Extraction of Solid Fermentation

All of the quinoa cultivating with T. brevicompactum NTU439 was soaked in the

methanol (MeOH) for one day. MeOH was evaporated under reducing pressure until it

dried out. The dry mixtures were then re-dissolved with ddH2O, followed by

partitioning with EtOAc. Lastly, we obtained the extracts from solid fermentation.

3.4 Chromatography

3.4.1 Column Chromatography

Sephadex LH-20 (Amersham Biosciences, Filial Sverige, Sweden) open column with

MeOH as the mobile phase. High performance liquid chromatography (HPLC) was

performed by Hitachi L-7000 (Tokyo, Japan) with UV detector (Hitachi L-7400, Tokyo,

Japan) or RI detector (Bischoff, Leonberg, Germany). Reversed-phase columns

including Phenomenex Luna PFP (5µm particles, 10mm×250mm, Torrance, CA, USA)

and Supelco Discovery C18 (5µm particles, 10mm×250mm, Bellefonte, PA, USA)

were able to purify complex mixtures with either MeOH (L.C., Avantor Performance

Materials Inc., PA, USA) or ACN (L.C., Avantor Performance Materials Inc., PA, USA)

as the mobile phase. In some cases, solvents were mixed with either 0.1% formic acid

(FA) or 0.03% trifluoroacetic acid (TFA) to increase the chromatographic resolution.

3.4.2 Thin Layer Chromatography

TLC (Silica gel 60 F254, 0.2mm, Merck, Darmstadt, Germany) and appropriate

condition of mobile phase (MeOH/DCM) were used to analyze the complex mixtures.

After separating, the spots were visualized via UV at 254nm and chromogenic reagent

(0.5g Vanillin in 250mL H2SO4/MeOH 1:4 v/v).

Purified compounds dissolved in the soluble deuterium solvents, including

CD3OD (Sigma-Aldrich, St. Louis, MO, USA) or DMSO-d6 (Sigma-Aldrich, St. Louis,

MO, USA), were loaded into 5 mm diameter NMR tubes. All of the NMR spectra were

acquired by Bruker AVIII-500MHz (Bruker, Ettlingen, Germany) in the Instrument

Center, National Taiwan University.

3.5.2 Mass Spectrometry

Each of the compounds was diluted to 5 ppm by MeOH prior to the acquisition of

mass data. High-resolution mass spectra were acquired by Q Exactive™ Hybrid

Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Fisher Scientific, Bremen,

Germany) in the Instrument Center, National Taiwan University. Tandem mass spectra

were acquired by LTQ Orbitrap XL™ Hybrid Ion Trap-Obritrap Mass Spectrometer

(Thermo Fisher Scientific, Bremen, Germany) in Genomics Research Center,

Academia Sinica.

3.5.3 Fourier Transform Infrared Spectroscopy

IR absorption spectra were measured by FT/IR 4100 Spectrometer (JASCO,

Tokyo, Japan).

Optical rotation angles were measured by P-2000 Digital Polarimeter (JASCO,

Tokyo, Japan).

3.6 In Vitro Bioactivity Test 3.6.1 Disk Diffusion Assays

Disk diffusion assays were used to determine which fractions contained the

bioactive substances. Forty microliter of collected fractions were added to each of the

8 mm discs without being quantified. Each of the discs was then lightly placed and

pressed down on the agar plates (1% yeast extract, 2% peptone, 2% dextrose and 2%

agar), which were mixed with target pathogens. After incubating overnight at 37°C, the

plates showed obvious inhibitory zones indicateing the antifungal activity. In these

experiments, Candidia albicans and Cryptococcus neoformans were chosen as target

pathogens. Ketoconazole was used to be a positive control.

3.6.2 Minimum Inhibitory Concentrations

The experiments of minimum inhibitory concentrations were cooperated with Dr.

Ying-Lien Chen, a professor of Department of Plant Pathology and Microbiology at

on YPD (1% yeast extract, 2% peptone and 2% glucose), defined medium 199

(Invitrogen), and yeast carbon base-bovine serum albumin. After incubating the

pathogens at 37°C for 24 hours, the concentration of pathogens was adjusted to 0.5

McFarland Standard for the following experiment. Purified compounds underwent

microdilution to 0.25-128 µg/mL with Mueller-Hinton broth in each well of 96-well

plate. The pathogens were then respectively added to each well. Mueller-Hinton broth

and pathogens with water were used as a blank and positive control. After 24 hours,

MIC values could be read by O.D. 600 nm.

3.6.3 Sulforhodamine B Assays

The experiments of Sulforhodamine B (SRB) assays were cooperated with Dr.

Shin-Wei Wang, a professor of Department of Medicine at Mackay Medical College.

Hepatocellular carcinoma cells (SK-Hep-1) were cultivated on RPMI-1640

medium containing 10% fetal bovine serum (FBS) and incubated at 37°C with 5% CO2.

The carcinoma cells were treated with purified compounds for 48 hrs, followed by

adding 0.04% (w/v) SRB solution containing 1% acetic acid. The values could be read

by ELISA reader.

4. RESULTS

4.1 Optimal Fermentation Condition for Trichoderma brevicompactum

T. brevicompactum (Figure 1) isolated from red algae was tested under different

cultivation conditions. Seven types of culture media, including MY, PDY, YES, MDS,

MEA, PDB, PDB+salt, and six types of cereals, including black sesame, oat, Job's tears,

wheat, brown rice and quiano, were supplied as the sources of carbon and nitrogen.

After cultivation for one week, each of the crude extracts from different conditions was

tested for its own bioactivities and 100 ppm ketoconazole was used as a positive control.

Based on the observations made in the sized of inhibition zones (Figure 2), MY (malt

and yeast extracts) and quiano were chosen for further large-scale liquid and solid

fermentations.

After mass cultivation (Figure 3), the crude extracts were first partitioned with

ethyl acetate (EA) to remove highly polar substances. Subsequently, the crude mixtures

were separated and purified by successive size exclusion chromatography (Sephadex

LH-20) and HPLC (reversed phase PFP) to obtain pure compounds.

Figure 1. Morphology of T. brevicompactum cultivated at PDA medium

Figure 2. Antimicrobial assay against C. albicans in different cultivation conditions of T. brevicompactum

(a) (b)

Figure 3. Mass cultivation of T. brevicompactum by using (a) MY for liquid

4.2 Flow Chart for Separation Process of Trichoderma brevicompactum 4.2.1 Flow Chart of Compound Purification from T. brevicompactum by Liquid Fermentation

Table 32. Antimicrobial activities of liquid-fermentation fractions

Fr. I Fr. II Fr. III Fr. IV Fr. V Fr. VI Fr. VII C. albicans – + + + + + + + + + + + + + + – Inhibition zone diameter: +++ 16-20 mm, ++ 10-15 mm, – no activity.

4.2.2 Flow Chart of Compound Purification from T. brevicompactum by Solid Fermentation

Table 33. Antimicrobial activities of solid-fermentation fractions

Fr. I Fr. II Fr. III Fr. IV Fr. V Fr. VI Fr. VII C. albicans – + + + + + + + + + + –

Inhibition zone diameter: +++ 16-20 mm, ++ 10-15 mm, – no activity.

4.3 Structure Elucidation of Compound 1

Compound 1 was obtained as a yellow-oily liquid. (+)-HR-ESI-MS (Figure 4)

showed a peak at m/z 309.1696 [M-H2O+H]⁺ , which was consistent with a molecular

formula of C17H25O5+ (calculated 309.1697). FTIR spectrum (Figure 5) showed

absorption bands at 1731 (cm^-1) and 3438 (cm^-1), which was an evidence for the

presence of ester carbonyl and hydroxyl group, respectively.

The ¹H NMR spectrum (Figure 6) revealed compound 1 had four methyl groups

at δH 0.90 (3H, s, H3-14), δH 0.94 (3H, s, H3-15), δH 1.76 (3H, s, H3-16) and 2.05 (3H,

s, H3-17). The several downfield signals at δH 3.81 (H, d, J=13.0 Hz, H-13)/ δH 3.94 (H,

d, J=13.0 Hz, H-13), δH 4.26 (H, d, J=5.2 Hz, H-2) and δH 5.28 (H, dd, J=7.8, 2.1 Hz,

H-4) were predicted to be connected with heteroatom. Due to the almost orthogonal

dihedral angle of the H-2 and the H!-3, the magnitude of the coupling constant was

O

in the form of doublet of doublet splitting pattern.

The ¹³C NMR (Figure 7) and HSQC (Figure 9) spectra revealed the presence of

17 carbon atoms, including four methylene carbons and six quaternary carbons. δC

172.1 (C-18) indicated a carbonyl carbon, δC 108.9 (C-10) and δC 117.6 (C-9) were

assigned as alkenyl carbons; δC 59.0 (C-13), δC 79.3 (C-4), δC 80.4 (C-2), δC 97.3

(C-12), and δC 108.9 (C-11) were speculated to be connected with oxygen atoms.

The ¹H-¹H COSY (Figure 8) revealed correlations of H-2 to H2-3, H2-3 to H-4 and

H2-7 to H2-8. Several deciding cross-peaks of H3-17 to C-18, H-4 to C-18, H2-13 to

C-12, H3-14 to C-5, 12, H3-15 to C-6, 7, 11 and H3-16 to C-8, 10 determined the overall

O C

H

H H

C 3 12 4

α β

90°

O

O OH O

H OH

OH

The ROESY spectrum (Figure 11) indicated the correlation of H-4 to H3-15; H2

-13 to H-2, H3-14, and thus the relative configurations were determined as (4R*,6R*)

and (2R*, 5S*, 12S*). By comparing with trichothecene-type skeleton, absolute

configurations were confirmed as mentioned above. The signal of the hydroxyl group

at C-11 was observed, when DMSO-d6 was used instead of CD3OD. Because the

hydroxyl group had the correlation to H-2, H2-13 and H-10 in ROESY spectrum, the

configuration of C-11 should be R form to make the hydrogens on the C-2, C-13 and

C-10 closer in spatial conformation.

By searching the SciFinder database, compound 1, named trichobrevidin A, was

identified to be a new trichothecene-type compound.

O

CH ₃ O OH O

H ₂ C H OH

OH H H ₃ C

H

Table 34. ¹H and ¹³C NMR (CD3OD, 500 and 125 MHz) for compound 1 Position δC, type δH (J in Hz) COSY HMBC 2 80.4, CH 4.26 (H, d, J=5.2) H-3 C-3, 4, 5 3 39.1, CH2 2.53 (H!, dd, J=16.6,

7.8)

H-2, 4 C-2, 4, 12

1.81-2.02 (H", ddd, J=

16.6, 5.2, 2.1)

4 79.3, CH 5.28 (H, dd, J=7.8, 2.1) H-3 C-2, 12, 18

5 54.3, C

6 49.9, C

7 31.5, CH2 1.46 (H, m) H-8 C-6, 8, 9, 11, 15 1.81-2.02 (H, m)

8 29.8, CH2 1.81-2.02 (H, m) H-7 C-6, 9, 10 2.14-2.21 (H, m)

9 146.2, C

10 117.6, CH 5.36 (H, m) H-16 C-8, 16

11 108.9, C 12 97.3, C

13 59.0, CH2 3.81 (H, d, J=13.0) C-2, 12 3.94 (H, d, J=13.0)

14 10.4, CH3 0.90 (3H, s) C-4, 5, 6, 7, 12

15 14.6, CH3 0.94 (3H, s) C-4, 5, 6, 7, 11

16 23.0, CH3 1.76 (3H, s) H-10 C-8, 9, 10

17 21.0, CH3 2.05 (3H, s) C-18

18 172.1, C=O

Figure 4. Mass spectrometry of compound 1

[M-H2O+H]⁺

Figure 6. ¹H NMR spectrum of compound 1

Figure 8. ¹H-¹H COSY spectrum of compound 1

Figure 10. HMBC spectrum of compound 1

4.4 Structure Elucidation of Compound 2

Compound 2 was obtained as a yellow-oily liquid. (+)-HR-ESI-MS (Figure 12)

showed a peak at m/z 289.2010 [M+H]⁺ , which was consistent with a molecular

formula of C15H29O5+ (calculated 289.2010). FTIR spectrum (Figure 13) showed

absorption bands at 1720 (cm^-1) and 3450 (cm^-1), which was an evidence for the

presence of carbonyl and hydroxyl group, respectively.

The ¹H NMR spectrum (Figure 14) revealed compound 2 had five methyl groups

at δH 0.87-0.91 (9H, m, H3-9; H3-10; H3-14), δH 1.14 (3H, d, J=7.1 Hz, H3-15) and δH

2.05 (3H, s, H3-12). The several downfield signals at δH 3.64 (H, m, H-3) and δH 4.68

(H, t, J=6.2 Hz, H-6) were predicted to be connected with heteroatom. The chemical

O

of H-6 moved to more downfield region.

The ¹³C NMR (Figure 15) and HSQC (Figure 17) spectra revealed the presence of

15 carbon atoms, including three methylene carbons and two quaternary carbons. δC

173.2 (C-11) and δC 179.3 (C-1) were assigned as carbonyl carbon; δC 74.7 (C-3) and

δC 83.2 (C-6) were speculated to be connected with oxygen atoms.

The ¹H-¹H COSY (Figure 16) revealed correlation of H-2 to H3-15, H-2 to H-3

and H-3 to H2-4. Several deciding cross-peaks of H-2 to C-1; H3-9 to C-7, 8; H3-10 to

C-7; H-6 to C-5, 7, 11, 13; H3-12 to C-11; H3-14 to C-5 determined the overall structural

connectivity in HMBC spectrum (Figure 18). In addition, long-range correlations (⁴JHC)

of H-6 to C-9 and C-14 were also observed.

By searching the SciFinder database, compound 2, named trichobrevidin B, was

O O

HO OH

O

Table 35. ¹H and ¹³C NMR (CD3OD, 500 and 125 MHz) for compound 2 Position δC, type δH (J in Hz) COSY HMBC 1 179.3, C=O

2 47.4, CH 2.49 (H, dt, J=7.1, 6.9) H-3, 15 C-1, 3, 4, 15

3 74.7, CH 3.64 (H, m) H-2, 4

4 32.6, CH2 1.22-1.27 (H, m) H-3 1.66-1.81 (H, m)

5 35.7, CH 1.66-1.81 (H, m)

6 83.2, CH 4.68 (H, t, J=6.2) C-5, 7, 8, 9, 11, 13, 14 7 37.0, CH 1.66-1.81 (H, m)

8 24.9, CH2 1.50 (H, m) C-7, 9, 10

1.04-1.12 (H, m)

9 16.1, CH3 0.87-0.91 (3H, m) C-7, 8

10 11.6, CH3 0.87-0.91 (3H, m) C-7, 8

11 173.2, C=O

12 20.9, CH3 2.05 (3H, s) C-11

13 28.3, CH2 1.04-1.12 (H, m) C-4, 7

1.66-1.81 (H, m)

14 17.0, CH3 0.87-0.91 (3H, m) C-5, 6, 13

15 13.8, CH3 1.14 (3H, d, J=7.1) H-2 C-1, 2, 3

Figure 12. Mass spectrometry of compound 2

[M+H]⁺ [M-H]⁺ [M-Ac-H]⁺

Figure 14. ¹H NMR spectrum of compound 2

Figure 16. ¹H-¹H COSY spectrum of compound 2

Figure 18. HMBC spectrum of compound 2

4.5 Structure Elucidation of Compound 3

Compound 3 was obtained as a brown-oily liquid. (+)-HR-ESI-MS (Figure 19)

showed a peak at m/z 293.1736 [M+H]⁺ , which was consistent with a molecular

3.7 Hz, H-4) were predicted to be connected with heteroatom. The chemical shifts of

1

ring which caused the chemical shifts more upfield. Moreover, that the protons of

epoxide ring were more upfield can be explained by the ring strain. In order to form the

triangular ring, the hybridized orbitals of carbons slightly changed, causing chemical

shift further upfield within 2.0-3.5 ppm, as well as the ¹³C NMR chemical shift of the

C-12 and C-13.

The ¹³C NMR (Figure 22) and HSQC (Figure 24) spectra revealed the presence of

17 carbon atoms, including four methylene carbons and five quaternary carbons.

Roughly, the ¹³C NMR spectrum of compound 3 was similar to compound 1 except the

carbons at δC 41.5 (C-6) and δC 71.8 (C-11), because the hydroxyl group of the eleventh

carbon was substituted by a hydrogen, making both C-6 and C-11 more shielded.

H

H

C

O C 13 12

5 H 2

H₃C14

O

O O

H

H-10 to H-11 and H2-7 to H2-8. Several deciding cross-peaks of H3-17 to C-18, H-4 to

C-18, H-2 to C-11, H2-13 to C-12, H3-14 to C-5, 12, H3-15 to C-6, 7 and H3-16 to C-8,

10 determined the overall structural connectivity in HMBC spectrum (Figure 25).

The ROESY spectrum (Figure 26) indicated correlations of H-4 to H-11, H3-15;

H2-13 to H-2, H3-14 and therefore the relative configurations were determined as (4R*,

6R*, 11R*) and (2R*, 5S*, 12S*). By comparing with trichothecene-type skeleton,

absolute configurations were confirmed as mentioned above.

By searching the SciFinder database, compound 3, named trichodermin, was

identified to be a known compound. The first isolation of trichodermin was from

Trichoderma viride in 1965¹⁰¹.

O

CH

₃

O O O

H

₂

C H

H

₃

C

H

Table 36. ¹H and ¹³C NMR (CD3OD, 500 and 125 MHz) for compound 3 Position δC, type δH (J in Hz) COSY HMBC 2 80.2, CH 3.72 (H, d, J=5.0) H-3 C-4, 5, 11, 12 3 37.3, CH2 2.50 (H!, dd, J=15.4,

7.9)

H-2,4 C-2, 4, 5, 12

1.87-2.02 (H", ddd, J=

15.4, 5.0, 3.7)

4 76.4, CH 5.58 (H, dd, J=7.9, 3.7) H-3 C-6, 12, 18

5 50.1, C

6 41.5, C

7 25.4, CH2 1.46 (H, m) H-8 C-6, 8, 9, 11, 15

1.87-2.02 (H, m)

8 28.9, CH2 1.87-2.02 (2H, m) H-7 C-6, 7, 9, 10

9 140.9, C

10 119.9, CH 5.37 (H, m) H-11, 16 C-6, 8, 16 11 71.8, CH 3.65 (H, d, J=5.6) H-10 C-7, 9, 10, 15 12 66.5, C

13 48.5, CH2 2.86 (H, d, J=4.0) C-12

3.07 (H, d, J=4.0)

14 6.2, CH3 0.71 (3H, s) C-4, 5, 6, 12

15 16.3, CH3 0.94 (3H, s) C-5, 6, 7, 11

16 23.4, CH3 1.71 (3H, s) H-10 C-8, 9, 10

17 21.0, CH3 2.05 (3H, s) C-18

18 172.3, C=O

Figure 19. Mass spectrometry of compound 3 [M+H]⁺

[M-Ac-H]⁺

Figure 21. ¹H NMR spectrum of compound 3

Figure 23. ¹H-¹H COSY spectrum of compound 3

Figure 25. HMBC spectrum of compound 3

4.6 Structure Elucidation of Compound 4

Compound 4 was obtained as a yellow powder. (+)-HR-ESI-MS (Figure 27)

showed two major fragment ions at m/z 1189.6949 and 774.4517, generated from

fragmenting the amide bond of Aib13-Pro14. The fragments of 1189.6949 and 774.4517

corresponded to N-terminal and C-terminal parts, respectively. Through

collision-induced dissociation (CID) and higher energy collisional dissociation (HCD), the

tandem mass results of 1189.6949 indicated the successive acylium ions assignable as

shown in Figures 28 and 29. The respective mass differences were attributed to Aib13,

Leu12, Gly11, Aib10, Val9, Aib8, Gln7, Ala6, Aib5, Ala4, Aib3, Pro2, Aib1+Ac. As well as

Aib = α-Aminoisobutyric acid

tandem mass spectra (Figures 30 and 31.). Thus, the sequence of m/z 774.4517 was

determined as Pro14-Val15-Aib16-Aib17-Gln18-Gln19-Penol20. FTIR spectrum (Figure 32)

showed absorption bands at 3321 (cm^-1) and 1658 (cm^-1), which was an evidence for

the presence of secondary amide (N-H) and amide carbonyl group, respectively.

The ¹H NMR spectra (Figures 33 and 34) revealed several signals in the chemical

shift range #H 6.5-9.0 ppm which is characteristic of amide protons, so the compound 4

was speculated to be a peptaibol.

The ¹³C NMR spectra (Figures 35 and 36) indicated the presence of 23 carbonyl

carbon atoms within #C 169-180 ppm. Two signals at #C 173.1 and #C 175.1 were

significantly higher in their intensities due to the overlap with other carbonyl carbons.

In TOCSY spectra (Figures 37 and 38), protons belonging to the same amino acid

monomer could be grouped together. Thereby, each vertical line between #H 6.90-8.20

ppm and #H 0.50-4.40 ppm indicated each amino acid. In sum, one glycine, one leucine,

one phenylalaninol, two alanine, two valine, and three glutamine residues were

identified. Although proline dose not possess amide proton, it could be divided into a

proton and the former carbonyl carbons in ²J coupling provided essential connective

clues, so three parts of the sequence could be pieced together as Ac-Aib1, Ala6-Gln7

-Aib8-Val9-Aib10-Gly11-Leu12-Aib13 and Gln18-Gln19-Phenol20. Unfortunately, the rest of

sequence were difficult to identify due to their severely overlapped signals.

Nonetheless, based on the information obtained from tandem mass fragmentation,

the overall sequence was completely deduced as Ac-Aib1-Pro2-AIb3-Ala4-Aib5-Ala6

-Gln7-Aib8-Val9-Aib10-Gly11-Leu12-Aib13-Pro14-Val15-Aib16-Aib17-Gln18-Gln19-Phenol20.

The absolute configuration of compound 4 could be referred to previous literature.

Compound 4, named alamethicin III, was reported in 1995 by the time-of-flight

correlation technique¹³³. The sequence of alamethicin III has been proved to be

Table 37. ¹H and ¹³C NMR (DMSO-d6, 500 and 125 MHz) for compound 4

Amino acid Position ¹³C ¹H

Acetyl (Ac) CH3 21.8 1.98

C’ (C=O) 169.7

!-Aminoisobutyric acid (Aib1) NH 8.79

C! (C) 55.1

!-Aminoisobutyric acid (Aib3) NH 7.55

C! (C) 55.1

!-Aminoisobutyric acid (Aib5) NH 7.83

C! (C) 55.0

Amino acid Position ¹³C ¹H C# (C=O) 172.6

&-NH2 7.16/6.73

C’ (C=O) 173.1

!-Aminoisobutyric acid (Aib8) NH 7.78

C! (C) 55.3

Amino acid Position ¹³C ¹H

&-NH2 7.15/6.76

C’ (C=O) 171.5

Glutamine (Gln19) NH 7.48

Amino acid Position ¹³C ¹H

&-NH2 7.05/6.60

C’ (C=O) 170.4

Phenylalaninol (Pheol20) NH 6.95

C! (CH) 51.8 3.89

C" (CH2) 36.2 2.54/2.90

C$ (C) 138.6

C# (CH) 128.7 7.17-7.23

C& (CH) 127.4 7.17-7.23

C' (CH) 125.3 7.10-7.13

Hydroxy (-ol) CH2 62.4 3.33

OH 4.59

Figure 27. Mass spectrometry of compound 4

Figure 29. HCD spectrum of the ion m/z 1189 within 100-700 of compound 4

Figure 31. HCD spectrum of the ion m/z 774 within 100-500 of compound 4

4.7 Structure Elucidation of Compound 5

Compound 5 was obtained as a yellow powder. By comparing ¹H NMR profile,

compounds 4 and 5 were extremely similar except a group of upfield signals appearing

at !H 0.70 ppm (Figure 42). Because it integrated to three protons and resonated at

shielding region, this signal was reasonably suspected to be a methyl group connected

to a methylene group, owing to its triplet splitting pattern.

(+)-HR-ESI-MS (Figure 43) showed two major fragment ions at m/z 1189.6947

and 788.4671. The fragment ion 788.4671 was 14 Da larger than the corresponding ion

NH

Aib = α-Aminoisobutyric acid Iva = Isovaline

determined by using tandem MS. According to tandem mass spectra, the extra methyl

group was assigned on the beta carbon of Aib17, forming an Iva17 residue. These mass

results corroborated with observations made from the NMR spectrum; therefore, the

structure of compound 5 was confirmed. Furthermore, absolute configuration of

compound 5 could be referred to the literature.

The compound 5 was identified to be a known compound named atroviridin B. It

was first isolated from Trichoderma atroviride in 2000¹¹⁵.

Figure 42. Comparison of ¹H NMR spectra of compounds 4 and 5

Figure 44. CID spectrum of the ion m/z 1189 within 600-1200 of compound 5

Figure 46. CID spectrum of the ion m/z 774 within 400-800 of compound 5

4.8 Structure Elucidation of Compound 6

Compound 6 was obtained as a yellow powder. (+)-HR-ESI-MS (Figure 48)

showed two major fragment ions at m/z 1203.7091 and 774.4510. The fragment ion

1203.7091 was 14 Da larger than the corresponding ion 1189.6949, observed from

compound 4, suggesting compound 6 possessed an extra methyl group. According to

tandem mass results, the Ala6was replaced by an Aib6residue. Furthermore, absolute

configuration of compound 6 could be referred to the literature.

The compound 6 was identified to be a known compound named alamethicin

F50/7. It was also named alamethicin F50/E and polysporin B.

NH

Aib = α-Aminoisobutyric acid

Figure 48. Mass spectrometry of compound 6

Figure 50. HCD spectrum of the ion m/z 1203 within 100-700 of compound 6

Figure 52. HCD spectrum of the ion m/z 774 within 100-500 of compound 6

4.9 Structure Elucidation of Compound 7

Compound 7 was obtained as a white powder. (+)-HR-ESI-MS (Figure 53)

showed two major fragment ions at m/z 1203.6976 and 775.4246. The latter ion is 1 Da

larger than the corresponding ion 774.4510, observed from compound 6, indicating the

substitution of an amino group by a hydroxyl group. According to tandem mass results,

the Gln18has been replaced by a Glu18residue. Furthermore, absolute configuration of

compound 7 could be referred to literature.

The compound 7 was identified to be a known compound named alamethicin II.

NH

Aib = α-Aminoisobutyric acid

Figure 53. Mass spectrometry of compound 7

Figure 55. HCD spectrum of the ion m/z 1203 within 100-700 of compound 7

Figure 57. HCD spectrum of the ion m/z 775 within 100-500 of compound 7

4.10 Structure Elucidation of Compound 8

Compound 8 was obtained as a white powder. (+)-HR-ESI-MS (Figure 58)

showed two major fragment ions at m/z 1189.6945 and 789.4514. By comparing with

the fragment ion 788.4671 observed from compound 5, compound 8 was larger than 5

by 1 Da, resulting from the substitution of NH2 group (m/z 16) by OH group (m/z 17).

According to tandem mass results, Gln18was replaced by a Glu18residue.

As for chiral configurations, most amino acids related to alamethicin sequences

existed in L-form except isovaline, which was naturally present in ethier (R) or (S)

enantiomer in peptiabols. Because the chemical shifts of C"-methyl protons were less

NH

Aib = α-Aminoisobutyric acid Iva = Isovaline

methylene protons was greater than !H 0.28 ppm, and the chemical shift of the

#-methylene carbon was less than !C 29 ppm (Table 38), the absolute configuration of

Iva17 was determined to be R form (Figure 63)¹³⁴.

By searching the SciFinder database, the structure of compound 8 was identified

to be a new compound named alamethicin F50/J.

Figure 58. Mass spectrometry of compound 8

Figure 60. HCD spectrum of the ion m/z 1189 within 100-700 of compound 8

Figure 62. HCD spectrum of the ion m/z 789 within 100-500 of compound 8

(a) (b)

Figure 63. (a) ¹H-NMR (b) ¹³C-NMR chemical shifts of Iva17 in DMSO-d6

Table 38. ¹³⁴NMR data used to assess the absolute configuration of Iva residues in a right-handed helical peptide^a

NMR data Value for (R)-Iva Value for (S)-Iva

1H-NMR Chemical shift of the "-methyl protons

! < 0.89 ppm ! > 0.91 ppm

Difference between the chemical shifts (Δ!) of the two #-methylene protons

Δ! > 0.28 ppm Δ! < 0.20 ppm

13C-NMR Chemical shift of the #-methylene carbon

! < 29 ppm ! > 33 ppm

aFor a left-handed helical peptide, the parameters should be reversed.

4.11 Structure Elucidation of Compound 9

Compound 9 was obtained as a white powder. (+)-HR-ESI-MS (Figure 64)

showed two major fragment ions at m/z 1189.6948 and 775.4359. The latter ion is 1 Da

larger than the corresponding ion 774.4517, observed from compound 4, indicating the

substitution of an amino group by a hydroxyl group. According to tandem mass results,

the Gln¹⁸ has been replaced by a Glu¹⁸ residue. Furthermore, absolute configuration of

compound 9 could be referred to literature.

The compound 9 was identified to be a known compound named alamethicin I.

NH

Aib = α-Aminoisobutyric acid

Figure 64. Mass spectrometry of compound 9

Figure 66. HCD spectrum of the ion m/z 1189 within 100-700 of compound 9

Figure 68. HCD spectrum of the ion m/z 775 within 100-500 of compound 9

4.12 Antimicrobial Activities of Compounds 1-9

MIC values indicated that both compounds 3 and 5 showed extraordinary

antifungal activities against C. albicans and C. neoformans, whereas compounds 8 and

9 only inhibited C. neoformans. Unfortunately, compounds 1, 2 and 4 had no sign of

activities against both of the two pathogenic fungi. Although compounds 4 (Aib17) and

5 (Iva17) are only different at the 17^th amino acid, they exhibit a huge difference in MIC

values. Therefore, structure activity relationships (SAR) could be further discussed.

Table 39. MIC values of compounds 1-9

MIC^a (µg mL^-1)

Compounds Candida albicans

SC5314

Crytococcus neoformans H99

Trichobrevidin A (1)* >64 >64

Trichobrevidin B (2)* >64 >64

Trichodermin (3) 4 4

Alamethicin III (4) >64 >64

Atroviridin B (5) 4 8

Alamethicin F50/7 (6) 64 8

Alamethicin II (7) 32 16

Alamethicin F30/J (8)* >64 32

Alamethicin I (9) >64 32

4.13 Cytotoxicity Activities of Compounds 1-2

The results of SRB assay showed that compound 2 exhibited a comparable

cytotoxicity activity at 30 µM. Therefore, it was believed that compound 2 could be a

potential lead compound in designing new drugs for the treatment of hepatocellular

carcinoma.

Table 40. Inhibition of cell proliferation by compounds 1 and 2 at different concentrations

SK-Hep-1^a

Compounds 10 µM 30 µM

Trichobrevidin A (1)* 91 ± 3% 61 ± 3%

Trichobrevidin B (2)* 60 ± 4% 10 ± 3%

*new compound. ^aHepatocellular carcinoma cells.

5. DISCUSSION

In this study, nine compounds were isolated and characterized from

marine-derived fungus Trichoderma brevicompactum NTU439. Among these compounds, six

out of nine belong to the peptaibol and exhibit antifungal activities except alamethicin

III (4). Compounds 1, 2 and 8 are hitherto unreported compounds. Although compound

2 does not respond to pathogenic fungi, it has shown anti-hepatocellular carcinoma

activity.

5.1 Review of Trichodermin

Trichodermin, which shows a strong antifungal activity against C. albicans and A.

fumigatus by inhibiting the peptidyl transferase center of eukaryotic ribosomes¹³⁵, was

first isolated from Trichoderma viride in 1965¹⁰¹. It was later isolated from T.

polysporum, T. sporulosum and T. reesei. In addition, some research has shown that the

presence of nonviable Rhizoctonia solani or Fusarium oxysporum mycelia could

increase the production of trichodermin by T. brevicompactum¹³⁶.

Trichodermin derivatives were synthesized to investigate the structure activity

relationships (SAR). The results indicated that the epoxide, double bond and ester group

plays an important role on bioactivities¹³⁷. Furthermore, trichodermin derivative,

(E)-3-(2-fluorophenyl)acrylate moiety connected to C8 position, showed outstanding

antifungal activity than the commercial fungicide prochoraz¹³⁸.

5.2 Review of Alamethicin

Alamethicin (ALM)¹³⁹, a channel-forming peptide¹⁴⁰, has been known to possess

versatile biology activities. ALM contains a large number of non-proteinogenic amino

acid residues. The N-terminal residue is linked to an acetyl group and the C-terminal

end is connected to a phenylalaninol. In 1967, the first ALM was found in the culture

broth of fungus Trichoderma viride¹⁴¹. Due to the complexity of structure and the

insufficient access to technologies, the primary structure of ALM was not revealed until

1977.

The crystal structure of ALM was determined by X-ray diffraction¹⁴². In

MeCN/MeOH mixture, ALM was predominantly in a helix conformation but slightly

distorted at Proline 14^th. Therefore, ALM was described as a proline-kinked $-helical

conformation. ALM could insert into the lipid bilayers and oligomerize to form a

在文檔中 海洋源真菌Trichoderma brevicompactum NTU439之二次代謝產物之生物活性及化學結構分析 (頁 57-0)