Variability in competitive fitness among environmental and clinical azole-resistant Aspergillus fumigatus isolates

ABSTRACT Azoles are the primary antifungal drugs used to treat infections caused by Aspergillus fumigatus. However, the emergence of azole resistance in A. fumigatus has become a global health concern despite the low proportion of resistant isolates in natural populations. In bacteria, antibiotic resistance incurs a fitness cost that renders strains less competitive in the absence of antibiotics. Consequently, fitness cost is a key determinant of the spread of resistant mutations. However, the cost of azole resistance and its underlying causes in A. fumigatus remain poorly understood. In this observation, we revealed that the 10 out of 15 screened azole-resistant isolates, which possessed the most common azole-targeted cyp51A mutations, particularly the presence of tandem repeats in the promoter region, exhibit fitness cost when competing with the susceptible isolates in azole-free environments. These results suggest that fitness cost may significantly influence the dynamics of azole resistance, which ultimately contributes to the low prevalence of azole-resistant A. fumigatus isolates in the environment and clinic. By constructing in situ cyp51A mutations in a parental azole-susceptible strain and reintroducing the wild-type cyp51A gene into the azole-resistant strains, we demonstrated that fitness cost is not directly dependent on cyp51A mutations but is instead associated with the evolution of variable mutations related to conidial germination or other unknown development-related processes. Importantly, our observations unexpectedly revealed that some azole-resistant isolates showed no detectable fitness cost, and some even exhibited significantly increased competitive fitness in azole-free environments, highlighting the potential risk associated with the prevalence of these isolates. IMPORTANCE Azole resistance in the human fungal pathogen Aspergillus fumigatus presents a global public health challenge. Understanding the epidemic trends and evolutionary patterns of azole resistance is critical to prevent and control the spread of azole-resistant isolates. The primary cause is the mutation of the drug target 14α-sterol-demethylase Cyp51A, yet its impact on competitive ability remains uncertain. Our competition assays revealed a diverse range of fitness outcomes for environmental and clinical cyp51A-mutated isolates. We have shown that this fitness cost is not reliant on cyp51A mutations but might be linked to unknown mutations induced by stress conditions. Among these isolates, the majority displayed fitness costs, while a few displayed enhanced competitive ability, which may have a potential risk of spread and the need to closely monitor these isolates. Our observation reveals the variation in fitness costs among azole-resistant isolates of A. fumigatus, highlighting the significant role of fitness cost in the spread of resistant strains.

immunocompromised individuals (1-3).Infections caused by A. fumigatus are currently treated with azoles as the first line of treatment, which prevent ergosterol biosynthe sis by targeting lanosterol 14-α-demethylase Cyp51A (4).Azole-resistant A. fumigatus isolates lead to a substantial burden of treatment failure and pose a grave threat to public health worldwide (5,6).One of the major mechanisms of resistance is associated with mutations in the cyp51A gene that involve tandem repeats (TR) in the promoter region in combination with point mutations in the coding sequence of the cyp51A gene (7,8).Among these mutations, TR 34 /L98H, which confers pan-azole resistance, is the most prevalent type found in clinical settings and environments (9)(10)(11).Drug-resistant isolates gain advantages over susceptible isolates in the presence of azole pressure.However, mutants that confer resistance tend to come with a fitness cost, and this fitness cost becomes a disadvantage in the absence of drug-selective pressure (12)(13)(14).Therefore, fitness plays a crucial role in determining the rate at which drug-resist ant isolates spread in natural populations (15)(16)(17).Understanding the fitness cost of azole-resistant A. fumigatus provides valuable insights into the distribution of drug-resist ant populations and their tendencies to expand in terms of frequency and geographical range.

