Keywords
Xanthoria parietina
- Teloschistaceae - parietin - aPDT - photoantimicrobial - EUCAST -
Candida auris
- Metschnikowiaceae
Abbreviations
aPDT:
antimicrobial photodynamic therapy
CDC:
Centers for Disease Control and Prevention
cdr1p:
Candida drug resistance protein 1
cdr1p:
Candida drug resistance protein 2
CLSM:
confocal laser scanning microscopy imaging
DLS:
dynamic light scattering
DMSO:
dimethyl sulfoxide
ECDC:
European Centre for Disease Prevention and Control
EUCAST:
European Committee on Antimicrobial Susceptibility Testing
HIV+:
human immunodeficiency virus positive
MB:
methylene blue
MDA:
malondialdehyde
mdr1p:
multidrug resistant 1 protein
MIC:
minimal inhibitory concentration
NCAC:
non-Candida albicans Candida species
NICD:
national Institute for Communicable Diseases
PAHO:
Pan American Health Organization
PS:
photosensitiser
RB:
rose bengal
RES:
reactive electrophilic species
ROS:
reactive oxygen species
TBO:
toluidine blue O
WHO:
World Health Organization
Introduction
Members of the genus Candida are the most common causative agents of fungal infections [1], [2], representing a severe risk to human health [2], [3]. A relatively new human pathogenic species of this genus, i.e., Candida auris, was first isolated from a patientʼs ear in Japan in 2009 [4]. Since then, reports of other C. auris infections rapidly occurred worldwide [5]. Infections are associated with crude mortality rates ranging from 28 to 66% [6], as physicians are confronted with a fungus that shows high rates of resistance
against first-line and emergency therapy [7]. Furthermore, treatments are impeded by diagnostic challenges in species identification.
Since 2016, governmental institutions (the Centers for Disease Control and
Prevention (CDC), European Centre for Disease Prevention and Control (ECDC), World
Health Organization (WHO), Pan American Health Organization (PAHO), and National Institute
for Communicable Diseases (NICD)) have stressed the clinical relevance of C. auris infections to healthcare facilities and issued interim guidelines for clinical management,
laboratory testing, and infection control [6], [8], [9]. However, the so-called “superbug” C. auris
[10] is not the only common human pathogen with reported resistance development [11], [12]. Also, non-Candida albicans Candida species (NCAC) such as Candida glabrata and Candida tropicalis
[1], [8], as well as the non-Candida yeast Cryptococcus neoformans, a
pathogenic fungus spread through airborne spores, raise concerns [8], [13]. Recently, C. auris, C. albicans, and Cryptococcus neoformans were listed in the critical priority group of the WHOʼs fungal priority pathogens
list to guide research, development, and public health action. C. glabrata and C. tropicalis were considered high-priority targets [8].
Increased use of antifungal drugs causes acquired resistance against these agents
on top of the inherent primary resistance, thus making Candida species highly resistant to existing antifungal agents [11]. The number of antifungal agents is limited compared to available antibacterial
agents, and most antifungal drugs are only fungistatic [12]. Although no improvement in survival rate was found, fungicidal therapy of invasive
Candida infections leads to a higher chance of therapeutic success and decreases the risk
of recurrent infections, compared to fungistatic therapy alone [14].
Antimicrobial photodynamic therapy (aPDT) is an alternative approach to antifungal
therapy with fungicidal properties. It is based on a different mechanism of action
than established antifungal drugs [15]: the antifungal substance is inactive under dark conditions, but when exposed to
visible light of corresponding wavelengths, it is activated and destroys pathogenic
target cells. This approach, based on the synergistic effect of light and a chromophore,
succeeds in killing multidrug-resistant microorganisms by causing multi-target damage
through reactive oxygen species (ROS) [16]. According to the current state of knowledge, it can be assumed that aPDT is unlikely
to promote relevant resistance in microorganisms [17], [18], [19], [20], making this approach a promising treatment option in the fight against
multidrug-resistant pathogens. With the help of a novel photoantimicrobial high-throughput
screening (HTS) based on the European Committee on Antimicrobial Susceptibility Testing
(EUCAST), our group recently reported the potential of fungal extracts containing
the photosensitiser parietin to eradicate Staphylococcus aureus and Candida albicans
[15]. Earlier reports showed parietinʼs photoantimicrobial effect [21].
