*5.1. Delitschiapyrone A*

Delitschiapyrone A (**40**) is a fungal polyketide bearing an unprecedented 6/6/7/5/6-fused pentacyclic ring system (Figure 7), and isolated from a solid culture of the leaf-associated

fungus *Delitschia* sp. FL1581 [52]. The absolute configuration of **40** was determined by spectroscopic analysis, X-ray crystallography data, and experimental and calculated ECD. A naphthalenone unit and an α-pyrone moiety were proposed to be linked together via the Diels–Alder addition followed by an α-ketol-type rearrangement to forge the pentacylic ring system of **40** (Figure 7), which suggested the biosynthetic pathway of **40** might be concerned with the crosstalk between two separate BGCs (one for naphthalenone biosynthesis and the other for α-pyrone biosynthesis). A bioinspired total synthesis of delitschiapyrone A has been achieved by simply stirring a heterogeneous mixture of two Diels−Alder substrates, which gave a hint that the intermolecular Diels−Alder reaction might be spontaneous. Recently, Houk and co-workers investigated the mechanisms and dynamics of biosynthetic formation of **40** by density functional theory (DFT) calculations and quasiclassical molecular dynamics simulations with DFT and xTB, and drew a conclusion that **40** is not formed from the proposed Diels−Alder/α-ketol rearrangement cascade but instead formed directly from a single cycloaddition reaction [53], which is of great significance to the subsequent study on the biosynthetic pathway of **40** through gene deletion, heterologous expression, and enzymatic assays. sociated fungus *Delitschia* sp. FL1581 [52]. The absolute configuration of **40** was determined by spectroscopic analysis, X-ray crystallography data, and experimental and calculated ECD. A naphthalenone unit and an α-pyrone moiety were proposed to be linked together via the Diels–Alder addition followed by an α-ketol-type rearrangement to forge the pentacylic ring system of **40** (Figure 7), which suggested the biosynthetic pathway of **40** might be concerned with the crosstalk between two separate BGCs (one for naphthalenone biosynthesis and the other for α-pyrone biosynthesis). A bioinspired total synthesis of delitschiapyrone A has been achieved by simply stirring a heterogeneous mixture of two Diels−Alder substrates, which gave a hint that the intermolecular Diels−Alder reaction might be spontaneous. Recently, Houk and co-workers investigated the mechanisms and dynamics of biosynthetic formation of **40** by density functional theory (DFT) calculations and quasiclassical molecular dynamics simulations with DFT and xTB, and drew a conclusion that **40** is not formed from the proposed Diels−Alder/α-ketol rearrangement cascade but instead formed directly from a single cycloaddition reaction [53], which is of great significance to the subsequent study on the biosynthetic pathway of **40** through gene deletion, heterologous expression, and enzymatic assays.

HtyC. Finally, the transaminase HtyB catalyzed the transamination of **38** to form **39**. In general, the biosynthesis of echinocandin B needs the coupling of two sperate BGCs, the *ecd* and *hty* BGCs. The *hty* BGC provides an important biosynthetic precursor L-homotyrosine which was recognized by the fourth A domain of the NRPS EcdA. Understanding the biosynthetic mechanism of echinocandin B will facilitate us to take advantage of synthetic biology techniques to bioengineer NRPSs to generate bioactive compounds [49–51].

**5. Representative Fungal NPs Might Be Biosynthesized by Two Separate Pathways** 

Delitschiapyrone A (**40**) is a fungal polyketide bearing an unprecedented 6/6/7/5/6 fused pentacyclic ring system (Figure 7), and isolated from a solid culture of the leaf-as-

*J. Fungi* **2022**, *8*, x FOR PEER REVIEW 10 of 16

**Figure 7.** The proposed biosynthetic pathway of delitschiapyrone A (**40**). **Figure 7.** The proposed biosynthetic pathway of delitschiapyrone A (**40**).

