Antifungal-resistant yeasts detected in recreational coastal waters of Rio de Janeiro, Brazil

Abstract

Background

Recreational coastal waters are widely used for leisure activities and require systematic microbiological monitoring to ensure the safety of bathers. However, such surveillance has traditionally focused on fecal indicator bacteria, whereas fungal communities remain largely neglected by regulatory and governmental agencies.

Aims

This study aimed to identify 18 yeasts isolated from three tourist beaches in Rio de Janeiro, Brazil, and to evaluate their antifungal susceptibility, biofilm-forming capacity and growth in saline conditions.

Methods

The molecular identification of the isolates was performed by sequencing the ITS1–5.8S–ITS2 rDNA region. Antifungal susceptibility was determined using the broth microdilution method. Biofilm formation was assessed in 96-well microtiter plates using crystal violet, safranin and XTT. Growth under different salt concentrations was also evaluated.

Results

Fungal isolates were identified as Candida parapsilosis species complex (n = 5), Candida tropicalis (n = 4), Nakaseomyces glabratus (formerly Candida glabrata; n = 2), Candidozyma haemuli species complex (formerly Candida haemulonii species complex; n = 2), Wickerhamiella infanticola (n = 2), and one isolate each of Candida palmioleophila, Candida ecuadorensis and Pichia manshurica. Resistance to azoles was observed in C. tropicalis, N. glabratus, C. haemuli and C. palmioleophila, with variable susceptibility to other antifungals. C. parapsilosis and C. ecuadorensis isolates were susceptible to all tested antifungals. W. infanticola isolates exhibited low MIC values against the antifungals tested, while P. manshurica exhibited elevated MICs to fluconazole and flucytosine. All yeasts formed biofilms on plastic and tolerated NaCl concentrations of up to 7.5%.

Conclusions

These findings demonstrate the presence of clinically relevant opportunistic yeasts in recreational seawater. The isolates exhibited antifungal resistance, salt tolerance, and the ability to form biofilms, all of which pose a potential public health concern.

Resumen

Antecedentes

Las aguas costeras recreativas se utilizan ampliamente para actividades de ocio y requieren un control microbiológico sistemático para garantizar la seguridad de los bañistas. Sin embargo, esta vigilancia se ha centrado tradicionalmente en bacterias indicadoras de contaminación fecal, mientras que el control de la presencia de hongos ha sido, en gran medida, desatendido por las agencias regulatorias y gubernamentales.

Objetivos

Este estudio tuvo como objetivo identificar 18 levaduras aisladas de tres playas turísticas de Río de Janeiro, Brasil, y evaluar su sensibilidad a los antifúngicos, su capacidad de formación de biopelícula y su crecimiento en condicionesde salinidad.

Métodos

La identificación molecular de los hongos aislados se realizó mediante la secuenciación de la región ITS1–5.8S–ITS2 del rDNA. La sensibilidad antifúngica se determinó mediante el método de microdilución en caldo. La formación de biopelículas se evaluó en placas de microtitulación de 96 pozos y el uso de cristal violeta, safranina y XTT. También se investigó el crecimiento bajo diferentes concentraciones de sal.

Resultados

Los aislamientos fúngicos se identificaron como Candida parapsilosis complejo de especies (n = 5), Candida tropicalis (n = 4), Nakaseomyces glabratus (anteriormente Candida glabrata; n = 2), Candidozyma haemuli complejo de especies (anteriormente Candida haemulonii complejo de especies; n = 2), Wickerhamiella infanticola (n = 2), y un aislamiento de cada una de las especies Candida palmioleophila, Candida ecuadorensis y Pichia manshurica. Se observó resistencia a los azoles en C. tropicalis, N. glabratus, C. haemuli y C. palmioleophila, con sensibilidad variable a otros antifúngicos. Los aislamientos de C. parapsilosis y C. ecuadorensis fueron sensibles a todos los antifúngicos evaluados. Los aislamientos de W. infanticola mostraron valores bajos de concentración minima inhibitoria frente a los antifúngicos probados, mientras que P. manshurica presentó valores de concentración minima inhibitoria elevados para el fluconazol y la flucitosina. Todas las levaduras formaron biopelículas sobre plástico y toleraron concentraciones de NaCl de hasta 7,5%.

Conclusiones

Estos hallazgos demuestran la presencia de levaduras oportunistas clínicamente relevantes en aguas marinas recreativas. Los aislamientos fueron resistentes a algunos antifúngicos, tuvieron tolerancia a ciertas concentraciones de sal y formaron biopelículas, cualidades que pueden suponer un potencial riesgo para la salud pública.

