Antifungal Susceptibility and cyp51A Gene Variation Analysis of Aspergillus fumigatus Isolated from Soils in Tea-Growing Areas of Guizhou, China

1. Introduction

Aspergillus fumigatus is a ubiquitous saprophytic and opportunistic pathogenic fungus with strong environmental adaptability, which widely colonizes soil, air, plant rhizosphere, and organic-rich habitats such as humus and compost [1,2,3,4]. The conidia of A. fumigatus are widely transmitted via the air [5,6]. In immunocompromised individuals, inhalation of airborne conidia readily causes invasive aspergillosis, with an overall clinical mortality rate ranging from 40% to 90% [7,9,10]. Clinical therapeutic and prophylactic options for A. fumigatus-associated aspergillosis are relatively scarce. Azole antifungal agents, including itraconazole, posaconazole and voriconazole, serve as the first-line core medications for the clinical prevention and treatment of such fungal infections [11,12].

In recent years, a growing number of studies have reported the isolation of ARAF strains from patients with aspergillosis [13,14]. While azole resistance can be induced during antifungal therapy [10,15,16], the widespread application of agricultural azole fungicides for crop protection has now been firmly linked to the emergence of azole resistance in environmental A. fumigatus populations [10,17,18]. The cyp51A gene encodes the target enzyme lanosterol 14α-demethylase, and mutations in this gene represent the predominant molecular mechanism of azole resistance in clinical and environmental A. fumigatus isolates [19,20,21,22]. Accumulated evidence has demonstrated that amino acid substitutions at the G54W/E/R/V, Y121F, G138C, P216L, F219C, M220K/T, A284T, Y431C, G432A, G434C and G448S loci serve as the predominant mutations closely associated with azole resistance in this pathogen [18,23,24,25,26,27,28,29].

Guizhou Province is located in Southwest China, with an extremely high proportion of karst landforms, mountains and hills. It has a fragile ecological environment and is widely distributed with underground karst caves, which significantly increases the difficulty and cost of local transportation construction. Restricted by such geographical conditions, Guizhou has long remained in a relatively closed and isolated state in history [30,31,32]. Our research team previously conducted a surveillance on the prevalence of ARAF in soil samples from vegetable gardens of farmers at 9 sites in Guizhou Province. The results showed that the frequency of azole resistance in that population of A. fumigatus was only 0.49% (1/206), among the lowest reported so far in the global environmental populations of A. fumigatus [4].

As a core green tea producing area in China, Guizhou has prominent advantages in the development of the tea industry. In 2025, the Tea-Growing Area of the province reached 4700 square kilometers, the tea output reached 3.26×10⁷ kg, and the comprehensive output value of the tea industry exceeded 14.6 billion US dollars. It has become a distinctive and advantageous pillar industry in the region, effectively driving farmers’ employment and increasing their income. To further understand the epidemiological characteristics of ARAF in environmental samples in Guizhou, this study conducted extensive sampling in Tea-Growing areas at nine distinct geographic locations in Guizhou Province, determined the drug susceptibility of the tested strains to two common triazole drugs used for clinical treatment of aspergillosis—itraconazole and voriconazole—and performed DNA sequencing and sequence alignment analysis of the triazole target gene cyp51A for all strains. The objectives of this study are: (1) to clarify the prevalence and distribution characteristics of ARAF in Tea-Growing areas soils in Guizhou; (2) to explore the genetic variation patterns of the azole target gene cyp51A; (3) to investigate the correlation between cyp51A sequence variations and triazole minimum inhibitory concentration (MIC).

2. Materials and Methods

2.1. Soil Sampling, Isolation and Identification of A. fumigatus

Soil sampling was carried out across nine tea plantations in Guizhou from August 29 to September 1, 2023. At each of the nine tea plantations, 100 topsoil samples weighing roughly 10 grams were taken at a depth of 0 to 5 cm, with one-meter intervals between adjacent sampling points [33]. Geographic details of sampling sites are illustrated in Figure 1. Each soil sample was stored in a separate sterile zipper bag. Isolation of A. fumigatus was performed following the protocol described previously [4,18]. Initial and final identification of the strains was conducted according to the methods reported in our prior studies [33,34].

