Biotransformation of Ursonic Acid by Aspergillus ochraceus and Aspergillus oryzae to Discover Anti-Neuroinflammatory Derivatives

Biotransformation of ursonic acid (1) by two fungal strains Aspergillus ochraceus CGMCC 3.5324 and Aspergillus oryzae CGMCC 3.407 yielded thirteen new compounds (4, 5, 7–10, and 13–19), along with five recognized ones. The structural details of new compounds were determined through spectroscopic examination (NMR, IR, and HR-MS) and X-ray crystallography. Various modifications, including hydroxylation, epoxidation, lactonization, oxygen introduction, and transmethylation, were identified on the ursane core. Additionally, the anti-neuroinflammatory efficacy of these derivatives was assessed on BV-2 cells affected by lipopolysaccharides. It was observed that certain methoxylated and epoxylated derivatives (10, 16, and 19) showcased enhanced suppressive capabilities, boasting IC50 values of 8.2, 6.9, and 5.3 μM. Such ursonic acid derivatives might emerge as potential primary molecules in addressing neurodegenerative diseases.


Introduction
Neurodegenerative diseases (NDDs), including Alzheimer's disease, Parkinson's disease, Huntington's disease, and multiple sclerosis, are an important global healthy problem due to an increase in the aging population [1].It brings a huge burden to patients and social sanitary systems all over the world.However, the pathologies of these diseases are not fully understood and some factors, such as genetic factors, oxidative stress, neuroinflammation, and environmental factors, are believed to have played a role in the development of NDDs.In clinics, there is no effective cure for these diseases [2,3].Therefore, it is a very challenging task to find innovative potential drugs for NDDs.
Ursonic acid (UNA, 1), an ursane-type compound with a five-ring triterpene structure, is commonly found in a variety of plants frequently used in traditional remedies [4][5][6].It exhibits an array of biological properties, encompassing anti-inflammatory, anti-cancer, growth-inhibitory, and anti-protozoan effects [7][8][9].Notably, being a primary oxo-derivative of ursolic acid (ULA), UNA is a crucial chemical precursor for developing potential drug candidates.However, its medicinal potential and the mechanisms driving its effects remain relatively underexplored [10].
Various semisynthetic derivatives of UNA have been chemically crafted, showcasing enhanced absorption in the digestive tract and amplified medicinal properties [11][12][13].Yet, these chemical methods have predominantly targeted only the activated substituents at C-3 and C-28 of the molecular framework.The potential for a wider range of structural variations of UNA is restrained due to its dearth of functional units.However, this limitation is difficult to overcome by conventional chemical synthesis.
Utilizing biotransformation emerges as a strategic method to attain structural variation, especially when dealing with intricate natural compounds [14][15][16][17].The significant improvements in instruments and experimental techniques have enabled the biotransformation process to be carried out in NMR tubes and in situ monitoring using NMR spectroscopy [18,19].Biotransformation offers a solution for targeting specific molecular sites that traditional chemical procedures may find challenging.In particular, microbial transformation stands out for its ability to provide precise and location-specific alterations in the structure of triterpenoids, thanks to its inherent stereo-and region-selective catalytic potential [20][21][22][23].Even though numerous microbial adaptations of ULA have been explored to yield derivatives with augmented solubility and therapeutic traits, such endeavors with UNA are not as prevalent [24].Hence, leveraging biotransformation to craft UNA derivatives is of immense significance.
Anti-neuroinflammatory effects of triterpenoids from medicinal plants are well reported [25].Nitric oxide (NO) is a molecule which is highly linked with immunity and inflammation [26].The anti-neuroinflammatory activity of some ULA derivatives has been evaluated with lipopolysaccharide (LPS)-induced BV-2 microglia [27].As ongoing research to find triterpenoid derivatives with anti-neuroinflammatory activity, in this study, we identified 13 previously undescribed derivatives of UNA derived from the biotransformation processes of two fungal varieties: Aspergillus ochraceus CGMCC 3.5324 and Aspergillus oryzae CGMCC 3.407.Additionally, we evaluated the anti-neuroinflammatory properties of these biotransformation derivatives to inhibit LPS-induced NO production in BV-2 cells.
The molecular structure of compound 19 was deduced as C 30 H 42 O 5 , as evidenced by the HR-ESI-MS displaying an [M + COOH] − ion at m/z 527.3017 (calcd.for C 31 H 43 O 7 , 527.3017).Three additional low-field proton signals were spotted at δ H 3.09, 3.30, and 3.42 in the 1 H NMR spectrum.Furthermore, in the 13