Recreational coastal waters play a crucial role in promoting leisure, tourism and well-being across the world. In this context, the city of Rio de Janeiro, Brazil, stands out for its extensive coastline, featuring numerous iconic beaches that attract not only local residents, but also millions of national and international tourists throughout the year.¹⁹ However, the intense and continuous use of these environments, combined with their close proximity to densely populated urban areas, subjects them to considerable anthropogenic pressure. Such impacts can compromise both the ecological balance and the microbiological quality of the sand and seawater.¹⁵ Traditionally, recreational water quality has been monitored based on the detection of classical fecal indicator bacteria, such as Escherichia coli and Enterococci, whose presence is associated with an increased risk of acquiring waterborne diseases through contact with contaminated water.²¹ Although this approach has proven effective for identifying bacterial contamination, it fails to account for other potentially harmful microorganisms, including opportunistic fungi, which are not currently considered in routine environmental assessments. As a result, a significant gap remains in the comprehensive evaluation of microbiological risks in these settings.²¹

The presence of fungi in recreational coastal waters has long been largely overlooked on a global scale. However, in recent years, growing attention has been directed toward the occurrence of fungi in these environments, particularly due to their potential impact on public health.^2,18,31 The emergence and spread of multidrug-resistant opportunistic fungal pathogens have further underscored the threat posed by fungal infections, especially in the context of medical advances that have extended the survival of individuals undergoing immunosuppressive therapies, organ transplants or chemotherapy.⁷ These individuals represent a growing at-risk population, highly susceptible to opportunistic fungal infections. Studies have reported the presence of opportunistic fungal pathogens in both sand and seawater samples taken from beaches in Brazil and other regions around the world.^{2,15,18,26,36} Among these, several current and former Candida species have been frequently identified. They are considered fecal contaminants, due to their presence in the gastrointestinal microbiota of humans and certain animals. This fecal origin, coupled with their environmental persistence, raises concerns regarding their potential for transmission and infection in exposed individuals. Clinically relevant species such as Candida tropicalis, Candida parapsilosis, Pichia kudriavzevii (formerly Candida krusei) and Nakaseomyces glabratus (formerly Candida glabrata) have been detected in the sand and seawater of Brazilian beaches, highlighting the importance of including fungi in routine environmental monitoring programs.^15,26,36 Worryingly, these studies have shown that C. tropicalis displays resistance to at least one antifungal agent and possesses the capacity to exhibit key virulence factors, including adhesion and biofilm formation, which are crucial for the initiation of an infectious process.^15,26,36

In response to growing concerns about fungal contamination in aquatic recreational environments, the Mycosands initiative was recently established by the joint working group of the International Society for Human and Animal Mycology (ISHAM) and the European Confederation for Medical Mycology (ECMM). The initiative aims to generate standardized data on the presence and diversity of fungi in beach sands, coastal waters and inland freshwater bathing sites across various European countries.² Findings from this large-scale study revealed that the most frequently isolated fungal genera in both sand and water samples included Aspergillus, Candida, Fusarium, and Cryptococcus, all of which encompass species known to cause opportunistic infections in humans. The widespread detection of these fungi in recreational settings underscores the need for a broader perspective in microbial water quality assessment. Notably, the results from the Mycosands initiative contributed to the consideration of fungi as a microbial taxon of interest in the Guidelines on Recreational Water Quality, published by the World Health Organization.³¹ This development marks an important step toward the inclusion of fungal indicators in future public health policies and monitoring protocols for recreational waters.

Considering all the aspects discussed above, the aim of the present study was to investigate the occurrence of yeast species in seawater from three recreational beaches in the city of Rio de Janeiro, Brazil. In addition, we evaluated their susceptibility to antifungal agents, their ability to grow under varying salt concentrations and their capacity to form biofilm on plastic support, which is an important virulence attribute closely associated with resistance to environmental stressors.

Material and methods

Microorganisms and growth conditions

In the present study, a total of 18 yeast isolates were obtained from seawater samples collected at three urban and touristic beaches in Rio de Janeiro city, each characterized by distinct levels of pollution. The sampling was conducted on June 20, 2017. The collection sites and their respective geographic coordinates (latitude, longitude) were: Botafogo (−22.944434 and −43.180056), Flamengo (−22.929164 and −43.170845) and Ipanema (−22.987266 and −43.205129). At each site, 2 L of seawater were collected 15 cm below the surface in a water column with a total depth of 100 cm. Samples were stored in sterile amber flasks, kept on ice, and processed within 8 h after collection. A volume of 500 mL from each sample was filtered using Millipore membranes (0.22-μm pore size), which were then placed on 90-mm culture plates containing Difco™ CHROMagar™ Candida (Becton, Dickinson and Company, Le Pont de Claix, France) supplemented with imipenem (1 μg/mL) to inhibit bacterial growth. The plates were incubated aerobically at 37 °C for 48 h. All colony-forming units (CFUs) were subsequently subcultured on Sabouraud dextrose agar (SDA; Sigma–Aldrich, St. Louis, MO, USA) under identical incubation conditions and further processed for molecular identification (as detailed below). For all subsequent experiments, yeast cells were cultured in Sabouraud dextrose broth (SDB; Sigma–Aldrich) at 37 °C for 48 h, and cell counts were estimated using a Neubauer chamber.

