Selective Culture Medium for Screening of Fosfomycin Resistance in Enterobacterales

ABSTRACT

A selective medium for screening fosfomycin (FOS)-resistant Enterobacterales was developed. Performances of this medium were first evaluated by using cultures of a collection of 84 enterobacterial clinical strains (42 FOS susceptible and 42 FOS resistant). The SuperFOS medium showed excellent sensitivity and specificity of detection (100%) in those conditions. Then, by testing spiked stool and spiked urine specimens, it revealed excellent performances, with lower limits of identification ranging from 10¹ to 10² CFU/ml. This screening medium allows easy and accurate detection of FOS-resistant isolates regardless of their resistance mechanisms.

INTRODUCTION

Fosfomycin (FOS) is a phosphonic antibiotic approved for the treatment of uncomplicated urinary tract infections (UTI) in the early 1970s, naturally produced by Streptomyces species such as Streptomyces fradiae, Streptomyces viridochromogenes, and Streptomyces wedmorensis and by some Pseudomonas species such as Pseudomonas syringae and Pseudomonas viridiflava (1–4). The mechanism of antibacterial action of FOS is based on the inhibition of the first steps of peptidoglycan synthesis. Its structure is analogous to phosphoenolpyruvate (PEP), which is used by the enzyme MurA (UDP-N-acetylglucosamine enolpyruvyl transferase). Hence, FOS binds to MurA and consequently induces cell lysis, eventually leading to bacterial death (5).

Three main mechanisms of resistance to FOS have been described as follows: (i) reduced permeability to FOS through mutations in the nutrient transport systems genes, namely, glpT and uhpT; (ii) modification of the target through the overexpression of the murA gene; and (iii) modification of the antibiotic through production of metalloenzymes that catalyze the opening of the epoxide ring by adding glutathione (so-called FosA1-FosA10, FosL1-2, and FosC2 enzymes), bacillithiol (FosB), and water (FosX) (5–15). A series of Gram-negative species have been shown to possess intrinsic FOS resistance genes in their chromosome, such as Klebsiella spp., Enterobacter spp., Serratia marcescens, and Pseudomonas aeruginosa, although they are absent in many others (Escherichia coli, Acinetobacter baumannii) (16). Chromosomal fos-like genes have been shown to significantly contribute to resistance to FOS in S. marcescens (16) or Klebsiella pneumoniae by mutations or/and overexpression (17).

In Enterobacterales, and particularly in E. coli, acquired FOS resistance genes are frequently located on plasmids within transposons or integrons, implying that they may disseminate quite easily. Moreover, plasmids carrying fos genes are commonly found to carry additional resistance determinants such as extended-spectrum β-lactamases (ESBLs, particularly of the CTX-M-type) or carbapenemases (particularly of the KPC type), enhancing their persistence and spread through coselection with β-lactam-based treatments (18).

Currently, FOS is commonly used for the treatment of urinary tract infections (UTI). Of note, the use of FOS in veterinary medicine is forbidden in most countries, including China and Europe; nevertheless, it is extensively used in many countries worldwide, especially in Africa, Asia, and South America, in pig and poultry production (19). In the last decade, the overuse of FOS led to an increase in FOS resistance rates. This increase has mainly been observed among ESBL-producing Enterobacterales, particularly in Asia, and recent reports tend to highlight a worldwide phenomenon (7, 20–22). The current emergence of FOS-resistant Enterobacterales worldwide therefore underscores a need for their rapid and accurate detection in order to implement measures for their control in term of dissemination, hence the development of screening tools. Thus, we aimed to develop a selective culture medium for screening FOS-resistant Enterobacterales regardless of the corresponding mechanism of resistance since such selective medium was not available so far.

MATERIALS AND METHODS

Susceptibility testing.

MICs were determined using AD fosfomycin agar dilution test (Liofilchem, Roseto degli Abruzzi, Italy) following the manufacturer’s instructions, and the results were interpreted according to the latest European Committee on Antimicrobial Susceptibility Testing (EUCAST) breakpoints for Enterobacterales (i.e., susceptibility [S], ≤32 mg/liter; resistance [R], >32 mg/liter) (Table 1) (23). It is important to note that the breakpoint used (32 μg/ml) is the EUCAST breakpoint for use of fosfomycin intravenously, and the breakpoint for oral treatment is 8 μg/ml, the latter applying only for E. coli (23).

TABLE 1.

