The impact of respiratory tract mimicking media on antifungal susceptibility testing

TL;DR

This study compares how different media affect antifungal susceptibility testing, finding that respiratory tract-mimicking media can change results for some fungi.

Contribution

The study introduces respiratory tract-mimicking media as a novel testing environment for antifungal susceptibility.

Findings

Candida species showed greater MIC variability in respiratory tract-mimicking media compared to RPMI1640.

Aspergillus terreus had higher MICs for certain antifungals in respiratory tract-mimicking media.

Respiratory tract-mimicking media may better reflect in vivo conditions for antifungal susceptibility testing.

Abstract

The rise in antifungal resistance and changing epidemiology underscores the need to improve antifungal susceptibility testing. This study tested the standard medium RPMI1640 and healthy lung and sinus media against Aspergillus, Mucor, and Candida species. Candida showed greater variability in minimum inhibitory concentrations (MICs) compared with molds, with higher MICs in respiratory tract-mimicking media than in RPMI1640. Aspergillus terreus was the most affected mold, showing higher MICs for certain antifungals in these media.

Tables

TABLE 1. Susceptibility profiles of amphotericin B (AmB), voriconazole (VRC), isavuconazole (ISA), and anidulafungin (ANF) against some mold representatives species in three different media; RPMI1640, Healthy Sinus Medium (HSM), and Healthy Lung Medium (HLM), according to the EUCAST methodb

Species	Antifungals	MIC-MEC [mg/L]-Media
		RPMI1640	HSM	HLM	Susceptibility categorization
A. fumigatus 1	AmBVRCISAANF	18>80.007	1>8>80.015	2>8>80.015
A. fumigatus 2	AmBVRCISAANF	28>80.007	1>8>80.015	28>80.031
A. fumigatus 3	AmBVRCISAANF	2110.007	1210.015	20.510.031
A. flavus 1	AmBVRCISAANF	8240.007	8220.015	4220.015
A. flavus 2	AmBVRCISAANF	2110.015	1220.031	2120.015
Mucor spp. 1	AmBVRCISAANF	0.25–a>8–	0.125–>8–	1–>8–
Mucor spp. 2	AmBVRCISAANF	0.5–>8–	0.5–>8–	0.25–>8–
A. niger 1	AmBVRCISAANF	1280.015	2480.007	4480.015
A. niger 2	AmBVRCISAANF	1120.031	1240.007	1240.007
A. terreus s.s 1	AmBVRCISAANF	2420.007	>88>80.015	>8880.031	Wt to Non-Wt
A. terreus s.s 2	AmBVRCISAANF	1420.003	4880.007	>8880.015	Wt to Non-Wt
A. terreus s.s 3	AmBVRCISAANF	2420.003	>8840.015	>8880.015	Wt to Non-Wt
A. terreus s.s 82	AmBVRCISAANF	>8220.007	>8480.015	>8440.015
A. terreus s.s 192	AmBVRCISAANF	>8110.007	>8480.015	>8240.015
					S to R
A. terreus s.s 9	AmBVRCISAANF	>8110.007	>8240.015	>8240.015
A. terreus s.s S2	AmBVRCISAANF	10.510.007	1120.015	0.5120.015
A. terreus s.s S3	AmBVRCISAANF	0.50.510.007	1120.015	0.5120.015
A. citrinoterreus 27	AmBVRCISAANF	>80.50.50.007	>8120.015	>8210.007
A. alabamensis 63	AmBVRCISAANF	8110.007	>8240.015	>8240.015
A. hortae 119	AmBVRCISAANF	810.50.007	>8240.007	>8220.015
A. hortae 142	AmBVRCISAANF	>80.510.007	>8240.015	>8240.015
A. citrinoterreus 220	AmBVRCISAANF	>8220.007	>8240.015	>8240.031
A. alabamensis 225	AmBVRCISAANF	>8220.003	>8220.007	>8220.015
A. fumigatus (ATCC 204305)	AmBVRCISAANF	210.50.015	1110.015	2210.015
A. flavus (ATCC 204304)	AmBVRCISAANF	2210.007	2220.031	4420.031

TABLE 2. Susceptibility profiles of amphotericin B (AmB), voriconazole (VRC), isavuconazole (ISA), and anidulafungin (ANF) against some yeast representatives species in three different media: RPMI1640, Healthy Sinus Medium (HSM), and Healthy Lung Medium (HLM), according to the EUCAST methodb

