*2.1. Synthesis of the Side-Chain Derivatives of Eurotiumide A*

Our synthetic plan is shown in Figure 2. We planned to introduce three types of functional groups: a hydrocarbon group, including hydrogen, alkyl, and aromatic rings (Type A); a heteroatom and heteroatom-containing alkyl group (Type B); and halogen atoms group (Type C). The derivatives of groups A and B could be derived from **2** by the cross-coupling reaction and functional group transformation. The halogenated derivatives (Type C) would be obtained from **3** by direct introduction of the halogen atoms. Although Wang et al. isolated the natural eurotiumide A (**1**) as a racemic form, they evaluated the antimicrobial activities of its enantiomers after separation by chiral HPLC and revealed that there was no significant difference between the enantiomers [23]. From the viewpoint of the efficiency of compound supply, we decided to make racemic compounds.

**Figure 2.** Synthetic plan of the side-chain derivatives of eurotiumide A (**1**).

First, we initiated the syntheses of the derivatives of group A (Scheme 1). The non-substituted derivative **4** was obtained from **3** by deprotection of the diMOM group with aqueous 6 M HCl in methanol at 40 ◦C in 79% yield. Catalytic hydrogenation of eurotiumide A (**1**) gave the isopentyl derivative **6** in quantitative yield. Methyl and vinyl groups were introduced by the Stille coupling reaction with **2** to afford methyl derivative **5a** and styrene derivative **7a** in 83% and quantitative yields, respectively. Phenyl derivative **9a** and biphenyl derivative **10a** were obtained from **2** by the Suzuki–Miyaura cross coupling reaction with the corresponding boronic acids in 75% and 77% yields, respectively. Deprotection of the diMOM group of derivatives **5a**, **7a**, **9a**, and **10a** then gave the corresponding desired products (**5**, **7**, **9**, and **10**). We tried to introduce the alkyne group by the Sonogashira coupling reaction; however, the desired alkyne product was obtained in only 12% yield. To improve the reaction yield, the Seyferth–Gilbert homologation using the Ohira–Bestmann reagent **21** was applied to the aldehyde derivative **12a** (vide infra) and afforded the desired alkyne **8a** in quantitative yield. After acidic treatment of **8a**, the alkyne derivative **8** was obtained in 68% yield.

With type A derivatives in hand, we turned our attention to preparing type B derivatives having heteroatom-containing side chains (Scheme 2). For the introduction of an alkyl group containing heteroatoms, we chose the styrene derivative **7a** as a starting point. Ozonolysis of the alkene moiety of **7a** afforded the diMOM-protected benzaldehyde **12a** in excellent yield. Acidic treatment of **12a** gave the desired deprotected benzaldehyde derivative **12** in 77%. On the other hand, reduction of the aldehyde moiety of **12a** with sodium borohydride to give the benzyl alcohol **11a** and the deprotection furnished the hydroxymethyl derivative **11** in moderate yield. To introduce a nitrogen group at the benzyl position of **11a**, the primary alcohol moiety was converted to a mesyl group (**22**) and a nucleophilic substitution reaction with sodium azide afforded diMOM-protected azide **13a** in good yield. Derivative **13a** was treated with aqueous 6 M HCl in MeOH to furnish the desired dihydroxy azide derivative **13**. We then tried to convert the azide into an amine functionality. After several attempts, we found that addition of triethylamine was crucial to keep the reaction clean and we succeeded to get **14a**. Then, deprotection of the diMOM group gave the desired aminomethyl derivative **14**.

**D**

**Scheme 1.** Synthesis of the hydrocarbon derivatives (type A).

**Scheme 2.** Synthesis of the derivatives having heteroatom-containing side chains (type B).

Next, a nitration reaction was conducted with non-substituted derivative **3** by adding HNO3 in AcOH to afford monoMOM-protected nitro derivative **15a** as a crude product; then it was deprotected under acidic condition to give the nitro derivative **15** (Scheme 3). After that, hydrogenation with Adam's catalyst produced the aniline derivative **16** from **15**.