Molecular identification

After growth in SDA, fungal DNA obtained from pure colonies was extracted with the Gentra® Puregene® Yeast and G+ Bacteria Kit (Qiagen, Germantown, MD, USA). The identification of yeasts was done by sequencing the ITS1–5.8S–ITS2 gene region of the rDNA,⁴ using the primers ITS1 (50–TCCGTAGGTGAACCTGCGG–30) and ITS4 (50–TCCTCCGCTTATTGATATGC–30). PCR amplification was carried out in a C1000 Touch thermocycler (Bio–Rad, USA) following the program: an initial denaturation at 95 °C for 5 min, followed by 30 cycles of denaturation at 95 °C for 1 min, annealing at 55 °C for 1 min, and extension at 72 °C for 1 min, with a final extension step at 72 °C for 5 min. Amplification products were purified using the QIAquick PCR Purification Kit (Qiagen, Germany) and sequenced in both directions. Sequencing was conducted at the DNA Sequencing Platform (ABI-3730; Applied Biosystems) (PDTIS/FIOCRUZ – Rio de Janeiro, Brazil). Sequences were edited using SeqMan software, version 7.0 (DNASTAR Inc., Madison, WI, USA) and compared by BLAST with sequences available from the NCBI/GenBank database.

Antifungal susceptibility assay

The antifungal susceptibility assay was performed according to the broth microdilution method described in the document M27Ed4 published by the Clinical & Laboratory Standards Institute (CLSI).⁷ The antifungal agents fluconazole (in concentrations ranging 0.125–64 mg/L), itraconazole (0.0313–16 mg/L), voriconazole (0.0313–16 mg/L), amphotericin B (0.0313–16 mg/L), caspofungin (0.015–8 mg/L), flucytosine (0.125–64 mg/L) and terbinafine (0.0313–16 mg/L) (Sigma-Aldrich) were used.⁷ Minimal inhibitory concentration (MIC) values of yeasts with species-specific breakpoints were categorized in accordance with the document CLSI M60Ed2.⁸ For Candida species and antifungals not covered by this protocol, MIC classification was based on the CLSI M27S3 protocol.⁶ The CLSI protocol has not determined breakpoints for terbinafine and amphotericin B. However, in general, clinical isolates of Candida presenting MIC values greater than 1 mg/L are considered resistant to amphotericin B. The reference strains P. kudriavzevii (formerly C. krusei) ATCC 6258 and C. parapsilosis ATCC 22019 were used for quality control, as recommended by CLSI guidelines. For yeast species other than Candida for which no CLSI clinical breakpoints are available, MIC values were reported without categorical interpretation.

Biofilm formation assay

Fungal cell suspensions (200 μL of SDB containing 1 × 10⁶ yeasts) were added to the wells of a flat-bottom 96-well polystyrene microtiter plate and incubated without agitation at 37 °C for 48 h. Control wells with only medium were also included in parallel. Afterwards, supernatant fluids were gently aspirated, and wells were washed three times with phosphate-buffered saline (PBS, pH 7.2) to remove non-adhered cells. Subsequently, biofilm parameters, including biomass, metabolic activity, and extracellular matrix content were quantified. Briefly, for biomass quantification, biofilms were fixed with methanol, stained with crystal violet solution (0.4%; Sigma–Aldrich), washed once with PBS to remove excess stain, decolorized with acetic acid (33%) and the absorbance was read at 590 nm (SpectraMax M3; Molecular Devices, Sunnyvale, CA, USA).^22,24 The biofilm metabolic activity was assessed by the metabolic reduction of 2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino) carbonyl]-2H-tetrazolium hydroxide (XTT; Sigma–Aldrich) to a brown formazan product soluble in water.^22,24 In this assay, a solution of XTT (2 mg in 10 mL of preheated PBS) and menadione (100 μL of 0.4 mM menadione solution in acetone) was added to the biofilms and incubated at 37 °C for 3 h in the dark. The colorimetric changes were read at 492 nm using a microplate reader (SpectraMax M3; Molecular Devices, San Jose, CA, USA).^22,24 To quantify the extracellular matrix (ECM) produced by biofilm-forming cells, unfixed biofilms were stained with safranin (0.1%; Sigma–Aldrich), the wells were washed with PBS once to remove excess stain, the ECM was decolorized with acetic acid (30%) and the absorbance was read at 530 nm (SpectraMax M3; Molecular Devices).^5,24