#### *5.2. Herpotrichone A 5.2. Herpotrichone A*

**Crosstalk**

*5.1. Delitschiapyrone A*

Herpotrichone A (**41**) is a fungal polyketide isolated from the isopod-associated fungus *Herpotrichia* sp. SF09 with an unprecedented pentacyclic 6/6/6/6/3 skeleton, and shows outstanding anti-neuro inflammatory activities in lipopolysaccharide (LPS)-induced BV-2 microglial cells (Figure 8) [54]. Interestingly, compound **41** is also an intermolecular [4 + 2] adduct that involves the coupling of epoxycyclohexenone and α-pyrone two building blocks via the Diels–Alder addition. Recently, the biosynthetic pathways of epoxycyclohexenone-derived fungal polyketides trichoxide, sordarial, and flavoglaucin have been investigated in detail, which provide new insights into the biosynthesis of the naphthalenone unit in **41** [55–57]. The biosynthesis of some fungal α-pyrone-linked NPs such as alternapyrones and citreoviridin have been reported. The α-pyrone moiety in alternapyrones and citreoviridin is indeed formed by the spontaneous intramolecular cyclization of PKS AlpA and CtvA, respectively [58,59]. Thus, herpotrichone A might share the similar strategy to forge the α-pyrone unit by utilization of an unidentified PKS. Overall, fungal Herpotrichone A (**41**) is a fungal polyketide isolated from the isopod-associated fungus *Herpotrichia* sp. SF09 with an unprecedented pentacyclic 6/6/6/6/3 skeleton, and shows outstanding anti-neuro inflammatory activities in lipopolysaccharide (LPS)-induced BV-2 microglial cells (Figure 8) [54]. Interestingly, compound **41** is also an intermolecular [4 + 2] adduct that involves the coupling of epoxycyclohexenone and α-pyrone two building blocks via the Diels–Alder addition. Recently, the biosynthetic pathways of epoxycyclohexenone-derived fungal polyketides trichoxide, sordarial, and flavoglaucin have been investigated in detail, which provide new insights into the biosynthesis of the naphthalenone unit in **41** [55–57]. The biosynthesis of some fungal α-pyrone-linked NPs such as alternapyrones and citreoviridin have been reported. The α-pyrone moiety in alternapyrones and citreoviridin is indeed formed by the spontaneous intramolecular cyclization of PKS AlpA and CtvA, respectively [58,59]. Thus, herpotrichone A might share the similar strategy to forge the α-pyrone unit by utilization of an unidentified PKS. Overall, fungal NPs with intermolecular Diels–Alder addition features usually have novel carbon skeletons. The unexpected architectures of these compounds may open an interesting topic such as the characterization of two separate BGCs crosstalk, and discovery of more fungal intermolecular Diels–Alderases.

#### *5.3. Citrifuran A*

Citrifuran A (**42**) is produced by the centipede intestine-associated *Aspergillus* sp. through solid fermentation (Figure 7), and showed moderate inhibitory activities against LPS-induced NO production in RAW 264.7 macrophages [60]. The novel skeleton of citrifuran A was constructed by coupling of azaphilone and furanone moieties via Michael addition (Figure 9). It is obvious that the two separate BGCs' crosstalk is indispensable to the biosynthesis of **42**. The furanone moiety was also contained in fungal polyketide gregatin A. For gregatin A biosynthesis, a single PKS GrgA with the aid of a *trans*-acting

enoylreductase GrgB could biosynthesize the C<sup>4</sup> and C<sup>11</sup> carbon chains. More interestingly, a predicted hydrolase GrgF is responsible for the fusion two carbon chains to produce the furanone skeleton of gregatin A [61]. This unusual chain-fusing reaction might be also suitable for the biosynthesis of furanone scaffold in **42** (Figure 9). The fungal polyketidederived mycotoxin citrinin also possesses azaphilone building block. The individual biosynthetic steps of citrinin have been studied by a combination of targeted gene knockout and heterologous gene expression in *A*. *oryzae* [62], which might provide new clues for investigation of the biosynthetic mechanism of **42**. Though the biosynthetic pathways for azaphilone and furanone have been investigated, the enzyme for catalyzation of Michael addition has not yet been identified, and needs further exploration. *J. Fungi* **2022**, *8*, x FOR PEER REVIEW 11 of 16 NPs with intermolecular Diels–Alder addition features usually have novel carbon skeletons. The unexpected architectures of these compounds may open an interesting topic such as the characterization of two separate BGCs crosstalk, and discovery of more fungal intermolecular Diels–Alderases. **Figure 8.** The proposed biosynthetic pathway of herpotrichone A (**41**). *5.3. Citrifuran A*  Citrifuran A (**42**) is produced by the centipede intestine-associated *Aspergillus* sp. through solid fermentation (Figure 7), and showed moderate inhibitory activities against LPS-induced NO production in RAW 264.7 macrophages [60]. The novel skeleton of citrifuran A was constructed by coupling of azaphilone and furanone moieties via Michael addition (Figure 9). It is obvious that the two separate BGCs' crosstalk is indispensable to the biosynthesis of **42**. The furanone moiety was also contained in fungal polyketide gre-

pathways for azaphilone and furanone have been investigated, the enzyme for catalyza-

NPs with intermolecular Diels–Alder addition features usually have novel carbon skeletons. The unexpected architectures of these compounds may open an interesting topic such as the characterization of two separate BGCs crosstalk, and discovery of more fungal

*J. Fungi* **2022**, *8*, x FOR PEER REVIEW 11 of 16

intermolecular Diels–Alderases.

ingly, a predicted hydrolase GrgF is responsible for the fusion two carbon chains to pro-**Figure 9.** The proposed biosynthetic pathway of citrifuran A (**42**). **Figure 9.** The proposed biosynthetic pathway of citrifuran A (**42**).

