Aureobasidin A

Research into the interactions of Aureobasidin A includes:

• Combination Therapy: Studies show that combining Aureobasidin A with other antifungals can enhance effectiveness against biofilms and resistant strains .
• Mechanistic Studies: Investigations into how Aureobasidin A alters cellular parameters and membrane permeability provide insights into its mode of action .
• Efflux Pump Inhibition: Studies indicate that Aureobasidin A may inhibit efflux pumps in fungi, further enhancing its antifungal activity when used alongside other agents .

Similar Compounds

Several compounds share structural or functional similarities with Aureobasidin A. Here are some notable examples:

Uniqueness of Aureobasidin A

Aureobasidin A is distinguished by its specific inhibition of inositol phosphorylceramide synthase, making it particularly effective against fungal pathogens that rely on sphingolipid biosynthesis for survival. This unique mechanism sets it apart from other antifungal agents that target different pathways within fungal cells .

Aureobasidin A represents a structurally complex cyclic depsipeptide with a well-defined molecular architecture [1] [2]. The compound possesses a molecular formula of C₆₀H₉₂N₈O₁₁ with a molecular weight of 1101.4 grams per mole, establishing it as a substantial macrocyclic natural product [2] [3]. The average mass is determined to be 1085.442 daltons, while the monoisotopic mass is precisely 1084.693641 daltons [1]. This molecular composition reflects the presence of eight amino acid units and one hydroxy acid unit, which together form the characteristic cyclic depsipeptide structure [4].

The structural complexity of Aureobasidin A is further emphasized by the presence of ten defined stereocentres out of eleven possible positions [1]. This high degree of stereochemical complexity contributes to the molecule's unique three-dimensional architecture and biological activity. The cyclodepsipeptide nature of the compound is established through the presence of both amide bonds connecting the amino acid residues and an ester bond linking the hydroxy acid component, creating a mixed peptide-ester macrocycle [4].

X-ray crystallographic analysis has revealed that Aureobasidin A crystallizes in an orthorhombic crystal system with space group P2₁2₁2₁ [5]. The unit cell dimensions are characterized by a = 21.643(3) Å, b = 49.865(10) Å, and c = 12.427(1) Å, with Z = 8, indicating eight molecules per unit cell [5]. The crystal structure contains two independent conformers per asymmetric unit, both adopting a similar arrowhead-like conformation that appears to be energetically favorable [5].

The amino acid sequence of Aureobasidin A has been definitively established through comprehensive structural analysis [4]. The sequence consists of 2(R)-hydroxy-3(R)-methylpentanoic acid at position 1, followed by N-methyl-L-valine at position 2, L-phenylalanine at position 3, N-methyl-L-phenylalanine at position 4, L-proline at position 5, allo-L-isoleucine at position 6, N-methyl-L-valine at position 7, L-leucine at position 8, and β-hydroxy-N-methyl-L-valine at position 9 [4]. This sequence was determined through acid hydrolysis of the antibiotic followed by detailed structural characterization of the individual components [4].

A particularly significant structural feature is the extensive N-methylation pattern observed in Aureobasidin A [5]. Four out of seven amide bonds in the molecule are N-methylated, specifically at positions 2, 4, 7, and 9 [5]. This high degree of N-methylation is crucial for the formation of the unique three-dimensional conformation and contributes significantly to the biological activity of the compound [5]. Energy minimization studies have demonstrated that reduction in the number of N-methylated amide bonds leads to conformers that differ substantially from the biologically active arrowhead form [5].

The stereochemical assignments for each residue have been rigorously established [4]. The hydroxy acid at position 1 possesses (R,R) configuration, while all amino acid residues exhibit L-configuration at their α-carbon centers [4]. The β-hydroxy-N-methyl-L-valine at position 9 contains an additional stereocenter at the β-position, which maintains the S-configuration [4]. These stereochemical features are essential for the proper folding and biological activity of the molecule.