Osmotic stress resistance assay

Fungal suspensions (200 μL of SDB containing 1 × 10⁴ yeasts) were added to 96-well polystyrene microtiter plates containing NaCl at concentrations ranging from 0.46 to 30%; plates were incubated at 37 °C for 48 h. The control of fungal growth was determined in the same conditions without the addition of NaCl. Fungal growth was determined at 530 nm (SpectraMax M3; Molecular Devices).²⁶

Statistics

The experiments were conducted in triplicate across three independent experimental sets. Statistical analysis was performed using One-Way Analysis of Variance (ANOVA) for comparisons involving three or more groups. Correlations were assessed using the Pearson correlation coefficient (r). All analyses were carried out using GraphPad Prism 9 software. P values ≤0.05 were considered statistically significant in all tests.

Results

Yeasts identification

The yeast isolates recovered from recreational waters in Rio de Janeiro city were identified using PCR followed by sequencing of the ITS1–5.8S–ITS2 region, which is considered the gold standard for fungal species identification. In this sense, among the 18 yeasts recovered from seawater samples, clinically relevant opportunistic Candida species were identified, such as C. tropicalis (n = 4, 22.23%), C. parapsilosis species complex (C. parapsilosis sensu stricto [n = 3, 16.70%]; Candida metapsilosis [n = 1, 5.55%] and Candida orthopsilosis [n = 1, 5.55%]), N. glabratus (n = 2, 11.11%), Candidozyma haemuli species complex (Candidozyma haemulid sensu stricto [formerly Candida haemulonii sensu stricto; n = 1, 5.55%] and Candidozyma haemuli var. vulneris [formerly Candida haemulonii var. vulnera; n = 1, 5.55%]) and Candida palmioleophila (n = 1, 5.55%). Three other environmental yeast species were also detected, namely Candida ecuadorensis (n = 1, 5.55%), Wickerhamiella infanticola (n = 2, 11.11%) and Pichia manshurica (n = 1, 5.55%). The sequences were deposited in GenBank database under the accession numbers described in Table 1. Half of the fungal isolates were recovered from Botafogo beach (n = 9; 50.0%), followed by Ipanema (n = 5; 27.8%) and Flamengo (n = 4; 22.2%) beaches.

Table 1.

Collection sites, CHROMagar™ Candida colony characteristics and ITS region-based molecular identification of fungal isolates recovered from recreational beaches in Rio de Janeiro city, Brazil.

Fungal isolate identification codes	Sequencing of ITS gene
	Local beaches	Colony color on CHROMagar™ Candida	Fungal identification	GenBank accession number
FL 01	Flamengo	Lilac	Candida parapsilosis	PV638871
FL 02	Flamengo	Blue	Candida tropicalis	PV638872
FL 03	Flamengo	Blue	Candida tropicalis	PV638873
FL 06	Flamengo	Cream/Lilac	Candida parapsilosis	PV638875
BT 01	Botafogo	Cream	Candida palmioleophila	PV638876
BT 02	Botafogo	Cream	Candidozyma haemuli var. vulneris	PV638877
BT 03	Botafogo	Lilac	Candida metapsilosis	PV638878
BT 04	Botafogo	Lilac	Candida parapsilosis	PV638879
BT 06	Botafogo	Cream	Candidozyma haemuli sensu stricto	PV638880
BT 07	Botafogo	Purple	Wickerhamiella infanticola	PV638881
BT 08	Botafogo	Cream	Nakaseomyces glabratus	PV638882
BT 09	Botafogo	Cream	Nakaseomyces glabratus	PV638883
BT 10	Botafogo	Lilac	Candida orthopsilosis	PV638884
IP 01	Ipanema	Cream	Candida ecuadorensis	PV638885
IP 03	Ipanema	Blue	Candida tropicalis	PV638886
IP 05	Ipanema	Blue	Candida tropicalis	PV638887
IP 06	Ipanema	Lilac	Pichia manshurica	PV638888
IP 07	Ipanema	Purple	Wickerhamiella infanticola	PV638889

