Modified Colorimetric Assay for Susceptibility Testing of Azole Antifungal Drugs against Candida Species

Abstract

We modified a rapid susceptibility assay (RSA) for antifungal susceptibility testing of azoles based upon glucose utilization. This modified RSA method provides quantitative endpoint readings in 6 h. In this study, the modified RSA and the National Committee for Clinical Laboratory Standards M27-A methods were used to determine the MICs of fluconazole and itraconazole for 118 Candida isolates. Yeast nitrogen base containing 0.12 g of glucose per liter was used for the modified RSA method. For fluconazole, agreement among assays within one or two twofold dilutions was 72.9 and 88.1%, respectively; for itraconazole, agreement within one or two twofold dilutions was 82.2 and 89.8%, respectively. These data suggest that the modified RSA method is a reliable and rapid method for azole antifungal susceptibility testing against Candida species.

The frequency of opportunistic fungal infections has increased over the past several years, and while several new antifungal drugs have been used in the treatment of these patients, the incidence of drug-resistant isolates, especially Candida species, has still increased dramatically (2, 7). Because systemic fungal infections generally occur in immunocompromised patients or in patients with severe underlying diseases, the prognosis for these patients is usually poor (1, 5, 10). As a result of these factors, the development of a reliable and rapid method to determine the susceptibility of isolates to antifungal drugs would therefore be helpful in the management of patients with invasive fungal disease.

While methods for in vitro susceptibility testing were developed as new antifungal drugs were identified, the reproducibility of these assays among laboratories was varied (3). In 1997 the National Committee for Clinical Laboratory Standards (NCCLS) published the reference method for broth dilution antifungal susceptibility testing of yeasts (6). This reference assay resulted in the standardization of key variables such as inoculum size, medium, and temperature and duration of incubation and also established MIC endpoints for several antifungal drugs. Application of the M27-A method has minimized the variation among laboratories (8). However, this method requires a 48-h culture before endpoints are measured for Candida species, and the interpretation of endpoints can sometimes be subjective. Therefore, a more rapid and convenient method is required for routine antifungal susceptibility testing of clinical isolates in diagnostic laboratories.

In 2000, Riesselman et al. (9) described a novel rapid susceptibility assay (RSA) method that uses colorimetric measurement for the determination of MICs of antifungal agents against Candida species. The rationale behind the development of this assay is that inhibition of growth by antifungal drugs also results in decreased uptake of exogenously added substrates such as glucose. By measuring the amount of residual glucose in the medium of drug-treated cultures and of untreated control cultures, the susceptibility of a fungal isolate to an antifungal drug can be determined.

Li et al. (4) used a similar method to test amphotericin B, 5-flucytosine, itraconazole, and fluconazole against 106 isolates of Candida species. The MICs obtained by RSA were compared with those obtained with the NCCLS M27-A method. The results indicated that after incubation for 6 h, the agreement between the two methods for determination of the MIC for amphotericin B and 5-flucytosine was high (100 and 98%, respectively), but that for fluconazole and itraconazole was low (63 and 61%, respectively). In order to improve the agreement between the two methods for assay of the MIC, cells required 19 h of incubation with itraconazole or fluconazole.

In this study, we modified the RSA method so that cells required only 6 h of incubation with fluconazole and itraconazole. The MICs of fluconazole and itraconazole for 118 Candida isolates obtained by the modified RSA method were compared to those obtained by the NCCLS M27-A method.