Parameter	Value
Molecular Formula	C₆₀H₉₂N₈O₁₁
Molecular Weight (g/mol)	1101.4
Average Mass (Da)	1085.442
Monoisotopic Mass (Da)	1084.693641
Number of Amino Acid Units	8
Number of Hydroxy Acid Units	1
Number of Defined Stereocentres	10 of 11
Crystal System	Orthorhombic
Space Group	P2₁2₁2₁
Unit Cell Dimensions (Å)	a = 21.643(3), b = 49.865(10), c = 12.427(1)
Z Value	8

Nuclear Magnetic Resonance-Based Conformational Analysis

Nuclear magnetic resonance spectroscopy has provided comprehensive insights into the solution-phase conformational behavior of Aureobasidin A [6] [7] [8]. The proton and carbon-13 nuclear magnetic resonance spectra have been analyzed using a variety of two-dimensional nuclear magnetic resonance techniques to elucidate the three-dimensional structure and conformational dynamics of the molecule [6].

A fundamental finding from nuclear magnetic resonance studies is the existence of two distinct conformational isomers of Aureobasidin A in solution [6] [7]. These conformers exist as an equilibrium mixture in deuteriochloroform, with the isomerism directly associated with cis-trans rotation of the amide bond between N-methylphenylalanine at position 4 and proline at position 5 [6]. This conformational equilibrium has been observed with population ratios of approximately 57:43 for the trans-proline to cis-proline conformers in deuteriochloroform [9].

The identification of these conformational states has been achieved through Nuclear Overhauser Effect experiments, which provide crucial information about spatial proximities between protons in the molecule [6]. These experiments have made possible the definitive identification of the conformation of the amide bonds throughout the Aureobasidin A structure [6]. The Nuclear Overhauser Effect data clearly demonstrate the presence of both cis and trans conformations around the critical MePhe⁴-Pro⁵ amide bond [6].

Recent advances in nuclear magnetic resonance-driven computational approaches have enabled the elucidation of a high-resolution three-dimensional structure of the cis-proline conformer of Aureobasidin A for the first time [7] [8]. This breakthrough was achieved using advanced conformational sampling techniques augmented by sparse distance and torsion angle constraints derived from nuclear magnetic resonance data [7] [8]. The obtained three-dimensional conformational structure has been validated using anisotropic residual dipolar coupling measurements, providing additional confidence in the structural assignments [7] [8].

Two-dimensional correlation spectroscopy experiments have been instrumental in establishing through-bond connectivity patterns within the molecule [6]. These techniques have provided complete assignments of the proton and carbon signals, enabling detailed analysis of the molecular framework [6]. The scalar coupling patterns observed in these experiments have confirmed the sequence of amino acid residues and have provided information about the conformational preferences of individual residues [6].

Two-dimensional nuclear Overhauser enhancement spectroscopy has revealed critical inter-residue spatial relationships that define the three-dimensional architecture of Aureobasidin A [6]. These experiments have identified specific proton-proton contacts that occur across space rather than through chemical bonds, providing essential constraints for three-dimensional structure determination [6]. The nuclear Overhauser enhancement patterns have been particularly valuable in defining the overall fold of the molecule and in distinguishing between the cis and trans conformational states [6].

The conformational analysis has revealed that the trans-proline conformer adopts a more extended conformation, while the cis-proline conformer exhibits a more compact, arrowhead-like structure [7] [8]. Both conformers appear to maintain the essential structural features necessary for biological activity, including the critical hydrogen bonding patterns and the overall macrocyclic architecture [7] [8].