Azole-resistant A. fumigatus isolates exhibit varied competitive fitness in vitro
TRs coupled with point mutations in the cyp51A gene are considered as the most common resistance mechanism (18).We therefore sought to dissect the impact of different types of TR repeat mutations on fitness.We first compared the colony growth and conidiation of the azole-resistant isolates with cyp51A mutations in solid minimal media and excluded those isolates with apparent morphological defects, which tended to harbor fitness costs when competing with susceptible isolates.To evaluate the overall trends of competitive fitness among azole-resistant isolates, we initially mixed eight isolates with TR 34 mutations and 16 isolates with TR 46 mutations, all showing no obvious colony defects, separately.Then, we compared these two mixtures (TR 34 mix and TR 46 mix) with a mixture of randomly selected susceptible isolates, all of which were obtained from different independent environmental and clinical sources (Fig. 1B).A competitive assay was employed to measure fitness in vitro (15,19).In this assay, the resistant isolates and susceptible strains were mixed at a 1:1 ratio and cocultured on solid media for 2 days.Then, total conidia were collected and plated onto media in the presence or absence of azole to determine the proportion of the resistant isolates and susceptible isolates (Fig. 1A).As shown in Fig. 1B, both mixed TR 34 and TR 46 isolates displayed significantly decreased competitive fitness when competing with the azole-susceptible isolates.These results suggested that azole-resistant A. fumigatus isolates from the environment and clinic may harbor fitness costs.However, it cannot rule out the possibility that at least one susceptible strain in the mixture of susceptible isolates may be more fit than the resistant strains and eventually dominate at the end of the competition assay.
To further investigate the factors contributing to fitness cost, 15 resistant isolates with no colony growth defects obtained from environment-, clinic-, and laboratory azoleinduced conditions were selected for one-to-one competition assay against correspond ing individual susceptible isolates in vitro (Fig. 1C; Fig. S1).These isolates contain different cyp51A mutations including TR 34 mutations (TR 34 /L98H and TR 34 /L98H/S297T/F495I), TR 46 mutations (TR 46 /Y121F/T289A), and point mutations (G54W and M220K and F46Y/ M172V/N248V/D255E/E427K) (Table S1).A total of 10 out of the 15 azole-resistant environmental isolates showed a significant reduction in colony-forming units (CFU) when cultured in azole-containing media over generations (Fig. 1B; Fig. S3), indicating that the 10 out of 15 resistant isolates exhibited a significant fitness cost when compet ing with the corresponding azole-susceptible isolates.Interestingly, we found that three TR 34 mutations ED-12L, ED-13L, and ED-29D did not manifest any discernible fitness cost despite harboring the same mutations as isolates ED-22G and CD-109, which did have a fitness cost.Notably, two isolates ED-T11 and CD-A77 with TR 34 mutations even exhibited increased fitness compared to the susceptible strains (Fig. 1C).Together, these results suggested that azole-resistant isolates exhibited varied competitive fitness in vitro.

