1 Introduction

The availability and accessibility of clean potable water, an essential resource for well-being and sustainable human living, in many developing countries, especially in rural communities remains grossly inadequate [1]. In rural settings, it is estimated that 8 out of 10 people are without access to potable water. This is occasioned by deteriorating condition of existing pipelines, lengthy years of misuse, mismanagement, overexploitation and contamination of freshwater sources, coupled with climate change [2]. This situation threatens Sustainable Development Goal 6 (clean water and sanitation for all) attainment by 2030, with an estimated annual death of 800,000, if the current trend progresses unchecked [3].

Water for human consumption must satisfy the threshold limits by different regulatory bodies to be considered safe [4]. Several studies reported microorganisms including bacteria, viruses and parasites as major waterborne contaminants [5, 6]. On the other hand, fungi are rarely mentioned [7]. The public health implications of fungal contaminated water are largely unknown, and there is paucity of knowledge on standards to employ in routine water quality checks [8]. Recently, fungi began to receive recognition as an emerging chronic water quality challenge [6].

Fungi are ubiquitous in diverse aquatic milieus, comprising surface, ground and tap water proposed for anthropogenic use [8]. Different species of Acremonium, Acrostalagmus, Alternaria, Arthrinium, Arthrographis, Ascochyta, Aspergillus, Asteroma, Astermella, Aureobasidium, Beauveria, Bionectria, Bipolaris, Cadophora, Candida, Cephalosporium, Chaetomium, Cytospora, Dactylaria, Discosporium, Epicoccum, Exophiala, Fusarium, Gibberella, Gliocladium, Kluyveromyces, Microsporum, Paecilomyces, Penicillium, Sarocladium, Talaromyces, Trichoderma, Mucor, Rhizopus etc. have been isolated from water sources in different countries of the world [8].Their distribution in aquatic milieu results in objectionable conditions in taste and odour [5], allergies and meningitis [9]. Some species produce mycotoxins or other secondary metabolites capable of causing respiratory disorders, cancer and opportunistic infections in immmunocompromised individuals [10, 11].

In the last two decades, potential health challenges associated with fungal water pollution, propelled investigations on bottled mineral water [12, 13], tap water [14,15,16], and water supply systems [17, 18] towards safeguarding public health. Of greater concern is the evolution of resistance to commercially available antifungal agents [19, 20], coupled with the limited discovery of new drugs. Antifungals, including azoles (ketoconazole and fluconazole), pyrimidine analog (flucytosine), and polyenes (amphotericin B) are important in modern clinical practice and extensive agriculture [21, 22], and their application brought appreciable progress to the treatment of animal, human and plant fungal infections [23]. Conversely, their efficacy is being confronted by the increasing rise in drug resistance [24]. In recent times, high level resistance i.e. 100% to ketoconazole and fluconazole [25], 75% to fluconazole [26], 47% to flucytosine [28], and 91% [27] and 67% [25] to amphotericin B have been reported.

In the most populous black African nation Nigeria, it is common for communities that do not have access to treated pipe-borne water to rely heavily on alternative sources, such as groundwater which is perceived clean and safe [2]. When groundwater is polluted with antimicrobial-resistant organisms, they impact health and financial burdens on individuals who rely on the water source for daily survival [2, 29]. In Osun State, Southwestern Nigeria, literatures reveal less represented information on antifungal resistance and dearth of data on the genetic relatedness of fungi in groundwater supplies [30, 31]. Previous studies documented mould contaminants using the phenotypic approach only in water sources in Osun State [30, 31]. In this light, the present investigation, being the first of its kind, investigated the antifungal resistance profile and phylogenetic relationship of moulds isolated from groundwater sources in a rural community of the state.