Anti-Neuroinflammatory Activities
To assess the potential capabilities of all modified products in counteracting neuroinflammation, we measured their suppressive effects on NO generation within LPS-triggered BV-2 cells using the Griess method.Table 5

Discussion
Earlier research on triterpenes reveals that microbial modifications possess a heightened catalytic propensity, resulting in a variety of hydroxylated and carbonylated byproducts [33]. A. ochraceus is prevalently found in the environment, commonly in soil and decaying plant matter.Historically, A. ochraceus has been employed as a biocatalyst in the hydroxylation processes of steroids, triterpenes, flavonoids, and coumarins [34][35][36][37].In our present investigation, we discerned that A. ochraceus primarily initiated hydroxylation, oxidation, lactonization, and epoxidation reactions on UNA.
The position and arrangement of hydroxyl and epoxyl groups on the UNA structure can influence their capacity to inhibit NO activity.Compound 2, which had a hydroxyl group at C-21β, demonstrated stronger inhibitory effects on NO generation compared to UNA itself.This finding indicated that introducing a hydroxyl group at C-21β could amplify the compound's inhibitory effect on NO production.On the contrary, compound 4, containing a carbonyl group at C-21, presented a notably reduced inhibition compared to compound 2, indicating the detrimental effect of carbonylation at C-21.Moreover, compounds 3 and 5, which had a hydroxyl group at C-7β, presented more potent inhibitory effects than that of compounds 2 and 4, respectively.These indicated that hydroxylation at C-7β could enhance the NO inhibitory activity.In a parallel fashion, the inhibitory impact on NO production by compounds 13-19, which possessed an epoxyl group at C-11β and C-12β, surpassed that of compounds 8 and 9.This finding pointed to the conclusion that epoxidation at C-11β and C-12β could significantly augment inhibitory effects on NO production.Meanwhile, compounds 8 and 9, carrying a lactone group at C-13β and C-28, did not exhibit any inhibitory activities, insinuating that having a lactone group at these positions could be detrimental to inhibiting NO production.Compounds 10, 16, and 19 showcased the strongest inhibitory potential, with IC 50 values of 8.23, 6.86, and 5.25 µM, respectively (Figure 6).Such results underlined the promise of these compounds as primary candidates for addressing neuronal injuries.
to compound 2, indicating the detrimental effect of carbonylation at C-21.Moreover, compounds 3 and 5, which had a hydroxyl group at C-7β, presented more potent inhibitory effects than that of compounds 2 and 4, respectively.These indicated that hydroxylation at C-7β could enhance the NO inhibitory activity.In a parallel fashion, the inhibitory impact on NO production by compounds 13-19, which possessed an epoxyl group at C-11β and C-12β, surpassed that of compounds 8 and 9.This finding pointed to the conclusion that epoxidation at C-11β and C-12β could significantly augment inhibitory effects on NO production.Meanwhile, compounds 8 and 9, carrying a lactone group at C-13β and C-28, did not exhibit any inhibitory activities, insinuating that having a lactone group at these positions could be detrimental to inhibiting NO production.Compounds 10, 16, and 19 showcased the strongest inhibitory potential, with IC50 values of 8.23, 6.86, and 5.25 μM, respectively (Figure 6).Such results underlined the promise of these compounds as primary candidates for addressing neuronal injuries.The biotransformation of UNA by A. ochraceus CGMCC 3.5324 and A. oryzae CGMCC 3.407 produced 18 derivatives, of which 13 were novel compounds (4, 5, 7-10, and 13-19).The principal reaction types observed were region-selective hydroxylation, epoxidation, lactonization, carbonylation, and transmethylation.Notably, A. ochraceus demonstrated the capability to concurrently catalyze the epoxidation at C-11 (12) and the lactonization at C-13(28).Additionally, the epoxidation and lactonization reactions were stereo-selective at C-11β, C-12β, and C-13β positions.Achieving such specific reactions through conventional chemical synthesis is challenging.On the other hand, A. oryzae facilitated hydroxylation, oxidation, and lactonization reactions and uniquely catalyzed the methoxylation reaction, resulting in two distinct products.Some of these biotransformed derivatives ex- The biotransformation of UNA by A. ochraceus CGMCC 3.5324 and A. oryzae CGMCC 3.407 produced 18 derivatives, of which 13 were novel compounds (4, 5, 7-10, and 13-19).The principal reaction types observed were region-selective hydroxylation, epoxidation, lactonization, carbonylation, and transmethylation.Notably, A. ochraceus demonstrated the capability to concurrently catalyze the epoxidation at C-11 (12) and the lactonization at C-13(28).Additionally, the epoxidation and lactonization reactions were stereo-selective at C-11β, C-12β, and C-13β positions.Achieving such specific reactions through conventional chemical synthesis is challenging.On the other hand, A. oryzae facilitated hydroxylation, oxidation, and lactonization reactions and uniquely catalyzed the methoxylation reaction, resulting in two distinct products.Some of these biotransformed derivatives exhibited significant inhibitory effects on NO production, positioning them as potential anti-neuroinflammatory agents.This research underscored the potential of biotransformation for the structural diversification of UNA, enabling the discovery of valuable derivatives.With the distinct biocatalytic capabilities of the fungi studied, a combination of microbial transformation and chemical semi-synthesis could be leveraged to produce an even broader array of UNA derivatives and analogs.