The competitive fitness is independent of cyp51A mutations but might be associated with mutations related to germination rates induced by long-term azole conditions
To further explore the potential relationship between mutations in the cyp51A gene and altered competitive fitness, we constructed in situ cyp51A mutations in the triazolesusceptible A. fumigatus reference strain ΔakuB KU80 by homologous recombination in a marker-free manner.These mutations included TR repeat mutations TR 34 /L98H, TR 34 / L98H/S297T/F495I, and TR 46 /Y121F/T289A and point mutations G54W and M220K (Fig. 2A).Phenotypic characterization indicated that laboratory-constructed azole-induced resistant A. fumigatus isolates exhibited no phenotypic defects (Fig. S2).The competitive assays showed that all laboratory-constructed cyp51A mutants did not show any fitness cost (Fig. 2B and C; Fig. S3) when competing with the parental strain ΔakuB KU80 .In addition, we reintroduced the wild-type cyp51A gene into the two azole-resistant isolates from the environment (ED-T11) and clinic (CD-109), respectively.The resultant strains were then competed with the corresponding parental azole-resistant isolates, showing that there was no fitness cost for the resulting strains (Fig. S4).These results indicated that the competitive fitness is independent of cyp51A mutations.Interestingly, although the colony growth and conidiation of the azole-resistant isolates induced from the environment, clinic, and laboratory were similar to those of the susceptible isolates, we observed notable differences in the conidial germination rates between the resistant and susceptible isolates (Fig. 2C).Specifically, we found that 10 out of 11 resistant isolates with decreased fitness displayed a slower germination rate compared to the susceptible strains, while those with increased fitness exhibited a faster germination rate than that of susceptible strains.Accordingly, those resistant isolates with no significant change in fitness displayed no notable difference in germination rates when compared to the susceptible strains.Collectively, these results suggested that mutations in the cyp51A gene are not associated with fitness costs.Instead, we concluded that the occurrence of fitness cost in naturally isolated resistant isolates might be associated with muta tions that affected conidial germination rates or other unknown development-related processes during long-term azole treatment (Fig. 2D).In summary, we comprehensively evaluated the competitive fitness of environmental, clinical, and laboratory-induced A. fumigatus cyp51A mutants in vitro (Fig. 2D).These mutants exhibited no apparent colony growth defects, yet we observed distinct fitness outcomes among these isolates.Although most isolates showed defective fitness (66.7%), we observed that some resistant isolates exhibited no fitness cost (20%), suggesting that the development of azole resistance in A. fumigatus does not necessarily incur a fitness cost or that the presence of compensatory mechanisms mitigate the fitness cost during evolution (Fig. 1C).On the other hand, azole-resistant isolates with higher competitive fitness (13.3%) are intriguing and deserve further special attention due to the high risk for the rapid spread of resistant strains (Fig. 1C).Interestingly, the introduction of cyp51A mutations into an isogenic background and reintroduction of the wild-type cyp51A gene into the azole-resistant isolates did not yield any fitness cost, group (eight strains) and TR 46 mutant group (16 strains) for in vitro competition.Experiments were performed at least in triplicate with each bar representing the mean ± standard deviation (SD).Statistical analysis was performed using one-way analysis of variance (ANOVA) with multiple comparisons tests.**P < 0.01.(C) In vitro fitness assay of environment-derived (ED), clinical-derived (CD), and laboratory-derived (LD) azole-resistant isolates.The resistant and susceptible isolates were mixed in a 1:1 ratio, and the culture was transferred every 48 h at 37°C.Two susceptible strains, used for competing with ED and CD isolates, are randomly selected from the environment and clinic, respectively.The susceptible strains used for competing with LD isolates is commonly used laboratory background strain ΔakuB KU80 .The proportion of resistant isolates was counted after 3rd, 5th, 8th, and 10th transfers.Experiments were performed at least in triplicate with each bar representing the mean ± SD.Statistical analysis was performed using one-way ANOVA with multiple comparisons tests.**P < 0.01; ns, not significant.implying that cyp51A mutations are not directly associated with fitness costs (Fig. 2A and B; Fig. S4).These results were consistent with the previous fitness study using barcode DNA sequencing analysis (20).Further analysis revealed that it is possible that the fitness costs observed in environmental and clinical isolates may be attributed to alterations in other genes related to conidial germination during long-term evolution that are independent of cyp51A (Fig. 2C).Our observation reveals the variation in competitive fitness among azole-resistant A. fumigatus isolates from the environment and clinic, emphasizing the significant role of fitness ability in dissemination of resistant isolates.Further investigation is needed to unravel the mechanisms underlying these observa tions.

FIG 1
FIG 1 Competition experiments reveal varied competitive fitness among the isolated cyp51A mutants.(A) Schematic of the in vitro fitness assay by colony-form ing unit (CFU) counts.(B) A mixture of multiple susceptible isolates (nine strains) was used as the susceptible group, which was compared with the TR 34 mutant (Continued on next page)

4 FIG 2
FIG 2The competitive fitness is not influenced by cyp51A mutations.(A) Schematic of the construction of in situ cyp51A mutation isolates.(B) In vitro fitness assay of laboratory-constructed azole-resistant cyp51A mutants with susceptible parental strain ΔakuB KU80 .A 1:1 mixture of the cyp51A mutants and parental strain ΔakuB KU80 was subjected to competition assay at 37°C, with transfers every 48 h and tested for resistance ratios in the 3rd and 5th generations.Experiments (Continued on next page)