2.2. Susceptibility of A. fumigatus Isolates and cyp51A Gene Sequencing

Two clinical azole drugs (itraconazole and voriconazole) commonly used for the treatment of aspergillosis were used to test the susceptibility of A. fumigatus isolated in this study following the methods described in the CLSI M38-A3 [35] and our previous studies [4,18,33,34]. The minimum inhibitory concentration (MIC) is defined as the lowest drug concentration that achieves complete (100%) inhibition of mycelial growth of the tested strains by visual observation. MIC₅₀ refers to the lowest concentration capable of inhibiting the growth of 50% of the tested strain population; MIC₉₀ represents the lowest concentration that inhibits the growth of 90% of the tested strain population. Two primer pairs, A7 (5′-TCATATGTTGCTCAGCGG-3′) and P450-A2 (5′ -CTGTCTCACTTGGATGTG- 3′) [27], were used for amplifying and sequencing the full-length cyp51A gene (encompassing coding and promoter regions) from all strains isolated in this study.

2.3. cyp51A Gene Mutation Analysis and Phylogenetic Analysis

Mutations of cyp51A gene and its promoter region were identified by comparing with the reference sequence of a wild-type azole-susceptible A. fumigatus strain under the accession number AF338659 in GenBank [18,36,37]. The polymorphism of the cyp51A gene was analyzed using DnaSP software [38]. Sequence alignment and phylogenetic tree construction of representative sequences were performed using MEGA 6.0 software [39]. To investigate whether the cyp51A sequences in our study were unique to Guizhou and evolutionarily clustered together, we compared our cyp51A gene sequences in A. fumigatus with those from diverse global geographical origins. Here, a total of 1303 cyp51A gene sequences of A. fumigatus were retrieved from the NCBI database for comparison. These sequences originated from 19 countries across the globe: China contributed 802 sequences, followed by France (195), Japan (71), Italy (61), the United Kingdom (36), the Republic of Korea (26), India (24), Portugal (19), Austria (18), Spain (14), Brazil (13), the United States (9), Canada (4), Australia (4), Kuwait (3), Peru (3), Denmark (2), and the Netherlands and Colombia each supplied one sequence. Phylogenetic analysis of the total dataset used the MEGA 6.0 software [39].

2.4. Data Analysis

Box plots were generated to illustrate the distribution of minimum inhibitory concentrations (MICs) of each antifungal agent across different geographical populations. The Kruskal-Wallis test was used to analyze differences in MIC values among geographical groups. All box plot construction and statistical analyses were performed using GraphPad Prism 10.6.0 (GraphPad Software, San Diego, California, USA). IBM SPSS Statistics 22.0 was used to analyze the potential correlations between cyp51A gene mutation sites and antifungal MIC values.

3. Results

3.1. Isolation and Susceptibility of A. fumigatus Isolates

In this study, 191 A. fumigatus strains were isolated and identified from 900 soil samples collected from 9 Tea-Growing areas in Guizhou Province (Figure 1). Among them, 22 strains were isolated from Guiyang and Qiandongnan respectively, and 21 strains were isolated from each of the other seven sampling sites. The isolation frequencies of A. fumigatus at each sampling site ranged from 20% to 21%, with similar isolation frequencies across sites. Antifungal susceptibility testing showed that all 191 A. fumigatus isolates in this study were susceptible to azole drugs. For itraconazole, the MIC values ranged from 0.015 to 0.5 μg/ml, the geometric mean MIC (GM-MIC) was 0.067 μg/ml, with a MIC₅₀ of 0.063 μg/ml and a MIC₉₀ of 0.25 μg/ml. The A. fumigatus population from Guiyang exhibited the highest GM MIC and MIC₅₀ values, which were 0.142 μg/ml and 0.125 μg/ml, respectively, while the highest MIC₉₀ value (0.5 μg/ml) was observed in the Liupanshui population (Table 1). For voriconazole, the MIC distribution spanned 0.031 to 0.25 μg/ml, yielding a GM-MIC of 0.086 μg/ml, a MIC₅₀ of 0.063 μg/ml, and a MIC₉₀ of 0.125 μg/ml. The A. fumigatus population from Qianxinan exhibited the highest GM MIC and MIC₉₀ values, which were 0.11 μg/ml and 0.25 μg/ml, respectively. The MIC₅₀ values for Zunyi, Qiandongnan, Qianxinan, Liupanshui, and Tongren were all 0.125 μg/mL (Table 1). Interestingly, statistical analysis results indicated that there were significant differences in the MIC distributions of itraconazole and voriconazole among A. fumigatus from some geographical populations. For instance, in the itraconazole susceptibility test, the A. fumigatus population in Guiyang was significantly different from all other geographical populations except that in Liupanshui, while the Liupanshui population showed significant differences from both the Qianxinan and Qiandongnan populations. In the voriconazole susceptibility test, the A. fumigatus population in Guiyang exhibited significant differences from those in Zunyi, Tongren, Liupanshui, Qianxinan, and Qiandongnan (Figure 2).

