The Synthesis and Biological Applications of the 1,2,3-Dithiazole Scaffold

The 1,2,3-dithiazole is an underappreciated scaffold in medicinal chemistry despite possessing a wide variety of nascent pharmacological activities. The scaffold has a potential wealth of opportunities within these activities and further afield. The 1,2,3-dithiazole scaffold has already been reported as an antifungal, herbicide, antibacterial, anticancer agent, antiviral, antifibrotic, and is a melanin and Arabidopsis gibberellin 2-oxidase inhibitor. These structure activity relationships are discussed in detail, along with insights and future directions. The review also highlights selected synthetic strategies developed towards the 1,2,3-dithiazole scaffold, how these are integrated to accessibility of chemical space, and to the prism of current and future biological activities.


Introduction
The 1,2,3-dithiazole core is a five membered heterocycle containing two sulfur atoms and one nitrogen atom. Despite the fact that the 1,2,3-dithiazole is not present in nature, similar to many other heterocycles, it does have a broad range of interesting biological activities. The 1,2,3-dithiazole moiety was first synthesized in 1957 by G. Schindler et al. [1]. This was followed two decades later by a report by J. E. Moore on behalf of Chevron Research Co. (San Ramon, CA, USA) where it showcased antifungal and herbicidal activity [2,3]. In 1985, Appel et al. reported the synthesis of 4,5-dichloro-1,2,3-dithiazolium chloride 1 (Appel's salt), a precursor which allowed access to the 1,2,3-dithiazole core within a single step [4,5].
The condensation of Appel salt 1 with primary anilines is well studied [4,5,15,36] and typically occurs by treatment with 1 equiv. of the aniline in the presence of pyridine (2 equiv.) as the base to give, in most cases, good yields of N-aryl-5H-1,2,3-dithiazol-5imines 4 (Scheme 5). Some limitations of this chemistry appear when using heterocyclic arylamines, such as aminopyridines. A recent study by Koutentis et al. highlighted that the reactions of the three isomeric aminopyridines with Appel salt 1 gave very different yields based on the position of the amino group. The 2-, 3-and 4-aminopyridines gave 69%, 24%, and 1% yields of the desired 1,2,3-dithiazole, respectively [37] (Scheme 6). Koutentis et al. suggested the low yield of 4-aminopyridine is likely attributed to the reduced nucleophilicity of the primary amine due to a contribution of its zwitterionic resonance form. The low reactivity of the amine leads to complex reaction mixtures due to side reactions. Scheme 6. Reaction of Appel salt 1 with aminopyridines.

Reactivity of C-4 and the Displacement of the Chloride
The less reactive C4 chlorine of neutral 5H-1,2,3-dithiazoles cannot be directly substituted by nucleophiles. However, utilizing an ANRORC-(Addition of the Nucleophile, Ring Opening, and Ring Closure)-style mechanism, nucleophilic substitution can occur on the C4 chlorine of the 1,2,3-dithazole. An example of this is where the N-Aryl-5H-1,2,3-dithiazol-5imines 4 react with an excess of dialkylamines to give 4-aminodithiazoles 9 in variable yields (Scheme 7). The reaction was found to proceed via an ANRORC-style mechanism [38,39] involving ring opening by nucleophilic attack on the S2 position to yield disulfides 10 and subsequent recyclization after amine addition on the cyano group [40]. In another report by Koutentis et al. [14], DABCO was reacted with neutral 5H-1,2,3-dithiazoles 4-6 to give N-(2-chloroethyl)piperazines 11 in good yields (Scheme 7). The chloroethyl group originating from chloride attack on the intermediate quaternary ammonium salt formed by the displacement of the C4 chloride by DABCO. Scheme 7. Displacement of the C4 chlorine of neutral 5H-1,2,3-dithiazoles.

Alternatives beyond Appel Salt Chemistry
A different way to access both monocyclic and ring fused 1,2,3-dithiazoles is by the reaction of oximes with disulfur dichloride. An example of the synthesis of a ring fused dithiazole is the reaction of benzoindenone oxime 12 to give dithiazole 13 in 81% yield [41,42] (Scheme 8). Acetophenone oximes 14 were reacted with disulfur dichloride to yield dithiazolium chlorides 2, which were subsequently converted to either imines 15, thiones 16, or ketone 17 [13] (Scheme 8). Insights in the mechanism of the oxime to dithiazole transformation were given by Hafner et al. [12], who isolated the dithiazole N-oxide, which is the intermediate in this reaction. Scheme 8. Synthesis of 1,2,3-dithiazoles from oximes.