2 Materials and methods

2.1 Description of the study area

Seke is a rural community located along the Dagbolu-Ikirun axis in Ifelodun Local Government Area of Osun State, Nigeria (Fig. 1). It has vast open land, and majority of the dwellers are farmers. They engage in cultivation of major food crops including cowpea, rice, maize, cassava and yam; and cash crops such as cocoa, kolanut and oil palm, amongst others. Some of Seke inhabitants are into poultry farming, and also hunt games. The community lacks basic infrastructures for potable water, forcing the rural settlers to depend on hand-dug wells to meet their daily water needs.

Fig. 1
figure 1

Map showing the location of Seke village, Osun State

Full size image

2.2 Sample collection

Grab water sampling technique was adopted in the collection of 17 hand-dug well water sources, once, during the dry (November 2022) and rainy (March 2023) seasons. All samples were designated A-Q. Water samples were aseptically collected in 1 L capacity germ-free plastic containers and conveyed to the laboratory on ice packs for fungal analysis in not more than 6 h of collection. A brand of hygienically packed bottled water served as a negative control.

2.3 Isolation and enumeration of moulds from groundwater samples

The pour-plate technique was used for fungal enumeration. Exactly 1 mL of each water sample was dispensed into sterile Petri dishes, and 20 mL of potato dextrose agar (PDA) medium (Oxoid, UK) supplemented with chloramphenicol was added, gently swirled and allowed to gel. The inoculated agar plates were incubated at 27 ± 2 °C for 5–7 days. Filamentous fungal colonies were counted and expressed as CFU/mL. Individual colonies were subcultured on PDA until pure cultures were obtained and stored on slants at 4 °C until needed. The occurrence (%) of isolates was calculated following Eq. 1.

$${\text{OC}}\% \, = \left( {a/b} \right)*{1}00\%$$
(1)

where, OC = occurrence of fungal isolates; a = number of fungal isolates; and b = total number of fungal isolates [2].

2.4 Phenotypic characterisation of filamentous fungal isolates

Morphological identification was done using macroscopic and microscopic approaches [32]. Macroscopic description of the isolates was done by observing the shape, size, colour and texture of pure fungal colonies on PDA. For microscopic characterisation, a pure colony of 5—7 days old was stained in lactophenol blue on a clean slide and examined under a microscope (X40). The fungal isolates were presumptively named by comparing observed features with those outlined [32].

2.5 Antifungal susceptibility testing

Susceptibility of the recovered fungal isolates to four antifungals comprising ketoconazole, 15 µg; fluconazole, 25 µg; flucytosine, 10 µg; and amphotericin B, 20 µg (Liofilchem, Italy) was determined following the standardized disc diffusion assay as described [29]. The 7-day-old fungal isolates were suspended in physiological saline and precisely 100 µl of 0.5 McFarland standard spore suspension was inoculated by spreading on PDA plates and allowed to dry. Antifungal discs were carefully positioned on the inoculated plates using a disc dispenser (Oxoid, UK) and incubated at 27 ± 2 °C for 2 days. The diameters of clear zones of inhibition were measured, recorded in the nearest millimetres and interpreted as susceptible (S), intermediate (I) and resistant (R) following standard interpretive guideline [33].

The frequency of antifungal-resistant isolates was estimated using Eq. 2.

$${\text{Frequency }} = \, \left( {{\text{E}}/{\text{F}}} \right) \, \times { 1}00$$
(2)

where ‘E’ is the total of number isolates resistant to a drug and ‘F’ is the total number of isolates involved in the study [2].

2.6 Multiple antifungal-resistant phenotypes and indexing of the isolates

Multiple antifungal-resistant phenotypes (MARP) for each sample were generated for isolates exhibiting resistance to three or more antifungals according to Wose et al. [34]. Multiple antifungal resistance indexes (MARI) for individual hand-dug wells were also assessed using Eq. 3.

$${\text{MARI }} = {\text{x}}/{\text{y}}$$
(3)

where ‘x’ represents the number of antifungal agents to which the isolate was resistant and ‘y’ is the number of antifungal agents tested against one isolate [2].

