Β-Aescin at Subinhibitory Concentration (Sub-MIC) Enhances Susceptibility of Candida Glabrata Clinical Isolates to Nystatin

Home , Aesculus hippocastanum, Sapindus

Medical Mycology, 2015, 53, 845–851 doi: 10.1093/mmy/myv035 Advance Access Publication Date: 19 June 2015 Original Article

Original Article β-Aescin at subinhibitory concentration (sub-MIC) enhances susceptibility of Candida glabrata clinical isolates to nystatin Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 Roman Franiczek1,∗, Michał Glensk´ 2, Barbara Krzyzanowska˙ 1 and Maciej Włodarczyk2

1Department of Microbiology and 2Department of Pharmacognosy, Wroclaw Medical University, Wroclaw, Poland

*To whom correspondence should be addressed. Roman Franiczek, Department of Microbiology, Wrocław Medical University, Chałubinskiego´ 4, 50-368 Wrocław, Poland. Tel: +4871 784 13 02; Fax: +4871 784 01 17; E-mail: [email protected]

Received 16 October 2014; Revised 2 April 2015; Accepted 22 April 2015

Abstract Aescin (escin) derived from the seeds of horse chestnut (Aesculus hippocastanum L.) is a natural mixture of triterpene saponins exhibiting a wide variety of pharmacological properties, including antiinﬂammatory, analgesic, and antipyretic activities. However, data concerning antifungal activities of these compounds are limited. This study aims to evaluate the in vitro antifungal susceptibility of Candida glabrata clinical isolates to α- aescin sodium, β-aescin crystalline and β-aescin sodium using the disk diffusion (DD) and broth microdilution (BMD) methods. Moreover, the inﬂuence of subinhibitory concentration (0.5×MIC) of β-aescins on the nystatin MIC was also studied. In general, the results obtained by the DD assay correlated well with those obtained by the BMD method. Both β-aescins effectively inhibited the growth of all 24 strains tested. The minimum inhibitory concentration (MIC) values ranging from 8 to 32 μg/ml for β-aescin crystalline, whereas those of β-aescin sodium were slightly lower and ranged from 4 to 16 μg/ml. In contrast, α-aescin sodium was found to be completely ineffective against the strains studied. MIC values of nystatin were reduced 2–16-fold and 2–4-fold in the presence of subinhibitory concentration of β-aescin crystalline and β-aescin sodium, respectively. Results of the present study may suggest the additive interaction between β-aescin and nystatin.

Key words: Candida glabrata, aescin, nystatin, minimum inhibitory concentration.

Introduction yeast belonging to the normal microbiota of the oral cavity, gastrointestinal and genital tracts, rarely associated with se- Candida albicans still remains the leading human fungal rious infections in humans. However, under certain circum- pathogen. In recent years, however, a progressive increase in stances, C. glabrata, like other yeasts, is capable of causing infections caused by non-albicans Candida species, such as a variety of diseases from mucocutaneous infections to se- Candida glabrata, has been reported worldwide [1–4]. His- vere and life-threatening invasive candidemia, especially in torically, C. glabrata has been recognized as a commensal

C The Author 2015. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology. 845 All rights reserved. For permissions, please e-mail: [email protected] 846 Medical Mycology, 2015, Vol. 53, No. 8 critically ill and immunocompromised patients [5,6]. The Chemicals epidemiology of fungal infections caused by this yeast α-aescin sodium (αAE-Na) was obtained from Roth (Karl- species varies depending on the geographical region. It has sruhe, Germany). β-aescin crystalline (βAE-C) was kindly been demonstrated that C. glabrata was one of the most fre- supplied by Noblius Ent (Warsaw, Poland), whereas β- quent causes of candidaemia in patients with hematologic aescin sodium (βAE-Na) by Labor (Wroclaw, Poland). malignancies in Texas during the period 1993–2003 [1]. NYT (151.26 mg = 106 units) was purchased from Cos- The results of retrospective study conducted by Asmunds- metics Pharma (Warsaw, Poland). The chemicals were pre- dottir et al. [4] revealed, that C. glabrata was the second pared as a 10 mg/ml stock solutions in dimethyl sulfox- most prevalent yeast species responsible for systemic infec- ide (DMSO) (Sigma-Aldrich, St. Louis, USA) and then tions in Iceland during a 12-year period (2000–2011). In aliquoted and stored at −80oC until used. Polish hospitals in years 2006–2007 the percentage of can-