In another example, the pyrazoleimino dithiazoles 20 were converted to 4-methoxypyrazolo [3,4-d]pyrimidines 21 in medium to good yields by treatment with sodium methoxide in methanol [16] (Scheme 10). The transformation occurs after addition of the methoxide on the nitrile followed by cyclisation onto the dithiazole C5 position that fragments losing S 2 and chloride to give the final pyrimidine 21. A similar example of ring transformations is that of 2-aminobenzyl alcohol dithiazoleimines 4e to 1,3-benzoxazines 22 and 1,3-benzothiazines 23 [44]. Treatment of imines 4e with sodium hydride in THF gave mixtures of benzoxazines 22 and benzothiazines 23, with the former as the main products (Scheme 11). The formation of the former involves deprotonation of the alcohol and cyclisation of the alkoxide onto the dithiazole C5 position. Subsequent fragmentation with loss of S 2 and chloride gave the final benzoxazine 22. Alternatively, treatment of imines 4e with Ph 3 P gave exclusively benzothiazines 23 in good yields (Scheme 11). Thiophilic attack on S1 ring opens the dithiazole ring and a second attack by Ph 3 P gives the intermediate alkene 24 that cyclizes to benzothiazine 23. Scheme 11. Synthesis of 1,3-benzoxazines 22 and 1,3-benzothiazines 23.
1,2,3-Dithiazole derivatives can also be converted to mercaptoacetonitriles by the removal of the S1 atom. One example of this are the 3-(1,2,3-dithiazolylidene)indololin-2ones 25 reacting with sodium hydride (2 equiv.) to yield the mercaptoacetonitrile products 26 in medium to good yields [45] (Scheme 12). Perhaps the most unstable 1,2,3-dithiazole is Appel salt itself, which, while relatively stable at ca. 20 • C under a desiccant, in its absence, Appel salt has a tendency to react with moisture. One study by Koutentis et al. revealed that simple stirring in wet MeCN gave elemental sulfur, dithiazole-5-thione 6, dithiazol-5-one 5, and thiazol-5-one 27 [46] (Scheme 13), assisting other scientists working with Appel salt, to identify these products. Interestingly, other dithiazolium salts have also been prepared with increased stability and lower sensitivity to moisture. A series of perchlorate salts of 1,2,3-dithiazoles were prepared by the anion exchange with perchloric acid allowing for more detailed characterization and study of the 1,2,3-dithiazole [29]. In another study by Rakitin et al., 4-substituted 5H-1,2,3-dithiazoles 16 and 17 were converted to 1,2,5-thiadiazoles 28 and 29 by treatment with primary amines [47] (Scheme 14). Mechanistically, the reaction occurs by addition of the amine to the C5 position followed by ring opening of the C-S bond and subsequent ring closing by loss of hydrogen sulfide. To summarize, 1,2,3-dithiazoles can be converted to other heterocyclic or ring opened derivatives. The six most common mechanisms involved in the transformations of 1,2,3dithiazoles to other systems are shown below (Scheme 15). These mechanisms begin via a nucleophile assisted ring opening of the dithiazole to disulfide intermediates that then can react either intermolecular or intramolecular with other nucleophiles via the six paths presented.

1,2,3-Dithiazoles in Medicinal Chemistry
3.1. Antimicrobial Activities of 1,2,3-Dithiazoles, including Antifungal, Herbicidal, and Antibacterial The first report of biological activity using the 1,2,3-dithiazole scaffold was published in a patent filed by J. E. Moore in 1977 on behalf of Chevron Research Co. [2,3]. The patent disclosed a series of novel 1,2,3-dithiazoles afforded in a 2-3 step sequence from N-aryl cyanothioformamide and sulfur dichloride. The main application of these compounds was the controlling of various fungal infections, leaf blights, invasive plant species, and mites.