3.2. cyp51A Polymorphism and Correlation of Mutation Sites with Triazole MICs

In this study, we successfully obtained the complete gene sequence of the cyp51A gene from 191 A. fumigatus strains. A total of 10 polymorphic sites (S) were detected and 6 haplotypes (h) were identified. The overall haplotype diversity (Hd) was 0.268, and the nucleotide diversity (Pi) was 0.00029 (Table 2). Sequence alignment comparisons with the reference sequence identified a total of 29 A. fumigatus strains carrying base substitutions in the cyp51A gene. The number of mutant strains and corresponding mutation frequencies in each geographic region were as follows: Bijie (6 strains, 28.57%, 6/21), Zunyi (2 strains, 9.52%, 2/21), Qiannan (2 strains, 9.52%, 2/21), Anshun (2 strains, 9.52%, 2/21), Tongren (3 strains, 14.29%, 3/21), Guiyang (3 strains, 13.64%, 3/22), Liupanshui (4 strains, 19.05%, 4/21), Qianxinan (4 strains, 19.05%, 4/21), and Qiandongnan (3 strains, 13.64%, 3/22). Among the 10 mutation sites located in exons, 4 were synonymous mutations (267G→A, 540G→A, 1074A→G, and 1362T→C), and 6 were non-synonymous mutations (137T→A, 514A→G, 743A→C, 744T→A, 765C→G, 1279G→A). The mutation at position 744 on the CDS had the highest frequency of 0.1204 (22/191), followed by positions 514 and 540, both with a frequency of 0.0209 (4/191) (Table S1). No mutations were found in the cyp51A gene of the remaining 162 A. fumigatus strains. Based on 10 polymorphic loci, the cyp51A gene of 191 A. fumigatus strains was classified into 6 genotypes. Genotype 6, which harbored wild-type strains, had the highest frequency at 0.839 (162/199). Genotype 5 was distributed across all 9 sampling sites. Genotype 1 was detected in Guiyang and Tongren, while Genotype 4 was found in Bijie and Qianxinan. Genotype 2 and Genotype 3 were identified exclusively in Bijie and Qiandongnan, respectively (Table 3). Phylogenetic analysis based on the cyp51A genotype revealed that Genotype 1 exhibited the greatest genetic divergence from the other six genotypes. This genotype harbored eight specific variant loci at positions 137, 267, 514, 743, 765, 1074, 1279, and 1362. Genotype 4 and Genotype 6 were closely genetically related, with only a single nucleotide difference detected at position 540 (Figure S1). The results of the statistical analysis demonstrated that all variant loci of the cyp51A gene (137, 267, 514, 540, 743, 744, 765, 1074, 1279, 1362) in this study showed no significant correlation with the minimum inhibitory concentration (MIC) values of itraconazole and voriconazole (p > 0.05) (Table 4).

3.3. cyp51A Amino Acid Substitution and Phylogenetic Analysis

Of the 29 A. fumigatus strains harboring cyp51A gene mutations, 25 exhibited amino acid substitutions at the protein level, including 4 strains from Bijie, 4 from Liupanshui, 3 from Tongren, Guiyang, and Qiandongnan each, and 2 from Zunyi, Qiannan, Anshun, and Qianxinan each. A total of 10 variant loci in the coding sequences (CDS) resulted in amino acid substitutions at 5 positions of the protein encoded by the cyp51A gene, namely F46Y, M172V, N248T/K, D255E, and E427K. Among these substitutions, N248T/K at position 248 exhibited the highest frequency and was distributed across all sampling sites, showing obvious geographical broad-spectrum characteristics, with an occurrence frequency of 0.1309 (25/191). The remaining four substitution loci (F46Y, M172V, D255E, and E427K) all had a frequency of 0.0105 (2/191) and were only detected in the Guiyang and Tongren sampling sites (Table 5).