didemia episodes caused by this yeast species was shown to Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 be 14.1% [7]. Moreover, C. glabrata clinical isolates often Antifungal susceptibility testing display intrinsic as well as acquired antifungal resistance Antifungal susceptibility to α-aescin sodium (αAE-Na), β- that reduce significantly therapeutic options and may lead aescin crystalline (βAE-C), β-aescin sodium (βAE-Na) and to treatment failure. According to Silva et al., [8] infections nystatin (NYT) was performed by the disk diffusion (DD) caused by C. glabrata are often associated with higher mor- and broth microdilution (BMD) methods. bidity and mortality rates than those caused by C. albicans. To overcome these problems, new antifungal agents target- ing drug–resistant yeasts are urgently needed. During the Disk diffusion (DD) testing past decades, many efforts have been devoted to isolating DD in vitro susceptibility tests were done in accordance compounds from natural sources, which could potentiate with Clinical and Laboratory Standards Institute (CLSI) the effect of conventional antifungal therapy, either when document M44-A [12]. Mueller Hinton agar (Becton Dick- administered alone or in combination with classical anti- inson, Sparks, USA) supplemented with 2% dextrose and fungals [9–11]. 0.5 mg/l methylene blue was used as the basic media. The The aim of the present study was to investigate the in yeast inocula were adjusted to the turbidity of a 0.5 Mc- vitro antifungal activity of three aescin compounds, includ- Farland standard (approximately 1–5 × 106 CFU/ml) in ing α-aescin sodium (αAE-Na), β-aescin crystalline (βAE- sterile saline (0.85%) and streaked onto agar plates us- C), and β-aescin sodium (βAE-Na) against clinical isolates ing sterile cotton swabs. The standard paper disks (6 mm of C. glabrata. In addition, the influence of subinhibitory in diameter; Becton Dickinson, Sparks, USA) were taken concentration (0.5 × MIC) of β-aescins on the nystatin in a blank petri dish under the laminar hood. Then disks (NYT) MIC was also investigated. were saturated with 10 μl of different aescin stock solutions (10 mg/ml) to obtain the desired concentration of 100 μg per disk. Dried disks were placed aseptically on the agar Material and methods surface. NYT (15 μg/disk) and free solvent (sterile saline) were used as positive and negative controls, respectively. Strains C. glabrata ATCC 90030 reference strain was included in A total of 24 C. glabrata clinical isolates were collected each susceptibility test for quality control and assessment of during a three-year period (2011–2013) from patients hos- reproducibility. The results were determined based on the ◦ pitalized in the Medical University Hospital in Wrocław, inhibition zone diameter after incubating at 35 C for 24 Poland. These isolates were recovered from various speci- h. All assays were performed in triplicate, and the results men types such as blood, urine, feces, throat swab, sputum, were expressed as means ± standard deviation (SD) of three catheter, bronchial aspiration, and vagina and were non- independent experiments. repetitive. The identity of the strains was confirmed using colony color characteristics of CHROMagar Candida (Ox- oid, Basingstoke, Hampshire, UK) and commercially avail- Minimum inhibition concentration (MIC) able ID32Ctest strips (bioMerieu´ Marcy–l’Etoile, France) determination according the manufacturer’s recommendations. Addition- The MIC values of α-andβ-aescins for 24 C. glabrata ally, C. glabrata ATCC 90030 was used as reference strain strains tested and C. glabrata ATCC 90030 reference for quality control and to monitor the reproducibility of in strain were determined by the two-fold broth microdilu- vitro susceptibility testing. tion (BMD) method according to the guidelines M27-A3, Franiczek et al. 847

Ta b l e 1 . In vitro susceptibilities of Candida glabrata isolates to α-aescin sodium (αAE-Na), β-aescin crystalline (βAE-C), and β-aescin sodium (βAE-Na) determined by disk diffusion (DD) and broth microdilution (BMD) methods.