A series of halogenated analogues 35-39 were then identified as active against the powdery mildew pathogen Erysiphe polygoni using bean seedlings with well-developed primary leaves. The (Z)-4-chloro-N-(4-chloro-2-methylphenyl)-5H-1,2,3-dithiazol-5-imine analogue (35) along with the corresponding 3-chloro 36 showed 100% protection at 250 ppm. The corresponding 5-chloro 37 and 4-bromo 38 both showed a small reduction in efficacy, 10% and 1%, respectively, while the 3,5-dichloro 39 was only net 76% effective ( Figure 3).  Initial screening was also carried out against necrotrophic fungus Botrytis cinerea on the well-developed primary leaves of a 4-6-week-old horsebean plant at a lower concentration (40 ppm). Only 1,2,3-dithiazole 35 was demonstrated to be effective with 92% inhibition ( Figure 3). However, after this initial result, screening was carried out on a broader panel of fungal ( Figure 4) and herbicidal strains ( Figure 5). The fungal panel included Botrytis cinerea, Rhizoctonia solani, Fusarium moniloforma, Phythium ultimum, and Aspergillus niger. The compounds 30, 33, and 39-48 were tested at 500 ppm and fungicidal activities were measured by the zone of inhibited mycelia growth ( Figure 4). Interestingly, the unsubstituted phenyl analogue (Z)-4-chloro-N-phenyl-5H-1,2,3-dithiazol-5-imine (40) was active on Botrytis cinerea at 0.33 µg/cm 2 . The addition of a 4-position methyl in analogue 41 reduced the activity against Botrytis cinerea by over 2-fold, but increased the activity against Rhizoctonia solani and Fusarium moniloforma.  The 1,2,3-dithiazoles were then screened at 33 ppm on a herbal panel that included wild oats (Awena fatua), watergrass (Echinochloa crusgall), crabgrass (Digitaria sanguinalis), mustard (Brassica arversis), pigweed (Amaranthus retroflexus), and lambsquarter (Cheropodium album) ( Figure 5). The first analogue (Z)-4-chloro-N-(p-tolyl)-5H-1,2,3-dithiazol-5-imine (41), showed good efficacy against Amaranthus retroflexus (90%) and total control of Brassica arvensis. Switching to the 4-fluoro analogue 49 increased coverage across all strains tested, including Avena fatua (40%), which was only weakly inhibited across the series and total control of Amaranthus retroflexus. The 4-chloro analogue 44 was 3-fold less effective against Digitaria sanguinalis and Avena fatua. The addition of a 2-position chloro 50 decreased strain coverage, but did mean total control of Brassica arvensis in addition to Amaranthus retroflexus, with additional high efficacy against Chenopodium Album (95%). The original 2-methyl 4-chloro analogue 35, while still showing efficacy across several strains, did not offer total or near total control for any of the strains tested. The 2-chloro analogue 50 showed total control for Chenopodium album and Brassica arvensis and near total for Amaranthus retroflexus (93%). However, 2-chloro 50 had a limited effect on Digitaria sanguinalis and Echinochloa crusgalli, with no impact on Avena fatua. The 3,5-dichloro analogue 39 demonstrated good efficacy against most strains, including total control of Amaranthus retroflexus, Chenopdium album, Brassica arvensis, and some activity against Avena fatua (35%). The 2-methyl, 5-chloro analogue 51 offered the highest efficacy across the series on Avena fatua (45%), total control of Amaranthus retroflexus and Brassica arvensis, with near total control of Chenopodium album (95%). The 3,4-dichloro analogue 45 had a potent but narrower band of activity with total control of Amaranthus retroflexus, Chenopdium album and Brassica arvensis, but weaker activity on the other three strains (30-55%). The 3-bromo analogue 38 has a similar profile to the 4-methyl 41, while the 2-naphthyl analogue 52 was the most potent in the screening for Echinochloa crusgalli (90%) and offered good control over Amaranthus retroflexus (85%) and total control over Brassica arvensis. In order to test for other pests, pinto bean leaves were treated with two spotted mites (Tetramuchus urticae). The mites were then allowed to lay eggs on the leaves, and after 48 h, the leaves were treated with 40 ppm of the test compound ( Figure 6). A series of halogenated phenyl-5H-1,2,3-dithiazol-5-imines were identified with activity against both Tetramuchus urticae and their eggs. The 3,5-dichloro analogue 39 showed a high