2.2.1. Terpenoid Phenazines

Terpenoid phenazines contain common structural feature of isoprenylated C or N side chains and most of them show moderate or weak antibacterial activity. Kondratyuk et al. isolated marine phenazines **9** and **10** (Figure 4) from *Streptomyces* sp. strain CNS284: **9** demonstrated inhibitory activity of NF-*κ*B and cyclooxygenase-2 (COX-2); **10** showed potent (sub-μM) inhibition activity of prostaglandin E2 (PGE2) production. However, these activities monitored did not have a strong correlation with each other. The mechanism of action was not apparent and needed to be further investigated [40]. Ohlendorf et al. isolated geranylphenazinediol (**11**, Figure 4) from a marine sediment *Streptomyces* sp. strain LB173 [41]: **11** bears geranylation at C-4 side. It showed weak antibacterial activity and grea<sup>t</sup> inhibitory activity toward human acetylcholinesterase (IC50 = 2.62 ± 0.35 μM). Phenaziterpenes A (**12**) and B (**13**, Figure 4) are structurally related to geranylphenazinediol, bearing *O*-geranylation. Song et al. isolated them from *Streptomyces lusitanus* SCSIO LR32. However, compounds **12** and **13** did not show antibacterial activity and cytotoxicity against tumor cell lines [42].

**Figure 4.** The chemical structures of compounds **9**–**19**.

Wu et al. isolated N-prenylated endophenazine **14** (Figure 4) from *Kitasatospora* sp. MBT66. **14** inhibited *B. subtilis* better than positive control drugs Ampicillin and Streptomycin [43]. Han et al. isolated several natural phenazines **15**–**19** (Figure 4) isolated from *Streptomyces* sp. NA04227; **15**–**19** showed moderate inhibitory activity against human acetylcholinesterase and moderate antibacterial activity against *Micrococcus luteus* (MIC = 4 μmol/L) [44].

## 2.2.2. Glycosylated Phenazines

A few natural glycosylated phenazines have so far been found and reported. The activity of glycosylated phenazines was not remarkable and needed to be further investigated. Rusman et al. firstly isolated deglycosylated phenazines (compounds **20**–**25**, Figure 5) from *Streptomyces* sp. Strain DL-93. None of the compounds exhibited any inhibitory activity against tested bacteria and fungi. However, **21**, **22** and **25** showed weak cytotoxicity against HCT-116 cancer cell line with EC50 values of 18 μM, 52 μM and 45 μM, respectively. In vitro

biological assay data demonstrated that the weak cytotoxic activity was not associated with DNA intercalations and topoisomerase inhibition. The mechanisms of action were uncertain and needed to be further investigated [45].

**Figure 5.** The chemical structures of compounds **20**–**30**.

Wu et al. also isolated glycosylated endophenazines (compounds **26**–**30**, Figure 5) from *Kitasatospora* sp. MBT66; **26** and **28** contain sugar moiety and the sugar is methylated at 2--O position. These two compounds are rare in nature and firstly reported; **26**–**30** all showed antibacterial activity to some extent against Gram-positive *B. subtilis*. In addition, **26**–**28** and **30** also showed antimicrobial activity against Gram-negative *E. coli*. Interestingly, they found that glycosylated **27** and **28**, compared with their corresponding aglycone, displayed enhanced activities against Gram-negative *E. coli* [43].

#### 2.2.3. Divergent Fused Phenazines

This class of phenazines contains more than one phenazine-derived moiety. There are a few divergent fused phenazines in nature that have been reported so far. Li et al. isolated diastaphenazine (**31**, Figure 6) from an endophytic *Streptomyces diastaticus* subsp. *ardesiacus*: **31** was a cytotoxic dimeric phenazine, showing antibacterial activity against *S*. *aureus* (MIC = 64 μg/mL). However, **31** was inactive against *E. coli* and *C. albicans* even at 128 μg/mL. Compared with positive control (adriamycin), **31** showed weak cytotoxicity against HCT116, BGC-823, HepG2, HeLa and H460 cell lines with IC50 values of 14.9 μM, 28.8 μM, 65.2 μM and 82.5 μM, respectively [46].