Microorganism and Substance
UNA (1) was procured from Push Bio-technology Co., Ltd., Chengdu, China.Its authenticity was confirmed by comparing its physical and spectroscopic data with previously reported values.The purity was verified to be above 98% through UV-HPLC analysis.All solvents used were of AR grade, sourced from Sinopharm Chemical Reagent Co., Ltd., Shanghai, China.The strains A. ochraceus CGMCC 3.5324 and A. oryzae CGMCC 3.407 were acquired from the China General Microbiological Culture Collection Center (CGMCC).
They were maintained on potato slants solidified with agar and stored at 4 • C. The BV-2 cell line was sourced from the Cell Bank of the Chinese Academy of Sciences.

Biotransformation Procedures
Biotransformation experiments were conducted in 1000 mL flasks, each containing 400 mL of liquid potato medium.These flasks were incubated on a rotary shaker at 26 • C with a shaking speed of 160 rpm.After a 24 h pre-culture period, 20 mg of the substrate dissolved in 2 mL of ethanol was added to each flask.Fermentation then proceeded for 7 days.In total, 2.0 g of UNA was used for A. ochraceus and 1.2 g for A. oryzae.After the 7-day incubation, the cultures from each flask were combined and filtered.The resulting filtrates underwent extraction with ethyl acetate (EtOAc) three times.The organic layers were then gathered, and after solvent evaporation, residues weighing 3.4 g and 2.3 g were obtained for A. ochraceus and A. oryzae, respectively.

Anti-Neuroinflammatory Activities
NO production was assessed indirectly using the Griess reaction by measuring nitrite concentration in a culture medium.BV-2 cells were cultured in Dulbecco's Modified Eagle Medium, supplemented with 10% FBS, 100 units/mL of penicillin, and 100 µg/mL of streptomycin.These cells were seeded in 96-well plates at a density of 8 × 10 4 cells/well and incubated for 24 h.Post incubation, they were exposed to 100 ng/mL of LPS alongside different compound concentrations for 48 h.A mixture of 100 µL culture supernatant and Griess reagent was left at room temperature for 10 min, with absorbance later read at 570 nm.L-NMMA served as the positive control.Cell viability was determined using the MTT assay, and experiments were conducted in triplicate.The IC 50 for NO production inhibition was computed using GraphPad Prism 7.00 software.

Figure 2 .
Figure 2. Key 1 H-1 H COSY, HMBC, and NOESY correlations for compounds 4, 5, and 7-10.For compound 5, its HR-ESI-MS data revealed a molecular formula of C 30 H 44 O 5 , as reflected by the [M − H] − ion at m/z 483.3126 (calcd.for C 30 H 43 O 5 , 483.3112).This was 30 amu heavier than UNA.The 1 H NMR spectrum showed a new signal at δ H 3.89, indicating the presence of a hydroxyl group.The 13 C NMR spectrum distinguished itself with an oxygenated methine at δ C 73.4 and a carbonyl signal at δ C 210.4 ppm.HMBC correlations between 26-CH 3 (δ H 0.81) and the oxygenated methine at δ C 73.4 suggested the attachment of the hydroxyl group to C-7 (Figure 2).The NOESY data further linked H-7 (δ H 3.87) and 27-CH 3 (δ H 1.08), affirming the β-orientation of the 7-OH group.The HMBC spectrum showcased H-22 (δ H 2.56 and 2.37) correlations with C-21 (δ C 210.4).The linkage of 30-CH 3 (δ H 0.99) with the carbonyl signal at δ C 210.4 confirmed the carbonyl's placement at C-21.Hence, compound 5 was determined to be 3,21-dioxo-7β-hydroxy-urs-12-en-28-oic acid.Compound 7, as determined by its HR-ESI-MS data, had a molecular formula of C 31 H 48 O 5 .This was supported by the [M − H] − ion at m/z 499.3433 (calcd.for C 31 H 47 O 5 , 499.3423), indicating that it was 46 amu heavier than UNA.In the 1 H NMR spectrum, two prominent signals at δ H 3.84 and 3.44 emerged.Additionally, the 13 C NMR and DEPT 135 spectra displayed new methine carbon signals at δ C 76.5 and 71.1.The oxygenated