degree of control with 90% of mites and 85% of eggs suppressed. This increased to almost total control with the 3,4-dichloro 45. Interestingly, the 2,4-dichloro analogue 31 demonstrated total mite control but had no effect on the eggs. The mono-substituted 2-chloro analogue had a similar profile with no effect on the mite eggs, but only 70% effective control of the mite. The 4-chloro, 2-methyl analogue 35 showed complete egg control and almost complete mite control (94%). The switch to the bromo 38 showed a similar profile, but with 70% mite control. The 2-methyl substituted match pair analogues 3-chloro 56 and 5-chloro 52 both demonstrated a high level of mite and egg control with the 3-position preferred.  This work was extended in a subsequent report by Pons et al. [50], where a focused library of 1,2,3-dithiazoles and related analogues were screened on a series of fungal targets. The 1,2,3-dithiazoles were the only compounds that showed antifungal activity, with most potent analogues identified as unsubstituted aromatic 40, the 2-methoxy 54, and 4-methoxy analogue 55 ( Figure 8). These three most potent analogues all had an MIC of 16 µg/mL on C. albicans, C. glabrata, C. tropicalis, L. orientalis, and an MIC of 8 µg/mL on C. neoformans.  [51], a company specializing in the manufacture of coatings. The disclosed innovation involved the use of 1,2,3-dithiazoles to rapidly inhibit microbial and algae growth for industrial applications. These included paints, coatings, treatments, and textiles, among others. The effective amount applied was between 0.1 to 300 ppm, with three main exemplar 1,2,3-dithiazoles highlighted ( Figure 9). This included 4-chloro-5H-1,2,3-dithiazol-5-one (5)  Subsequently in 1998, more detailed screening and structure activity relationships (SAR) were published from Pons et al. related to the antimicrobial properties of the 1,2,3dithiazole scaffold [52,53]. These two studies tested activity against bacteria: S. aureus, E. faecalis, S. pyogenes, and L. monocytogenes, and fungi: C. albicans, C. glabrata, C. tropicalis, and I. orientalis. This screening supported earlier work on the 1,2,3-dithiazole scaffold, and broadened the scope of this inhibition to several new fungal and bacteria strains ( Figure 10). The compounds showed antibacterial activity against Gram-positive bacteria, but as previously described [15], there was no activity against Gram-negative bacteria. A switch to fused heterocycles including quinolines and naphthalene was maintained rather than increased overall potency and coverage. The quinolin-6-yl substituted analogue 62 showed potency against C. glabrata and C. albicans (both MIC = 16 µg/mL), while the quinolin-5-yl 63 was only active against the C. albicans. The naphthalen-1-yl 65 and hydroxy substituted naphthalen-1-yl 66 had the same profile with coverage against all four bacteria tested (S. aureus, E. faecalis, S. pyogenes, and L. monocyotogenes all MIC = 16 µg/mL) and C. albicans. The hydroxy substitution of 65 to afford 66 did not provide any potency advantage, but did demonstrate there was an ability to alter physicochemical properties without affecting potency.
More recently, a 2020 study by our group reported a set of 1,2,3-dithiazoles and matched pair 1,2,3-thiaselenazoles as antimicrobials [55]. The rare 1,2,3-thiaselenazoles were synthesized by sulfur extrusion and selenium insertion into 1,2,3-dithiazoles [55,56]. This work was part of the Community for Antimicrobial Drug Discovery (CO-ADD) project to develop new lead compounds for priority targets with an unmet clinical need [57]. The compounds were screened against S. aureus, A. baumannii, C. albicans, and C. neoformans var. grubii. with a toxicity counter screen in HEK293 cells and an additional hemolysis assay (Hc10) (Figure 20). These strains are considered by the World Health Organization (WHO) to be the highest priority to develop novel antibiotics for control of these bacteria and fungi [58].       [1,4]oxazine (120) were similar with antifungal activity against C. albicans and C. neofromans (all MIC = ≤0.25 µg/mL, apart from 120, C. neofromans = 2 µg/mL). Taken together these results demonstrate an ability for the 1,2,3-dithiazole/ 1,2,3-thiaselenazole to inhibit a broad range of challenging and clinically relevant bacteria and fungi [55,58].