**Figure 6.** The chemical structures of compounds **31**–**38**.

Baraphenazines A−C (**32**–**34**, Figure 6) are fused 5-hydroxyquinoxaline/alpha-keto acid amino acid compounds. Baraphenazines D and E (**35** and **36**, Figure 6) are special diastaphenazine-type compounds. In addition, baraphenazines F and G (**37** and **38**, Figure 6) are phenazinolin-type compounds. Wang et al. isolated them from *Streptomyces* sp. PU-10A and investigated their anticancer activity. Only **36** displayed appreciable activity against A549 and PC3 cell lines with IC50 values of 2.4 μM and 4.7 μM, respectively.Structure-activity relationship (SAR) indicated that the group of amide on **36** was important to the anticancer activity. On the contrary, the group of free acid on **32**−**35**, **37** and **38** was not benefit to the antiproliferative activity. These bioactivity data could explain a general toxicity-based mechanism of action [47].

#### 2.2.4. Biological Activity of New Phenazines

Kennedy et al. isolated 5-methyl phenazine-1-carboxylic acid (**39**, Figure 7) from a rhizosphere soil bacterium. It showed selective cytotoxicity against A549 and MDA MB-231 cell lines in a dose-dependent manner, with IC50 values of 488.7 nM and 458.6 nM, respectively. It exhibited antiproliferative activity by inhibiting cell viability, DNA synthesis and induced G1 cell cycle arrest and apoptosis in cancer cell lines. It was mediated by mitochondrial apoptotic pathway via activation of caspase-3 and down regulation of Bcl-2 expression [32].

**Figure 7.** The chemical structures of compounds **39**–**45**.

Lee et al. isolated compounds **40**–**42** (Figure 7) from yeast-like fungus *Cystobasidium larynigs* IV17-028. These compounds, except for compound **41**, could also inhibit the production of NO, thereby showing inhibitory activity against lipopolysaccharide (LPS)- induced murine macrophage RAW 264.7 cells with EC50 values of 18.10 mg/mL (46.8 μM) and 6.15 mg/mL (22.9 μM), respectively [38]. Deng et al. isolated bioactive compound **43** (Figure 7) from *Streptomyces lomondensis* S015, which inhibited *Pythium ultimum*, *Rhizoctonia solani*, *Septoria steviae* and *Fusarium oxysporum* f. sp. *Niveum*, with MIC values of 16 μg/mL, 32 μg/mL, 16 μg/mL and 16 μg/mL, respectively. These biological data showed the potency of **43** as a promising hit for the further development as a biopesticide [48]. Cha et al. isolated compounds **44** and **45** (Figure 7) from *Streptomyces* sp. UT1123. These two compounds had a unique methylamine linker rather than common methyl ether. Additionally, **44** and **45** showed neuronal protective activity on HT-22 mouse hippocampal neuronal cells even in a low concentration [14].

#### **3. The Progresses of Biosynthetic Pathways of Phenazines**

McDonald et al. found that 2-amino-2-deoxyisochorismic acid could be completely converted into **1**. These compounds were mainly extracted from *Pseudomonas* spp. *PhzB*, *phzD*, *phzE*, *phzF*, *phzG* and so on, which belong to the *phz* gene family and they were proved to play important roles in phenazine synthesis [49]. Chorismic acid (**56**) was not only a common precursor for many primary and secondary metabolism but also the first substrate in biosynthetic pathway towards natural phenazines. Many important phenazines could be produced from microorganisms by this biosynthetic pathway. Combining with previous reports of Xu et al. and Blankenfeldt et al., the classical biosynthetic pathway towards strain-specific phenazines starting from chorismic acid is shown in Figure 8 [5,50].

Normally, the biosynthetic pathway in *Pseudomonas* mainly focused on simple modification of phenazine cores. Shi et al. reported a different biosynthetic pathway of various complex phenazines from the entomopathogenic bacterium *Xenorhabdus szentirmaii*. By modifying the core structure of phenazine, such as electron-rich aromatic rings, reduced form nitrogen(s) and carboxylic acid, a variety of natural phenazine derivatives can be generated. The synthesis of compound **59** is controlled by the typical *phz* operon in *X. szentirmaii* similar to classical biosynthetic pathway. Further modification of **59** was diversified by the enzymes from two discrete biosynthetic gene clusters. This progress of biosynthetic pathway was involved in multiple enzymatic and non-enzymatic reactions (Figure 9) [51].

Guo et al. developed a biosynthetic pathway to synthesize phenazine N-oxides in *Pseudomonas chlororaphis* HT66 (Figure 10). They used three enzymes, a monooxygenase (*phzS*), a monooxygenase (*phzO*) and the N-monooxygenase (*naphzNO1*). Additionally, *naphzNO1* only catalyzed the conversion of **80**, but failed to convert into **81** in vitro. This study also provided a promising method for the synthesis of aromatic N-oxides by *naphzNO1* [52].