**Scheme 3.** Synthesis of nitro and aniline derivatives.

Finally, we tried to synthesize the halogenated derivatives (Scheme 4). Chloro and iodo groups were introduced to treat **3** with *N*-chlorosuccinimide and *N*-iodosuccinimide in DMF to afford the chloro derivative **18a** and the iodo derivative **20a**, respectively. The diMOM groups of **18a** and **20a** were then deprotected under acidic conditions to afford the desired **18** and **20**. Bromo derivative **19** was obtained from **2** in 97% yield by acid treatment to cleave the diMOM group. However, despite several efforts to introduce fluorine to the aromatic ring from **3**, we could not get the desired fluoro derivative **17**. We also tried the Sandmeyer reaction with **16** but did not obtain the desired **17**.

**Scheme 4.** Synthesis of halogenated derivatives (type C).

#### *2.2. Antimicrobial Evaluation of Synthesized Derivatives*

After the initially set derivatives of eurotiumide A were synthesized, the first antimicrobial activity screening was conducted against the Gram-positive MSSA and MRSA as well as the Gram-negative *P. gingivalis* in 10 μM solutions of the synthesized derivatives to narrow down the promising antimicrobial candidates. The results are depicted in Figure 3. (+/−)-Eurotiumide A (**1**) exhibited mild antimicrobial activity against MSSA at this concentration (Figure 3a). While most of the derivatives did not show antimicrobial activity against this strain, the isopentyl derivative **6** and the iodo derivative **20** exhibited more potent antimicrobial activity than **1**. Next, we tested the same screening against MRSA (Figure 3b). Most of the derivatives that displayed good activity against MSSA showed no antimicrobial activity against MRSA. Even natural product **1** and the iodo derivative **20** also did not show good antimicrobial activity against MRSA. Surprisingly, only the isopentyl derivative **6**, which was a reduced derivative of **1**, was found to have good antimicrobial activity against MRSA. We also conducted antimicrobial screening against *P. gingivalis* (Figure 3c). Unlike the case with *S. aureus*, many derivatives, specifically eurotiumide A (**1**), isopentyl derivative **6**, vinyl derivative **7**, aniline derivative **16**, and three halogenated derivatives (**18**, **19**, **20**), were effective against *P. gingivalis*.

**Figure 3.** Initial screening of antimicrobial activity against (**a**) methicillin-susceptible *S. aureus*, (**b**) methicillin-resistant *S. aureus*, and (**c**) *P. gingivalis*. The terminal concentration was 10 μM.

Since we acquired promising agents against all three strains, we determined the IC50 values of these candidates (Table 1). The IC50 values of the isopentyl derivative **6** and the iodo derivative **20** against MSSA were 5.6 μM (2.0 μg/mL) and 9.0 μM (3.7 μg/mL), respectively. Moreover, the IC50 value of **6** against MRSA was 4.3 μM (1.5 μg/mL), which is the same level of activity against MSSA. The IC50 values of these seven candidates (**1**, **6**, **7**, **16**, **18**, **19**, and **20**) against *P. gingivalis* ranged from 2.0 to 7.0 μM. We also checked the cytotoxicity of three compounds (**1**, **6**, and **20**) against the A549 cell line, and these three compounds were non-toxic in 10 μM.

**Table 1.** The IC50 values (μM) of the selected side chain derivatives against methicillin-susceptible *S. aureus* (MSSA), methicillin-resistant *S. aureus* (MRSA), and *P. gingivalis*. Vancomycin (VCM) was used as a positive control against MSSA and MRSA. Cefcapene pivoxyl (CFPN-PI) was used as a positive control against *P. gingivalis*.


In this study, we discovered that the isopentyl derivative **6**, which is a one-point modified compound of natural product **1**, and the iodo derivative **20** have superior antimicrobial activity to **1** against MSSA and *P. gingivalis*. Although **20** did not exhibit good efficacy against MRSA, **6** was found to maintain antimicrobial activity against these three strains, including MRSA. These results indicate that *S. aureus* is sensitive to changes in the side chain of the aromatic ring and that MRSA can distinguish the subtle difference between prenyl and isopentyl moieties. Moreover, the weak antimicrobial activity of **1** against MRSA suggests a binding affinity between **1** and the penicillin binding protein 2' [26], which is the main resistance mechanism of MRSA against antibiotics. The inhibition of cell wall synthesis seems to be the mode of action of **1**, although a more detailed study is needed to clarify the mode of action of **6** and **20**. On the other hand, we found that several compounds having alkyl and halogenated side chains well suppressed the increase in *P. gingivalis*.
