Evaluation of EDTA and Dipicolinic Acid in Broth Microdilution with Polymyxin B as a Phenotypic Test to Detect the mcr -1 Gene

Abstract

Polymyxins (colistin and polymyxin B) have recently regained significant importance as last-line drugs to treat infectious diseases due to multidrug-resistant gram-negative bacteria. However, resistance to polymyxins has increased, and the recognition of plasmid-mediated resistance (by the mcr gene) has led to an epidemiological concern. We aimed to evaluate the reduction of the polymyxin B minimum inhibitory concentration (MIC) in the presence of EDTA or dipicolinic acid (DPA) by using the broth microdilution (BMD) method for phenotypic screening of acquired polymyxin resistance mediated by the mcr-1 gene. Overall, 94 Enterobacterales (48 polymyxin-resistant and 46 polymyxin-susceptible) were evaluated: 47 mcr-1 positive (36 Escherichia coli, 2 Klebsiella pneumoniae, and 9 Salmonella spp.) and 47 mcr-1 negative (3 E. coli and 44 K. pneumoniae—27 isolates with MIC from ≤0.125 to 8 μg/mL and 20 isolates with MIC from 16 to 64 μg/mL). Results were categorized as positive when the chelator decreased the original BMD MIC by ≥2 logs. The majority (95.7%) of mcr-1 positive isolates displayed at least a 3 log dilution decrease in the MIC of polymyxin B with EDTA or DPA. The EDTA-based BMD assay detected 45 mcr-1-positive isolates, with only one false-positive among the mcr-1-negative isolates (sensitivity [SN], 95.7%; specificity [SP], 97.9%), whereas the DPA-based BMD assay detected 44 mcr-1-positive isolates (SN, 93.6%; SP, 95.7%), with two false-positive results. The accuracy of EDTA- and DPA-based BMD assays were 97% and 95%, respectively. The EDTA- and DPA-based assays were demonstrated to be reliable methods to detect mcr-1 positive isolates with excellent accuracy.

Introduction

Colistin (polymyxin E) and polymyxin B belong to a group of polypeptide antibiotics that have recently regained significant interest as a consequence of the increasing incidence of infections due to multidrug-resistant gram-negative bacteria. Currently, polymyxins are used as last-line drugs for treating severe infections due to carbapenem-resistant Enterobacterales. However, polymyxin resistance has increased gradually during the past few years; this has led to a large number of studies regarding these multifaceted resistance mechanisms.^1–3

Although not all the mechanisms underlying resistance to polymyxins have been elucidated, it is usually attributed to lipopolysaccharide modification through diverse routes. Resistance to polymyxins is normally related to chromosomal mutations, which involve modulation of the two-component regulatory systems leading to modification of lipid A, which reduces the action of polymyxins.^3,4 However, in November 2015, Liu et al. described for the first time a gene (mcr-1) responsible for the horizontal transfer of resistance to polymyxins.⁵ The production of MCR-1, the protein encoded by the mcr-1 gene, leads to an increase in the minimum inhibitory concentration (MIC) for polymyxins and is a major worldwide health problem, as it can disseminate rapidly.³

To date, several molecular methods to detect mcr-1-based polymyxin resistance in isolates from pure cultures and/or clinical samples have been developed, but most of them are expensive. With regard to phenotypic tests, the reference method used to establish the susceptibility of an isolate to polymyxins is the broth microdilution (BMD), which has been demonstrated to be reliable and accurate.^6,7 The rapid polymyxin NP test is a recently developed phenotypic method that can be used as an alternative to BMD to detect resistant isolates; this test requires a maximum of 2 hours to produce results.^8–11 However, none of these phenotypic tests can distinguish polymyxin resistance mediated by the mcr-1 gene from resistance due to chromosomal mechanisms.

Recent studies have demonstrated that the structure of the catalytic site of phosphoethanolamine transferase, MCR-1, is composed of a zinc metalloprotein. Therefore, it is thought that the addition of a chelator such as EDTA or dipicolinic acid (DPA) in the BMD of polymyxins may reduce the colistin/polymyxin B MICs for isolates expressing the mcr-1 gene.^7,12–14

Owing to the increasing emergence of polymyxin resistance, the development of reliable and cost-effective screening methods to determine susceptibility and resistance to polymyxins is an important need for clinical microbiology laboratories to prevent dissemination of the mcr-1 gene. In this study, we evaluated the effect of chelators (EDTA or DPA) on the MIC of polymyxin B by BMD.