The proposed mechanism of action was modelled on previously experimental reports ( Figure 22) [59][60][61][62]. Zinc ejection from nucleocapsid protein starts with Zn 2+ coordinated to cysteine thiol(ate)s reacting with the disulfide of the 1,2,3-dithiazole core to generate a transient intermediate disulfide. This complex then rearranges to form an intramolecular protein disulfide, which has a consequent reduction in zinc ion affinity. This results in the zinc being ejected from the protein in a similar mechanism as previously reported for the HIF1alpha/P300 interaction triple zinc finger [59]. To indirectly prove the mechanism in addition to computational modelling,    This idea was followed up in 2019 by our group [63], investigating the same inhibitors later reported as antimicrobials [55]. The key rationale behind this subsequent work was the further investigation of the disulfide bridge involvement on antiviral efficacy with a matched pair side by side comparison between the 1,2,3-dithiazoles and the 1,2,3-thiaselenazole scaffold. Where the weaker S-Se vs. S-S bond should assist in increasing the antiviral efficacy. This followed on from a previous report of a successful selenide isosteric replacement to several literature nucleocapsid protein inhibitors, including DIBA-4 to DISeBA-4 HIV inhibitors, resulting in good potency and only a very limited associated toxicity [64]. The antiviral efficacy of the 1,2,3-dithiazole scaffold was tested using FL-4 cells, but in this study an additional toxicity assay was preformed directly on the FL-4 cells ( Figure 24). The 8-phenylindeno [1,2- [1,4]oxazine (128) and selenium analogue 120 was even more pronounced with an almost 17-fold increase in potency. These results highlight the advantages of including selenium in the 1,2,3-dithiazole scaffold. More recently, a further extension of investigation of the 1,2,3-dithiazoles in 2022 by our group evaluated a further series of 1,2,3-dithiazoles against FIV as a model for HIV infection [36]. The rationale of this investigation was to find a tractable series of 1,2,3-dithiazoles with consistently high potency and lower toxicity to further advance the scaffold. The antiviral screening was performed using FL-4 cells, with a direct toxicity assay on FL-4 cells in addition to CrFK and feline embryo cell line (FEA) cells (Figures 25-27).    substituted analogues 4a, 90, 112 and 132, at a similar level to the earlier analogues but with a divergent SAR profile ( Figure 26). The most promising compound identified in this work was the pyrazole (Z)-4-chloro-N-(3-methyl-1H-pyrazol-5-yl)-5H-1,2,3-dithiazol-5-imine (133) that showed good antiviral potency EC 50 = 0.083 µM with very limited toxicity ( Figure 27).
The proposed mechanism of action on the nucleocapsid protein of this 4-chloro-1,2,3dithiazol-5-imine series is different to the C4 substituted version previously reported ( Figure 22). The DFT calculations and previous ANRORC-style rearrangement reported on this scaffold suggest that there will be a ring opening and chloride elimination. An outline mechanism would be a Zn 2+ -coordinating cysteine thiol(ate) reacts with 2-S of the 1,2,3-dithiazole core mediated by water to generate a transient trisulfide. This is then followed by a rearrangement to a more thermodynamically stable cyano functionality, resulting in the loss of HCl and water from the system. The, disulfide then rearranges to form an intramolecular protein disulfide with consequent reduction in zinc ion affinity. The zinc ion is then ejected to form a stable complex, with or without adducts ( Figure 28). In addition to the literature rearrangement examples [38][39][40], we also provided extensive computational modelling to support the mechanistic rationale provided.