Phylogenetic analysis was performed using 1494 A. fumigatus cyp51A sequences, including 1303 sequences downloaded from the NCBI database and 191 sequences obtained from isolates collected in Guizhou. The resulting phylogenetic tree was divided into two clusters (A and B). Cluster A comprised 1484 sequences with wide geographical representation, whereas Clade B contained 10 sequences originating from the USA (9 sequences) and Spain (1 sequence). All the six sequence types from our current study belonged to cluster A. While there were some geographic-specific subclusters within cluster A, the six sequence types from this study were broadly distributed across the cluster A portion of the cyp51A gene tree (Figure S2)..

4. Discussion

4.1. Extremely Low Prevalence of Triazole Resistance and Absence of Resistant A. fumigatus Strains

In this study, we systematically conducted azole antifungal susceptibility testing on 191 strains of A. fumigatus isolated from the soil of 9 tea plantations in Guizhou Province, the core green tea-producing region of China. Among them, the susceptibility test results for itraconazole and voriconazole showed that the MIC of all tested strains was below 1.0 µg/ml, and no azole-resistant strain was detected. This indicates that A. fumigatus isolates in the soil of Guizhou tea plantations are highly susceptible to the above two azole drugs. In an earlier study, we conducted a similar investigation on ARAF using the soil of 9 rural vegetable gardens across Guizhou Province as the research object. The results showed that the incidence of azole resistance was only 0.49% (only 1 out of 206 strains was resistant) [4]. The findings of these two studies suggest that the prevalence of ARAF in agricultural soil samples from Guizhou Province is extremely low, much lower than the findings of previous investigations conducted in the neighboring Yunnan Province [18,33,34] and other regions of China [40,41]. Globally, although a similarly low triazole resistance frequency of 0.27% has been reported in soil A. fumigatus populations in Canada [42], azole resistance rates in A. fumigatus from most regions worldwide are generally much higher. The resistance rate reaches up to 50% in some areas outside China [43,44,45] and reaches nearly 80% in some domestic regions [18,41]. In the previous study on vegetable garden soil, we have initially speculated that the low incidence of ARAF in Guizhou Province may be mainly attributed to four aspects: first, the strict control of agricultural pesticide use by the local government, which effectively restricts the abusive use of azole fungicides; second, the long-term adherence of local farmers to traditional green planting models, which reduces the dependence on chemical pesticides; third, the unique karst landforms and the perennial low-temperature and humid climatic characteristics of Guizhou Province, which may have a certain inhibitory effect on the growth and reproduction of A. fumigatus and its azole resistance mutations; fourth, the limited gene flow of A. fumigatus among different geographical populations, making it difficult to achieve the wide spread of azole resistance genes [4]. The sampling sites of soil samples collected from tea-growing areas in this study were highly consistent with those of previous vegetable garden soil samples, both located in typical agricultural planting areas in Guizhou Province. Therefore, it is speculated that the core reasons for the low incidence of ARAF are consistent with those in the previous study. In addition, through in-depth communication with local tea planting experts, it was learned that the overall incidence of fungal diseases in Guizhou tea planting is relatively low; and for occasional tea fungal diseases, local farmers generally adopt the prevention and control measure of “directly destroying diseased plants in the early stage of infection”. This measure can eliminate the spread and transmission of fungal pathogens from the source, thereby reducing the usage of various chemical pesticides, including triazole fungicides, further reducing the probability of A. fumigatus being exposed to azole fungicides, and providing additional guarantee for the maintenance of its low azole resistance level.

4.2. Non-Azole-Resistance Mutation Sites in the cyp51A Gene

In the present study, the cyp51A genes of 191 environmental A. fumigatus isolates were sequenced and analyzed. A total of 29 isolates were identified to harbor different base substitution mutations, with an overall mutation frequency of 15.19% (29/191). Combined with the in vitro antifungal susceptibility data, further analysis showed that these cyp51A base substitutions did not lead to a notable increase in the MIC of itraconazole and voriconazole, and no significant changes were observed in the drug-susceptible phenotypes of the isolates. This finding is consistent with the molecular epidemiological results of A. fumigatus reported worldwide [29,36,46,47,48], suggesting that abundant non-resistance-related genetic variations commonly exist in the cyp51A gene of natural A. fumigatus populations. Most gene mutations without phenotypic differences in drug susceptibility are silent or neutral variations, which only reflect genetic polymorphisms at the nucleotide level and cannot directly mediate the emergence of triazole-resistant phenotypes [29,47]. At present, the evolutionary pressure, environmental selection factors and molecular regulatory mechanisms responsible for the high prevalence of non-resistant mutations in the cyp51A gene remain unclear. Further investigations with expanded sample sizes, combined with regional environmental features and molecular evolutionary analysis, are therefore needed to systematically clarify the underlying causes and biological significance of these mutations.