#### duce the furanone skeleton of gregatin A [61]. This unusual chain-fusing reaction might *5.4. Acautalide A 5.4. Acautalide A*

*5.4. Acautalide A*

be also suitable for the biosynthesis of furanone scaffold in **42** (Figure 9). The fungal polyketide-derived mycotoxin citrinin also possesses azaphilone building block. The individual biosynthetic steps of citrinin have been studied by a combination of targeted gene knockout and heterologous gene expression in *A*. *oryzae* [62], which might provide new clues for investigation of the biosynthetic mechanism of **42**. Though the biosynthetic pathways for azaphilone and furanone have been investigated, the enzyme for catalyzation of Michael addition has not yet been identified, and needs further exploration. The fungal polyketide acautalide A (**43**) is produced from the solid-state cultivation of isopod *Armadillidium vulgare*-associated *Acaulium* sp. H-JQSF on rice medium, and exhibits neuroprotective bioactivity with antiparkinsonic potential in the 1-methyl-4-phenylpyridinium-challenged nematode model [63]. The architectural features of 43 indicated that two biosynthetic precursors 10-keto-acaudiol and octadeca-9,11,13-trienoic acid intertwined together through intermolecular Diels–Alder cycloaddition (Figure 10). The 10-keto-acaudiol unit is proposed to be an early-stage precursor in the biosynthesis of The fungal polyketide acautalide A (**43**) is produced from the solid-state cultivation of isopod *Armadillidium vulgare*-associated *Acaulium* sp. H-JQSF on rice medium, and exhibits neuroprotective bioactivity with antiparkinsonic potential in the 1-methyl-4-phenylpyridinium-challenged nematode model [63]. The architectural features of 43 indicated that two biosynthetic precursors 10-keto-acaudiol and octadeca-9,11,13-trienoic acid intertwined together through intermolecular Diels–Alder cycloaddition (Figure 10). The 10-keto-acaudiol unit is proposed to be an early-stage precursor in the biosynthesis of acaulide and acaulins, two fungal macrodiolides previously isolated from the *Acaulium* sp. H-JQSF [64,65]. However, the biosynthetic mechanism of 10-keto-acaudiol remains unclear. The octadeca-9,11,13-trienoic acid motif might be derived from the biosynthesis of fungal polyunsaturated fatty acids, either biosynthesized from the fungal desaturation of octadecanoic acids in rice. Interestingly, the obtained **43** was in enantiomerically pure form, thus the Diels–Alder cycloaddition might be truly enzymatic. A fungal Diels–Alderase could be expected to catalyze the intermolecular [4 + 2]-cycloaddition in the assembly line of **43** [63].

The fungal polyketide acautalide A (**43**) is produced from the solid-state cultivation of isopod *Armadillidium vulgare*-associated *Acaulium* sp. H-JQSF on rice medium, and ex-

nylpyridinium-challenged nematode model [63]. The architectural features of 43 indicated that two biosynthetic precursors 10-keto-acaudiol and octadeca-9,11,13-trienoic acid intertwined together through intermolecular Diels–Alder cycloaddition (Figure 10). The 10-keto-acaudiol unit is proposed to be an early-stage precursor in the biosynthesis of

**Figure 9.** The proposed biosynthetic pathway of citrifuran A (**42**).

**Figure 10.** The proposed biosynthetic pathway of acautalide A (**43**). **Figure 10.** The proposed biosynthetic pathway of acautalide A (**43**).

#### **6. Discussion 6. Discussion**

bly line of **43** [63].

One reason that fungi endow great potentials to produce natural products with complex structures and excellent biological activities might be ascribed to the complex metabolic regulatory networks and interaction of different biosynthetic gene clusters. Understanding how simple precursors are synthesized and assembled together to construct complex natural products in organisms may promote the development of new combinational and synthetic biology strategies to create new molecules. One reason that fungi endow great potentials to produce natural products with complex structures and excellent biological activities might be ascribed to the complex metabolic regulatory networks and interaction of different biosynthetic gene clusters. Understanding how simple precursors are synthesized and assembled together to construct complex natural products in organisms may promote the development of new combinational and synthetic biology strategies to create new molecules.

acaulide and acaulins, two fungal macrodiolides previously isolated from the *Acaulium* sp. H-JQSF [64,65]. However, the biosynthetic mechanism of 10-keto-acaudiol remains unclear. The octadeca-9,11,13-trienoic acid motif might be derived from the biosynthesis of fungal polyunsaturated fatty acids, either biosynthesized from the fungal desaturation of octadecanoic acids in rice. Interestingly, the obtained **43** was in enantiomerically pure form, thus the Diels–Alder cycloaddition might be truly enzymatic. A fungal Diels–Alderase could be expected to catalyze the intermolecular [4 + 2]-cycloaddition in the assem-