Antifungal susceptibility profile

The antifungal susceptibility profile of the fungal isolates was assessed using the broth microdilution method, following the CLSI M27-Ed4 guidelines.⁷ The interpretation of the results was based on the breakpoints published in CLSI documents M60-Ed2⁸ and M27-S3.⁶ All C. tropicalis isolates were resistant to fluconazole, itraconazole, voriconazole and caspofungin, and showed elevated MIC values for terbinafine (MIC >16 mg/L). However, these C. tropicalis isolates remained susceptible to amphotericin B and flucytosine. Among the N. glabratus isolates, one was resistant to fluconazole and itraconazole, while the other showed dose-dependent susceptibility to these azoles. Additionally, both N. glabratus isolates were resistant to caspofungin and exhibited high MICs to terbinafine (16 and >16 mg/L, respectively), but were susceptible to voriconazole, amphotericin B and flucytosine. The two isolates of the C. haemuli species complex, C. haemuli sensu stricto and C. haemuli var. vulneris, were resistant to the three azoles tested. C. haemuli sensu stricto was also resistant to amphotericin B, whereas C. haemuli var. vulneris was susceptible to this polyene antifungal. Moreover, both C. haemuli isolates were susceptible to caspofungin and flucytosine and exhibited high MICs to terbinafine (MIC >16 mg/L). The C. palmioleophila isolate was resistant to fluconazole, showed dose-dependent susceptibility to itraconazole, and was fully susceptible to voriconazole, amphotericin B, caspofungin and flucytosine. All isolates of the C. parapsilosis species complex were susceptible to the antifungal agents tested. Similarly, the C. ecuadorensis isolate was susceptible to all antifungals. The same pattern was observed in one W. infanticola isolate, with the other isolate exhibiting higher MIC to itraconazole in comparison with MIC breakpoints available for Candida species (1 mg/L). P. manshurica showed high MIC values for fluconazole (>64 mg/L), flucytosine (16 mg/L), and terbinafine (>16 mg/L), suggesting reduced susceptibility to these agents. Lower MIC values were observed for itraconazole, voriconazole, amphotericin B, and caspofungin (Table 2).

Table 2.

Antifungal susceptibility profile of fungal isolates recovered from recreational beaches in Rio de Janeiro city, Brazil.

Fungal species (isolates)	MIC (mg/L)^a
	FLC	ITC	VRC	AMB	CSF	5-FC	TEB
C. tropicalis
FL 02	>64 (R)	>16 (R)	>16 (R)	1 (S)	>8 (R)	0.5 (S)	>16
FL 03	>64 (R)	>16 (R)	>16 (R)	0.5 (S)	>8 (R)	0.5 (S)	>16
IP 03	64 (R)	2 (R)	4 (R)	0.5 (S)	4 (R)	<0.12 (S)	>16
IP 05	>64 (R)	>16 (R)	>16 (R)	0.5 (S)	>8 (R)	1 (S)	>16
C. parapsilosis
FL 01	1 (S)	0.12 (S)	<0.03 (S)	0.5 (S)	2 (S)	0.5 (S)	4
FL 06	2 (S)	<0.03 (S)	<0.03 (S)	0.5 (S)	2 (S)	0.5 (S)	2
BT 04	2 (S)	<0.03 (S)	0.06 (S)	0.12 (S)	2 (S)	0.25 (S)	2
C. metapsilosis
BT 03	2 (S)	0.06 (S)	0.06 (S)	0.5 (S)	1 (S)	1 (S)	0.5
C. orthopsilosis
BT 10	8 (S)	0.5 (SDD)	0.25 (S)	0.5 (S)	2 (S)	2 (S)	4
N. glabratus
BT 08	64 (R)	>16 (R)	1 (S)	0.5 (S)	1 (R)	0.12 (S)	>16
BT 09	32 (SDD)	0.25 (SDD)	0.5 (S)	<0.03 (S)	1 (R)	1 (S)	16
C. haemulid
BT 06	>64 (R)	>16 (R)	>16 (R)	2 (R)	0.25 (S)	0.5 (S)	>16
C. haemuli var. vulneris
BT 02	64 (R)	4 (R)	8 (R)	1 (S)	0.12 (S)	0.5 (S)	>16
C. palmioleophila
BT 01	64 (R)	0.5 (SDD)	0.5 (S)	0.12 (S)	1 (S)	2 (S)	8
C. ecuadorensis
IP 01	4 (S)	0.12 (S)	0.5 (S)	0.25 (S)	1 (S)	<0.12 (S)	1
W. infanticola
BT 07	8	1	0.12	0.12	1	0.5	2
IP 07	4	0.12	0.06	<0.03	2	<0.12	0.25
P. manshurica
IP 06	>64	0.5	0.5	0.12	0.25	16	>16

a

MIC, minimal inhibitory concentration; FLC, fluconazole; ITC, itraconazole; VRC, voriconazole; AMB, amphotericin B; CSF, caspofungin; 5-FC, flucytosine; TEB, terbinafine; R, resistant; SDD; susceptible-dose dependent; S, susceptible. CLSI protocols have not established breakpoints for terbinafine and amphotericin B; in general clinical isolates of Candida with MIC values >1 mg/L are considered resistant to amphotericin B.