4.3. Widespread Distribution of the CYP51A Mutations

In this study, five amino acid substitution mutations were identified in the cyp51A gene, including F46Y, M172V, N248T/K, D255E, and E427K. Among all mutation variants, N248K presented the highest overall detection frequency with a mutation rate of 0.1205 (23/191). This mutation was widely distributed across all nine geographical sampling areas surveyed in this study, and its spatial distribution pattern was highly consistent with previous investigations of natural A. fumigatus populations across different regions of China [2,18,49,50]. Accumulating clinical studies have demonstrated that the N248K variation can reduce the susceptibility of A. fumigatus to azole antifungal agents and weaken their inhibitory activity, thereby exerting adverse impacts on the clinical management of fungal infections [48,51]. Nevertheless, numerous epidemiological investigations and molecular phenotypic analyses have further confirmed that environmental A. fumigatus strains carrying only the single N248K mutation remain phenotypically susceptible to conventional azoles. This single-site variation cannot independently mediate the development of drug resistance and shows no significant statistical correlation with the resistant phenotype [29,49,50]. The above research conclusions are completely consistent with the results of drug susceptibility testing and genetic variation analysis obtained in the present study.

In this study, a total of two A. fumigatus isolates carrying five amino acid substitution mutation sites, namely F46Y, M172V, N248T, D255E and E427K, were identified from the populations in Tongren and Guiyang. These five mutations occurred in linkage. The distribution pattern of this compound mutation is consistent with previous studies on A. fumigatus populations in other regions of China [18,49,52] and across the globe [29,47,53], which further confirms the prevalence of such multi-site combined mutations in environmental strains. Antifungal susceptibility tests showed that the two above-mentioned isolates, derived from Tongren and Guiyang respectively, had itraconazole minimum inhibitory concentration (MIC) values of 0.125 µg/mL and 0.25 µg/mL, and both exhibited a voriconazole MIC of 0.063 µg/mL. All MIC values were lower than the clinical resistance breakpoints for azole drugs, and the isolates remained phenotypically susceptible with no azole resistance observed. The results indicate that this multi-site compound mutation pattern cannot induce azole resistance in A. fumigatus, which is highly consistent with the findings of existing domestic [49,50] and international studies [29,54,54].

Of the six cyp51A sequence types identified here, three (genotypes 2, 3, and 4) were newly reported in this study and found so far only in Guizhou. Two of these three genotypes (genotypes 2 and 3) were singletons, represented by only one strain each, while genotype 4 was represented by three strains. This result is consistent with at least some unique genetic diversity of cyp51A in A. fumigatus in Guizhou. In contrast, the more common in Guizhou, genotypes 1, 5, and 6, were shared broadly with strains from other regions in China as well as from outside of China, consistent with the Guizhou population of A. fumigatus being part of the global metapopulation. Interestingly, cyp51A genotypes 1, 5, and 6 were not clustered together on the phylogenetic tree but are broadly distributed, indicating recent or ongoing genetic exchanges among geographic populations of A. fumigatus between Guizhou and those from outside of Guizhou.

5. Conclusions

In this study, 191 A. fumigatus strains were isolated from tea plantation soils across nine geographical regions of Guizhou Province. Antifungal susceptibility to triazoles was determined, and the mutational characteristics of the drug target gene cyp51A were systematically analyzed. The extensive sampling coverage is representative of the baseline genetic characteristics of natural A. fumigatus populations in Guizhou tea-growing ecosystems. Antifungal susceptibility tests indicated that all isolates were susceptible to triazole antifungals, with MIC values no more than 0.5 μg/mL. Non-resistance-related mutations in the cyp51A gene were detected in 29 strains, yielding a mutation frequency of 15.18% (29/191). Notably, the nucleotide mutation at position 744 in the CDS region of cyp51A was prevalent among all sampling sites, with a frequency of 12.04% (23/191). This study further demonstrates that environmental A. fumigatus strains harbor abundant naturally occurring genetic variations that are not associated with triazole resistance.