A total of 118 Candida isolates were obtained from the Research Center of Medical Mycology, Peking University. Candida albicans (45 isolates), C. glabrata (22 isolates), C. krusei (10 isolates), C. tropicalis (17 isolates), and C. parapsilosis (24 isolates) were tested. C. krusei (ATCC 6258) and C. parapsilosis (ATCC 22019) were included for quality control. Isolates were cultured on Sabouraud dextrose agar plates at 35°C and passaged twice at a 48-h interval before use. RPMI 1640 (Invitrogen, Carlsbad, Calif.) containing 2 g of glucose per liter was used for the NCCLS M27-A method. Yeast nitrogen base (YNB; Difco, Kansas, Mo.) containing 0.12 g of glucose per liter as the sole carbohydrate was used for the modified RSA. Itraconazole and fluconazole were generously provided by Janssen Pharmaceutica and Shanghai Sanve Pharmaceutica, respectively. The colorimetric reagent used to detect residual glucose in the culture medium contained 100 mmol of sodium phosphate (pH 7.0) per liter, 13 mmol of 4-aminoantipyrine per liter, 30 mmol of phenol per liter, 6 U of horseradish peroxidase per ml, and 13 U of glucose oxidase per ml.

The NCCLS M27-A method of antifungal susceptibility testing was performed by following the NCCLS recommendations (6). Briefly, Candida cells were suspended in RPMI 1640 medium, and the turbidity was adjusted to a McFarland unit of 0.5 by using a Densimat spectrophotometer (bioMérieux Italia, S.p.A). The suspension was further diluted in RPMI 1640 medium to yield an inoculum concentration of approximately 1 × 10³ to 5 × 10³ cells/ml. One hundred microliters per well was inoculated into wells containing dilutions of each drug. The final concentration ranges after inoculation were 0.03 to 16 μg/ml for itraconazole and 0.125 to 64 μg/ml for fluconazole. Drug-free purity controls and growth controls were included for each preparation. The plates were incubated at 35°C, and the results were determined after 24 and 48 h by visual inspection. As recommended by the NCCLS (6), the MIC was defined as the lowest drug concentration at which the turbidity was 20% or less than that of the control without drug.

For the modified RSA in this study, Candida cells were suspended in YNB medium containing 0.12 g of glucose per liter to a turbidity of a McFarland unit of 0.5. The suspension was further diluted in YNB with 0.12 g of glucose per liter to yield an inoculation concentration of approximately 1.25 × 10⁵ to 6.25 × 10⁵ cells/ml. One hundred microliters of this suspension per well was inoculated into wells containing various dilutions of drug. The final concentration ranges after inoculation were 0.03 to 16 μg/ml for itraconazole and 0.125 to 64 μg/ml for fluconazole, and the final glucose concentration in YNB was 0.12 g per liter. After the plates were incubated at 35°C for 6 h, 80 μl of freshly prepared colorimetric reagent was added to each well. The plates were then incubated at room temperature for 15 min to allow color development. The optical density of each well was determined spectrophotometrically with a microplate reader (Bio-Rad, Hercules, Calif.) at 490 nm. The optical density values, which reflect the residual glucose concentrations, were plotted against drug concentrations. The MIC was defined as the lowest drug concentration on the plateau preceding a 10% or greater drop in optical density. Controls included wells without drug but with a yeast strain (growth control) and wells without drug and without a yeast strain (blank control).

The MICs obtained by the two methods are shown in Table 1. The deviation of MICs obtained by the modified RSA method from those measured by the NCCLS M27-A method is presented as the change in MIC for the number of isolates (Table 2). The tested isolates were divided into susceptible, susceptible dose dependent, and resistant groups by MICs obtained with the modified RSA method and the NCCLS M27-A method. In Table 3, we compare the identities and differences between the two methods. From these data, we can see that the MICs obtained with the modified RSA method are in general compatible with those obtained from the NCCLS M27-A method both for the drugs and for most of the isolates tested.

TABLE 1.

