization of 4-allyl-3-hydroxy-9H-xanthen-9-one. All compounds were tested for antiproliferative activity towards human ovarian adenocarcinoma 2008 cell line, and cisplatin-resistant subline C13* [147]. It was found that compounds 119 and 128 presented good ability to inhibit 2008 cell line [148]. Most of the other compounds only presented cytotoxic activity at the highest tested concentration [147].
In the same study, Gobbi et al. [147] also described another 12 XAA derivatives (135-146) (Figure 10), specifically the intermediates for synthesis of the analogues 119-134; however, they were not tested for cytotoxic activity.
2008 cell line, and cisplatin-resistant subline C13* [147]. It was found that compounds 119 and 128 presented good ability to inhibit 2008 cell line [148]. Most of the other compounds only presented cytotoxic activity at the highest tested concentration [147].
In the same study, Gobbi et al. [147] also described another 12 XAA derivatives (135-146) ( Figure  10), specifically the intermediates for synthesis of the analogues 119-134; however, they were not tested for cytotoxic activity. In 2007, eight new analogues of DMXAA (2) and XAA (104) bearing azido, nitro and amino moieties, compounds 147-154 (Figure 11), were reported by Palmer [148]. All compounds were tested In 2007, eight new analogues of DMXAA (2) and XAA (104) bearing azido, nitro and amino moieties, compounds 147-154 (Figure 11), were reported by Palmer [148]. All compounds were tested for their cytotoxicity on HECPP murine endothelial cells, as well as their ability to induce hemorrhagic necrosis in mice with colon 38 tumors [148]. It was found that compounds 147 and 148 caused profound necrosis on the tested tumors, when compared to the carboxyxanthone derivative 2 [148]. Compound 147 was able to bind specifically to cellular proteins through photoreaction, which could be a useful tool to identify the receptors of DMXAA (2) [148]. for their cytotoxicity on HECPP murine endothelial cells, as well as their ability to induce hemorrhagic necrosis in mice with colon 38 tumors [148]. It was found that compounds 147 and 148 caused profound necrosis on the tested tumors, when compared to the carboxyxanthone derivative 2 [148]. Compound 147 was able to bind specifically to cellular proteins through photoreaction, which could be a useful tool to identify the receptors of DMXAA (2) [148].  Moreover, additional efforts aiming to identify derivatives with improved activity than DMXAA (2) are under investigation. Recently, DMXAA-pyranoxanthone hybrids were reported to enhance inhibition activity against human cancer cells with multi-target functions [150]. Moreover, additional efforts aiming to identify derivatives with improved activity than DMXAA (2) are under investigation. Recently, DMXAA-pyranoxanthone hybrids were reported to enhance inhibition activity against human cancer cells with multi-target functions [150].
Compounds 270-274 were studied as antagonists of leukotriene B 4 receptor (LTB 4 ) [163,164]. These compounds were shown to be, in general, good antagonists of LTB 4 by blocking the up-regulation of the CD11b/CD18 receptor, being compounds 271, 272 and 274 the most active LTB 4 antagonists. It is also important to highlight that compound 274 presented strong binding abilities to human neutrophils and guinea pig lung membranes, being one of the most potent antagonists [163,164]. Table 1. Structure of 9-oxo-9H-xanthene-2-carboxylic acid (162) and analogues
In our group, carboxyxanthone derivative 289 has been used as a building block to obtain diverse chiral derivatives with potential biological activities [167,169,173], as well as chiral selectors for analytical liquid chromatography application [21,22].

Conclusions
During several years, diverse carboxyxanthone derivatives have been obtained either from natural sources or by synthetic methods. Nature afforded more complex structures, but synthetic methodologies could furnish a large variety of carboxyxanthone derivatives for biological activity and structure-activity relationship studies, enlarging the chemical/biological space. For the synthesis of carboxylated xanthone derivatives, diverse methods can be applied if using suitable building blocks. The biological and pharmaceutical significance of these compounds in different areas have been highlighted in this review. Some of them revealed promising activities including antibacterial, antifungal, antiviral, antitumor, antiallergic, anti-inflammatory, diuretic and uricosuric activities as well as inhibitory activity against aldose reductase and as antagonists of leukotriene B4 receptor. Their application as suitable chemical substrates to obtain new bioactive derivatives was also demonstrated. It is anticipated that data compiled in this review will not only update researchers about the pharmacologic significance of carboxyxanthones, but also guide the design for the synthesis of new bioactive xanthone derivatives with improved medicinal properties.