Methods

Isolates

A total of 94 Enterobacterales (48 polymyxin-resistant and 46 polymyxin-susceptible) were tested (Supplementary Table S1). Among these 94 isolates, 47 carried the mcr-1 gene (36 Escherichia coli, 2 Klebsiella pneumoniae, and 9 Salmonella enterica), including 29 isolates categorized as polymyxin-susceptible (MIC ≤2 μg/mL). The remaining isolates (n = 47) were mcr-1 negative (3 E. coli and 44 K. pneumoniae) and comprised 30 isolates resistant to polymyxins by other mechanisms and 17 polymyxin-susceptible isolates. All isolates included in this study were obtained from clinical samples (85 isolates: 39 E. coli and 46 K. pneumoniae), as well as from food-producing animals (9 S. enterica) obtained from previous surveillance studies in southern Brazil from 2013 to 2018.¹⁵ All the isolates were identified by the VITEK^® MS Plus Biomérieux.

Identification of the mcr-1 gene

The presence of the mcr-1 gene was evaluated by submitting bacterial colonies to DNA extraction followed by polymerase chain reaction (PCR). In brief, thermal lysis was carried out using pure colonies in an extraction buffer (TE) subjected to heating at 80°C for 20 minutes, followed by freezing at −20°C for 20 minutes. The extracted DNA was submitted to PCR with specific primers for the mcr-1 gene (forward, 5′-CGGTCAGTCCGTTTGTTC-3′; and reverse, 5′-CTTGGTCGGTCTGTAGGG-3′).⁵

Subinhibitory concentrations of EDTA and DPA

Before performing BMD with polymyxin B and EDTA or DPA, a BMD assay without antibiotic was carried out by varying the chelator concentration with 25 representative isolates: E. coli (n = 7), K. pneumoniae (n = 8), Salmonella spp. (n = 9), and one control (E. coli ATCC 25922). This was necessary to establish the lowest subinhibitory concentration of each chelator to be used in this study. The assays used 10 concentrations of EDTA from 18,700.8 μg/mL (64 mM) to 36.5 μg/mL (0.125 mM) and 4 concentrations of DPA (900, 675, 450, and 225 μg/mL). These assays indicated that concentrations of 584.4 μg/mL (2 mM) EDTA and 225 μg/mL DPA did not inhibit the growth of the isolates. These concentrations were used in all BMD experiments to evaluate the effect of the chelators on the polymyxin B MIC.

EDTA- and DPA-based BMD assays

MICs of polymyxin B were determined by BMD according to the ISO 20776-1, and the results were interpreted according to the European Committee on Antimicrobial Susceptibility Testing (EUCAST) guidelines, that is, isolates with MICs >2 μg/mL were considered resistant.¹⁶ In brief, a bacterial suspension prepared by suspending colonies from an overnight culture in saline was used to inoculate (final inoculum of 10⁵ CFU/mL) a 96-well microtiter plate containing cation-adjusted Mueller–Hinton broth (CAMHB; BioMerieux, Heidelberg, France) and polymyxin B at concentrations from 0.125 to 128 μg/mL. Reduction of the polymyxin B MIC was evaluated by the addition of EDTA or DPA to BMD at fixed concentrations of 584.4 (2 mM) and 225 μg/mL, respectively.

All microdilution tests were performed in duplicate. E. coli strain ATCC 25922 was used for quality control. A decrease in the polymyxin B MIC of at least 2 logs in the presence of EDTA or DPA was considered a significant effect of the chelator (significant decrease in the MIC).

Sensitivity (SN) and specificity (SP) were determined for each assessed chelator-based BMD assay using as a standard reference the results of molecular characterization (presence of the mcr-1 gene). When the chelator-based phenotypic test performed with a strain possessing mcr-1 presented a negative result, it was considered a false negative, and when an mcr-1-negative strain presented a positive result, it was considered a false positive. The accuracy values of the EDTA- and DPA-based BMD assays were determined according to the number of true-positive values plus the number of true-negative values divided by their sum plus the number of false-negative and false-positive results.

Results

A total of 45 out of 47 mcr-1-positive isolates (SN, 95.7%) displayed a significant decrease in the MIC values of polymyxin B in the presence of EDTA. Indeed, two mcr-1 positive isolates (one E. coli and one K. pneumoniae) presented a nonsignificant decrease in the MIC (1 log) with EDTA. In contrast, only one mcr-1-negative isolate (a K. pneumoniae in which the MIC was reduced by 3 logs) presented a significant difference in MIC values when EDTA was added to polymyxin B in the BMD assay (SP, 97.9%) (Table 1).