Assessment of biofilm formation

Biofilm formation by the yeast species was assessed on a plastic (polystyrene) surface by analyzing three parameters: (i) total biomass, as determined by crystal violet staining of methanol-fixed cells; (ii) ECM production, evaluated via safranin staining of non–fixed cells; and (iii) metabolic activity, measured by the conversion of XTT to formazan in non-fixed biofilm-forming cells. All fungal isolates recovered from recreational coastal waters demonstrated the ability to form biofilms, albeit to varying extents. Biofilm biomass absorbance values ranged from 0.189 to 1.277 (overall mean 0.525 ± 0.311), ECM production ranged from 0.172 to 2.036 (mean 0.719 ± 0.523) and metabolic activity ranged from 0.288 to 1.506 (mean 0.912 ± 0.390) (Fig. 1). Although no statistically significant differences were observed among the species for the evaluated parameters, isolates of C. tropicalis and C. parapsilosis exhibited higher values across all three biofilm-related measures. On the other hand, C. orthopsilosis and one isolate of N. glabratus displayed high levels of biofilm–associated metabolic activity. Remarkably, P. manshurica showed prominent ECM production and metabolic activity, while the metabolic activity of C. ecuadorensis was also comparatively elevated. Notably, a significant positive correlation was observed between biofilm biomass and ECM (p = 0.0008; r = 0.7197). However, no correlations were observed between biofilm biomass and metabolic activity (p = 0.6129; r = 0.1279), nor between ECM and metabolic activity (p = 0.0618; r = 0.4487). Fig. 2 shows representative images of the biofilms formed, illustrating the variability in the extent and resistance of the biofilms among the different species.

Fig. 1.

Biofilm formation by the fungal isolates obtained from recreational coastal waters of Rio de Janeiro. Biofilms were developed on a polystyrene surface at 37 °C for 48 h and subsequently prepared for quantification of fungal biomass by measuring absorbance at 590 nm, extracellular matrix at 530 nm and metabolic activity at 492 nm. The results are expressed as absorbance values (ABS) per isolate studied. The results represent means ± standard deviation of three independent experiments. The codes on the X-axis of the graph represent each of the 18 yeast isolates used.

Fig. 2.

Representative micrographs of biofilms formed by the fungal isolates obtained from recreational coastal waters of Rio de Janeiro after 48 h of incubation at 37 °C in contact with polystyrene. The biofilms were visualized by means of crystal violet staining, to highlight the biomass, and by safranin staining, to highlight the extracellular matrix (ECM). The isolates used were FL03 and IP05 of C. tropicalis (Ct), BT04 of C. parapsilosis (Cp), BT03 of C. metapsilosis (Cm), BT10 of C. orthopsilosis (Co), BT09 of N. glabratus (Cg), BT06 of C. haemuli (Ch), BT02 of C. haemuli var. vulneris (Chv), BT01 of C. palmioleophila (Cpalm), IP01 of C. ecuadorensis (Ce), BT07 of W. infanticola (Wi) and IP06 of P. manshurica (Pm).

Resistance to osmotic stress

Since the yeast isolates used herein were recovered from marine waters, we investigated their resistance to osmotic stress using NaCl. In this sense, all yeast isolates were able to grow in concentrations up to 7.5% NaCl, except for the isolate of C. ecuadorensis that grew up to 3.75% (Fig. 3).

Fig. 3.

Resistance to osmotic stress of the fungal isolates obtained from recreational coastal waters of Rio de Janeiro. Fungal cells in SDB (1 × 10⁴ cells in 200 μL) containing different concentrations of NaCl (0–30%) were incubated in polystyrene microtiter plates at 37 °C for 48 h. Fungal growth was read at 530 nm and the results are expressed as absorbance values (ABS) per isolate studied. The results represent mean ± standard deviation of three independent experiments. (A) C. tropicalis isolates (FL02, FL03, IP03 and IP05), (B) C. parapsilosis species complex isolates (C. parapsilosis sensu strictu: FL01, FL06 and BT04; C. metapsilosis: BT03; C. orthopsilosis: BT10), (C) N. glabratus (BT08 and BT09), C. haemuli species complex (C. haemuli sensu stricto: BT06; C. haemuli var. vulneris: BT02) and C. palmioleophila (BT01), (C) C. ecuadorensis (IP01), W. infanticola (BT07 and IP07) and P. manshurica (IP06).