Nuclear Magnetic Resonance Technique	Application	Key Findings
¹H Nuclear Magnetic Resonance	Proton assignment and chemical shift analysis	Two conformational isomers in CDCl₃ solution
¹³C Nuclear Magnetic Resonance	Carbon framework characterization	Complete carbon skeleton assignment achieved
2D Correlation Spectroscopy	Scalar coupling connectivity determination	Through-bond connectivity established
2D Nuclear Overhauser Enhancement Spectroscopy	Spatial proximity identification	Inter-residue spatial relationships determined
Nuclear Overhauser Effect	Conformational state identification of amide bonds	cis-trans isomerism at MePhe⁴-Pro⁵ amide bond
Residual Dipolar Coupling	Structural validation of cis-proline conformer	High-resolution 3D structure of cis-Pro conformer

Comparative Analysis with Aureobasidin Derivatives

The structural characterization of Aureobasidin A derivatives has provided valuable insights into structure-activity relationships and conformational requirements for biological activity [10] [11] [12]. Over twenty aureobasidin congeners have been isolated and characterized, each representing specific modifications to the parent Aureobasidin A structure [10] [13] [14].

Aureobasidin B represents a significant derivative where the 2(R)-hydroxy-3(R)-methylpentanoic acid residue at position 1 is replaced with D-hydroxyisovaleric acid [10] [11]. This modification results in approximately two-fold enhanced activity compared to the parent compound [11]. The structural analysis indicates that this derivative maintains the characteristic arrowhead-like conformation observed in Aureobasidin A, suggesting that the overall three-dimensional architecture is preserved despite the structural modification [11].

Aureobasidin C features a replacement of the allo-L-isoleucine residue at position 6 with L-valine [10] [11]. This relatively conservative amino acid substitution also leads to approximately two-fold enhanced biological activity [11]. The retention of the overall molecular fold in this derivative demonstrates the tolerance of the position 6 site for structural modifications while maintaining the essential conformational features required for activity [11].

Aureobasidin D incorporates a modification at the critical position 9, where the β-hydroxy-N-methyl-L-valine residue is replaced with γ-hydroxy-N-methyl-L-valine [10] [11]. This derivative also exhibits enhanced activity, approximately two-fold greater than Aureobasidin A [11]. The modification affects the γ-turn region of the molecule, yet the essential structural elements appear to be maintained [11].

Aureobasidin E presents a unique structural variant where β-hydroxy-N-methyl-phenylalanine replaces the N-methyl-L-phenylalanine at position 4 [6] [15]. Nuclear magnetic resonance analysis of Aureobasidin E has revealed that almost all of the molecule exists as a single conformer in deuteriochloroform, with the amide bond between the β-hydroxy-N-methylphenylalanine and proline maintained in the cis-conformation [6]. This represents a significant difference from Aureobasidin A, which exists as a mixture of cis and trans conformers [6]. Despite this conformational restriction, Aureobasidin E maintains comparable antifungal activity to the parent compound [15].

The comparative analysis extends to synthetic derivatives where specific amino acid positions have been systematically modified [16] [12]. Derivatives with alkyl chain modifications at positions 6, 7, and 8 have been synthesized and evaluated [16] [12]. The introduction of L-glutamic acid residues at these positions, followed by esterification with various alkyl chains, has provided insights into the hydrophobic requirements of the binding site [16] [12]. Shorter alkyl chains (C4 or C6) generally maintain or enhance activity, while longer chains (C14) result in complete loss of antifungal activity [16] [12].

Structure-activity relationship studies have revealed that the P-glycoprotein inhibitory activity and antifungal activity do not always correlate directly [10] [11]. Some derivatives with reduced antifungal activity demonstrate enhanced P-glycoprotein inhibitory effects, indicating that different structural requirements govern these two biological activities [10] [11]. The [2,3-dehydro-MeVal⁹]-Aureobasidin A derivative represents the most potent P-glycoprotein inhibitory aureobasidin, being 13-fold more potent than Aureobasidin A and 19-fold more potent than cyclosporin A [11].