Anticancer Activities of 1,2,3-Dithiazoles
Initial reports of anticancer activity with the 1,2,3-dithiazole scaffold were reported in 2002 by Baraldi et al. [16]. A set of ten 1,2,3-dithiazoles were prepared and screened across multiple antimicrobial and anticancer therapeutic targets. Several of the compounds 134-136 showed low signal digit micromolar potency against the leukemia cell lines L1210 and K562 (Figure 29). While the overall SAR within the series was flat, this first phenotypic report showed tractable activity across both cell lines. The antibacterial screen showed limited activity, but the antifungal screening identified 135 and 136 as having some activity against Aspergillus niger at MIC 50 = 10 µM. This further supports the overall tractability of this scaffold as an antifungal.  Subsequent to these phenotypic reports of 1,2,3-dithiazoles as anticancer compounds in various cell lines, a just over forty compound 1,2,3-dithiazole library was screened by Indiveri et al. against a transporter target over-expressed in various cancers, the glutamineamino acid transporter ASCT2 in 2012 [65]. Interactions with scaffold proteins and posttranslational modifications regulate the stability, trafficking, and transport activity of ASCT2 [66]. The expression of ASCT2 has been shown to increase in cells with rapid proliferation, including stem cells and inflammation, this enables delivery of the increased glutamine requirements [67]. This same mechanism can be hijacked by cancer promoting pathways to fulfill glutamine demand and facilitate rapid growth by over-expression of ASCT2 [68]. In addition to being described as an anticancer target, ASCT2 also has the ability to traffic virions to infect human cells [69]. A series of 1,2,3-dithiazoles were synthesized and evaluated as transporter inhibitors. While many compounds were in-active at 30 µM, six compounds showed activity at IC 50 =~10 µM or below ( Figure 31). These compounds potently inhibited the glutamine/glutamine transport catalyzed by ASCT2. Figure 31. Anticancer 1,2,3-dithiazoles ASCT2 transport inhibitors.
The inhibition was shown to be non-competitive. The inhibition was also reversed by addition of dithiothreitol (DTE), indicating the reaction with protein Cys formed adducts, indicating that the reaction was likely going via an ANRORC-style rearrangement. Modelling, including molecular and quantum mechanical studies (MM and QM, respectively) and Frontier Orbital Theory (FOT) on 1,2,3-dithiazole models showed pathway (ii) was more likely, which is also supported by previous reports on the 1,2,3-dithiazole ( Figure 32) [38][39][40]. The ASCT2 report was followed up by a screening of just over fifty 1,2,3-dithiazoles by Indiveri et al. against the LAT1 transporter in 2017 [70]. ASCT2 and LAT1 are both amino acid transporters that are overexpressed in cancer [71]. Subsequently, a number of inhibitors have been reported against both ASCT2 and LAT1, with one LAT1 inhibitor JPH203 used in a recent phase 1 clinical trial [72]. The results of the library screen were eight compounds with inhibition of >90% at 100 µM. The two most potent compounds were 144 and 145 with an IC 50 = <1 µM (Figure 33). The inhibition kinetics, performed on the two best inhibitors (144 and 145), indicated a mixed type of inhibition with respect to the substrate. The inhibition of LAT1 was still present after removal of the compounds from the reaction mixture, indicating irreversible binding. However, this effect could be reversed by the addition of dithioerythritol, a S-S reducing agent, which supports the rationale of the formation of disulfide(s) bonds between the compounds and LAT1. Molecular modelling of 144 and 145 on a homology model of LAT1, highlighted the interaction with the substrate binding site and the formation of a covalent bond with the residue C407. This was further supported by a more detailed study reported in 2021 by Marino et al., which also highlighted the need for a molecule of water in the reactive pathway [73].
Subsequent to this screening, a series of C-4 substituted analogues were evaluated resulting in a trend of activity against breast cancer (MCF7) (Figure 36) [14].

Melanin Synthesis Inhibitors
In 2015, a phenotypic screen was carried out using Xenopus laevis embryos by Skourides et al. [19]. This led to the identification of a series of 1,2,3-dithiazoles, which caused loss of pigmentation in melanophores and the retinal pigment epithelium (RPE) of developing embryos ( Figure 37). This effect was independent of the developmental stage of initial exposure and was reversible. While the target was not elucidated, SAR of the series indicated that the presence of the mesmerically electron-donating methoxy group was important for pigment loss. Compounds with inductive and/or mesmerically electron-withdrawing groups had no effect on pigment loss. The (Z)-4-chloro-N-(4-methoxyphenyl)-5H-1,2,3-dithiazol-5-imine (154) analogue demonstrated complete pigment loss at 10 µM and moderate at 5 µM. Extension of the methoxy to propyloxy 155 or butyloxy 156 reduced the potency, with the addition of a methyl group in the 2-position had the same effect. The formation of a 3,4-fused methyl catacol 157 increased potency, but did not match the activity of the 4-position methoxy 154. The extension to form that benzyloxy analogue 158 did boost the potency of 154, resulting in 158 having complete pigment loss at 5 µM.