**Figure 8.** Classical biosynthetic pathway towards strain-specific phenazines starting from chorismic acid.

**Figure 9.** Complex phenazines **4**, **8** and **59**–**77** and their biosynthetic pathway in *X. szentirmaii*.

**Figure 10.** The designed biosynthetic pathway for phenazine N-oxides **80** and **81**.

#### **4. Synthetic Phenazine Derivatives**

Although natural phenazines possess a variety of biological activities, most of which show moderate or weak activity, thus lacking the possibility to be used as drugs. Structural modification and total synthesis are used to achieve some phenazine derivatives which show notable activity. Normally, the researchers focus on enhancing one special biological activity. Here, synthetic phenazine derivatives will be classified into the following categories in detail, with the perspective of biological activities and functional groups connected to phenazine core.

## *4.1. Antimicrobial Activity*

#### 4.1.1. Halogenated Phenazine Derivatives

According to related reports, bacterium would stop growth in MIC of 2–4 μg/mL and die in MIC of 2 μg/mL [48]. Halogenated phenazines derivatives are tested as antibacterial agents which could target multiple persistent bacterial phenotypes effectively and show negligible toxicity against mammalian cells. Antibacterial effect of halogenated phenazine derivatives could be attributed to membrane disruption, interference with redox cascades or electron-flow and the production of ROS [13,53]. Halogenated phenazine derivatives needed to be further developed by chemists due to the grea<sup>t</sup> antibacterial activity.

Conda-Sheridan et al. synthesized a series of phenazines derivatives inspired by some natural halogenated phenazines. They found *N*-(methylsulfonyl) amide group in the position of C-4 and halogenated group in the position of C-6 would remarkably improve the activity against MRSA. Compounds **82** and **83** (Figure 11) showed stronger antibacterial activity in these synthetic halogenated phenazines compared to positive drug vancomycin (MIC = 2 μg/mL). The mechanism of action of the most active compound was also investigated, but various tested biological data indicated that **83** did not have correlations with major reported antibacterial mechanisms. The in vitro IC50 values of **80** and **81** (Figure 11) against HaCaT cells (immortal keratinocytes) were 118 mM (**82**) and 193 mM (**83**), respectively; **83** seemed to be a promising molecule for the development of MRSA drugs. Additionally, the application of computational methods such as quantitative structure–activity relationship (QSAR) and the prediction of LogP would promote the development of antibacterial drugs [54].

**Figure 11.** The chemical structures of compounds **82** and **83**.

Garrison et al. synthesized a series of phenazines derivatives modified in the positions of C-2, C-4, C-7 and C-8. Compound **84** (Figure 12) proved to be the most potent biofilm-eradicating agen<sup>t</sup> ( ≥99.9% persister cell killing) against Methicillin-resistant *Staphylococcus aureus* (minimal biofilm eradication concentration (MBEC) < 10 μM), Methicillinresistant *Staphylococcus epidermidis* (MBEC = 2.35 μM) and vancomycin-resistant *Enterococcus* (MBEC = 0.20 μM) biofilms, while compound **85** (Figure 12) demonstrated antibacterial activity against *M. tuberculosis* (MIC = 3.13 μM) [55]. Yang et al. explored a series of halogenated phenazines derivatives modified in the positions of C-4, C-6 and C-8. They discovered that 6-substituted halogenated phenazines derivatives could enhance biofilm eradication and antibacterial activities against Methicillin-resistant *Staphylococcus aureus*, Methicillin-resistant *Staphylococcus epidermidis* and vancomycin-resistant *Enterococcus*. In addition, Yang et al. synthesized a polyethylene glycol (PEG)-carbonate phenazine derivative **86** (Figure 12). Its water solubility was improved and demonstrated 30- to 100-fold

enhancement of antibacterial activities against Methicillin-resistant *Staphylococcus aureus* strains, likely through a prodrug mechanism [53].

**Figure 12.** The chemical structures of compounds **84**–**86** and related activity data.

Borrero et al. synthesized several halogenated phenazine derivatives which were inspired by natural halogenated phenazine **87** (Figure 13); **87** was selected as a lead antibiotic which displayed grea<sup>t</sup> inhibitory activity against *S. aureus* (MIC= 1.56 μM). For example, the activity of **88** increased two folds by systematic structural diversification and the SAR was discussed as shown in Figure 13 [56].