The analysis of aureobasidin derivatives has confirmed the importance of the β-hydroxy-N-methyl-L-valine residue at position 9 for biological activity [5] [10] [11]. This residue is consistently positioned at the tip of the characteristic arrowhead-like conformation and appears to be indispensable for antifungal activity [5]. Modifications to this residue generally result in significant changes in biological activity, either enhancement or reduction, depending on the specific chemical alteration [10] [11].

Aureobasidin Derivative	Structural Difference	Conformation	Relative Activity
Aureobasidin A	Reference compound	Two conformers (cis/trans-Pro)	100% (reference)
Aureobasidin B	[D-Hiv¹]-AbA replacement	Similar arrowhead conformation	~2-fold enhanced activity
Aureobasidin C	[Val⁶]-AbA replacement	Retained overall fold	~2-fold enhanced activity
Aureobasidin D	[γHOMeVal⁹]-AbA replacement	Modified γ-turn region	~2-fold enhanced activity
Aureobasidin E	β-hydroxy-N-methyl-phenylalanine at position 4	Predominantly cis-Pro single conformer	Comparable antifungal activity

The biosynthesis of Aureobasidin A is encoded by the aba1 gene, which represents one of the most extensively characterized nonribosomal peptide synthetase complexes in fungal secondary metabolism [1]. The aba1 gene consists of a single, intronless open reading frame spanning 34,980 base pairs, encoding an 11,659 amino acid protein with a calculated molecular mass of 1,286,254 daltons [1]. This massive enzyme complex operates through a modular assembly-line mechanism that directly correlates with the structural composition of the final cyclic depsipeptide product.

The aba1 gene encodes a nonribosomal peptide synthetase composed of nine distinct biosynthetic modules, each responsible for the incorporation and modification of specific amino acid or hydroxy acid residues [2] [1]. Eight of these modules contain adenylation domains with recognizable amino acid specificity-conferring code elements, while the first module lacks such recognizable specificity elements, consistent with its role in incorporating the unique (R)-2-hydroxy-4-methylpentanoic acid residue [2] [1]. The gene organization reflects the linear assembly-line nature of nonribosomal peptide synthesis, where the sequential arrangement of modules determines the order of monomer incorporation into the growing peptide chain [1].

A remarkable feature of the aba1 gene is the extraordinarily high degree of sequence similarity among eight of the nine biosynthetic modules at both nucleotide and amino acid levels [1]. The majority of modules share better than 70% nucleotide identity with other modules in the complex, with modules having identical amino acid adenylation specificity sharing up to 95% identity [1]. This unprecedented level of similarity among fungal nonribosomal peptide synthetase genes suggests a unique evolutionary origin and provides insights into the mechanisms of modular enzyme evolution in secondary metabolite biosynthesis.

The aba1 gene contains four embedded methylation domains located within modules that incorporate methylated amino acid residues [1]. These domains are responsible for the N-methylation of valine residues at positions 2, 4, 7, and 8 in the Aureobasidin A structure, contributing to the compound's unique chemical properties and biological activity [1]. The presence of these methylation domains correlates precisely with the methylated amino acids found in the final product, demonstrating the high fidelity of the modular assembly system.

Functional validation of the aba1 gene was achieved through targeted gene disruption experiments using hygromycin B phosphotransferase gene cassettes [1]. Insertion of the resistance cassette in place of module 4 sequence completely abolished Aureobasidin A production, confirming that the isolated gene encodes the complete biosynthesis complex [1]. This genetic evidence established aba1 as the sole determinant for Aureobasidin A biosynthesis in Aureobasidium pullulans strain BP-1938.