Species	Antifungals	MIC [mg/L] -Media
		RPMI1640	HSM	HLM	Susceptibility categorization
C. albicans 1	AmBVRCISAANF	0.50.0080.0080.008	0.25>4>40.015	0.25>4>40.015
					S to R
C. albicans 2	AmBVRCISAANF	0.510.0150.015	0.25>40.1250.008	0.25>40.1250.008
C. tropicalis 1	AmBVRCISAANF	0.50.060.0080.015	0.520.060.007	0.50.250.030.007
					S to R
C. tropicalis 2	AmBVRCISAANF	0.540.0080.03	>4>40.060.007	1>4>40.007	S to R
					S to R
N. glabratus 1 (formerly Candida glabrata)	AmBVRCISAANF	0.50.1250.0150.015	–a – – –	0.50.1250.060.03
N. glabratus 2 (formerly Candida glabrata)	AmBVRCISAANF	0.50.060.030.008	– – – –	10.1250.060.03
P. kudriavzevii 1 (formerly Candida krusei)	AmBVRCISAANF	1 0.5 0.25 0.03	>4 >4 >4 >4	>4 >4 >4 >4	S to R
					Wt to Non-Wt
					S to R
P. kudriavzevii 2 (formerly Candida krusei)	AmBVRCISAANF	10.250.0080.03	10.50.50.06	40.50.50.06
K. marxianus 21146779 (Candida kefyr)	AmBVRCISAANF	10.0150.0150.015	0.50.1250.250.06	– – – –
C. parapsilosis(ATCC 22019)	AmBVRCISAANF	10.030.0081	0.50.1250.060.5	10.060.031
C. krusei (ATCC 6258)	AmBVRCISAANF	10.250.0080.03	0.510.50.03	10.50.50.03

Full text

INTRODUCTION

Invasive fungal infections (IFIs) have been rising in recent decades, primarily due to the growing population of immunosuppressed patients (1), which has led to increased reliance on systemic antifungal drugs, such as echinocandins, polyenes, and triazoles (2). However, the widespread use of these drugs has contributed to developing antifungal resistance, posing a significant challenge to treatment efficacy (3). Consequently, antifungal susceptibility testing (AFST) has become increasingly crucial for optimizing therapeutic strategies, guiding patient management, and monitoring antifungal resistance (4, 5). AFST is recommended in situations, such as confirmed or suspected IFIs, suspected cases of acquired resistance, and for patients experiencing refractory, relapsing, or breakthrough infections (6, 7). Several methods of AFST are currently used or under development (5). The main objective behind the standardization and periodic optimization of reference guidelines, such as those from the Clinical and Laboratory Standards Institute (CLSI) and the European Committee on Antimicrobial Susceptibility Testing (EUCAST), is to minimize the effect of various factors on AFST results, including inoculum size, incubation time, temperature, shaking conditions, and the composition of the medium (5, 8, 9). The growth medium is presumably one of the most profoundly influential factors of AFST results among the parameters noted, affecting the minimum inhibitory/effective concentration (MIC/MEC) value (8, 10, 11). The significance of host-mimicking environments in accurately reflecting pathogen behavior is widely recognized. However, RPMI_1640_, commonly used as a standard medium in reference AFST, may inadequately replicate the chemical and immunological complexity of the host environment, which underscores the need for further investigation into more representative testing media.

To address this gap, we evaluated the effect of culture medium on in vitro AFST by comparing artificial respiratory tract media, Healthy Sinus Medium (HSM) (12) and Healthy Lung Medium (HLM) (12), with standard RPMI_1640_ medium (13). The activity of frequently used representative antifungal agents from three major classes, polyene (amphotericin B; AmB, Sigma-Aldrich, A4888), azoles (voriconazole; VRC, Sigma-Aldrich, PZ0005; isavuconazole; ISA, Sigma-Aldrich, SML2357), and echinocandins (anidulafungin; ANF, Sigma-Aldrich, SML2288), commonly utilized in clinical diagnostic laboratories, was assessed against selected mold and yeast pathogens typically associated with infections of the lung and sinuses (14–17), including Aspergillus fumigatus (n = 3), Aspergillus flavus (n = 2), Mucor spp. (n = 2), Aspergillus niger (n = 2), Aspergillus terreus sensu stricto (s.s) (n = 8), Aspergillus citrinoterreus (n = 2), Aspergillus alabamensis (n = 2), Aspergillus hortae (n = 2), Candida albicans (n = 2), Candida tropicalis (n = 2), Nakaseomyces glabratus (formerly Candida glabrata) (n = 2), Pichia kudriavzevii (formerly Candida krusei) (n = 2) and Kluyveromyces marxianus (formerly Candida kefyr) (n = 1), according to the EUCAST guideline (13, 18). A. fumigatus (ATCC 204305), A. flavus (ATCC 204304), C. krusei (ATCC 6258), and C. parapsilosis (ATCC 22019) were used as control strains. The minimum inhibitory concentrations (MICs), defined as the concentration at which no hyphal growth was detected, were determined for amphotericin B and azoles. For echinocandins, the minimal effective concentrations (MECs), which represent the concentration at which hyphal growth was significantly altered and blunted colonies formed, were assessed. Susceptibility testing was performed in duplicate, with any significant variations in MIC or MEC defined as changes greater than two log₂ dilutions.