Skourides et al. extensively investigated the structural features driving the phenotypic effects observed with 159. An analogue of 159 was synthesized in two steps via the oxime route (Scheme 8) [12,13], where the 4-chlorine substituent was replaced with a phenyl group to give compound 160 (Figure 38). A second analogue of 159 was furnished where the nitrogen of the 1,2,3-dithiazole was replaced with a chlorocarbon in one step from Boberg salt [75,76] to give compound 161 ( Figure 38) [75,77]. The replacement of the 4-chlorine substituent yielded compound 160, which showed no phenotypic affect. This supports the idea of an ANRORC-style ring opening mechanism, as the chlorine is a good nucleofuge that facilities the ring opening mechanism [40]. The second analogue 161, showed some mild activity at 5 µM, while at 10 µM, mild toxicity and developmental defects were observed. This again pointed towards a ring opening ARONOC style mechanism, but more work needs to be done to establish the exact mechanism of action [19].
In 2005, Ananthanarayanan et al. screened a Maybridge compound library against Heat shock protein 47 (Hsp47), which, at the time, had no known inhibitors. Hsp47 is a collagen-specific molecular chaperone whose activity has been implicated in the pathogenesis of fibrotic diseases. The regulation of both Hsp47 and collagen expression has been implicated in several different disease indications where changes in the collagen expression are found. These diseases include fibrotic diseases of the liver [78], kidney [79], lung [80], and skin [81], in addition to atherosclerosis [82] and cancer [83]. The screen resulted in a primary hit rate of 0.2%, with 4 out of 2080 compounds being shown to be inhibitors of Hsp47. Secondary screening confirmed 162 (Figure 39), as the most potent compound (IC 50 = 3.1 µM).

Arabidopsis Gibberellin 2-Oxidase Inhibitors
In 2010, screening a commercial library of starting points against to Arabidopsis gibberellin 2-oxidases identified compound 162 (Figure 40) [20]. The screening aimed to identify an inhibitor that could both promote Arabidopsis seed germination and seedling growth. Compound 162 was able to do both, without having broad spectrum activity similar to Prohexadione (PHX), which is a broad-spectrum inhibitor of all three 2-oxoglutarate dependent dioxygenase's (2ODD) that were involved in Gibberellin (GA) production (GA 2-oxidase (GA2oxs), GA 3-oxidase (GA3oxs), and GA20-oxidase (GA20oxs)) [84,85]. The 1,2,3-dithiazole 162 was shown to have inhibition GA2oxs with a high degree of specificity, but not on other 2ODDs. The selective inhibition of GA2oxs activity could potentially lead to the delay of GA catabolism in plants, and hence, extend the life of endogenous GA.
The first screening was carried out by Chevron Research Co. in 1977 [2,3], this relatively detailed study has been the foundation of the phenotypic biology observed on this scaffold. It described detailed work on a series of herbicidal effects and anti-mite efficacy, in addition to antifungal activities. This work was followed up in the late 1990s and 2000s by a series of groups extending the understanding of the antifungal and antibacterial SAR scope of the 1,2,3-dithiazole scaffold [15,[50][51][52][53]. Interestingly, a patent in 1997 by Rohm & Haas Co. [51] highlighted a potential coating application for the 1,2,3-dithiazole with the discovery of potent antialgae and Gram-negative bacteria inhibition. In 2012, a patent filed by Bayer Cropscience AG presented a much broader library of heteroaromatic derivatives [54], highlighting a wider range of antifungal activities, with high degrees of control of commercially important fungi for crop protection. More recently, the antifungal and antibacterial screening has focused on clinically relevant hospital derived infections with good efficacy [86], in part aided by a series of matched pair 1,2,3-thiaselenazoles [55].
More recently, several other phenotypic observations have been reported. These include antiviral efficacy against FIV as a model for HIV, where modelling and mechanistic rationale point to cystine containing nucleocapsid protein (NCp) as the target for the 1,2,3-dithiazole [17,36,63]. Anticancer effects against a broad range of cancer cell lines have also been reported with limited off-target toxicity [13,16,74]. These were also supported by modelling and mechanistic rationale highlighting ASCT2 [65] and LAT1 [70] as potential targets responsible. This rationale has been further supported by a series of mechanism of action experiments [65,70,73].