Table 1: aba1 Gene Structure and Module Organization

Module	Amino Acid Specificity	Adenylation Domain	Methylation Domain	Function
Module 1	(R)-2-hydroxy-4-methylpentanoic acid	No recognizable code	No	Hydroxy acid incorporation
Module 2	L-N-methyl-valine	Recognized	Yes	Methylated valine incorporation
Module 3	L-phenylalanine	Recognized	No	Phenylalanine incorporation
Module 4	L-N-methyl-phenylalanine	Recognized	Yes	Methylated phenylalanine incorporation
Module 5	L-proline	Recognized	No	Proline incorporation
Module 6	L-allo-isoleucine	Recognized	No	Allo-isoleucine incorporation
Module 7	L-leucine	Recognized	No	Leucine incorporation
Module 8	L-N-methyl-valine	Recognized	Yes	Methylated valine incorporation
Module 9	L-hydroxy-N-methyl-valine	Recognized	Yes	Hydroxylated methylated valine

Enzymatic Modules for Hydroxy Acid Incorporation

The incorporation of (R)-2-hydroxy-4-methylpentanoic acid into the Aureobasidin A structure represents a critical and unique aspect of the biosynthetic pathway [2] [1]. This hydroxy acid component is incorporated through Module 1 of the aba1 nonribosomal peptide synthetase, which exhibits distinct characteristics compared to the other eight modules in the complex. The adenylation domain of Module 1 lacks recognizable amino acid specificity-conferring code elements, reflecting its specialized function in activating and incorporating this non-proteinogenic hydroxy acid substrate [2] [1].

The enzymatic mechanism for hydroxy acid incorporation involves several distinct catalytic domains working in coordinated fashion [3]. The adenylation domain of Module 1 catalyzes the ATP-dependent activation of (R)-2-hydroxy-4-methylpentanoic acid to form the corresponding acyl-adenosine monophosphate intermediate [3]. This activated intermediate is subsequently transferred to the phosphopantetheine prosthetic arm of the adjacent thiolation domain, also known as the peptidyl carrier protein domain [3]. The thioesterification reaction results in the formation of a covalent thioester linkage between the hydroxy acid and the carrier protein, enabling subsequent enzymatic transformations.

The condensation domain associated with Module 1 catalyzes the formation of the depside bond between the hydroxy acid and the amino acid incorporated by Module 2 [3]. This condensation reaction represents the formation of the characteristic ester linkage that distinguishes depsipeptides from conventional peptides [4]. The specificity of this condensation domain for hydroxy acid substrates rather than amino acid substrates reflects the evolutionary adaptation of the nonribosomal peptide synthetase machinery to accommodate non-standard building blocks.

Post-assembly modifications play a crucial role in generating the final hydroxy acid-containing structure of Aureobasidin A [2]. The hydroxylation of L-valine in Module 9 to produce L-hydroxy-N-methyl-valine occurs through a post-assembly mechanism catalyzed by a hydroxylase enzyme not yet fully identified [2]. This post-synthetic modification demonstrates the complexity of secondary metabolite biosynthesis, where the linear assembly of monomers is followed by additional enzymatic transformations to generate the final bioactive compound.

The thioesterase domain located at the C-terminus of the aba1 enzyme complex catalyzes both the cyclization and release of the completed nonapeptide product [2] [1]. This domain exhibits dual functionality, first catalyzing the intramolecular cyclization reaction to form the macrocyclic structure, and subsequently hydrolyzing the thioester bond to release the mature Aureobasidin A molecule from the enzyme complex [2]. The cyclization mechanism involves nucleophilic attack of the terminal amino group on the thioester-linked hydroxy acid, forming the characteristic cyclic depsipeptide structure.

Table 2: Enzymatic Modules for Hydroxy Acid Incorporation

Module Component	Function	Specificity	Position in Pathway
Module 1 Adenylation Domain	Activates (R)-2-hydroxy-4-methylpentanoic acid	No recognizable specificity code	Initial substrate activation
Carrier Protein Domain	Carries activated hydroxy acid substrate	Phosphopantetheine prosthetic group	Substrate transport
Condensation Domain	Forms peptide bond with next module	Hydroxy acid to amino acid linkage	Chain elongation
Hydroxylase Enzyme	Post-assembly hydroxylation of valine	L-valine to hydroxy-methyl-valine	Post-assembly modification
Thioesterase Domain	Cyclization and product release	Cyclization of nonapeptide	Product formation and release