Detection of FOS susceptibility/resistance using the SuperFOS screening medium

Strain	Species	Origin	MIC of FOSa (mg/liter)	FOS resistance determinantb	FOS susceptibility/resistancec	Isolate detected at 101 and/or 102 CFU/ml in:
						Saline	Stools	Urine
N6e	Escherichia coli	Switzerland	1	NAd	S	No	No	No
N8e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N9e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N11e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N13e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N21e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N23e	Escherichia coli	Switzerland	4	NA	S	No	No	No
N122e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N125e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N126e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N132e	Escherichia coli	Switzerland	1	NA	S	No	No	No
N137e	Escherichia coli	Switzerland	4	NA	S	No	No	No
R101e	Escherichia coli	France	1	NA	S	No	No	No
R147e	Escherichia coli	France	1	NA	S	No	No	No
R159e	Escherichia coli	France	1	NA	S	No	No	No
R722e	Escherichia coli	Lebanon	1	NA	S	No	No	No
R978e	Escherichia coli	Unknown	1	NA	S	No	No	No
R979	Escherichia coli	Unknown	1	NA	S	No
R980	Escherichia coli	Unknown	1	NA	S	No
R984	Escherichia coli	Unknown	1	NA	S	No
R989	Escherichia coli	Unknown	1	NA	S	No
R1014e	Escherichia coli	Unknown	8	NA	S	No	No	No
N522e	Klebsiella pneumoniae	Switzerland	8	NA	S	No	No	No
N580e	Klebsiella pneumoniae	Switzerland	16	NA	S	No	No	No
N644e	Klebsiella pneumoniae	Switzerland	32	NA	S	No	No	No
N628e	Klebsiella pneumoniae	Switzerland	32	NA	S	No	No	No
N732	Klebsiella pneumoniae	Switzerland	16	NA	S	No
N829	Klebsiella pneumoniae	Switzerland	8	NA	S	No
N835	Klebsiella pneumoniae	Switzerland	16	NA	S	No
N896	Klebsiella pneumoniae	Switzerland	32	NA	S	No
N906	Klebsiella pneumoniae	Switzerland	8	NA	S	No
N956	Klebsiella pneumoniae	Switzerland	8	NA	S	No
N583e	Klebsiella pneumoniae	Switzerland	16	NA	S	No	No	No
N608e	Klebsiella pneumoniae	Switzerland	8	NA	S	No	No	No
N646e	Klebsiella pneumoniae	Switzerland	2	NA	S	No	No	No
N723	Klebsiella pneumoniae	Switzerland	16	NA	S	No
N1179	Klebsiella pneumoniae	Switzerland	4	NA	S	No
N1185	Klebsiella pneumoniae	Switzerland	16	NA	S	No
N1297	Klebsiella pneumoniae	Switzerland	2	NA	S	No
N1447	Klebsiella pneumoniae	Switzerland	32	NA	S	No
N1563	Klebsiella pneumoniae	Switzerland	8	NA	S	No
N1699	Klebsiella pneumoniae	Switzerland	4	NA	S	No
N174	Klebsiella pneumoniae	Switzerland	8	NA	S	No
N292e	Escherichia coli	Switzerland	>256	fosA3	R	Yes	Yes	Yes
Ne104	Escherichia coli	Nepal	>256	fosA3	R	Yes
Ne89	Escherichia coli	Nepal	>256	fosA3	R	Yes
R2013	Escherichia coli	Unknown	>256	fosA3	R	Yes
R2742	Escherichia coli	South Africa	>256	fosA3	R	Yes
R2745e	Escherichia coli	South Africa	>256	fosA3	R	Yes	Yes	Yes
R3021	Escherichia coli	China	>256	fosA3	R	Yes
R4759	Escherichia coli	Switzerland	>256	fosA3	R	Yes
R3022	Escherichia coli	China	>256	fosA3	R	Yes
Un195	Escherichia coli	Switzerland	>256	fosA3	R	Yes
Un515	Escherichia coli	Switzerland	>256	fosA3	R	Yes
Un82	Escherichia coli	Switzerland	>256	fosA3	R	Yes
Un95	Escherichia coli	Switzerland	>256	fosA3	R	Yes
N140e	Escherichia coli	Switzerland	64	Unknown	R	Yes	Yes	Yes
N181e	Escherichia coli	Switzerland	128	Unknown	R	Yes	Yes	Yes
N279	Escherichia coli	Switzerland	>256	Unknown	R	Yes
Un376	Escherichia coli	Switzerland	>256	Unknown	R	Yes
Un421	Escherichia coli	Switzerland	>256	Unknown	R	Yes
Un454	Escherichia coli	Switzerland	>256	Unknown	R	Yes
Un50	Escherichia coli	Switzerland	>256	Unknown	R	Yes
Un546	Escherichia coli	Switzerland	>256	Unknown	R	Yes
FE7.2e	Escherichia coli	Egypt	>256	fosA4	R	Yes	Yes	Yes
FE19.1e	Escherichia coli	Egypt	>256	fosA4	R	Yes	Yes	Yes
FE31.1e	Escherichia coli	Egypt	>256	fosA3	R	Yes	Yes	Yes
FE48.3e	Escherichia coli	Egypt	>256	fosA6	R	Yes	Yes	Yes
FE49.1e	Escherichia coli	Egypt	>256	fosA3	R	Yes	Yes	Yes
FE50.1e	Escherichia coli	Egypt	>256	fosA4	R	Yes	Yes	Yes
FE50.2e	Escherichia coli	Egypt	>256	fosA4	R	Yes	Yes	Yes
FE53.2e	Escherichia coli	Egypt	>256	fosA3	R	Yes	Yes	Yes
FE55.1e	Escherichia coli	Egypt	>256	fosA4	R	Yes	Yes	Yes
FE56.1e	Escherichia coli	Egypt	>256	fosA3	R	Yes	Yes	Yes
N1345e	Escherichia coli	Switzerland	>256	fosA8	R	Yes	Yes	Yes
R4390e	Escherichia coli	Switzerland	>256	fosA8	R	Yes	Yes	Yes
R4880e	Escherichia coli	Kuwait	>256	fosA4	R	Yes	Yes	Yes
R4750e	Escherichia coli	Switzerland	>256	fosL1	R	Yes	Yes	Yes
N148e	Klebsiella pneumoniae	Switzerland	>256	NA	R	Yes	Yes	Yes
N164e	Klebsiella pneumoniae	Switzerland	>256	NA	R	Yes	Yes	Yes
N1554e	Klebsiella pneumoniae	Switzerland	>256	NA	R	Yes	Yes	Yes
R3182e	Klebsiella pneumoniae	China	>256	fosA3	R	Yes	Yes	Yes
R3183e	Klebsiella pneumoniae	China	>256	fosA3	R	Yes	Yes	Yes
R4881e	Klebsiella pneumoniae	Kuwait	>256	NA	R	Yes	Yes	Yes
R4901e	Klebsiella pneumoniae	Switzerland	128	NA	R	Yes	Yes	Yes