αAE-Na βAE-C βAE-Na

Strain IZD (mm) MIC (μg/ml) IZD (mm) MIC (μg/ml) IZD (mm) MIC (μg/ml)

973 0.0 ± 0.0 >2048 21.8 ± 0.27 8 28.1 ± 0.22 4 2122 0.0 ± 0.0 >2048 21.9 ± 0.25 8 28.1 ± 0.17 4 2124 0.0 ± 0.0 >2048 14.0 ± 0.16 32 20.3 ± 0.24 16 2129 0.0 ± 0.0 >2048 18.1 ± 0.14 16 24.5 ± 0.41 8 2136 0.0 ± 0.0 >2048 18.5 ± 0.08 16 24.2 ± 0.30 8 2165 0.0 ± 0.0 >2048 18.2 ± 0.21 16 23.9 ± 0.14 8 2222 0.0 ± 0.0 >2048 21.6 ± 0.25 8 28.5 ± 1.10 4 Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 2228 0.0 ± 0.0 >2048 14.4 ± 0.31 32 20.5 ± 0.56 16 2232 0.0 ± 0.0 >2048 18.3 ± 0.24 16 24.0 ± 0.41 8 2267 0.0 ± 0.0 >2048 14.0 ± 0.87 32 20.1 ± 0.14 16 2270 0.0 ± 0.0 >2048 17.8 ± 0.82 16 24.0 ± 0.58 8 2272 0.0 ± 0.0 >2048 17.5 ± 0.41 16 24.0 ± 0.41 8 2278 0.0 ± 0.0 >2048 13.8 ± 0.87 32 20.0 ± 0.61 16 2283 0.0 ± 0.0 >2048 14.1 ± 0.14 32 20.0 ± 0.41 16 2304 0.0 ± 0.0 >2048 17.7 ± 0.56 16 24.2 ± 0.47 8 2306 0.0 ± 0.0 >2048 22.0 ± 0.41 8 28.2 ± 0.24 4 2308 0.0 ± 0.0 >2048 18.1 ± 0.14 16 24.3 ± 0.24 8 2310 0.0 ± 0.0 >2048 18.4 ± 0.27 16 23.7 ± 0.24 8 2311 0.0 ± 0.0 >2048 14.0 ± 0.41 32 20.3 ± 0.24 16 2313 0.0 ± 0.0 >2048 18.1 ± 0.14 16 24.3 ± 0.00 8 2317 0.0 ± 0.0 >2048 21.9 ± 0.29 8 28.5 ± 1.10 4 2319 0.0 ± 0.0 >2048 13.8 ± 0.87 32 20.2 ± 0.47 16 2330 0.0 ± 0.0 >2048 22.1 ± 0.14 8 28.7 ± 0.47 4 2445 0.0 ± 0.0 >2048 18.2 ± 0.85 16 24.1 ± 0.14 8 ATCC 90030 0.0 ± 0.0 >2048 18.6 ± 0.13 16 24.2 ± 0.20 8