Table 1.

Bacterial Isolates Tested and Polymyxin B Minimum Inhibitory Concentration Values by Broth Microdilution in the Presence of 584.4 μg/mL EDTA or 225 μg/mL Dipicolinic Acid

Species	Mechanism of polymyxin resistance	Polymyxin MIC (μg/mL)	Log change with chelator	No. of isolates (EDTA-BMD assay)^a	No. of isolates (DPA-BMD assay)^a
Escherichia coli	mcr-1 positive	2	0 or 1	1	2
(n = 39)	(n = 36)		2	9	2
			3	15	10
			4	1	12
		4	2	—	1
			3	5	3
			4	5	—
			5	—	6
	mcr-1 negative	≤0.125 to 2	0 or 1	2	2
	(n = 3)	4–8	0 or 1	1	1
Klebsiella. pneumoniae	mcr-1 positive	4	4	1	—
(n = 46)	(n = 2)		5	—	1
		8	0 or 1	1	1
	mcr-1 negative	≤0.125 to 2	0 or 1	15	15
	(n = 44)	4–8	0 or 1	9	9
		≥16	0 or 1	19	20
			3	1	—
Salmonella spp.	mcr-1 positive	2	2	2	2
(n = 9)			3	1	1
		4	3	3	3
			4	2	1
			5	—	1
		8	3	—	1
			4	1	—

For a detailed description of tested strains and results, see Supplementary Table S1.

Dashes in empty cells indicate no isolate presenting those results.

BMD, broth microdilution; DPA, dipicolinic acid; MIC, minimum inhibitory concentration.

For the DPA-based BMD assay, a total of 44 mcr-1-positive isolates (SN, 93.6%) presented a significant reduction in the MIC. Two mcr-1-positive isolates presented a (nonsignificant) reduction of only 1 log in the presence of DPA. Noteworthy, the latter were the same isolates that presented a nonsignificant reduction in the MIC with EDTA. One mcr-1-positive isolate did not present a change in the MIC value in the presence of DPA. Among the mcr-1-negative isolates (n = 47), two strains of K. pneumoniae showed a significant MIC reduction (2 logs) (SP, 95.7%).

The EDTA- and DPA-based BMD assays presented accuracies of 97% and 95%, respectively.

Discussion

Our results indicate that either EDTA or DPA added to polymyxin B in the BMD assay can significantly decrease the MIC for the vast majority of isolates presenting the mcr-1 gene. This phenotypic assay could, therefore, be used for the presumptive detection of MCR-1-producing isolates in diagnostic microbiology laboratories, as it presented high SN for both chelators. Likewise, the MIC values of polymyxin B did not change significantly for almost all mcr-1-negative isolates, which confers high SP of either EDTA or DPA as inhibitors of the MCR-1 protein. In fact, we found only 1 and 2 false-positive results with EDTA and DPA assays, respectively, out of 47 isolates tested.

The majority of mcr-1-positive isolates displayed a decrease of at least 3 logs in MIC values of polymyxin B with EDTA or DPA but, apparently, DPA presented an increased inhibitory capacity of the MCR enzyme, since 21 mcr-1-positive isolates presented a decrease of at least 4 logs in the DPA-based BMD assay, compared with only 10 isolates in the EDTA-based BMD assay.

Since it was shown that the mcr-1 gene encodes for zinc-dependent phosphoethanolamine transferases, and consequently, metallo-chelators such as EDTA and DPA could theoretically improve polymyxin activity, some studies have evaluated the effect of EDTA or DPA on the inhibition of MCR-1 by zinc deprivation.^13,17,18 In a study by Esposito et al.,¹² a concentration of 80 μg/mL of EDTA in the EDTA-based BMD test was evaluated for the detection of colistin-resistant mcr-1-positive isolates. They obtained encouraging results for the detection of mcr-1-possessing E. coli strains, as they found a 4-log decrease in the MIC of colistin in the presence of EDTA. In another study, inhibition of MCR-1 by DPA was reported as a useful method for phenotypic screening of mcr-1-positive colistin-resistant E. coli strains.¹⁴