^aFOS, fosfomycin. MICs of FOS were determined using agar dilution supplemented with glucose-6-phosphate following the EUCAST recommendations.

^bfosA5/fosA6 genes intrinsic to the K. pneumoniae species were not considered.

^cR, resistant; S, susceptible.

^dNA, not applicable. Fosfomycin-susceptible strains were not investigated for the presence of fos genes.

^eIsolate used in spiked stool test.

PCR amplification and sequencing.

DNA of the isolates was recovered using the QIAamp DNA minikit and the QIAcube workstation (Qiagen, Courtaboeuf, France) according to the manufacturer’s instructions. PCR amplification was performed to detect the plasmid-mediated fosA1-6 genes with the primers described (22). Additionally, fosA7-8 and fosL1-2 genes were screened by using the following primers: fosA7_Fw (5′-ATACTGGCGCTTACCTTACC-3′), fosA7_Rv (5′-TATTTTCCAGGCCCGAAACG-3′), fosA8_Fw (5′-AACCATCTGACCCTTGCTGT-3′), fosA8_Rv (5′-CGAGAAAATAGAACGACGCC-3′), fosL_Fw (5′-AGTCTTCGTGAAGGTAGGCG-3′), and fosL_Rv (5′-GATGGCTTCCGACCCAATGC-3′). The PCR conditions were 98°C for 1 min, followed by 35 cycles of 98°C for 1 s, 55°C for 30 s, and 72°C for 30 s, with a final extension of 72°C for 5 min. Control DNA was used for all fos genes tested. To discriminate between different fos genes amplified with the same couple of primers, positive PCRs were further sequenced by Microsynth (Balgach, Switzerland) and results analyzed with Clone Manager Professional (Sci Ed Software, Denver, CO, USA).

Selective medium for fosfomycin resistance.

CHROMagar orientation medium (reference RT412; CHROMagar, Paris, France), which is commonly used as a differential medium for the isolation and differentiation of common urinary tract pathogens, was used for optimal screening as described (24). The optimal final concentration of FOS (Sigma-Aldrich, St. Louis, MO, USA) was 16 μg/ml in combination with glucose-6-phosphate (G-6-P) (AppliChem GmbH, Darmstadt, Germany) at 25 μg/ml. During the development of the medium (named the SuperFOS medium), the possible contamination by Gram-positive bacteria and fungi was also considered. Therefore, vancomycin (Duchefa Biochemie, Haarlem, Netherlands) was supplemented in the medium at a final concentration of 20 μg/ml to prevent growth of Gram-positive bacteria such as Enterococcus spp., Streptococcus spp., and Staphylococcus spp. Moreover, amphotericin B (Acros Organics, NJ, USA) was also added as an antifungal at a final concentration of 5 μg/ml. Table 2 shows how the stock solutions of FOS, G-6-P, vancomycin, and amphotericin B were prepared. CHROMagar orientation powder was diluted in distilled water and autoclaved at 121°C for 15 min. The antibiotic stock solutions were added when the medium reached 56°C (Table 2). Prepared plates of this SuperFOS medium were stored at 4°C and were protected from direct light exposure before use for as long as 1 week.

TABLE 2.

Preparation of the SuperFOS medium

Compound	Stock solution	Quantity or vol addeda	Final concn
CHROMagar orientation		13.2 g	3.3%
Distilled water		400 ml
Fosfomycin	50 mg/ml in 1 M HCl	128 μl	16 μg/ml
Glucose-6-phosphate	50 mg/ml in water	200 μl	25 μg/ml
Vancomycin	50 mg/ml in water	160 μl	20 μg/ml
Amphotericin B	10 mg/ml in DMSOb	200 μl	5 μg/ml

^aThe volume of 400 ml of SuperFOS medium was for 20 plates.

^bDMSO, dimethyl sulfoxide.

Evaluation assay.

Starting with an optical density of a 0.5 McFarland standard (an inoculum of 1.5 × 10⁸ CFU/ml), serial 10-fold dilutions were made in 0.85% saline solution, and 100-μl aliquots of each dilution were plated onto the SuperFOS selective medium. To quantify the viable bacteria in each dilution step, tryptic soy agar plates were inoculated concomitantly with 100 μl of each suspension and were incubated overnight at 37°C. Viable colonies were counted the following day. When no growth was observed after 18 h, incubation was extended up to 48 h in order to definitely assess the negativity of the culture. The lower limit of detection for the strains tested was determined based on the results obtained with the SuperFOS medium. The specificity (that corresponds to the proportion of FOS-susceptible isolates that are correctly identified) and the sensitivity (corresponding to the proportion of FOS-resistant isolates being correctly identified) were calculated.

Spiking experiments.

With the aim to evaluate the selective medium by mimicking in vivo stool colonization, spiked stools were also tested with 50 representative isolates from the collection of FOS-resistant (n = 25) and -susceptible (n = 25) Gram-negative bacteria using this selective culture medium. To carry out this experiment, stool suspensions were obtained by suspending 6 g of freshly pooled feces from healthy volunteers in 60 ml of distilled water as described previously (24). Spiked fecal samples were made by adding 100 μl of serial-fold bacterial dilutions to 900 μl of stool suspension. Aliquots (100 μl) of the spiked stool suspension were inoculated onto the SuperFOS medium. Additionally, aliquots (100 μl) of stool suspensions (nonspiked) were plated onto the SuperFOS medium as negative controls. The sensitivity and specificity were determined using the same cutoff value set at 10³ CFU/ml. The same approach was performed using spiked urine specimens obtained from healthy volunteers.

Additionally, a total of 21 clinical specimens obtained from 21 patients hospitalized at the Fribourg hospital (11 urine and 10 fecal specimens) were tested using the SuperFOS medium. A loopful (10 μl) of the urine and diluted fecal specimens was inoculated onto the SuperFOS medium and were incubated overnight at 37°C. Colonies of different morphology, size, and color from each plate were selected for further experiments. The grown colonies were tested for FOS susceptibility using AD fosfomycin agar dilution test (Liofilchem, Italy) following the manufacturer’s instructions in order to confirm their resistance patterns, and the results were interpreted according to the latest EUCAST breakpoints for Enterobacterales (23).

RESULTS

The performances of the SuperFOS medium were evaluated using a collection of 84 isolates of worldwide origin. According to the MIC results, this collection included 42 FOS-susceptible isolates (21 K. pneumoniae and 21 Escherichia coli) and 42 FOS-resistant isolates (21 K. pneumoniae and 21 E. coli) (Table 1).

All resistance isolates were screened at the molecular level for the presence of fos genes by PCR and sequencing (Table 1). Among the 42 FOS-resistance isolates, fosA3 (n = 19), fosA4 (n = 6), fosA5-6 (n = 6), 2 fosA8 (n = 2), and fosL1 (n = 1) genes encoding acquired resistance genes were detected (the intrinsic fosA5-6 genes of K. pneumoniae was not considered).

The sensitivity and specificity cutoff values for the detection of FOS-resistant Enterobacterales were set at 1 × 10³ CFU/ml, meaning that positive results were only considered when isolates could indeed be recovered onto the SuperFOS selective medium when plated at concentrations corresponding to <1 × 10³ CFU/ml (24). Actually, all the FOS-resistant isolates could be recovered within 24 h on SuperFOS medium plates by using an inoculum below the cutoff value of 1 × 10³ CFU/ml (1 × 10¹ to 1 × 10² CFU/ml) (Table 1). In contrast, all the FOS-susceptible isolates did not grow under those conditions, giving rise to sensitivity and specificity values of 100%. Additionally, in order to evaluate this medium at higher inoculum, 10³ and 10⁴ CFU/ml inoculum concentrations were tested. All the FOS-susceptible isolates did not grow under those new conditions, showing 100% of sensitivity and specificity at higher inoculum.

Spiked stool and urine specimens were also tested with a representative collection of FOS-resistant and -susceptible Gram-negative bacteria using this selective culture medium (Table 1). An identical lower limit of detection was found for all FOS-resistant strains, ranging from 10¹ to 10² CFU/ml (below the cutoff value at 1 × 10³ CFU/ml), whereas the FOS-susceptible strains did not grow under those conditions. Sensitivity and specificity were found to be 100%.

The SuperFOS selective medium was also clinically evaluated using a set of 21 clinical specimens (urine and fecal specimens). Among the 11 urine samples analyzed, 4 gave positive cultures with the SuperFOS medium, corresponding to fosfomycin-resistant Gram-negative isolates (E. coli [n = 1], P. aeruginosa [n = 1], and Proteus mirabilis [n = 2]). The fosfomycin resistance phenotype was further confirmed by the agar dilution method and showed MICs ranging from 64 to >256 μg/ml. On the other hand, 6 out of the 10 fecal samples gave positive cultures with the SuperFOS medium, corresponding to a diversity of fosfomycin-resistant isolates that showed MICs ranging from 64 to 128 μg/ml, namely, Hafnia alvei (n = 1), Morganella morganii (n = 1), Klebsiella pneumoniae (n = 1), Serratia liquefaciens (n = 1), and E. cloacae (n = 2), thus indicating an excellent agreement between SuperFOS selective medium and the MIC results of the standard agar dilution method. Noteworthy, after 24 h of incubation, no competing flora did grow on the SuperFOS medium, highlighting the high specificity of this selective medium and its high specificity in selecting FOS-resistant bacteria.

The shelf life of the SuperFOS selective plates was determined using Candida albicans (fungus) and Staphylococcus aureus Gram-positive strains, as well as the FOS-susceptible E. coli ATCC 25955 reference strain, which were subcultured daily onto the SuperFOS selective plates from a single batch of plates stored at 4°C. For at least a 7-day period, no growth could be observed.

DISCUSSION

Here, we have developed the SuperFOS selective medium, which is the very first screening medium that may detect FOS-resistant enterobacterial isolates. This selective medium may allow screening of FOS-resistant Enterobacterales regardless of their mechanism of FOS resistance. In particular, isolates producing acquired fosA3, fosA4, fosA5, fosA6, fosA8, and fosL1 genes were evaluated, corresponding to the most frequent Fos determinants in Enterobacterales. Along with the rapid fosfomycin/E. coli NP test that we previously developed (25), this medium constitutes a useful tool combination for identification of FOS-resistant Enterobacterales strains and possibly prevents further dissemination and outbreaks by screening patients potentially colonized with such resistant strains. While the rapid fosfomycin/E. coli NP test is useful for rapid categorization of E. coli strains as FOS susceptible or resistant from isolated colonies, the screening medium allows direct screening from samples that contained mixed bacterial population, such as rectal swabs usually collected for infection control purposes. Such medium may also be useful for epidemiological studies for human medicine and also for veterinary medicine. According to our evaluation, the selectivity of the SuperFOS medium for detecting fosfomycin-resistant bacteria was not impacted by mixed flora that may be found in a high number of clinical specimens. This selective medium covers a need since FOS has become the first option against uncomplicated UTIs caused by Gram-negative bacteria in many places, and a rising trend in FOS-resistant Gram-negative bacteria is being reported around the world (21, 26–28). This medium, which is easy to be prepared, inexpensive, and shows excellent performance, adds to the diagnostic toolbox dedicated to rapid detection of antibiotic resistance. Further clinical evaluations of this medium will now be needed in daily clinical practice to further assess its usefulness.