MIC50 >2048 MIC50 = 16 MIC50 = 8 MIC90 >2048 MIC90 = 32 MIC90 = 16 G-MIC >2048 G-MIC = 16.45 G-MIC = 8.22

Note: αAE-Na, α-aescin sodium; βAE-C, β-aescin crystalline; βAE-Na, β-aescin sodium; DD, disk diffusion method; BMD, broth microdilution method; IZD, inhibition zone diameter; MIC, minimum inhibitory concentration; MIC50, MIC for 50% of the isolates; MIC90, MIC for 90% of the isolates, G-MIC, MIC geometric mean; ATCC 90030, Candida glabrata reference strain. recommended by the Clinical and Laboratory Standards Results Institute [13]. The MIC of NYT was determined alone The in vitro antifungal activities of αAE-Na, βAE-C and or in the presence of a fixed subinhibitory concentration βAE-Na were determined against 25 C. glabrata strains, (0.5×MIC) of βAE-C or βAE-Na. The testing was carried including 24 clinical isolates and one C. glabrata ATCC out in sterile flat-bottomed 96-well microtiter plates with an 90030 reference strain. The results of the disk diffusion inoculum concentration of 0.5 × 103 to 2.5 × 103 CFU/ml (DD) and broth microdilution (BMD) methods are shown and RPMI 1640 medium buffered to pH 7.0 with 0.165 M in Table 1 and Figure 1. Generally, results obtained by morpholinepropanesulfonic acid (Sigma-Aldrich, St. Louis, DD assay were consistent with those determined by BMD USA). The final antifungal agents concentrations ranged method; an increase of MIC values was accompanied by from 512 to 0.125 μg/ml. The minimal inhibitory concen- a decrease of inhibition zone diameters and vice versa.In trations (MICs) were determined visually following 24 h of our study, αAE-Na did not exhibit any antifungal activity incubation at 35◦C and were defined as the lowest concen- against the yeasts tested. Lack of activity of this compound tration of the drug that inhibited the growth of yeast cells, correlated very well with the absence of inhibition zones as indicated by the absence of turbidity. Drug-free purity in DD method. By contrast, both β-aescins (βAE-C and controls and growth controls were included for each exper- βAE-Na) effectively inhibited the growth of all C. glabrata iment. MIC values were calculated as the average means of strains with a zone diameter ranging from 13.8 to 22.1 mm three independent experiments. 848 Medical Mycology, 2015, Vol. 53, No. 8 Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021

Figure 1. Disk diffusion results of α-aescin sodium (αAE-Na), β-aescin crystalline (βAE-C), β-aescin sodium (βAE-Na) and nystatin (NYT) for two selected Candida glabrata isolates. for βAE-C and from 20.0 to 28.7 mm for βAE-Na. MIC proved their antimicrobial activities when used alone or in values of βAE-C ranged from 8.0 to 32.0 μg/ml, while MICs combination with traditional antibiotics [10,17]. Saponins at which 50% and 90% of the isolates tested were inhibited synthesized constitutively by different plants represent one

(MIC50 and MIC90) were 16 and 32 μg/ml, respectively. On of the most common and widespread phytoanticipins in- the other hand, βAE-Na displayed slightly stronger anti- volved in plant defense through their antimicrobial po- candidal activity with MICs ranged from 4.0 to 16.0 μg/ml tency [18–20]. According to the literature, saponins ex- and MIC50 and MIC90 of 8 and 16 μg/ml, respectively. The tracted from different plants display important antifungal geometric mean MIC values were 16.45 and 8.22 μg/ml activity against both yeasts and filamentous fungi [17,21– for βAE-C and βAE-Na, respectively. The further studies 25]. In a study of Zhang et al., [17] saponins extracted were designed to evaluate the influence of subinhibitory from Tribulus terrestris L. were found to be very ef- concentrations (0.5×MIC) of βAE-C or βAE-Na on the fective against fluconazole-resistant pathogenic yeasts, in- nystatin MIC for the each strains tested. As shown in Ta- cluding C. glabrata. Similar results were reported by Du ble 2,thein vitro MIC values of NYT alone varied from et al., [23] who demonstrated antifungal efficacy of triter- 1.0 to 32.0 μg/ml, while those determined in the presence penoid saponins with hederagenin or oleanolic acid as agly- of sub-MIC of β-aescins were slightly lower ranging from cones against various yeast species, including C. glabrata 0.125 to 8.0 μg/ml for βAE-C and from 0.25 to 8.0 μg/ml as well as dermatophytes such as Microsporum canis and for βAE-Na depending on the strains tested. The results ob- Trichophyton mentagrophytes. The findings reported by tained indicate that sub-MIC of βAE-C reduced 2- to 16- Tsuzuki et al. [24] confirmed the antifungal activity of fold MICs of NYT. On the other hand, sub-MIC of βAE-Na saponins derived from Sapindus saponaria against non- resulted in 2- to 4-fold reduction of nystatin MICs. Candida albicans species, especially C. parapsilosis.The studies mentioned above inspired us to investigate the antifungal activities of natural compounds extracted from Discussion the seeds of horse chestnut (Aesculus hippocastanum L.) The prevalence of infections caused by antibiotic-resistant known under the collective name aescin. It is a mixture microorganisms has increased dramatically over the past of triterpenoid saponins consisting of two forms α-and few decades. The failure of the conventional antimicro- β, which vary mainly by solubility in water, haemolytic bial treatment is now the most important challenge to re- index, melting point and specific rotation [26]. The main searchers and pharmaceutical industry. Therefore, there is component of the horse chestnut extract is β-aescin (Fig. 2) an urgent need for new drugs and antimicrobial strate- that was shown to exhibit a broad spectrum of pharmaco- gies. More and more studies have focused on the possi- logical properties, including antiinflammatory, analgesic, bility of using natural plant-derived substances in antimi- and antipyretic activities. Extract from seeds and bark of crobial therapy. In fact, plant extracts are regarded as the the Aesculus hippocastanum L. (Hippocastanaceae) is to- invaluable sources of secondary metabolites endowed with day used as herbal medicine in Europe against chronic ve- a variety of pharmacological properties [14–16]. A consid- nous insufficiency (CVI; i.e., varicose veins accompanied erable number of plant originated compounds have already with pain, oedema, pruritus, and a sense of heaviness). Franiczek et al. 849