Discussion

The use of recreational waters for leisure activities is of recognized importance worldwide, and especially in Brazil, a country internationally known for the beauty of its coastal regions. These environments are of significant tourist interest, attracting both domestic and international visitors, which is a key factor in the Brazilian economy. In this sense, the city of Rio de Janeiro stands out as tourist destination due to its natural beauty, including several beaches distributed throughout its territory. The Brazilian legislation classifies the quality of seawater for primary contact based solely on the presence of fecal indicators bacteria, while the presence of other microorganisms, including fungi, remains neglected.²¹ Recently, the World Health Organization included fungi as a taxon of interest for monitoring recreational water quality,³¹ based on the results of the studies developed by the European working group called Mycosands.² This working group investigated the occurrence of fungi in beach sand and seawater of 13 different European countries, identifying as the most frequently detected fungal genera yeasts like Candida and Cryptococcus, and molds like Aspergillus and Fusarium.² The authors found a median of site–blinded total fungal of 89 CFUs/g and suggested that this fungal concentration could be utilized as a baseline for beaches in Europe where analytical data are unavailable.²

In the present study, we recovered both environmental yeast species and clinically relevant opportunistic yeasts from recreational coastal waters of Rio de Janeiro city. Among the opportunistic yeasts identified, C. tropicalis, C. parapsilosis species complex and N. glabratus are notable etiological agents of invasive fungal infections, especially in critically ill patients.³² Indeed, these fungal species are among the most frequently isolated Candida species in candidemia cases in Brazilian hospitals, second only to C. albicans, which remains the predominant species recovered from these patients.¹⁴ Other studies have also demonstrated the high occurrence of C. tropicalis in both sand and seawater samples of different Brazilian beaches.^15,26,36 Additionally, the occurrence of C. parapsilosis species complex and N. glabratus in recreational coastal waters of Brazil and other countries, such as Iran and Qatar, has already been reported.¹⁵ With regard to the antifungal susceptibility of C. tropicalis, this species has been shown to exhibit prominent resistance to azoles, This finding is corroborated by the results of the present study, which revealed that all isolates of C. tropicalis were resistant to the three azole antifungals tested. Similar results were observed in C. tropicalis isolates obtained years prior in these same beaches.²⁶ Other studies also reported high resistance rates to azole in C. tropicalis isolates obtained from coastal environments.^15,36 All the isolates of C. tropicalis tested in this study exhibited resistance to caspofungin, a finding that differs from the literature on this species, which has been classified as susceptible to this echinocandin.^26,36

In the present study, isolates within C. haemuli species complex, an emergent multidrug-resistant group of yeasts, were also recovered from the recreational coastal waters of Rio de Janeiro city. The origin of species within the C. haemuli species complex remains a subject of ongoing debate, with increasing speculation regarding a possible environmental reservoir. Interestingly, the species C. haemuli was first isolated from the gut of the fish Haemulon scirus, from the water of Biscayne Bay, Florida, and from the seawater of the coast of Lisbon, Portugal, being initially called Torulopsis haemulonii.²⁹ Members of the C. haemuli species complex have recently been isolated from sand and seawater samples collected from different Brazilian beaches, a finding that corroborates the aforementioned hypothesis.¹⁵ The supposition that coastal environments may serve as ecological niches and potential sources for the emergence and dissemination of these opportunistic multidrug-resistant yeasts is thus reinforced. Species belonging to the C. haemuli species complex have emerged over the past decades as clinically significant multidrug–resistant pathogens in hospitals worldwide, including Brazil.^23,25 These yeasts have been increasingly associated with a broad spectrum of infections, ranging from superficial colonization to severe invasive disease, particularly among immunocompromised and critically ill patients.^13,23,25 The antifungal susceptibility profile of the species within this yeast group isa cause of concern, as its members are typically resistant to azoles and amphotericin B, yet generally susceptible to echinocandins,^4,23,25 as evidenced by the results of the present study. Their intrinsic reduced susceptibility to multiple classes of antifungal agents and their frequent misidentification by conventional diagnostic methods further complicate the clinical management and may contribute to unfavorable therapeutic outcomes.