**Figure 13.** The chemical structures of compounds **87**–**96** and their inhibitory activity against *S. aureus*.

4.1.2. Derivatives of Clofazimine

Although clofazimine (Figure 14) is an antibiotic against multidrug-resistant *M. tuberculosis*, the clinical utility of this agen<sup>t</sup> is limited by its poor physical and chemical properties and the possibility of skin discoloration. TBI-1004 and B4100, modified at different positions (Figure 14), showed stronger anti-*M. tuberculosis* activity than clofazimine. Zhang et al. designed and synthesized a series of riminophenazine derivatives which

contained a pyridyl group at the C-3 position of the phenazine core, inspired by previous investigations about the developments of TBI-1004 and B4100. Among these derivatives, compound **97** (Figure 14) demonstrated similar activity against *M. tuberculosis*. Additionally, reduced the possibility of skin discoloration in an experimental mouse infection model as compared to clofazimine. In addition, physicochemical properties and pharmacokinetic profiles of **97** were improved [57].

**Figure 14.** The chemical structures of clofazimine, B4100, TBI-1004, 97 and the data of related biochemical properties (MIC, IC50 and ClogP).

Tonelli et al. also synthesized a series of riminophenazine derivatives which contained quinolizidinylalkyl and pyrrolizidinylethyl moieties. These riminophenazine derivatives were tested against *M. tuberculosis* strains H37Rv and H37Ra, six clinical isolates of *M. avium* and *M. tuberculosis* and three mammalian cell lines (HMEC-1, MT-4 and Vero 76). The best compounds **98**–**101** (Figure 15) showed grea<sup>t</sup> inhibition against all strains of *M. tuberculosis* (MIC = 0.82–0.86 μM), **98** showed grea<sup>t</sup> inhibition against *M. avium* (MIC = 3.3 μM). The MIC values for clofazimine were 1.06 μM against *M. tuberculosis* and 4.23 μM against *M. avium*, respectively; **98** demonstrated a selectivity index (SI = 5.23) against the human cell line MT-4 comparable with clofazimine (SI = 6.4). Toxicity of **98** against mammalian Vero 76 cell line was quite low (SI = 79) [58].

**Figure 15.** The chemical structures of compounds **98**–**101**.

#### 4.1.3. Derivatives of Phenazine-1-carboxylic acid

Exemplified by compounds **1** and **2**, simple natural phenazines possess grea<sup>t</sup> fungicidal activities. Many groups continued to investigate their derivatives, hoping to discovery new eco-friendly agrochemicals. Niu et al. designed and synthesized derivatives of phenazine-1-carboxylic acid (**1**) linking with different amino-acid esters. Compounds **102**–**109** (Figure 16) showed greater activity than **1** (EC50 = 66 μg/mL) with EC50 values between 5.35 to 8.85 μg/mL. Particularly, **107** (EC50 = 6.47 μg/mL) and **108** (EC50 = 5.35 μg/mL) showed the best fungicidal activities against *Rhizoctonia solani* Kuhn and none of them had phloem mobility [59].

**Figure 16.** The chemical structures of compounds **102**–**109**.

Taking compound **2** as the lead compound, Zhu et al. designed and synthesized a series of phenazine-1-carboxylic acid (**1**) diamide derivatives. The fungicidal activities were tested by using the inhibitory ratio under 0.2 mmol/L (%) against six phytopathogenic fungi *Rhizoctonia solani*, *Fusarium graminearum*, *Alternaria solani*, *Fusarium oxysporum*, *Sclerotinia sclerotiorum* and *Pyricularia oryzac*. Although all derivatives had fungicidal activities to some degree, the inhibitory activities of most derivatives were lower than control (compound **1**). Compounds **121**–**124** (Scheme 1) demonstrated inhibitory rates more than 50% against *R. solani* and *A. solani*. Particularly, **121** showed the most potent fungicidal activity against *R. solani*, with the inhibitory rate of 72.7%. Compound **121** demonstrated the strongest fungicidal activity against *P. oryzae* with the inhibitory rate of 82.0% [60].