Compound 9 '
s molecular formula was inferred to be C 30 H 44 O 5 from its HR-ESI-MS results ([M + COOH] − m/z 529.3171, calcd.for C 31 H 45 O 7 529.3165), a 16 amu increment from metabolite 8, pointing to an extra oxygen atom.In the 1 H NMR reading, two more low-field protons, δ H 3.87 and 3.39, were present.The 13 C NMR, DEPT 135, and HSQC results displayed two fresh oxygenated methine signals at δ C 72.6 and 71.7.Contrasted with 8's NMR findings, the 13 C NMR readings were largely aligned, excluding the B ring.HMBC correlations between the distinct 26-CH 3 (δ H 1.07) and the new oxygenated methine signal at δ C 72.6 were evident (Figure 2).Furthermore, NOESY correlations between H-7 (δ H 3.87) and 27-CH 3 (δ H 1.15) confirmed the β-orientation of the 7-OH group.Thus, compound 9 was identified as 3-oxo-7β,21β-dihydroxy-urs-11-en-13β,28β-lactone.For compound 10, its molecular formula was concluded to be C 31 H 48 O 3 based on HR-ESI-MS, showing a [M − H] − ion at m/z 467.3534 (calcd.for C 31 H 47 O 3 , 467.3525), a 14 amu increase from UNA.The 1 H NMR reading for compound 10 displayed an extra vinyl proton at δ H 4.39 and an oxygenated methine at δ H 3.47.Their respective carbon signals at δ C 89.6 and δ C 54.2 appeared in the HSQC reading.The 13 C NMR and DEPT 135 findings disclosed a new seasonal double-bond carbon signal at δ C 160.6, and when compared to UNA's NMR readings, metabolite 10's keto carbonyl signal at C-3 vanished.The carbon signal at δ C 89.6 and the new seasonal double-bond carbon signal at δ C 160.6 were likely paired due to the HMBC's vinyl proton (δ H 4.39) connections with the seasonal double-bond carbon signal at δ C 160.6 (Figure 2).Moreover, the HMBC connections between 23-CH 3 (δ H 1.04), 24-CH 3 (δ H 0.93), and the new seasonal double-bond carbon signal (δ C 160.6) hinted at the double bond's positioning at C-2 and C-3.Additionally, the oxygenated methine group should connect to C-3 based on the HMBC correlation of C-3 (δ C 160.6) with the oxygenated methine signal at δ H 3.47.The 2D structure of compound 10 was further endorsed by suitable crystal X-ray crystallography [Cu Ka; Flack parameter: −0.4(5); CCDC: 2266165] (Figure

Figure 6 .
Figure 6.Preliminary structure-activity relationship of biotransformation products.

Figure 6 .
Figure 6.Preliminary structure-activity relationship of biotransformation products.

δ C δ H (J in Hz) δ C δ H (J in Hz) δ C δ H (J in Hz) 18
H 3.39 in the 1 H NMR reading for 8, with the corresponding carbon reading at δ C 71.8 evident in the HSQC reading.The HMBC revealed clear connections between the distinct 30-CH 3 (δ H 1.01) and the new oxygenated methine reading at δ C 71.8 (Figure2).NOESY correlations between H-21 (δ H 3.39) and H-19 (δ H 1.82) confirmed the β-orientation of the 21-OH group.In the 1 H NMR reading, two vinyl protons appeared at δ H 5.92 and 5.51, linking with two sp 2 methine readings at δ C 133.3 and 128.8 in the HSQC reading.
Figure 4. Key 1 H-1 H COSY, HMBC, and NOESY correlations for compounds 13-18 and 17-19.a H and C were measured at 600 and 150 MHz, respectively, in Pyridine-d5.b H and C were measured in CD3OD.
a H and C were measured at 400 and 100 MHz, respectively, in CDCl 3 .

Table 5 .
Inhibitory effects of transformed products on NO production induced by LPS in BV-2 cells (mean ± SD, n = 3).
a L-NMMA as a positive control.