Fermentation Optimization in Aureobasidium pullulans

The optimization of fermentation conditions for Aureobasidin A production in Aureobasidium pullulans represents a critical aspect of biosynthetic pathway enhancement [5]. Recent studies have demonstrated that systematic optimization of culture parameters can significantly improve the yield of this bioactive cyclic depsipeptide. The strain Aureobasidium pullulans PA-2 has emerged as a particularly promising producer, with optimized fermentation conditions resulting in Aureobasidin A yields of 920 mg/L, representing a 51% increase over non-optimized conditions [5].

Temperature control emerges as one of the most critical parameters affecting Aureobasidin A production [5]. The optimal temperature for fermentation has been established at 26°C, which promotes the yeast-like morphological form of Aureobasidium pullulans that is primarily responsible for secondary metabolite production [5]. This temperature optimum reflects the biological requirement for maintaining the appropriate cellular morphology, as the yeast-like phase exhibits higher metabolic activity for nonribosomal peptide synthesis compared to the hyphal form [6]. Deviation from this optimal temperature range significantly reduces both biomass accumulation and Aureobasidin A production.

The pH of the fermentation medium exerts a profound influence on both enzyme activity and overall biosynthetic efficiency [5]. An initial pH of 7.0 has been identified as optimal for Aureobasidin A production, with this neutral to slightly alkaline condition supporting maximum expression and activity of the aba1 nonribosomal peptide synthetase complex [5]. The pH stability throughout the fermentation process is crucial, as fluctuations can affect the conformational stability of the large multi-domain enzyme complex and subsequently impact product formation rates.

Aeration and agitation parameters require careful optimization to ensure adequate oxygen transfer while maintaining appropriate hydrodynamic conditions [5]. The optimal rotation speed of 216 rpm provides sufficient mixing for oxygen transfer and nutrient distribution without creating excessive shear stress that could damage the fungal cells [5]. Interestingly, non-stirred fermentation conditions have also shown promise, particularly when combined with controlled aeration rates, offering potential cost savings for large-scale production processes [6].

Inoculum size and timing significantly affect the fermentation kinetics and final product yield [5]. An optimal inoculum size of 6.8% (v/v) with an optical density (OD600) of 0.25 provides the appropriate cell density to initiate rapid growth while avoiding early nutrient depletion [5]. The fermentation time course studies indicate that maximum Aureobasidin A production occurs at 48 hours of cultivation, coinciding with the transition to stationary growth phase when secondary metabolite production is typically enhanced [5].

Carbon and nitrogen source optimization plays a crucial role in maximizing both biomass formation and Aureobasidin A biosynthesis [5]. Sucrose at a concentration of 3% (w/v) serves as the optimal carbon source, providing sustained energy and carbon building blocks for both primary and secondary metabolism [5]. Ammonium sulfate at 0.5% (w/v) concentration represents the preferred nitrogen source, supporting adequate protein synthesis for enzyme production while not inhibiting secondary metabolite formation [5]. The carbon-to-nitrogen ratio must be carefully balanced to promote nonribosomal peptide synthetase expression without triggering nitrogen catabolite repression mechanisms.

Table 3: Fermentation Optimization Parameters for Aureobasidium pullulans

Parameter	Optimal Range	Effect on Production	Reference Condition
Temperature	25-26°C	Critical for yeast-like morphology	Optimized PA-2 strain
pH	7.0	Affects enzyme activity	Initial pH adjustment
Inoculum Size	6.8% (v/v)	Influences biomass formation	OD600 = 0.25
Rotation Speed	216 rpm	Oxygen transfer optimization	Non-stirred alternative possible
Culture Time	48 hours	Maximum production at stationary phase	Production plateau after 48h
Carbon Source	Sucrose (3%)	Primary carbon substrate	Glucose alternative available
Nitrogen Source	Ammonium sulfate (0.5%)	Essential nitrogen source	Yeast extract alternative

1: Tan HW, Tay ST. The inhibitory effects of aureobasidin A on Candida planktonic and biofilm cells. Mycoses. 2013 Mar;56(2):150-6. doi: 10.1111/j.1439-0507.2012.02225.x. Epub 2012 Aug 12. PubMed PMID: 22882276.
2: Alqaisi AQI, Mbekeani AJ, Llorens MB, Elhammer AP, Denny PW. The antifungal Aureobasidin A and an analogue are active against the protozoan parasite Toxoplasma gondii but do not inhibit sphingolipid biosynthesis. Parasitology. 2018 Feb;145(2):148-155. doi: 10.1017/S0031182017000506. Epub 2017 May 10. PubMed PMID: 28486997; PubMed Central PMCID: PMC5964465.
3: Takesako K, Kuroda H, Inoue T, Haruna F, Yoshikawa Y, Kato I, Uchida K, Hiratani T, Yamaguchi H. Biological properties of aureobasidin A, a cyclic depsipeptide antifungal antibiotic. J Antibiot (Tokyo). 1993 Sep;46(9):1414-20. PubMed PMID: 8226319.
4: Wuts PG, Simons LJ, Metzger BP, Sterling RC, Slightom JL, Elhammer AP. Generation of Broad-Spectrum Antifungal Drug Candidates from the Natural Product Compound Aureobasidin A. ACS Med Chem Lett. 2015 Apr 23;6(6):645-9. doi: 10.1021/acsmedchemlett.5b00029. eCollection 2015 Jun 11. PubMed PMID: 26101567; PubMed Central PMCID: PMC4468416.
5: Katsuki Y, Yamaguchi Y, Tani M. Overexpression of PDR16 confers resistance to complex sphingolipid biosynthesis inhibitor aureobasidin A in yeast Saccharomyces cerevisiae. FEMS Microbiol Lett. 2018 Feb 1;365(3). doi: 10.1093/femsle/fnx255. PubMed PMID: 29240942.
6: Alqaisi AQI, Mbekeani AJ, Llorens MB, Elhammer AP, Denny PW. The antifungal Aureobasidin A and an analogue are active against the protozoan parasite Toxoplasma gondii but do not inhibit sphingolipid biosynthesis - Corrigendum. Parasitology. 2018 Feb;145(2):156. doi: 10.1017/S0031182017000877. Epub 2017 Jun 9. PubMed PMID: 28595661; PubMed Central PMCID: PMC5964473.
7: Munusamy K, Vadivelu J, Tay ST. A study on Candida biofilm growth characteristics and its susceptibility to aureobasidin A. Rev Iberoam Micol. 2018 Apr - Jun;35(2):68-72. doi: 10.1016/j.riam.2017.07.001. Epub 2018 Mar 12. PubMed PMID: 29544734.
8: Wang XH, Guo XJ, Li HY, Gou P. Characteristics of inositol phosphorylceramide synthase and effects of aureobasidin A on growth and pathogenicity of Botrytis cinerea. J Gen Appl Microbiol. 2015;61(4):108-16. doi: 10.2323/jgam.61.108. PubMed PMID: 26377130.
9: Endo M, Takesako K, Kato I, Yamaguchi H. Fungicidal action of aureobasidin A, a cyclic depsipeptide antifungal antibiotic, against Saccharomyces cerevisiae. Antimicrob Agents Chemother. 1997 Mar;41(3):672-6. PubMed PMID: 9056012; PubMed Central PMCID: PMC163770.
10: Ikai K, Shiomi K, Takesako K, Kato I, Naganawa H. NMR studies of aureobasidins A and E. J Antibiot (Tokyo). 1991 Nov;44(11):1199-207. PubMed PMID: 1761416.
11: Takahashi T, Ohara Y, Sawatari M, Sueno K. Isolation and characterization of sake yeast mutants with enhanced isoamyl acetate productivity. J Biosci Bioeng. 2017 Jan;123(1):71-77. doi: 10.1016/j.jbiosc.2016.07.002. Epub 2016 Jul 27. PubMed PMID: 27475923.
12: Cerantola V, Guillas I, Roubaty C, Vionnet C, Uldry D, Knudsen J, Conzelmann A. Aureobasidin A arrests growth of yeast cells through both ceramide intoxication and deprivation of essential inositolphosphorylceramides. Mol Microbiol. 2009 Mar;71(6):1523-37. doi: 10.1111/j.1365-2958.2009.06628.x. Epub 2009 Feb 2. PubMed PMID: 19210614.
13: Aoki K, Uchiyama R, Yamauchi S, Katayama T, Itonori S, Sugita M, Hada N, Yamada-Hada J, Takeda T, Kumagai H, Yamamoto K. Newly discovered neutral glycosphingolipids in aureobasidin A-resistant zygomycetes: Identification of a novel family of Gala-series glycolipids with core Gal alpha 1-6Gal beta 1-6Gal beta sequences. J Biol Chem. 2004 Jul 30;279(31):32028-34. Epub 2004 May 20. PubMed PMID: 15155728.
14: Gostinčar C, Ohm RA, Kogej T, Sonjak S, Turk M, Zajc J, Zalar P, Grube M, Sun H, Han J, Sharma A, Chiniquy J, Ngan CY, Lipzen A, Barry K, Grigoriev IV, Gunde-Cimerman N. Genome sequencing of four Aureobasidium pullulans varieties: biotechnological potential, stress tolerance, and description of new species. BMC Genomics. 2014 Jul 1;15:549. doi: 10.1186/1471-2164-15-549. PubMed PMID: 24984952; PubMed Central PMCID: PMC4227064.
15: Tiberghien F, Kurome T, Takesako K, Didier A, Wenandy T, Loor F. Aureobasidins: structure-activity relationships for the inhibition of the human MDR1 P-glycoprotein ABC-transporter. J Med Chem. 2000 Jun 29;43(13):2547-56. PubMed PMID: 10891114.
16: Kuroda M, Hashida-Okado T, Yasumoto R, Gomi K, Kato I, Takesako K. An aureobasidin A resistance gene isolated from Aspergillus is a homolog of yeast AUR1, a gene responsible for inositol phosphorylceramide (IPC) synthase activity. Mol Gen Genet. 1999 Mar;261(2):290-6. PubMed PMID: 10102364.
17: Tani Y, Funatsu T, Ashida H, Ito M, Itonori S, Sugita M, Yamamoto K. Novel neogala-series glycosphingolipids with terminal mannose and glucose residues from Hirsutella rhossiliensis, an aureobasidin A-resistant ascomycete fungus. Glycobiology. 2010 Jan;20(4):433-41. doi: 10.1093/glycob/cwp190. Epub 2009 Dec 9. PubMed PMID: 20007186.
18: Ogawa A, Hashida-Okado T, Endo M, Yoshioka H, Tsuruo T, Takesako K, Kato I. Role of ABC transporters in aureobasidin A resistance. Antimicrob Agents Chemother. 1998 Apr;42(4):755-61. PubMed PMID: 9559778; PubMed Central PMCID: PMC105537.
19: Kurome T, Inoue T, Takesako K, Kato I. Syntheses of antifungal aureobasidin A analogs with alkyl chains for structure-activity relationship. J Antibiot (Tokyo). 1998 Mar;51(3):359-67. PubMed PMID: 9589073.
20: Shimizu T, Kinoshita H, Nihira T. Development of transformation system in Monascus purpureus using an autonomous replication vector with aureobasidin A resistance gene. Biotechnol Lett. 2006 Jan;28(2):115-20. PubMed PMID: 16369695.