MIC determined by the NCCLS M27-A method and modified RSA method

Candida species (no. of isolates)	Druga	MICb (μg/ml)
		NCCLS M27-A			Modified RSA
		Range	MIC50	MIC90	Range	MIC50	MIC90
C. albicans (45)	FLU	0.5-64	2	64	0.25-64	4	64
	ITRA	0.03-16	0.5	1	0.06-16	0.5	1
C. glabrata (22)	FLU	1-64	64	64	1-64	64	64
	ITRA	0.125-16	2	16	0.25-16	2	16
C. krusei (10)	FLU	8-64	32	64	8-64	64	64
	ITRA	0.25-2	1	2	1-16	1	16
C. parapsilosis (24)	FLU	0.5-64	4	64	2-64	16	64
	ITRA	0.125-16	0.25	16	0.125-16	0.5	16
C. tropicalis (17)	FLU	1-32	4	16	2-32	8	16
	ITRA	0.03-2	0.5	2	0.25-1	0.5	1
Total (118)	FLU	0.5-64	4	64	0.25-64	16	64
	ITRA	0.03-16	0.5	1	0.06-16	0.5	2

^aFLU, fluconazole; ITRA, itraconazole.

^bMIC₅₀, MIC for 50% of strains; MIC₉₀, MIC for 90% of strains.

TABLE 2.

MICs obtained by the modified RSA method that deviated from those measured by the NCCLS M27-A method

Candida species (no. of isolates)	Druga	No. of isolates with MIC difference (fold):				Agreement (%)
		0	1	2	>2	±1	±2
C. albicans (45)	FLU	18	19	4	4	82.2	91.1
	ITRA	18	21	3	3	86.7	93.3
C. glabrata (22)	FLU	10	3	5	4	59.1	81.8
	ITRA	10	6	2	4	72.7	81.8
C. krusei (10)	FLU	3	6	0	1	90.0	90.0
	ITRA	4	3	1	2	70.0	80.0
C. parapsilosis (24)	FLU	3	10	8	3	54.2	87.5
	ITRA	11	9	2	2	89.4	91.7
C. tropicalis (17)	FLU	6	8	1	2	82.4	88.2
	ITRA	12	3	1	1	88.2	94.1
Total (118)	FLU	40	46	18	14	72.9	88.1
	ITRA	55	42	9	12	82.2	89.8

^aSee Table 1, footnote a.

TABLE 3.

Comparison of identities and differences determined by the modified RSA and NCCLS M27-A methods^a

Candida species (no. of isolates)	No. of isolates
	Fluconazole				Itraconazole
	Identical	Major difference		Minor difference	Identical	Major difference		Minor difference
		S to R	R to S			S to R	R to S
C. albicans (45)	39	2	0	4	29	1	1	14
C. glabrata (22)	12	0	4	6	20	1	0	1
C. krusei (10)	5	1	0	4	7	0	0	3
C. parapsilosis (24)	6	1	0	17	15	1	0	8
C. iropicalis (17)	13	0	0	4	15	0	0	2

^aS, susceptible; S-DD, susceptible, dose dependent; R, resistant. Major difference, two full steps; minor difference, only one step.

Pathogens must utilize exogenous nutrients for their growth and reproduction. When drug-sensitive Candida strains are exposed to a medium containing an antifungal drug, their glucose uptake from the external milieu is impaired. The decrease in glucose consumption should precede the inhibition of their reproduction and can be measured earlier. On the other hand, drug-resistant Candida strains should still have the ability to take up glucose from the medium in the presence of drug, and glucose consumption was only slightly reduced. Therefore, measurement of the residual glucose in the medium is an alternative and time-saving way to determine the MIC.

In the RSA as well as other MIC assay methods, several factors, such as the density of inoculation, the growth rate of the strain, the period of incubation, and the antifungal mechanism of the drug used, influence the accuracy of the MIC in different ways. Specifically, the dependency on glucose and the consumption rate of glucose in different Candida species considerably affect the MIC obtained with the RSA method.

In the RSA method reported previously (4, 9), a higher concentration of glucose (0.5 g per liter) was used in the medium. At a higher concentration of glucose, passive diffusion of glucose into the cells may occur and is not driven by energy consumption. When the glucose concentration in the medium is low, as we used in this study, active transportation against the concentration gradient, which requires energy, may happen. Since energy production is compromised in susceptible cells cultured in medium containing the drug, MIC determination by the modified RSA method with a lower concentration of glucose should be more sensitive and can be performed in a relatively short period.