In addition to these reports, a series of other studies also highlighted other activities of the 1,2,3-dithiazole scaffold. These included an anti-melanin phenotype in Xenopus laevis embryos, where active potent (>5 µM) non-toxic compounds were identified in an in vivo model [19]. An ANRONC style mechanism of action was proposed supported by a series of chemical modifications to the scaffold [38][39][40]. Finally, two reports of high-throughput screens identified hit compounds against antifibrotic collagen specific chaperone hsp47 [18] and Arabidopsis Gibberellin 2-Oxidase [20].
The full potential of the 1,2,3-dithiazole scaffold has yet to be realized. Key areas of biological activities have been identified with preliminary work in the literature showing encouraging results. These included activities as antifungal [2], herbicidal [2], antibacterial [15], anticancer [16], antiviral [17], antifibrotic [18], and being a melanin [19] and Arabidopsis gibberellin 2-oxidases [20] inhibitors. These results provide a prospective to the versatility as to what is possible with this scaffold. In addition to these interesting reported biology applications, there are potentially significant untapped chemical biology opportunities towards targeting cystine reactive sites [21][22][23][24][25]; using the ANRORC-style 1,2,3-dithiazole chemistry as a latent functionality ( Figure 41). [54], highlighting a wider range of antifungal activities, with high degrees of control of commercially important fungi for crop protection. More recently, the antifungal and antibacterial screening has focused on clinically relevant hospital derived infections with good efficacy [86], in part aided by a series of matched pair 1,2,3-thiaselenazoles [55].
More recently, several other phenotypic observations have been reported. These include antiviral efficacy against FIV as a model for HIV, where modelling and mechanistic rationale point to cystine containing nucleocapsid protein (NCp) as the target for the 1,2,3dithiazole [17,36,63]. Anticancer effects against a broad range of cancer cell lines have also been reported with limited off-target toxicity [13,16,74]. These were also supported by modelling and mechanistic rationale highlighting ASCT2 [65] and LAT1 [70] as potential targets responsible. This rationale has been further supported by a series of mechanism of action experiments [65,70,73].
In addition to these reports, a series of other studies also highlighted other activities of the 1,2,3-dithiazole scaffold. These included an anti-melanin phenotype in Xenopus laevis embryos, where active potent (>5 μM) non-toxic compounds were identified in an in vivo model [19]. An ANRONC style mechanism of action was proposed supported by a series of chemical modifications to the scaffold [38][39][40]. Finally, two reports of highthroughput screens identified hit compounds against antifibrotic collagen specific chaperone hsp47 [18] and Arabidopsis Gibberellin 2-Oxidase [20].
The full potential of the 1,2,3-dithiazole scaffold has yet to be realized. Key areas of biological activities have been identified with preliminary work in the literature showing encouraging results. These included activities as antifungal [2], herbicidal [2], antibacterial [15], anticancer [16], antiviral [17], antifibrotic [18], and being a melanin [19] and Arabidopsis gibberellin 2-oxidases [20] inhibitors. These results provide a prospective to the versatility as to what is possible with this scaffold. In addition to these interesting reported biology applications, there are potentially significant untapped chemical biology opportunities towards targeting cystine reactive sites [21][22][23][24][25]; using the ANRORC-style 1,2,3dithiazole chemistry as a latent functionality ( Figure 41).

Conclusions
Taken together, the chemistry and biology of the 1,2,3-dithiazoles chemotype has shown a lot of exciting potential. The ANRORC-style rearrangements potentially affording a new route for potential chemical tools and relative cystine within proteins pockets, while the sub-micro molar phenotypic potencies against a series of diverse targets demonstrate potential for further development. Many of these diseases and pathogens have limited treatment options and need new therapies with novel mechanisms of action. The

Conclusions
Taken together, the chemistry and biology of the 1,2,3-dithiazoles chemotype has shown a lot of exciting potential. The ANRORC-style rearrangements potentially affording a new route for potential chemical tools and relative cystine within proteins pockets, while the sub-micro molar phenotypic potencies against a series of diverse targets demonstrate potential for further development. Many of these diseases and pathogens have limited treatment options and need new therapies with novel mechanisms of action. The identification of starting points and defined SAR provides the foundation to define a medicinal chemistry trajectory towards optimized inhibitors and potential new treatments for a broad range of diseases.

Conflicts of Interest:
The authors declare no conflict of interest.
Sample Availability: Not applicable.