Ta b l e 2 . MIC values of nystatin (NYT) alone or in the presence of subinhibitory concentration (0.5×MIC) of β-aescin crystalline (βAE-C) or β-aescin sodium (βAE-Na) for Candida glabrata isolates.

MIC (μg/ml)

NYT+βAE-C (fold-change NYT+βAE-Na (fold-change Strain NYT alone relative to MIC of NYT alone) relative to MIC of NYT alone)

973 2 0.5 (4) 1 (2) 2122 2 0.5 (4) 1 (2) 2124 1 0.125 (8) 0.25 (4) 2129 8 0.5 (16) 2 (4) 2136 2 0.5 (4) 0.5 (4) 2165 16 4 (4) 8 (2) Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 2222 2 0.5 (4) 0.5 (4) 2228 1 0.125 (8) 0.25 (4) 2232 1 0.5 (2) 0.5 (2) 2267 32 8 (4) 8 (4) 2270 4 1 (4) 2 (2) 2272 1 0.5 (2) 0.5 (2) 2278 2 1 (2) 1 (2) 2283 2 0.5 (4) 0.5 (4) 2304 4 2 (2) 1 (4) 2306 1 0.125 (8) 0.5 (2) 2308 16 8 (2) 4 (4) 2310 4 2 (2) 2 (2) 2311 2 0.5 (4) 1 (2) 2313 2 0.5 (4) 0.5 (4) 2317 1 0.125 (8) 0.5 (2) 2319 4 1 (4) 1 (4) 2330 16 4 (4) 8 (2) 2445 2 0.5 (4) 0.5 (4) ATCC 90030 1 0.25 (4) 0.5 (2)

Note: NYT, nystatin; βAE-C, β-aescin crystalline; βAE-Na, β-aescin sodium; MIC, minimum inhibitory concentration.

Figure 2. The structure of β-aescin.

Administration is oral or topical. Recommended oral dose β-aescin sodium against 24 clinical isolates of C. glabrata is 50 mg aescin twice a day [26–28]. and C. glabrata ATCC 90030 reference strain. The suscep- However, studies focusing on the antifungal activities tibility examination was carried out using two methods the of α-andβ-aescins, as well as on their interactions with disk diffusion (DD) and broth microdilution (BMD). Im- drugs routinely used in the treatment of fungal diseases portantly, there was very good agreements between CLSI are very scarce [29,30]. Therefore, the present study aimed DD and BMD assays. Our results showed that α-aescin to investigate the antifungal effects of three aescin com- sodium was completely ineffective against the strains stud- pounds, including α-aescin sodium, β-aescin crystalline and ied. On the other hand, both β-aescins tested were found to 850 Medical Mycology, 2015, Vol. 53, No. 8 inhibit the growth of the yeasts tested. Moreover, β-aescin 4. Asmundsdottir LR, Erlendsdottir H, Gottfredsson M. National sodium exhibited stronger anticandidal activity compared study of candidemia, antifungal use, and antifungal drug re- with β-aescin crystalline. sistance in Iceland, 2000 to 2011. J Clin Microbiol 2013; 51: Previous studies have shown that polyene antibiotics, 841–848. 5. Bodey GP, Mardani M, Hanna HA et al. The epidemiology of such as nystatin and aescin share a common mode of ac- Candida glabrata and Candida albicans fungemia in immuno- tion. Both compounds exert their effects by interacting with compromised patients with cancer. Am J Med 2002; 112: 380– sterol molecules within the fungal cell membrane, lead- 385. ing to the formation of transmembrane pores [31]. This 6. Pfaller MA, Moet GJ, Messer SA et al. Candida bloodstream in- process disturbs the cellular semi-permeability and leads fections: comparison of species distributions and antifungal re- to the death of the fungal cell. Nystatin is able to form sistance patterns in community-onset and nosocomial isolates in small barrel-shaped pores with a radius of approximately the SENTRY Antimicrobial Surveillance Program, 2008–2009. Antimicrob Agents Chemother 2011; 55: 561–566.

0.42 nm in fungal membrane whose permeability is limited Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 7. Nawrot U, Pajaczkowska M, Fleischer M et al. Candidaemia to monovalent cations and small molecules [32]. By con- in Polish hospital – a multicenter survey. Mycoses 2013; 56: trast, aescin pores, with a radius of 2.56 nm, are permeable 576–581. to larger molecules, such as nucleotides and drugs [31]. In 8. Silva S, Negri M, Henriques M et al. Candida glabrata, Can- the present study, β-aescin crystalline as well as β-aescin dida parapsilosis and Candida tropicalis: biology, epidemiol- sodium at their subinhibitory concentrations decreased the ogy, pathogenicity and antifungal resistance. FEMS Microb Rev nystatin MIC values for all C. glabrata isolates tested. It is 2012; 36: 288–305. interesting to note that nystatin MIC reduction was greater 9. Martins CVB, da Silva DL, Neres ATM et al. Curcumin as a promising antifungal of clinical interest. J Antimicrob for β-aescin crystalline although this compound displayed Chemother 2009; 63: 337–339. lower antifungal potency against the strains studied rela- 10. Amber K, Aijaz A, Immaculata X et al. Anticandidal effect of β tive to -aescin sodium. Our findings revealed that MICs of Ocimum sanctum essential oil and its synergy with fluconazole nystatin–β-aescin combination were lower compared with and ketoconazole. Phytomedicine 2010; 17: 921–925. those obtained for the individual compounds. These results 11. Damke E, Tsuzuki JK, Cortez DA et al. In vivo activity of Sapin- may confirm a synergistic or additive effect of nystatin and dus saponaria against azole-susceptible and -resistant human β-aescin against the C. glabrata strains tested, however, vaginal Candida species. BMC Complement Altern Med 2011; further studies are needed to explain the exact interaction 11, 35. 12. Clinical and Laboratory Standards Institute. Method for anti- between these agents. fungal disk diffusion susceptibility testing of yeasts. Approved In conclusion, the findings of the present study authenti- guideline, M44-A. Wayne, PA: CLSI; 2004. β cated the effective in vitro anticandidal activity of -aescin. 13. Clinical and Laboratory Standards Institute. Reference method In addition, nystatin–β-aescin combination may be a useful for broth dilution antifungal susceptibility testing of yeasts: Ap- strategy to potentiate the effect of conventional antifungal proved guideline, M27-A3. Wayne, PA: CLSI; 2008. therapy. 14. Lee MW, Kim SU, Hahn DR. Antifungal activity of modified hederagenin glycosides from the leaves of Kalopanax pictum var. chinense. Biol Pharm Bull 2001; 24: 718–719. Declaration of interest 15. Abad MJ, Ansuategui M, Bermejo P. Active antifungal substances from natural sources. Arkivoc 2007; 7: 116–145. The authors report no conflicts of interest. The authors 16. Abreu AC, McBain AJ, Simoes˜ M. Plants as sources of new alone are responsible for the content and the writing of the antimicrobials and resistance-modifying agents. Nat Prod Rep paper. 2012; 29: 1007–1021. 17. Zhang JD, Cao YB, Xu Z et al. In vitro and in vivo antifungal activities of eight steroid saponins from Tribulus terrestris L. with References potent activity against fluconazole-resistant fungal pathogens. 1. Hachem R, Hanna H, Kontoyiannis D et al. The changing epi- Biol Pharm Bull 2005; 28: 2211–2215. demiology of invasive candidiases: Candida glabrata and Can- 18. Potter DA, Kimmerer TW. Inhibition of herbivory on young dida krusei as the leading causes of candidemia in hematologic holly leaves: evidence for the defensive role of saponins. Oecolo- malignancy. Cancer 2008; 112: 2493–2499. gia 1989; 78: 322–329. 2. Pereira GH, Muller¨ PR, Szeszs MW et al. Five-year evaluation of 19. Hostettmann KA, Marston A. Saponins. Chemistry and Pharma- bloodstream yeast infections in a tertiary hospital: the predomi- cology of Natural Products. 1st ed. Cambridge, UK: Cambridge nance of non-C. albicans Candida species. Med Mycol 2010; 48: University Press, 1995. 839–842. 20. Osbourn AE. Saponins in cereals. Phytochemistry 2003; 62:1–4. 3. Oberoi JK, Wattal C, Goel N et al. Non-albicans Candida species 21. Segal R, Schlosser¨ E. Role of glycosidases in the membran- in blood stream infections in a tertiary care hospital at New lytic, antifungal action of saponins. Arch Microbiol 1975; 22: Delhi, India. Indian J Med Res 2012; 136: 997–1003. 147–150. Franiczek et al. 851

22. Gruiz K. Fungitoxic activity of saponins: practical use and of postoperative adhesions. World J Surg 2005; 29: 1614– fundamental principles. Adv Exp Med Biol 1996; 404: 1620. 527–534. 28. Persson IAL, Persson K. Horse chestnut (Aesculus hippocas- 23. Du Z, Zhu N, Ze-Ren-Wang-Mu N et al. Two new antifungal tanum L.). Recent Prog Med Plants 2010; 28: 159–171. saponins from the Tibetan herbal medicine Clematis tangutica. 29. Krause H, Wienert V. Fungistatic effect of aescin on dermato- Planta Med 2003; 69: 547–551. phytes in vitro. Arzneimittelforschung 1970; 5: 703–705 [in Ger- 24. Tsuzuki JK, Svidzinski TI, Shinobu CS et al. Antifungal activity man]. of the extracts and saponins from Sapindus saponaria L. An 30. Omelkova J, Slavikova E, Vadkerti A. Inhibiting effect of beta- Acad Bras Cienc 2007; 79: 577–583. escine on the growth of yeasts and yeast microorganism. Biologia 25. Coleman JJ, Okoli I, Tegos GP et al. Characterization of plant- 1990; 45: 925–929 [in Slovak]. derived saponin natural products against Candida albicans. ACS 31. Morton MJ, Main MJ. Use of escin as a perforating agent on Chem Biol 2010; 5: 321–332. the Ion Works quattro automated electrophysiology platform. J 26. Sirtori CR. Aescin: pharmacokinetics and therapeutic proﬁle. Biomol Screen 2013; 18: 128–134. Pharmacol Res 2001; 44: 183–193. 32. Akaike N, Harata N. Nystatin perforating patch recording and Downloaded from https://academic.oup.com/mmy/article/53/8/845/981745 by guest on 29 September 2021 27. Fu F, Hou Y, Jiang W et al. Escin: inhibiting inﬂammation its applications to analyses of intracellular mechanisms. Jpn J and promoting gastrointestinal transit to attenuate formation Physiol 1994; 44: 433–473.