The inoculation density that we used was 1.25 × 10⁵ to 6.25 × 10⁵ CFU/ml, a concentration much higher than that used in the NCCLS M27-A method. The higher inoculation density will accelerate the utilization of glucose in the medium by the cells, so that the MIC can be measured after 6 h. Furthermore, a higher concentration is apparently needed for the slow-growing Candida species.

In this study, the optical densities at 490 nm of the wells containing Candida cells were measured directly after addition of the colorimetric reagent. To examine the impact of the cells in the wells on measurement of the MIC, MICs of fluconazole and itraconazole were measured for eight isolates of Candida in the modified RSA method. After incubation for 6 h, the optical densities at 490 nm were determined before addition of the colorimetric reagent as the blanks and after color development by addition of the colorimetric reagent. Optical densities subtracted from their respective blanks were obtained, and the calibrated MICs of the drugs against the isolates were calculated. As shown in Table 4, there were no differences between noncalibrated and calibrated MICs except those for an isolate of C. glabrata, for which the noncalibrated and calibrated MICs of itraconazole were 8 and 16 μg/ml, respectively. Therefore, optical density measurement with cells in wells is convenient and accurate enough for the MIC assay.

TABLE 4.

Noncalibrated and calibrated MICs for eight isolates

Isolate	MIC (μg/ml)
	Fluconazole		Itraconazole
	Non- calibrated	Calibrated	Non- calibrated	Calibrated
C. albicans isolate 1	64	64	16	16
C. albicans isolate 2	8	8	8	8
C. glabrata isolate 1	64	64	8	16
C. glabrata isolate 2	16	16	1	1
C. krusei	16	16	1	1
C. parapsilosis isolate 1	2	2	0.5	0.5
C. parapsilosis isolate 2	8	8	2	2
C. tropicali	4	4	0.5	0.5

As shown in Table 2, MICs obtained by the modified RSA method and the NCCLS M27-A method were fairly comparable for the Candida species and drugs tested. For the MICs of fluconazole obtained by the two methods, the difference was within one dilution in 72.9% of isolates and within two dilutions in 88.1% of isolates. As for itraconazole, the difference was within one dilution in 82.2% of isolates and within two dilutions in 89.8% of isolates. In the previous report on the RSA method compared with the NCCLS M27-A method, however, MIC differences within two dilutions were found in only 63 and 61% of isolates for fluconazole and itraconazole, respectively (4). This suggests that our modified RSA method is preferable, possibly due to the lower concentration of glucose used in the medium.

The resistance endpoints for fluconazole and itraconazole were clearly defined (6). Consequently, isolates can be divided into susceptible, susceptible dose dependent, and resistant groups by using the MICs from the modified RSA and the NCCLS M27-A methods. In Table 3, we compare the identities and differences in the results from the two methods. For C. albicans, C. glabrata, C. krusei, and C. tropicalis assessed by the two methods, most of the isolates tested showed identical results, and only a few isolates had very different results. These results indicate that the modified RSA is reliable in the assessment of susceptibility to fluconazole and itraconazole for these Candida species. For the 24 isolates of C. parapsilosis, however, only 6 isolates had identical results for fluconazole, and most of the different results were minor. C. parapsilosis grows slowly, and this may account for the inadequate results with the modified RSA method. Therefore, the modified RSA method used in this study may need to be further improved for Candida species that grow slowly, such as C. parapsilosis.

In conclusion, for most of the Candida species, the results for susceptibility to fluconazole and itraconazole obtained by the modified RSA method were in good compliance with those obtained by the NCCLS M27-A method, but the endpoints can be determined in 6 h when the former method is used, versus the 48 h necessary for the latter method. Susceptibility results provided in a short time may sometimes be very helpful in the treatment of critical patients infected with Candida species.