Over the course of evolution, fungi have evolved different strategies to increase the diversity of their NPs to protect themselves and acclimatize to the surrounding ecological environment [66]. For example, a great progress has been made in the discovery and identification of fungal NPs with homo-dimeric or hetero-dimeric skeletons, which effectively expand structural diversity of NPs and accelerates the occurrence of new biological activities. For fungal homodimer NPs, the building blocks are mostly biosynthesized by one single BGC, and catalyzed by the crucial enzymes including cytochrome P450 enzymes, intermolecular Diels–Alderases, and multicopper oxidases to afford the homo-dimeric skeletons, such as the rugulosin A, bisorbicillinol, and viriditoxin biosynthesis [67–70]. On the contrary, a rare but intriguing phenomenon is that two different building blocks, usually produced by two separate BGCs, were coupled together to generate fungal heterodimers NPs. However, the underlying mechanisms including how two structural units are biosynthesized, and whether the two BGCs crosstalk process is enzymatic or spontaneous are still mysterious and need further exploration. The coupling reactions between the two different building blocks are various. The Diels–Alder reaction and Michael addition reaction have been reported to splice the separate biosynthetic precursors together [14,52]. To in-depth understand the biosynthetic process of two separate BGC crosstalk, we should characterize the biosynthetic gene function in two BGCs and acquire crucial biosynthetic precursors through gene knockout and heterologous expression. With the important intermediates in hand, we can further investigate whether this hybridization process is spontaneous or enzymatic. However, the determination of which enzymes responsible for the two separate BGC crosstalk process is sometimes challenging, because the corresponding biosynthetic genes might not be located within the two gene clusters, and distributed elsewhere in the genome of targeted strain. Moreover, these enzymes may be hypothetical proteins, and it is difficult to be identified through bioinformatic analysis. Over the course of evolution, fungi have evolved different strategies to increase the diversity of their NPs to protect themselves and acclimatize to the surrounding ecological environment [66]. For example, a great progress has been made in the discovery and identification of fungal NPs with homo-dimeric or hetero-dimeric skeletons, which effectively expand structural diversity of NPs and accelerates the occurrence of new biological activities. For fungal homodimer NPs, the building blocks are mostly biosynthesized by one single BGC, and catalyzed by the crucial enzymes including cytochrome P450 enzymes, intermolecular Diels–Alderases, and multicopper oxidases to afford the homo-dimeric skeletons, such as the rugulosin A, bisorbicillinol, and viriditoxin biosynthesis [67–70]. On the contrary, a rare but intriguing phenomenon is that two different building blocks, usually produced by two separate BGCs, were coupled together to generate fungal heterodimers NPs. However, the underlying mechanisms including how two structural units are biosynthesized, and whether the two BGCs crosstalk process is enzymatic or spontaneous are still mysterious and need further exploration. The coupling reactions between the two different building blocks are various. The Diels–Alder reaction and Michael addition reaction have been reported to splice the separate biosynthetic precursors together [14,52]. To in-depth understand the biosynthetic process of two separate BGC crosstalk, we should characterize the biosynthetic gene function in two BGCs and acquire crucial biosynthetic precursors through gene knockout and heterologous expression. With the important intermediates in hand, we can further investigate whether this hybridization process is spontaneous or enzymatic. However, the determination of which enzymes responsible for the two separate BGC crosstalk process is sometimes challenging, because the corresponding biosynthetic genes might not be located within the two gene clusters, and distributed elsewhere in the genome of targeted strain. Moreover, these enzymes may be hypothetical proteins, and it is difficult to be identified through bioinformatic analysis.

In general, more endeavors are needed to carry out the research for the discovery of fungal heterodimer NPs constructed by two building blocks, which not only provides an In general, more endeavors are needed to carry out the research for the discovery of fungal heterodimer NPs constructed by two building blocks, which not only provides an outstanding opportunity for investigation of the currently underestimated hidden biosynthetic crosstalk, but also facilitates the discovery of new BGCs, new regulatory mechanisms, and enzyme catalysts with novel catalytic mechanisms.

**Author Contributions:** Conceptualization, G.D. and X.B.; writing—original draft preparation, G.D.; writing—review and editing, G.D. and X.B.; discussion of the contents, G.D.; Q.S.; Y.Z., and X.B.; funding acquisition, G.D. and X.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by National Key R&D Program of China (2019YFA0905700), National Natural Science Foundation of China (32100040, 32070060), the Shandong Provincial Natural Science Foundation, China (ZR2019JQ11), and Qingdao Postdoctoral Application Research Project (62450070311109).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **Abbreviations**


### **References**