The MIC and MEC values for all tested mold and yeast species are presented in [Tables 1 and 2](#T1 T2), respectively. Among the mold species, most A. terreus s.s. isolates and one A. hortae isolate predominantly showed higher MICs (more than two log2 dilutions), particularly for AmB and occasionally ISA, in HSM and HLM media compared with RPMI_1640_ (Table 1). In comparison, among the yeast species, a broader range of Candida species—including C. albicans, C. tropicalis, P. kudriavzevii (formerly Candida krusei), and K. marxianus (Candida kefyr)—exhibited higher MIC values in respiratory-mimicking media than in RPMI_1640_, primarily for azoles (Table 2). Positive growth controls indicated that each of the three media tested successfully supported fungal growth at the designated reading times. Overall, tested Aspergillus species were generally less influenced by the tested media compared to Candida species, with results indicating more variability among individual strains rather than a consistent response pattern. Selecting the appropriate test medium that supports fungal growth without interfering with drug activity is essential (5), as fungal isolates grow at different rates on various media, leading to variable MIC values (5, 19). It has been demonstrated by some studies (20–22) that the susceptibility of fungi depends on the growth stage, with some indicating increased susceptibility in the exponential phase compared to the lag phase; this may explain how nutrient-limited media can lead to a prolonged lag phase for the organism and consequently affect the potential efficacy of the applied antifungal, underscoring the importance of the testing medium’s impact on AFST results. While no single culture medium can support the growth of all clinically relevant fungi, RPMI_1640_, a fully defined synthetic formula with minimal batch variation, was selected as the standard medium for AFST due to its high interlaboratory agreement, specifically for yeast (23, 24), although there was no evidence that this medium was suitable for filamentous fungi (25). However, certain disadvantages raise concerns about RPMI_1640_ as a suitable medium for all fungi and antifungals, including its nutrient-limited composition, which can impact fungal growth (19), and its inability in some cases to differentiate between amphotericin B-resistant or tolerant isolates and susceptible ones (9, 11, 26). Therefore, the opinion that RPMI_1640_ medium is universally suitable for all fungal species and antifungals warrants careful reconsideration, and consistent results for quality control strains should not be considered as definitive evidence of its effectiveness for all fungi and antifungals. In vitro drug susceptibility tests are usually performed under artificial growth conditions that supply the organism with readily metabolized carbon, such as glucose, and nitrogen, predominately glutamine and arginine, usually at non-physiological levels (12). Such conditions are known to inhibit the expression of virulence-associated genes in vitro (27).

To address these limitations, HSM and HLM were examined to more accurately resemble the nutrient composition and physiological environment of the healthy human respiratory tract. The choice of “healthy” conditions is intended to establish a controlled, consistent, and standardized baseline that reflects normal host–pathogen interactions, in the absence of inflammation or immunosuppression, which can vary between individuals. While both HSM and HLM media incorporate host-like carbon and nitrogen sources, the sinus medium is more nutritionally limited, particularly in carbon and nitrogen availability (12). Both media also contain ionic profiles, protein concentrations, host antimicrobial peptides and enzymes, mucin, and extracellular DNA, components that are absent in RPMI_1640_ (12). These physiologically relevant conditions can influence fungal metabolism and growth dynamics, potentially impacting AFST, particularly for agents whose activity is modulated by the fungal metabolic state and membrane composition (19, 20, 29). Further investigation is warranted to assess the growth kinetics of representative mold and yeast isolates in airway-mimicking media compared with RPMI_1640_, to clarify the potential influence of growth stages and metabolic state across media on antifungal efficacy. With further validation, HSM and HLM may serve as more physiologically relevant and standardized alternatives to RPMI_1640_ for AFST, helping to bridge the gap between overly simplistic in vitro conditions and the complex in vivo airway environment, particularly for fungal pathogens associated with the sinuses and lungs.