Likewise, the antifungal resistance index (ARI) for each sampling location was evaluated using Eq. 4.

$${\text{ARI }} = {\text{ A}}/{\text{N}}\left( {\text{Y}} \right)$$
(4)

where ‘A’ is the total number of resistant isolates recorded, ‘N’ is the number of isolates and ‘Y’ represents the total number of antifungals tested [29].

2.7 Molecular confirmation of isolates

2.7.1 DNA extraction

Fungal DNA was extracted from a 5-day-old broth culture using the ZR Fungal/Bacterial DNA kit™ (Zymo Research, USA) according to the manufacturer’s instructions. Quantification and purity of DNA were done using NanoDrop™ One spectrophotometer (Thermo Scientific, UK).

2.7.2 Polymerase chain reaction and gel electrophoresis

The nuclear ribosomal internal transcribed spacer (ITS) region was amplified by PCR with previously described universal primers ITS4 (5′-TCC TCC GCT TAT TGA TATGC-3) and ITS5 (5′-GGA AGT AAA AGT CGT AAC AAGG -3′) [35]. PCR amplicons were analysed using 1.5% (w/v) agarose gel electrophoresis. The gel image was visualized with a Vilber Lourmat E-Box gel documentation imaging to confirm the size of PCR products. The expected amplicon size of the primers is between 300 and 600 bp.

2.7.3 Amplicon sequencing, deposition into GenBank

PCR products were cleaned up using an enzymatic method (ExoSAP-IT™ PCR Product Cleanup Reagent (Thermofisher, UK)) according to the manufacturer’s instructions. Sanger sequencing was performed using the Nimagen, Brilliant Dye™ Terminator Cycle Sequencing Kit V3.1, following the manufacturer’s instructions. Sequence similarities of the generated ITS region were calculated using the Basic Local Alignment Search Tool (BLASTn), for homology to identify the fungal isolates [36]. Sequences were deposited in the GenBank® and for accession numbers assignment.

2.7.4 Phylogenetic analysis

Evolutionary history was inferred between 44 nucleotide sequences using the Neighbor-Joining method [37, 38]. The evolutionary distances were computed using the Maximum Composite Likelihood method [39] and are in the units of the number of base substitutions per site. All ambiguous positions were removed for each sequence pair (pairwise deletion option). There were a total of 1166 positions in the final dataset. Evolutionary analyses were conducted in MEGA X [40].

3 Results

3.1 Enumeration of moulds

The frequency of moulds in the samples ranged from 0 to 4 CFU/ml across both seasons. Moulds were detected in all the samples except A, D and E in the rainy season. Overall, 29 and 27 moulds were recovered from dry and rainy seasons respectively (Table 1). The negative control showed no growth.

Table 1 Counts of mould recovered from the water samples
Full size table

3.2 Presumptive identification of the isolates

Morphologically, 6 genera of fungi belonging to 10 species were identified as follows: Aspergillus flavus, A. fumigatus, A. niger, A. welwitschiae, Fusarium equiseti, F. falciforme, Mucor sp., Penicillium citrinum, Rhizopus delemar and Trichoderma longibrachiatum following macroscopic and microscopic description. Representative pictures of the isolates are shown in Fig. 2.

Fig. 2
figure 2

Pictures (macroscopy and microscopy) of representative fungal isolates from Seke groundwaters. a Aspergillus flavus b A. fumigatus c A. welwitschiae d A. niger e Fusarium equiseti f F. falciforme g Mucor sp. h Penicillium citrinum i Trichoderma longibrachiatum j Rhizopus delemar

Full size image

3.3 Occurrence of fungal species in the groundwater samples

The prevalence of moulds in the water samples is as follows: Penicillium citrinum (13;21.43%), Trichoderma longibrachiatum, Aspergillus welwitschiae (11;19.64%), Aspergillus niger (8;14.29%), Fusarium falciforme (4;7.14%), Rhizopus delemar, Aspergillus flavus, Fusarium equiseti (2;3.57%), and Aspergillus fumigatus, Mucor sp. (1;1.79%) (Table 2).

Table 2 Frequency of moulds in the groundwater samples
Full size table

3.4 Antifungal susceptibility pattern

Antifungal susceptibility testing was performed on representative 56 isolates, randomly selected across the sampling points (Fig. 3). Resistance of the isolates to the different antifungal agents approximately ranged from 55 to 98%. The highest resistance was against fluconazole 55(98%), followed by flucytosine 49(88%), ketoconazole 45(80%) and amphotericin B 31(55%). Susceptibility was in descending order amphotericin 25(45%), flucytosine 5(9%) ketoconazole 4(7%) and fluconazole 1(2%) (Fig. 3).

Fig. 3
figure 3

Antifungal susceptibility patterns of moulds. KCA—ketoconazole; FLU—fluconazole; AFY—flucytosine; and AMB—amphotericin B

Full size image

3.5 Multiple antifungal resistance phenotypes and indices of moulds from groundwaters

The results generated for MARP are presented in Table 3. Forty-six (46) of the 56 isolates were multidrug resistant ranging from 3 to 4 drugs. Other isolates (10) were resistant to 1 or 2 antifungal agent(s). In terms of percentage frequency, 23(50%) each of the isolates were resistant to 3 and 4 drugs, respectively (Table 3). In the same vein, the antifungal resistance index (ARI) ranged from 0.17 to 1.00, whereas the multiple antifungal resistance index (MARI) evaluated was 1 except sample D with 0.75 (Table 4).

Table 3 Patterns of multiple antifungal resistance phenotypes (MARPs) of mould contaminants
Full size table
Table 4 Predominant antifungal resistance patterns of mould contaminants from groundwater samples
Full size table

3.6 Molecular confirmation of fungal isolates

3.6.1 Amplicon size obtained after PCR

The gel electrophoresis image, confirming the amplification of ITS 4 and 5 of the isolates is in Fig. 4. The amplicon sizes ranged between 400 and 600 bp.

Fig. 4
figure 4

Gel electrophoresis of PCR amplified ITS region of the representative isolates. L—Molecular weight marker; 1—Trichoderma longibrachiatum; 2—Aspergillus welwitschiae; 3—A. flavus; 4—Penicillium citrinum; 5—A. niger; 6—Mucor sp.; 7—Fusarium equiseti; 8—F. falciforme; 9—A. fumigatus; 10—Rhizopus delemar

Full size image

3.6.2 BLASTn analysis

BLASTn analysis of the ITS region of 10 fungal strains from this study was obtained and identified accordingly (Table 5). Our isolates showed between 90% and 100% similarity with their reference species. Aspergillus flavus, A. fumigatus, A.niger, Fusarium equiseti, F. falciforme and Penicillium citrinum had 100% identity with reference strains MK091395, OW985950, MN474007, MN559437, OW985200, and MN069574 respectively. Likewise, Aspergillus welwitschiae, Mucor sp. BAB-3377 and Rhizopus delemar recorded 99% similarity with MK450669, KU504304, and LC514303. Only Trichoderma longibrachiatum had 90% with GenBank strain OQ402425 (Table 5).

Table 5 Summary of BLASTn result
Full size table

3.6.3 Phylogenetic analysis

The evolutionary relationship of taxa as revealed by MEGA-X is shown in Fig. 5. The result revealed the presence of 10 clades. All the investigated strains in this work clustered with selected reference strains from the GenBank (Fig. 5). Aspergillus flavus (OR350638), A. niger (OR743924), A. welwitschiae (OR743925), A. fumigatus (OR74392), Penicillium citrinum (OR743923), Trichoderma longibrachiatum (OR345452), Fusarium falciforme (OR350641), F. equiseti (OR350640), Mucor sp. (OR350639), Rhizopus delemar (OR350642) exhibited close genetic relationship with A. flavus (MT447545), A. aculeatus (KY859793), A. welwitschiae (MK450669), A. fumigatus (OM802836), P. citrinum (MN634544), Trichoderma longibrachiatum (MF967326) and Trichoderma sp. (MK645995), F. oxysporum (EU888922), F. equiseti (MW812252), M. indicus (MT993845) and M. nederlandicus (MZ433254) and Rhizopus sp. (OR304274), respectively (Fig. 5).

Fig. 5
figure 5

Neighbour-joining tree showing the phylogenetic position of mould contaminants based on the ITS region gene sequence

Full size image

4 Discussion

Fungi occur ubiquitously in diverse environments, including aquatic milieus [41, 42]. In this present study, 29 and 27 moulds were isolated from groundwater samples across dry and rainy seasons respectively, suggesting the unsuitability of the water sources for drinking, with its attendant threat to public health [43]. Previous studies isolated 35 and 2 moulds from bottled mineral and tap waters, respectively [44] and 55 from the water distribution system [17]. The widespread distribution of fungi in water has been attributable to factors including location, solar radiation, temperature, ion composition and pH [8]. Additionally, the existence of carbon-based matter, dissolved oxygen concentration, type of water treatment, and biofilm formation influence occurrence of fungi in water environment [8].

Globally, fungal pathogens account for at least 13 million infections resulting in about 1.5 million mortalities annually [45, 46].The moulds detected in the groundwater samples (Table 2) are of a public health concern as they are causative agents of respiratory, mucosal, rhinocerebral, cutaneous and subcutaneous infections [47]. Earlier investigations documented species belonging to Alternaria, Aspergillus, Aureobasidium, Beauveria, Botrytis, Chaetomium, Cladosporium, Epicoccum, Exophiala, Fusarium, Paecilomyces, Penicillium, Purpureocillium, Sarcocladium, Scopulariopsis, Stachybotrys, Trichoderma, Mucor and Rhizopus from different water sources including groundwaters [8, 17, 48, 49]. In the current study, the genera Aspergillus was the most predominant (Table 2), agreeing with previous investigations [50,51,52]. Aspergillus plays important roles in the maintenance of the freshwater environment through the degradation of dead plant litters and animal parts; however, some exist as pathogens or endophytes of aquatic lives [51].

The existence of antifungal resistance in environmental samples is well reported [53,54,55,56,57], but its incidence is scarcely reported in groundwater samples. In the present study, resistance observed amongst isolates ranged from 55 to 98%. The detection of drug resistance suggests the contamination of the environment with antifungal agents. Generally, the natural environment is the main reservoir for moulds, but antifungals used in agriculture may drive resistance development [58]. Seke village is characterized by intense agricultural activities, almost all the hand-dug wells sampled are either/both rarely treated or/and uncovered. This exposes humans to antifungal-resistant pathogens and poses a risk for patients who acquire an infection from environmental antifungal-resistant strains [58].

Microorganisms develop resistance to drugs in a quest to outcompete and persist in their natural ecological systems [60]. The greater than 80% resistance observed for fluconazole, flucytosine and ketoconazole (Fig. 1) was comparable with Shittu et al. [11] who reported resistance to fluconazole (76.19%), ketoconazole (73.80%), clotrimazole (92.86%), griseofulvin (88.09%) and nystatin (100%). This suggests indiscriminate and inappropriate use of the drug in that environment. Antifungal compounds are widely used as therapeutics in clinical and/or veterinary medicine; as antimycotic in body care products [60]; and in agricultural settings as fungicides for plant protection [61]. Once antifungals are in the environment, some accumulate and persist for long periods [62] encouraging the development of resistant strains. Most of the villagers are farmers and regular users of chemicals on their farmlands for prophylaxis purposes. The direct use of fungicides in crops serves as a route of antifungal entry to the environment [63], and subsequently through agricultural runoff [64].

The highest susceptibility recorded in amphotericin B 25(44.64%) suggests it is the therapeutic option in treating fungal infections arising from contaminated groundwater. Generally, amphotericin B resistance is rare unlike the azoles and flucytosine [65]. Currently, there are few therapeutic antifungal alternative drug targets. Antifungal agents consistently damage cell structure and rigidity by interacting with either cell wall or cell membrane constituents. As a result, resistance to one or more classes of antifungals drastically reduces viable treatment options for potentially life-threatening infections [63]. Amphotericin B is the most known member of the antifungal class, polyene. Although the class acts synergistically with some of the most effective antifungal compounds, they are associated with severe nephrotoxicity [66] and low water solubility [67], which limits their use as antifungal drugs [68].

In vitro and in vivo multiple drug resistances are of great therapeutic consequences [29]. The current study elucidated multiple antifungal resistance patterns and found MARI ranging from 0.75–1.00 across both seasons. Fundamentally, diverse human activities and other associated events within the village might be responsible for the observed MARI values (0.75 to 1.00) which denote a high-risk source contamination of the groundwater samples. This reveals that the isolates originated from sources where antifungals are frequently applied [2], possibly from the farmlands in Seke Village where fungicides are frequently applied. Overuse of antifungals in either agriculture or humans exert pressure on fungi, selecting those that have developed resistance, encouraging their survival and proliferation [68, 70]. However, the environment may drive the evolution of antifungal resistance, and may also play a role in human exposure to drug-resistant opportunistic fungal pathogens [71].

The mechanism of resistance in fungi could either be intrinsic or acquired [72]. While intrinsic resistance is inherent, acquired resistance is influenced by antimicrobial selective pressures, and may be caused by mutations, genome rearrangements and/or overexpression of resistance gene products, changes in ergosterol synthesis and aneuploidy [73,74,75,76,77,78].

Molecular characterization provides a way out of misidentification associated with conventional phenotypic identification [79]. Our current investigation employed genetic markers ITS 4 and 5, with amplicon sizes ranging between 400 and 600 bp (Fig. 4). Ezeonuegbu et al. [80] reported an amplicon size of 600 bp in fungi isolated from industrial wastewater. On the other hand, Fujita et al. [81] noted PCR products have band sizes that range from 350–880 bp. Generally, in fungal identification, variance of fungal ITS region and primer combination employed in PCR account for variability in band size [81, 82]. In this study, our isolates recorded 99–100%identity with GenBank strains. Previous works noted 97–99% similarity in fungal isolates from soil and plant parts [83], and 100% in fungal isolates from dumpsite soils [84]. Phylogenetic analysis (Fig. 5) revealed that all the investigated strains are genetically related with previously deposited sequences in the GenBank, including those of environmental sources: soils (MK450669, MW812252, MZ433254), farmland (MF967326), roots (MT447545, KY859793, EU888922), drinking water (MN634544), plant (MK645995), aquaculture water (OR304274), drainage water and sediments (OM802836), oil mill effluent (MT993845)[85].

5 Conclusion

Phenotypic detection and genotypic confirmation of moulds in any groundwater source is indicative of waters not suitable for drinking. The findings of this study signify a high prevalence of drug resistance in the mould contaminants towards standard antifungals. Multiple antifungal-resistant phenotypes and indices assessed imply the presence of antifungal-resistant moulds that can lead to deleterious health complications for individuals who rely heavily on the hand-dug well waters for their domestic purposes. Amphotericin B was identified as a choice drug for infections arising from ingestion of contaminated waters. Groundwaters will continue to experience a rise in the incidence of resistant fungal strains unless indiscriminate use of antifungals in medical (human and veterinary practices) and agricultural fields is restricted. This can be achieved by public awareness and education, legislation and restriction of prescription and dispensing of antifungals to strictly trained and certified specialists. Unrestricted access to fungicides through uncertified retailers should also be discouraged. Good intervention programmes targeted at monitoring antifungal resistance patterns in groundwater is recommended for the Seke rural inhabitants to safeguard public health.