Han et al. designed and synthesized a series of phenazine-1-carboxylic (compound **1**) piperazine derivatives. Most phenazine-1-carboxylic piperazine derivatives showed fungicidal activities in vitro. Particularly, compound **125** (Figure 17) showed inhibitory activity against all tested pathogenic fungi (*R. solani*, *A. solani*, *F. oxysporum*, *F. graminearum* and *P. oryzac*) with EC50 values of 24.6 μM, 42.9 μM, 73.7 μM, 73.8 μM and 34.2 μM, respectively [61]. Lu et al. designed and synthesized the derivatives based on the skeleton of **1**, which contained a series of 1,3,4-oxadiazol-2-yl thioether derivatives. The results of biological assay demonstrated that target compounds possessed moderate to good fungicidal activities against *R. solani*, *S. sclerotioru* and *P. oryzac Cavgra*. Compounds **126** and **127** showed more than 90% inhibitory rate against *S. sclerotioru*. The EC50 values of **126** and **127** were 11.16 μM and 30.47 μM, respectively. In addition, the EC50 value of **127** against *S. sclerotioru* was 10.49 mM, similar as that of compound **1** [62].

**Scheme 1.** Synthesis of compounds **121**–**124**. Reagents and conditions: (**a**) Oxalyl, CH2Cl2, DMF, reflux, 8 h; (**b**) MeOH, rt, 1 h; (**c**) EDA, MeOH, rt, 0 ◦C to reflux, 2 h; (**d**) SOCl2, reflux, 6 h; (**e**) Intermediates **117**–**120**, 0 ◦C, 1 h.

**Figure 17.** The chemical structures of compounds **125**–**127**.

Li et al. also designed and synthesized the derivatives of compound **1** containing substituted groups of triazole. Most 3-benzyl mercapto-1,2,4-triazol derivatives demonstrated fungicidal activity against one or multiple plant pathogens in vitro and in vivo. Compounds **137**–**140** (Scheme 2) displayed better inhibitory activity against rice blast (*P. oryzae*) than **1**. These results provided valuable references for further studies [63].

**Scheme 2.** Synthesis of compounds **137**–**140**. Reagents and conditions: (**a**) Oxalyl chloride, CH2Cl2, DMF, reflux, 8 h; (**b**) NaOH (4%), EtOH, reflux, 20 min; (**c**) pyridine, C4H8O2, 0 ◦C, 2 h.

4.1.4. Water-Soluble Triazole Phenazine Derivatives

Hayden et al. evaluated water-soluble triazole phenazine derivatives, which were synthesized previously. Compounds **141**–**143** (Figure 18) showed high antimicrobial activity at tested concentrations without cytotoxicity against human epithelial cells and tested biological data suggested that **141**–**143** could interrupt metabolic electron-transfer cascades thereby exhibiting cytotoxicity against *E. coli*, rather than production of ROS [64].

**Figure 18.** The chemical structures of compounds **141**–**143**.

## *4.2. Insecticidal Activity*

Podophyllotoxin is a natural product used as the lead compound for the preparation of insecticidal agents. It contains A, B, C, D and E rings. Zhi et al. designed and synthesized a series of podophyllotoxin-based phenazine derivatives modified in the C, D and E rings. In addition, the insecticidal activity of target compounds was investigated which showed insecticidal activity against *Mythimna separata* Walker in vivo. Compounds **148** and **149** (Scheme 3) were phenazine derivatives of 4-acyloxypodophyllotoxin modified in the E ring. They demonstrated stronger insecticidal activity than toosendanin [65]. Then, they designed and synthesized a series of oxime derivatives of podophyllotoxin-based phenazines modified in the C, D and E rings. Compounds **153**–**157** (Scheme 4) exhibited equal or higher insecticidal activity than toosendanin. The combination of podophyllotoxin and phenazine was proved to enhance insecticidal activity [66].

**Scheme 3.** Synthesis of compounds **148** and **149**. Reagents and conditions: (**a**) NaIO4, HOAc, rt, 17 h; (**b**) CH3Cl, rt, 0.5 h; (**c**) RCOOH, DIC, DMAP, DCM, rt, 0.5–7 h.

**Scheme 4.** Synthesis of compounds **153**–**157**. Reagents and conditions: (**a**) NaIO4, HOAc, rt, 17 h; (**b**) CH3Cl, rt, 0.5 h; (**c**) 10% aq. NaOAc/EtOH, reflux, 15 h; (**d**) CrO3, pyridine, CH2Cl2; (**e**) NH4Cl, pyridine, EtOH, reflux, 54 h; (**f**) RSO2Cl, NaH, THF, rt, overnight.