The rare opportunistic species C. palmioleophila was also isolated in our study. This species has been associated with different marine ecosystems, including oligotrophic hypersaline coastal waters of the Arabian Gulf,¹¹ a hydrographic basin in Minas Gerais state, Brazil, especially during the dry season¹⁷ and wastewaters in Niteroi city, Rio de Janeiro, Brazil, during COVID-19 pandemic.^9,10C. palmioleophila was first isolated from soil samples,²⁰ being further associated with catheter-related fungemia.²⁸ In recent years, the isolation frequency of C. palmioleophila from clinical cases has increased, particularly in Europe.^1,3 This species is commonly resistant to fluconazole,^10,34 as demonstrated in this study. Indeed, a case of C. palmioleophila infection with a fatal outcome was reported in China.³³ Conversely, C. palmioleophila is usually susceptible to the other antifungal agents, a finding that was corroborated by the present study's results.⁹

Environmental yeasts such as C. ecuadorensis, W. infanticola and P. manshurica were also recovered from seawater samples in the present study. The species W. infanticola has also been isolated from recreational beaches of Paraná state, Brazil, being susceptible to fluconazole and amphotericin B, and susceptible dose-dependent to itraconazole.¹⁵ The two isolates of W. infanticola of our study exhibited low MIC values for all the antifungals tested, and would be considered susceptible according to the breakpoints established for Candida in CLSI document M27-S3,⁶ except for one isolate, which would be classified as resistant to itraconazole.

Biofilm, as a syntrophic community, is widely regarded as the preferred way of life for microorganisms.²⁷ Biofilms are complex structured communities of microorganisms attached to a surface or interface, covered by an ECM produced by the microorganisms themselves, being important during infection processes and in the environment colonization.²⁷ All environmental isolates used in our study were able to form biofilm on a plastic (polystyrene) surface. In this sense, biofilm formation by clinically relevant Candida species, including C. tropicalis, C. parapsilosis species complex, C. haemuli species complex, N. glabratus and C. palmioleophila, have been described in the literature.¹⁶ Indeed, catheter-related fungemia caused by Candida species in hospital settings represents a huge problem for public health worldwide, since these fungal pathogens are able to adhere to different types of catheters, forming biofilms, which facilitates the access to the bloodstream of patients. Fungal cells forming biofilms exhibit different phenotypes in comparison to their planktonic counterparts, being generally more resistant to antifungal treatments, aggravating the clinical situation of patients.¹⁶ Biofilms also play a significant role in many different environmental niches, including recreational coastal waters. In this sense, biofilms can protect microorganisms from challenging conditions, such as high salinity, ultraviolet radiation, poor nutrients, extreme pH and temperatures, acting as “protective clothing”.³⁵ Additionally, microorganisms can form biofilms on microplastics and, interestingly, microplastics with biofilms exhibit greater pollutant sorption capacity than virgin counterparts, ultimately serving as vectors capable of introducing contaminants and microbial agents to aquatic habitats.³⁰ The environmental yeast species tested in the present study were able to form biofilms on polystyrene surface similarly to the opportunistic yeasts, corroborating the relevance of this virulence attribute regardless of whether the yeast is pathogenic or not.

Given that the yeasts used in this study were isolated from marine environments, we assessed their ability to grow under high NaCl concentrations. Except for C. ecuadorensis that grew up to 3.75% NaCl concentration, all the other yeast isolates were able to grow up to 7.5% NaCl. This finding is in agreement with our prior research, in which isolates of C. tropicalis and C. parapsilosis recovered from recreational coastal waters of Rio de Janeiro were also able to grow in concentrations up to 7.5% NaCl.²⁶ Additionally, several C. tropicalis isolates obtained from sand samples of beaches in Northeast Brazil were able to grow in higher salt levels (15% NaCl).³⁶

Taken together, our results reveal the presence of clinically relevant opportunistic Candida species, as well as traditionally environmental yeasts, in seawater samples from beaches in Rio de Janeiro. Notably, except for the isolates within the C. parapsilosis species complex, all clinically important Candida species, and the environmental yeast P. manshurica, exhibited a concerning antifungal resistance profile. Furthermore, all isolates exhibited some virulence attributes, such as biofilm formation on plastic surface and growth under high salinity, which are associated with environmental persistence and pathogenic potential. Altogether, these findings highlight the presence of potentially pathogenic and drug-resistant yeasts in recreational coastal waters of Rio de Janeiro. To better assess the public health risk, future studies should include seasonal and long-term monitoring of fungal communities in marine environments, incorporate quantitative assessments of fungal load, and investigate the potential of yeast transmission to humans through recreational water exposure. The integration of fungal surveillance into routine water quality programs, alongside bacterial indicators, is also recommended to provide a more comprehensive understanding of microbial risks in coastal areas.

Funding

This study was supported by grants from the Brazilian Agencies: Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação Carlos Chagas Filho de Amparo à Pesquisa no Estado do Rio de Janeiro (FAPERJ) and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES – Financial code 001). Lívia S. Ramos was supported by FAPERJ #E-26/203.487/2023.

Conflict of interest

The authors report no conflict of interest.

Uncited reference

12.