Recently, Büdel et al. evaluated and compared EDTA- and DPA-based BMD tests for the detection of mcr among 92 isolates of Enterobacteriaceae.⁷ In contrast to our study, they found 22 false-positive results when EDTA was added to the BMD test (54.2% of SP), indicating that the assay is not sufficiently accurate to identify the MCR producers among the overall collection of colistin-resistant Enterobacteriaceae tested. Intriguingly, Büdel et al. used a lower concentration of EDTA (100 μg/mL) in comparison with our experiments (584.4 μg/mL), which could lead to a lower SN but not to a lower SP. However, we used a CAMHB (as recommended by CLSI and EUCAST to test polymyxins),¹⁹ which is supplemented with calcium and magnesium, whereas Büdel et al. used a nonsupplemented MH broth. According to Esposito et al., the addition of calcium and magnesium would impair the inhibitory activity of EDTA, as the nonspecific binding of the chelator to the excess of calcium and magnesium could reduce the concentration of free EDTA needed to chelate zinc ions required for MCR-1 activity.¹² Therefore, it is possible to consider that the concentration of free EDTA was at least comparable between ours and Büdel's experiments.

In contrast, Büdel et al.⁷ demonstrated that the DPA-based BMD assay presented good accuracy but only for E. coli isolates, with no significant MIC reduction when DPA was added to the BMD assay for K. pneumoniae and Salmonella spp. isolates carrying the mcr-1 gene. Conversely, we used DPA at a concentration four times lower than Büdel et al., but we found 95% accuracy using this chelator in the BMD assay, regardless of the species tested (E. coli, K. pneumoniae, and Salmonella spp.).

Isolates carrying mcr-1 may present MICs for polymyxins as low as 2 μg/mL, which characterize them as susceptible. Although the clinical relevance of these isolates may be controversial, the epidemiological concern exists that they could transfer the gene horizontally. In this context, the EDTA- and DPA-based BMD assays are important tools to aid in the detection of mcr-1-positive isolates in laboratories that do not have access to molecular techniques.

We observed the inhibitory activity of the chelators against mcr-1-positive isolates presenting borderline polymyxin MICs, that is, 4 and 8 μg/mL. It would be important to test mcr-1-positive isolates with higher MICs of polymyxins to confirm the inhibition of MCR-1 by a metalloenzyme chelator; however, mcr-1 isolates with high MICs are rare. In contrast, we tested mcr-1-negative isolates with low and high MICs (27 mcr-1 negative isolates presenting MICs ranging from ≤0.125 to 8 μg/mL and 20 mcr-1 negative isolates presenting MICs ranging from 16 to 64 μg/mL), and we observed only one and two false-positive isolates in the EDTA- and DPA-based BMD assays, respectively, confirming the SP of EDTA and DPA as inhibitors of the MCR-1 protein.

In conclusion, our results indicate that inhibition of MCR-1 by EDTA or DPA is a reliable phenotypic approach that demonstrates excellent accuracy, which can be used for the presumptive detection of MCR-1-producing bacteria using the BMD.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by INPRA—Instituto Nacional de Pesquisa em Resistência Antimicrobiana—Brazil (INCT/CNPq: 465718/2014-0 and INCT/FAPERGS: 17/2551-0000514-7) and by Fundo de Incentivo a Pesquisa e Eventos do Hospital de Clínicas de Porto Alegre (FIPE/HCPA; Project No. 16-0559).

Supplementary Material

References

Falagas

M.E.

, Kasiakou

S.K

, and Saravolatz

L.D.

. 2005. Colistin: the revival of polymyxins for the management of multidrug-resistant Gram-negative bacterial infections. Clin. Infect. Dis. 40:1333–1341.

Kaye

K.S.

, Pogue

J.M

, Tran

T.B

, Nation

R.L

, and Li

. 2016. Agents of last resort: polymyxin resistance. Infect. Dis. Clin. North Am. 30:391–414.

Poirel

, Jayol

, and Nordmann

. 2017. Polymyxins: antibacterial activity, susceptibility testing, and resistance mechanisms encoded by plasmids or chromosomes. Clin. Microbiol. Rev. 30:557–596.

Olaitan

A.O.

, Morand

, and Rolain

J.M.

. 2014. Mechanisms of polymyxin resistance: acquired and intrinsic resistance in bacteria. Front. Microbiol. 5:643.

Liu

Y.Y.

, Wang

, Walsh

T.R

, Yi

L.X

, Zhang

, Spencer

, Doi

, Tian

, Dong

, Huang

, Yu

L.F

, Gu

, Ren

, Chen

, Lv

, He

, Zhou

, Liang

, Liu

J.H

, and Shen

. 2016. Emergence of plasmid-mediated colistin resistance mechanism MCR-1 in animals and human beings in China: a microbiological and molecular biological study. Lancet Infect. Dis. 16:161–168.