The aim of this study is to evaluate the potential of parietin as a photosensitiser
(PS) against human pathogenic yeasts and to draw a comprehensive picture of its physicochemical
and antimicrobial properties.
Results
Parietin (CAS 521 – 61 – 9) was isolated from freeze-dried samples of Xanthoria parietina by sonification in acetone and purified via a short sequence of chromatographic methods.
The metabolite was obtained as yellow needles with a yield of 0.043% (see web-only
Supplementary Figs. 1S and 2S).
Frozen stock solutions of parietin in DMSO in lower concentrations undergo a decay
over 29 days, while it is stable in higher concentrations: at a concentration of c = 31.25 mg/L
(109.94 µM), the highest decay (21.5%) was found. The lowest decay (7.39%) was determined
for a concentration of c = 125.00 mg/L (439.74 µM) (see web-only Supplementary Fig. 3S). Photostability experiments (see web-only Supplementary Fig. 4S) revealed a half-life of parietin (2.50 mg/L, 8.80 µM, DMSO) under light irradiation
(λ = 430 nm, 0.6785 mW) of t1/2 = 133 min (H = 5.41 J/cm2). A singlet oxygen photoyield (φΔ
) of 94% was determined for parietin in deuterated methanol (see web-only Supplementary
Fig. 5S).
Initial light toxicity experiments revealed a stronger effect of blue light alone
on the viability of C. auris and C. tropicalis (60 – 80%), while the effect on other species was moderate (20 – 40%, [Fig. 1]). Intriguingly, C. auris showed a distinctive intra-strain susceptibility: strain B1 showed the highest sensitivity
(82% reduced population); strain B2 instead was only moderately affected (23%) by
blue light (λ = 428 nm, H = 30 J/cm2).
Fig. 1 Graphical visualisation of the light toxicity, meaning the effect of light irradiation
without PS on the growth of microorganisms (λ = 428 nm, H = 30 J/cm2).
The average effect of parietin applied in the dark was negligible (< 10% growth inhibition,
cmax = 1.250 mg/L, [Fig. 2]). However, when activated by light, parietin induced a strong inhibition of growth
(> 90%) in all five yeast species at a concentration of c = 1.250 mg/L (4.397 µM)
([Table 1], [Fig. 2]). For most strains, even smaller concentrations were sufficient (see web-only Supplementary
Fig. 6S). For C. glabrata, a concentration of c = 0.625 mg/L (2.199 µM) led to an inhibition of growth of 91.1%
(see web-only Supplementary Fig. 7S), and for C. auris, a parietin concentration of c = 0.313 mg/L (1.099 µM) killed 92.8% of the population
(see web-only Supplementary Fig. 6S). Concentrations of c = 0.157 mg/L (0.550 µM) inhibited growth by 93.2% and 96.0%
in C. tropicalis and Cryptococcus neoformans, respectively (see
web-only Supplementary Figs. 6S and 7S). Two of the Candida albicans strains (A1 and A2) showed the highest survival rates of all tested microorganisms
(see web-only Supplementary Figs. 8S and 9S). Interestingly, for strain A1, only a growth inhibition of 82.6% was obtained with
concentrations up to c = 1.250 mg/L (4.397 µM) (see web-only Supplementary Figs. 8S and 9S). Concentrations of c = 2.500 mg/L (8.795 µM) showed an adverse effect against all
tested yeast populations besides strains of Cryptococcus neoformans and C. tropicalis (see web-only Supplementary Figs. 9S-13S).
Fig. 2 Inhibition of growth achieved by parietin-based aPDT (1.250 mg/L, 4.397 µM) against
five different species of yeast, averaged from three different strains per species.
The grey bars represent the growth of the controls treated with parietin in the dark;
blue bars denote parietin treatment in combination with light (λ = 428 nm, H = 30 J/cm2, PI = 30 min).
Table 1 Overview of the antifungal effect of parietin under irradiation (λ = 428 nm, H = 30 J/cm2) against the tested strains, *For Candida albicans (A1) with a concentration of 1.250 mg/L (4.40 µM), only an inhibition of growth of
83% was observed.
Work ID
|
Species
|
Parietin [mg/L] for aPDT inhibition of growth > 90%
|
Parietin [µM] for aPDT inhibition of growth > 90%
|
A1
|
Candida albicans
|
> 1.250*
|
> 4.397*
|
A2
|
0.625
|
2.199
|
A3
|
0.625
|
2.199
|
B1
|
Candida auris
|
0.157
|
0.550
|
B2
|
0.625
|
2.199
|
B3
|
0.313
|
1.099
|
C1
|
Candida tropicalis
|
0.313
|
1.099
|
C2
|
0.039
|
0.137
|
C3
|
0.039
|
0.137
|
G1
|
Candida glabrata
|
1.250
|
4.397
|
G2
|
0.625
|
2.199
|
G3
|
0.313
|
1.099
|
K1
|
Cryptococcus neoformans
|
0.078
|
0.275
|
K2
|
0.156
|
0.550
|
K3
|
0.078
|
0.275
|
Experiments with C. albicans (A1) showed that a longer pre-irradiation time of t = 60 min could not further increase
the uptake of parietin ([Fig. 3]). HPLC-DAD quantification of excess, wash, and uptake fractions allowed detailed
tracking of parietin distribution.
Fig. 3 Uptake of parietin (1.25 µg/mL/4.40 µM) by Candida albicans (A1) after pre-irradiation times of 1/10/20/30/60 min. Parietin uptake is shown in
yellow; parietin excess is shown in white. Turquoise sections represent the amount
of parietin removable through washing in PBS.
CLSM imaging of C. albicans (A3) treated with 0.625 mg/L (2.199 µM) parietin (RPMI, 0.5% DMSO) confirmed that
the cells associated with parietin ([Fig. 4 a, b]). In addition, in a burst cell and remaining cell debris, parietin was located in
the outer layers ([Fig. 4 c, d]).
Fig. 4 Confocal laser scanning microscope (CLSM) imaging of Candida albicans (A3) treated with 0.625 mg/L (2.199 µM) parietin (RPMI, 0.5% DMSO). (a) and (c) bright field images. (b) and (d) excitation was generated with an argon laser (λ = 466 nm) and emission collected above λ>505 nm and false-colorised red to visualise parietin distribution. (a) and (b) intact cell; (c) and (d) burst cell.
Quantification of reactive electrophilic species (RES) and aldehydes, both typical
breakdown products of lipid peroxides, after irradiation experiments (λ = 428 nm, H = 30 J/cm2, PI = 30 min) against Candida albicans (A3) with the PS parietin (0.625 mg/L, 2.199 µM, in H2O w/0.5% DMSO) showed a 10.6 × 102-fold increase in 2-pentenal, compared to dark samples. Furthermore, increased values
for acrolein (5.2 × 102-fold), butyraldehyde (2.8 × 102-fold), hexanal (2.7 × 102-fold), malondialdehyde (MDA, 4.2 × 102-fold), 2-hexenal (3.5 × 102-fold), 2-nonenal (5.4 × 102-fold), and 4-hydroxynonenal (5.7 × 102-fold) were found (see web-only Supplementary Figs. 15S and 16S).
DLS experiments were performed in settings simulating the aPDT experiment (i.e., similar
concentration range and medium). Up to the concentration of 1.25 mg/L (4.40 µM), supramolecular
aggregates of 2068.6 nm ± 14.0% were found. Higher concentrations of parietin in RPMI
with 0.5% DMSO led to larger particles (10.00 mg/L, 29 060.0 nm ± 7.7%) (see web-only
Supplementary Fig. 17S). Rodlike parietin formations could be observed by fluorescence microscopy starting
at concentrations of 1.25 mg/L (4.40 µM) (see web-only Supplementary Fig. 18S).
Discussion
Candida species are opportunistic pathogens causing vaginitis, oral candidiasis, cutaneous
candidiasis, candidemia, and systemic infections especially in immunosuppressed, diabetic,
or HIV+ patients, but they can also affect healthy people [11]. Local infections, such as mucosal candidiasis, are generally easier to treat than
systemic infections such as candidemia, which is one of the most common bloodstream
infections in hospitals [22]. However, increasing numbers of drug-resistant Candida spp. infections are hampering the treatment of local infections, particularly in
immunosuppressed patients. Photodynamic inhibition, or aPDT, is a topical treatment
that has been shown to be successful and safe in clinical trials [23]. For example, in a recent study, the combined treatment of methylene blue, potassium
iodine, and light was shown to improve the clinical course of oral
Candida infections in AIDS patients [24]. In this clinical trial, up to 600 µM of the drug methylene blue (MB, 191.91 mg/L)
and a pre-irradiation time of five minutes were utilised [24].
The pre-irradiation time is thought to have no effect on the efficiency of aPDT treatments
against bacteria [25]. For yeasts, however, the pre-irradiation time seems to be relevant and is considered
to correlate with the intracellularisation of the photosensitiser [15], [26]. Thus, the uptake of parietin was studied here in detail to determine the optimal
pre-irradiation time (tPI,opt) against C. albicans. Based on the observation that the internal content of parietin reaches its steady
state after 30 minutes ([Fig. 3]), an optimal pre-irradiation time (tPI,opt) of 30 minutes was defined. The identified tPI,opt correlates to earlier reports [15], [27], where tPI,opt was determined based on the maximum log reduction reached. There are many putative
reasons
for this time dependency [27]. For example, the involvement of efflux pumps (Cdr1p and Cdr2p) has been proposed
[11]. However, according to this rationale, a clear difference between susceptible and
resistant strains at the defined tPI should be observed. In this study, as in others [28], no such difference was observed: the fluconazole-resistant C. albicans strain A1 (ATCC 64 550) was as susceptible as the non-resistant strain A2 (ATCC 90 028);
see Fig. 8S (Supporting Information). Next to efflux pumps, the drug import may be mediated through
transporters [29] or passive diffusion. However, the number of studies on drug uptake in fungal pathogens
is limited. One described drug transporter is the mdr1p transporter, which is, however,
also involved in fluconazole resistance [30].
Applying the identified tPI,opt and the modified EUCAST protocol [15], promising PhotoMIC90 values were defined for parietin: for the most robust species, Candida albicans, inhibition of growth occurred at c = 1.250 mg/L (4.397 µM), whilst up to a concentration
of c = 400 mg/L (1407.162 µM) [31], no activity was reported under dark conditions. Photoantimicrobial inhibition (0.156 mg/L,
0.549 µM) against C. tropicalis was over 600-fold smaller than reported under standard conditions (100 mg/L, 351.79 µM,
[31]), emphasising the synergistic effect of light and PS. For Cryptococcus neoformans (12.50 mg/L, 43.97 µM, [31]), the photodynamic effect of parietin under irradiation (0.16 mg/L, 0.55 µM) led
to an 80-fold amplification of antimicrobial activity. For C. auris and C. glabrata, no published data about the
antimicrobial activity of parietin could be found. In this study, we could show that
under light irradiation > 90%, growth inhibition was observed at c = 0.313 mg/L (1.10 µM)
against Candida auris and at c = 0.625 mg/L (2.20 µM) against Candida glabrata, while in the dark and with the same concentrations, no activity > 10% was recorded.
The EUCAST protocol [32] defines 0.5% DMSO as the final concentration for antimicrobial susceptibility testing,
thus limiting the maximum concentration and therewith the possibility to determine
an MIC under dark condition in this study due to solubility issues.
Parietin is a potent singlet oxygen producer (φΔ
= 94%, d4-MeOH), thus acting as a PDT type II photosensitiser, while slowly degrading
under light irradiation (Fig. 4S, Supporting Information). Due to the lipophilic nature of parietin, it likely localised
to the plasma membrane, in which the release of 1O2 could lead to lipid peroxidation. Confocal laser scanning microscopy imaging indicated
a cellular accumulation of parietin at the border layers of the cell ([Fig. 4]). Unharmed, normal-shaped Candida albicans cells in the dark control samples were found with a size of 6 µm (see web-only Supplementary
Fig. 14S). Another anthraquinone-like photosensitiser (i.e., aloe-emodin) was reported to
cause photodynamic damage through ROS to the cell envelope of C. albicans
[33], corroborating the hypothesis of impairment of the cell membrane function as a mode
of
photoantimicrobial action. To test the hypothesis, analysis of 1O2-induced photooxidation products, so-called reactive electrophilic species (RES) and
aldehydes, after aPDT irradiation were performed in this study. Oleic acid (26.0 ± 3.0%),
linoleic acid (30.0 ± 3.0%), and linolenic acid (8.0 ± 3.0%) are among the most relevant
unsaturated fatty acids in C. albicans and together make up 64% of total fatty acids in plasma membranes of yeast [34]. Their photooxidation products acrolein, 2-pentenal, hexanal, 2-hexenal, and 2-nonenal
(see web-only Supplementary Figs. 15S and 16S) were indeed increased after PDT treatment. Furthermore, the common lipid oxidation
markers [34], [35], [36], [37] MDA (4.22 × 102-fold increase), 4-hydroxynonenal (5.66 × 102-fold increase), and
butyraldehyde (2.77 × 102-fold increase) were significantly enhanced, confirming the hypothesis of the lipid
oxidation being the mode of action.
Other groups have reported on the putative anticancer activity of parietin; for example,
apoptosis and autophagy were induced by parietin in cells of a cervical cancer cell
line (i.e., HeLa) with an EC50 between 80 and 160 µM (22.73 and 45.45 mg/L, respectively) in the dark [38]. In contrast, a PhotoEC50 of c = 30 µM was found for treatment (1 h) with parietin under blue light irradiation
(λ = 405 nm, 3.08 J/cm²) against IGROV-1, another human ovarian cancer cell line [39]. However, as these reports are based on monolayer cell cultures, they are of limited
relevance in estimating potential adverse effects of, e.g., skin lesion treatments.
For example, a related chemical entity (i.e., the anthraquinone aloe-emodin) has recently
been shown to be a potent photoantifungal agent against Trichophyton rubrum in an in vivo guinea pig model [40].
Investigations of the treated skin revealed no harmful side effects [40], despite several reports of aloe emodinʼs in vitro photocytotoxicity (e.g., [41], [42]). Thus, a straightforward estimation without a skin model based on the available
literature data is not possible.
The observed high photoactivity of parietin against the pathogenic species Candida auris is, nevertheless, of special interest. Recently, another group described the photoantimicrobial
effects of the well-established photosensitisers toluidine blue O (TBO, 400 µM/122.33 mg/L),
methylene blue (MB, 400 µM/127.94 mg/L), and rose bengal (RB, 400 µM/389.48 mg/L)
against C. auris biofilms, reporting smaller inhibition of growth while using higher concentrations
of the photosensitiser and stronger irradiation power (240 J/cm2) [43]. However, in the present study, the inhibition against planktonic cells was determined.
Nevertheless, a concentration of c = 0.625 mg/L (2.199 µM) killed 98.5% of a C. auris population (λ = 428 nm, H = 30 J/cm2, PI = 30 min) (see web-only Supplementary Fig. 6S), while 8 mg/L (25.01 µM) of MB were needed against planktonic cells (λ = 635 nm, 12 J/cm2,
PI = 60 min) [44]. Therewith, parietin is a promising new candidate to explore as a photosensitiser
against C. auris. Local infections of the ear, as caused by C. auris
[45], can be, for example, selectively treated utilising state-of-the-art light fibres,
activating parietin solely in the spatial area of the light beam [46].
In conclusion, the present research shows that parietin concentrations in the low
micromolar range (0.55 – 4.40 µM/0.16 – 1.25 mg/L) are effective in killing different
strains of the tested Candida spp. and Cryptococcus neoformans when combined with blue light (λ = 428 nm, H = 30 J/cm2). Furthermore, we present an experimental setup for the evaluation of the ideal pre-irradiation
time. For Candida albicans with parietin, an assumed optimal pre-irradiation time of tPI = 30 min could be confirmed. RES quantification showed the fungicidal effect of aPDT
through the apparent damage of cell membranes. The photosensitising properties and
aggregation behaviour of parietin in conditions of EUCAST antifungal susceptibility
testing were characterised. Our work supports the search for urgently needed novel
antimicrobials in the global fight against spreading resistance and shall offer new
options for photosensitiser selection and research
for aPDT. A major drawback of parietin is its limited solubility; however, future
studies will investigate liposomal formulations.
Materials and Methods
Parietin was isolated from thalli of Xanthoria parietina through acetone extraction, crystallisation, size exclusion chromatography, and preparative
HPLC (see web-only Supplementary Table 1S – 4S). Stability tests in DMSO were performed over 29 days in 7-day intervals. Photostability
was tested via kinetic full spectrum measurement with a photometer over 24 h. Near-infrared
emission measurements determined the singlet oxygen (1O2) yield (φΔ
). Dynamic light scattering of parietin samples were measured in concentrations up
to c = 10.00 mg/L (35.18 µM, RPMI with 0.5% DMSO). Photoantimicrobial broth microdilution
experiments were performed as previously described [15], based on the adjusted EUCAST protocol for antifungal susceptibility testing [32] using a set of five yeast species (i.e., Candida albicans, C. auris, C. glabrata, C. tropicalis, and Cryptococcus
neoformans; sources and further details are listed in Table 5S, Supporting Information), with three separate strains for each species. Uptake experiments
were undertaken by incubating C. albicans cells (McFarland standard No. 1.5) with parietin (c = 1.25 mg/L, 4.40 µM, RPMI, 0.5%
DMSO) at five different incubation times (1 min, 10 min, 20 min, 30 min, and 60 min).
The final quantification of parietin was done via HPLC-DAD measurements. For fluorescence
and confocal laser scanning microscopy, C. albicans cultures treated with parietin concentrations from c = 0.625 mg/L (2.20 µM) up to
c = 10.00 mg/L (35.18 µM) were prepared. For quantifying reactive electrophile species
(RES) and aldehydes after photoantimicrobial treatment, C. albicans suspensions adjusted to McFarland standard No. 2 and treated with parietin c = 1.25 mg/L
(4.40 µM, water, 0.5% DMSO) were utilised. After irradiation (λ = 428 ± 15 nm, H = 30 J/cm2,
tPI = 30 min), samples were analysed by HPLC-DAD/MS.
Contributorsʼ Statement
J. F. isolated and characterised parietin, performed all antifungal tests, contributed
significantly to all other experiments, analysed the data, and wrote the first draft;
T.R performed and analysed the RES measurements; A. H. performed and analysed the
CLSM experiments; Y. H. performed and analysed the singlet oxygen yield measurements;
FH contributed to the characterisation of parietin; L. D. contributed significantly
to the isolation of parietin; E. S. A. contributed to DLS measurement; H. S. developed
the irradiation device for the photostability measurements; S. B. supervised the singlet
oxygen measurements; F. L. supervised the DLS experiments; I. K. authenticated X. parietina, designed, supervised; ML designed and supervised the antimicrobial experiments,
discussed the photoantimicrobial activity experiments, and edited the manuscript;
BS designed and supervised, discussed, analysed data, and wrote and edited the manuscript.