Humphries, R.M. 2015. Susceptibility testing of the polymyxins: where are we now? Pharmacotherapy, 35:22–27.

Büdel

, Clément

, Bernasconi

O.J

, Principe

, Perreten

, Luzzaro

, and Endimiani

. 2018. Evaluation of EDTA- and DPA-based microdilution phenotypic tests for the detection of MCR-mediated colistin resistance in Enterobacteriaceae. Microb. Drug Resist. 25:494–500.

Nordmann

, Jayol

, and Poirel

. 2016. Rapid detection of polymyxin resistance in Enterobacteriaceae. Emerg. Infect. Dis. 22:1038–1043.

Jayol

, Nordmann

, Lehours

, Poirel

, and Dubois

. 2018. Comparison of methods for detection of plasmid-mediated and chromosomally encoded colistin resistance in Enterobacteriaceae. Clin. Microbiol. Infect. 24:175–179.

10.

Poirel

, Larpin

, Dobias

, Stephan

, Decousser

J.W

, Madec

J.Y

, and Nordmann

. 2018. Rapid polymyxin NP test for the detection of polymyxin resistance mediated by the mcr-1/mcr-2 genes. Diagn. Microbiol. Infect. Dis. 90:7–10.

11.

Dalmolin

T.V.

, G.Á. Dias, de Castro

L.P

, H. Ávila, Magagnin

C.M

, Zavascki

A.P

, de Lima-Morales

, and Barth

A.L.

. 2019. Detection of Enterobacterales resistant to polymyxins using rapid polymyxins NP test. Braz. J. Microbiol. 50:425–428.

12.

Esposito

, Fernandes

M.R

, Lopes

, Muñoz

, Sabino

C.P

, Cunha

M.P

, Silva

K.C

, Cayô

, W. Martins

M.B.S

, Moreno

A.M

, Knöbl

, Gales

A.C

, and Lincopan N

. 2017. Detection of colistin resistant MCR-1-positive Escherichia coli by use of assays based on inhibition by EDTA and zeta potential. J. Clin. Microbiol. 55:3454–3465.

13.

Hinchliffe

, Yang

Q.E

, Portal

, Young

, Li

, Tooke

C.L

, Carvalho

M.J

, Paterson

N.G

, Brem

, Niumsup

P.R

, Tansawai

, Lei

, Li

, Shen

, Wang

, Schofield

C.J

, Mulholland

A.J

, Shen

, Fey

, Walsh

T.R

, and Spencer

. 2017. Insights into the mechanistic basis of plasmid-mediated colistin resistance from crystal structures of the catalytic domain of MCR-1. Sci. Rep. 7:39392.

14.

Coppi

, Cannatelli

, Antonelli

, Baccani

, Di Pilato

, Sennati

, Giani

, and Rossolini

G.M.

. 2017. A simple phenotypic method for screening of MCR-1-mediated colistin resistance. Clin. Microbiol. Infect. 24:201.e1–201.e3.

15.

Dalmolin

T.V.

, Wink

P.L

, de Lima-Morales

, and Barth

A.L.

. 2018. Low prevalence of the mcr-1 gene among carbapenemase-producing clinical isolates of Enterobacterales. Infect. Control Hosp. Epidemiol. 40:263–264.

16.

European Committee on Antimicrobial Susceptibility Testing (EUCAST). 2018. Breakpoint tables for interpretation of MICs and zone diameters. Version 8.0. Available at http://www.eucast.org/clinical_breakpoints/ (accessed March 8, 2019).

17.

Stojanoski

, Sankaran

, Prasad

B.V

, Poirel

, Nordmann

, and Palzkill

. 2016. Structure of the catalytic domain of the colistin resistance enzyme MCR-1. BMC Biol. 14:81.

18.

Sun

, Xu

, Gao

, Lin

, Wei

, Srinivas

, Li

, Yang

R.S

, Li

X.P

, Liao

X.P

, Liu

Y.H

, and Feng

. 2017. Deciphering MCR-2 colistin resistance. MBio, 8:e00625-17.

19.

European Committee on Antimicrobial Susceptibility Testing (EUCAST). 2018. Recommendations for MIC determination of colistin (polymyxin E) As recommended by the joint CLSI-EUCAST Polymyxin Breakpoints Working Group. Available at http://www.eucast.org/fileadmin/src/media/PDFs/EUCAST_files/General_documents/Recommendations_for_MIC_determination_of_colistin_March_2016.pdf (accessed March 8, 2019).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB