Dielectrophoresis System for Testing Antimicrobial Susceptibility of

Subscriber access provided by University of Newcastle, Australia

Article

A dielectrophoresis system for testing antimicrobial susceptibility of Gram-negative bacteria to beta-lactam antibiotics I-Hsiu Su, Wen-Chien Ko, Chung-Hsin Shih, Fang-Hao Yeh, YungNien Sun, Jung-Chih Chen, Po-Lin Chen, and Hsien-Chang Chang Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b00220 • Publication Date (Web): 17 Mar 2017 Downloaded from http://pubs.acs.org on March 20, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure 1. Electric field distribution in a quadruple electrode array (QEA) and the dielectrophoretic antimicrobial susceptibility testing device. Obj, objective lens of optical microscopy; CCD, charge coupled device. 97x111mm (600 x 600 DPI)

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2. Serial morphological changes of E. coli ATCC 25922 treated by four β–lactam antibiotics at minimal inhibitory concentrations in a dielectrophoretic test. Arrowheads indicate elongation of bacteria and arrows indicate swelling of bacteria. 60x21mm (600 x 600 DPI)


Page 2 of 14

Page 3 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. The morphological changes of bacteria strains, K. pneumoniae ATCC 700603, P. aeruginosa ATCC 27853, and A. baumannii ATCC 19606, observed under light microscope after treatment with antimicrobials for 90 min. Arrowheads indicate cell elongation, arrows indicate cell swelling, and circles indicate no morphological change. 41x10mm (600 x 600 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. Common morphological categorization of Gram-negative bacteria treated by β–lactam antibiotics in a dielectrophoretic antimicrobial susceptibility testing. Bacteria susceptible to antibiotics showed morphological changes: (A) elongation, (B) lysis, (C) swelling. No morphological change was found in resistant isolates (D) 141x236mm (600 x 600 DPI)


Page 4 of 14

Page 5 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5. Serial changes of length of Gram-negative standard strains treated with minimal inhibitory concentrations of four β-lactam antibiotics: (A) E. coli ATCC 25922, (B) K. pneumoniae ATCC 700603, (C) P. aeruginosa ATCC 27853, (D) A. baumannii ATCC 19606. Areas of cell swellings induced by doripenem was also measured. The symbol “x” on the X-axis indicates cell lysis. *** indicates a P value of less than 0.0001, as compared with time zero. 210x523mm (600 x 600 DPI)



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 14

A dielectrophoresis system for testing antimicrobial susceptibility of Gram-negative bacteria to beta-lactam antibiotics I-Hsiu Su, Wen-Chien Ko, Chung-Hsin Shih, Fang-Hao Yeh, Yung-Nien Sun§, Jung-Chih Chen∥, Po Lin Chen *, Hsien-Chang Chang *

Department of Biomedical Engineering, National Cheng Kung University, Tainan 70101, Taiwan

Department of Internal Medicine, National Cheng Kung University Hospital, Tainan 70101, Taiwan

§

Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan 70101, Taiwan

∥

Institute of Biomedical Engineering, National Chiao Tung University, Hsinchu 30010, Taiwan

ABSTRACT: Gram-negative bacteria (GNBs) are common pathogens causing severe sepsis. Rapid evaluation of drug susceptibility would guide effective antibiotic treatment, and promote life-saving. A total of 78 clinical isolates of 13 gram-negative species collected between April 2013 and November 2013 from two medical centers in Tainan were tested. Bacterial morphology changes in different concentrations of antibiotics were observed under the electric field of a quadruple electrode array using light microscopy. The minimal inhibitory concentrations (MICs) of four antimicrobial agents, namely cefazolin, ceftazidime, cefepime, and doripenem, were determined by the dielectrophoretic antimicrobial susceptibility testing (dAST) and by the conventional broth dilution testing (BDT). The antibiotics at the concentration of 1x MIC induced obvious morphological changes in susceptible GNBs, including cell elongation, cell swelling, or lysis, at 90 min. In contrast, resistant strains remained unchanged. The MIC results measured by dAST were in good agreement with those of BDT (essential agreement 95.6%). The category agreement rate was 89.2%, and the very major errors rate for dAST was 2.9%. In conclusion, dAST could accurately determine drug susceptibility within 90 min. Comprehensive tests by dAST for more drugs against more GNB species are possible in the future.

Early use of appropriate antibiotics and the avoidance of antibiotic overuse for patients with severe sepsis present a clinical dilemma. Administration of effective antibiotics in time greatly relies on the support of a clinical laboratory, which provides data on pathogen identification and drug susceptibility. Rapid susceptibility tests are able to increase appropriateness of antibiotic therapy, improve patient survival1 and reduce medical costs.2 Conventional drug susceptibility methods, such as broth dilution, disc diffusion, or E-test, require a time period of 16-24 h. Commercial testing, such as the Vitek 2 system (bioMérieux, Marcy-l’Étoile, France), shortens the testing time but still needs 6–12 h.3 Rapid tests have been designed to determine antimicrobial susceptibility based on the observation of bacterial growth, biochemical or morphological changes induced by antibiotics. For example, antimicrobial susceptibility can be measured by bacterial numbers in a microfluidic channel,4 rotation rates of magnetic beads inversely proportional to bacteria mass,5,6 bacterial fluctuation associated with bacterial metabolism,7 or fluorescent signals form metabolic activity of bacteria in droplets;8 or morphological analysis of bacterial cells in response to antibiotics.9 Correct identification of pathogens in clinical samples using Raman microspectroscopy has been gained increasing attention over the recent years.10 Unique Raman spectral fingerprint generated by inelastic scattering of photons due to intrinsic molecular bonds as a laser beam probes the cells helps identification of microbial species.11 Moreover, the impact of antibiotic treat-

ment on metabolic state can be observed12,13 and profiled based on different class of antimicrobial agents by Raman microspectroscopic measurement in single-cell level.14 A pilot study showed that surface-enhanced Raman scattering-based AST could determine MICs of antibiotics for bacteria in two hours.15 In contrast to tests determining the growth or metabolism of bacteria, Choi et al. proposed that single cell morphological change can accurately predict antimicrobial susceptibility.9 In their studies, susceptible bacteria under different antibiotic concentrations presented varied morphological changes. For example, the cases of filamentary formation or swelling were regarded as being susceptible to the tested drug. In contrast, cell division was considered to be resistant as bacteria can grow in number. The turnaround time of such a rapid diagnostic system could be reduced from 66 h to 52 h, which was that of conventional methods.9 More rapid techniques, such as PCR genotyping to detect resistance genes within 2 h after primary cultures,16 cannot determine minimal inhibitory concentrations (MICs) and unknown resistance genes.17 Our previous work laid the ground for a rapid susceptibility test based on the changes in dielectrophoretic (DEP) behaviors related to β-lactam-induced elongation of gram-negative bacteria (GNBs) on a quadruple electrode array (QEA).18,19 DEP is the force applied on dielectric particles in a non-uniform electric field. After the bacteria is treated by in vitro active drugs, they become more polarizable than their surrounding medium, and are attracted to the strong region of an electric


Page 7 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


field along its gradient lines (positive DEP, pDEP). In contrast, if cells treated by inactive drugs or without drugs Table 1. Species distribution of 78 gram-negative isolates. Species

dAST, and were compared with those derived from the conventional broth dilution testing (BDT).

Number

Non-Enterobacteriaceae Acinetobacter baumannii

8

Pseudomonas aeruginosa

7

Enterobacteriaceae Citrobacter koseri

2

Enterobacter aerogenes

1

Enterobacter cloacae

7

Escherichia coli

12

ESBL-producing Escherichia coli

9

Carbapenem-resistant Escherichia coli

3

Klebsiella oxytoca

1

Klebsiella pneumoniae

6

ESBL-producing Klebsiella pneumoniae

2

KPC-producing Klebsiella pneumonia

7

Morganella morganii

2

Proteus mirabilis

5

Proteus vulgaris

1

Salmonella enteritidis group B

2

Salmonella enteritidis group D

1

Serratia marcescens

2

Total

78

ESBL = extended-spectrum β-lactamase. KPC = Klebsiella pneumoniae carbapenemase.

become less polarizable than the surrounding medium, they will be repelled toward the weak region of an electric field along its gradient lines (negative DEP, nDEP). In addition to morphological changes, the bacterial numbers decreased after incubation with active antibiotics. In contrast, the bacterial number remained unchanged or increased in the cases of antibiotic-resistant bacteria. Thus, we could rapidly assess the susceptibility of β-lactam antibiotics for GNBs. Our method was able to determine the MICs of cefazolin for Escherichia coli (E. coli) and Klebsiella pneumoniae (K. pneumoniae) within 90 min. This result is consistent with those obtained from the broth dilution method.18,19 In addition, the dielectrophoretic antimicrobial susceptibility testing (dAST) system can be applied to clinical microbiology laboratories, as the procedure is simple. In the present study, we optimized our previously designed QEA, which can collect bacteria cells and observe morphological changes rapidly. In order to validate the efficacy and efficiency of the dAST technology, we accessed the in vitro responses of four standard strains (E. coli, K. pneumoniae, Pseudomonas aeruginosa [P. aeruginosa], and Acinetobacter baumannii [A. baumannii]), which were representative GNBs in the community and hospitals.20,21 Secondly, dAST was validated with 78 clinical isolates belonging to 13 important GNBs. The MICs of four drugs, namely cefazolin, ceftazidime, cefepime and doripenem, were determined by

Figure 1. Electric field distribution in a quadruple electrode array (QEA) and the dielectrophoretic antimicrobial susceptibility testing device. Obj, objective lens of optical microscopy; CCD, charge coupled device.

EXPERIMENTAL METHODS Bacteria and antibiotics. Seventy-eight isolates of 13 species were cultivated from clinical specimens (i.e. blood, sputum, bile, urine, and pus) at the Clinical Microbiology Laboratory of National Cheng Kung University Hospital, a medical center in southern Taiwan, from January 2004 to April 2011, and stored at -80℃ until use (Table 1). Three reference strains, including E. coli ATCC 25922, P. aeruginosa ATCC 27853, and K. pneumoniae ATCC 700603, which were commonly used as quality control strains in clinical microbiology laboratories with known MIC ranges of studied antibiotics, were included. As for A. baumannii ATCC 19606, a reference strain commonly used for susceptibility studies for A. baumannii isolates was also included for the dAST. The supplied drugs included cefazolin (Fluka, Sigma-Aldrich, USA), ceftazidime (Sigma, USA), cefepime (Sigma, USA), and doripenem (Sigma, USA). The study was ethically approved by the Institutional Review Board of National Cheng Kung University Hospital (IRB no. B-ER-101-146). Dielectrophoretic antimicrobial susceptibility testing and broth dilution testing. The QEA plate was fabricated by standard photolithography techniques on an indium tin oxide glass slide (76 x 26 x 1 mm) and modified by a previous pattern.18,19 The schematic configuration of dAST and the electric field were demonstrated in Figure 1, and the bare electrode gap was 30 µm. In the device, the electric field on QEA was



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 14

produced by a function generator. The chip had no coating on the electrode surface. The suspension directly contacted to the bare electrodes. The morphological change of GNBs was observed by optical microscopy, in conjunction with a charge coupled device (CCD) camera and a monitor screen. Bacteria with morphological change induced by β-lactam antibiotics

Figure 2. Serial morphological changes of E. coli ATCC 25922 treated by four β–lactam antibiotics at minimal inhibitory concentrations in a dielectrophoretic test. Arrowheads indicate elongation of bacteria and arrows indicate swelling of bacteria.

Figure 3. The morphological changes of bacteria strains, K. pneumoniae ATCC 700603, P. aeruginosa ATCC 27853, and A. baumannii ATCC 19606, observed under light microscope after treatment with antimicrobials for 90 min. Arrowheads indicate cell elongation, arrows indicate cell swelling, and circles indicate no morphological change.

were attracted to the electrode edge, i.e. pDEP. In contrast, control bacteria and drug-resistant bacteria were repelled toward the electrode center (nDEP). The dAST procedures were as follows: bacteria isolates were grown overnight in TSB medium at 37°C with shaking, and then were adjusted to 0.5 McFarland standards. Two-fold dilution of 100 µl of bacterial suspensions and 1000 µl of antibiotic solution were then mixed in a tube and incubated in a shaking incubator at 37°C for 0, 30, 60, 90, 120 and 150 min treatments. After treatment, the suspension was centrifuged at 5000 rpm for 5 min, and then the supernatant was decanted. Two hundred fifty µl of 0.2 M HEPES buffer (σ = 0.3 S/m) was added for the dAST experiments. After shaking, 11 µl of suspension was pipetted on the chip, as shown in Figure 1B. The function generator was connected with the QEA and ran with voltage 10 Vpp and frequency of 400 kHz. Time-lapse images of bacteria were taken every 30 s for 3 min, using a camera (DS12611, Canon, 40x objective) connected to an inverted microscope (CK40, Olympus), controlled by the software DSLR Remote Pro. Each test has a drug-free control with the same DEP condition. The cell length and area were calculated by the ImageJ software. The BDT for GNBs followed the procedure recommended by the Clinical and Laboratory Standard Institution (CLSI) M100-

S24.22 The criteria of susceptibility suggested by the CLSI are shown in the Table S1. Comparison of results obtained from dAST and BDT. According to CLSI M100-S24, three to four antibiotics concentrations were adopted for dAST. Concentrations of cefazolin relevant to identify the cut-off value in determining drug susceptibility were 2, 4 and 8 µg/ml; 4, 8, and 16 µg/ml for ceftazidime; 2, 4, 8 and 16 µg/ml for cefepime; 1, 2, and 4 µg/ml for doripenem. Each test has its non-treated control group. The MIC in dAST was defined as the minimal drug concentration inducing the typical morphological changes (elongation, swelling or lysis) after treatment for 90 min. Cephalosporin susceptibility was determined by changes of cell length, and in cases of doripenem, by changes of cell areas. The presence or absence of morphological changes was determined at 90 min by an observer using light microscopy. The comparative data of dAST and BDT were expressed according to the recommendations proposed by U.S. Food and Drug Administration.23 Contradicting findings were classified as “very major errors” (VME), if susceptible by dAST but resistant by BDT, and “major errors” (ME), if resistant by dAST but susceptible by BDT. Other discrepancies were de-


Page 9 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


fined as “minor errors” (mE). Essential agreement (EA) is achieved if the MIC of dAST reaches exact agreement or within ± one two-fold dilution with BDT. Category agreement (CA) is determined as interpretive agreement of the susceptibility, intermediate or resistant category between dAST and BDT. Statistical analysis. Statistical analysis was performed using SPSS 17.0. Paired sample T-test was used for analyzing the change of cell length and area in time. RESULTS Morphological changes in the presence of different betalactams. The morphological changes of E. coli ATCC 25922 contrast, cells became swollen at 90 min of doripenem treatment, and underwent lysis between 120 and 150 min (Figure 2D). For three other GNB strains (K. pneumoniae ATCC 700603, P. aeruginosa ATCC 27853, and A. baumannii ATCC 19606), varied types of morphological changes, such as elongation, swelling, and lysis of cells, were observed in susceptible strains treated by antibiotics for 90 min (Figure 3). In contrast, resistant strains were concentrated at the central area of QEA without morphological changes (Figure 3). Bacteria had no morphological change in every antibiotic-free control. In summary, typical morphological changes of antibioticsusceptible or -resistant bacterial strains were illustrated in Figure 4. Regarding susceptible strains, elongated cells

Figure 4. Common morphological categorization of Gramnegative bacteria treated by β–lactam antibiotics in a dielectrophoretic antimicrobial susceptibility testing. Bacteria susceptible to antibiotics showed morphological changes: (A) elongation, (B) lysis, (C) swelling. No morphological change was found in resistant isolates (D)

treated by cefazolin, ceftazidime, cefepime, and doripenem at 1 x MIC concentration for 30, 60, 90,120, and 150 min were shown in Figure 2. With cefazolin, ceftazidime, or cefepime treatment, bacterial length extended with time. In the elongated bacteria, cell length reached 30 µm, the width of electrode gap in the chip, between 90 and 120 min (Figure 2A-2C). In



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 5. Serial changes of length of Gram-negative standard strains treated with minimal inhibitory concentrations of four βlactam antibiotics: (A) E. coli ATCC 25922, (B) K. pneumoniae ATCC 700603, (C) P. aeruginosa ATCC 27853, (D) A. baumannii ATCC 19606. Areas of cell swellings induced by doripenem was also measured. The symbol “x” on the X-axis indicates cell lysis. *** indicates a P value of less than 0.0001, as compared with time zero.

Table 2. Discrepancy of drug susceptibility between dAST and BDT for 78 clinical Gram-negative isolates.


Page 10 of 14

Page 11 of 14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Antibiotics

Analytical Chemistry Isolate number (%) Minor errors

Major errors

Very major errors

Category agreement

Cefazolin (n=63)

8 (12.7)

0

0

55 (87.3)

60 (95.2)

Ceftazidime (n=78)

3 (3.8)

0

0

75 (96.2)

78 (100.0)

Cefepime (n=78)

Essential agreement

12 (15.4)

0

1 (4.8)

65 (83.3)

71 (91.0)

Doripenem (n=78)

6 (7.7)

0

2 (9.6)

70 (89.7)

75 (96.2)

Total (n=297)

29 (9.8)

0

3 (2.9)

265 (89.2)

284 (95.6)

became attached to the electrode edge on a QEA after incubation with β-lactam. However, Doripenem, a kind of carbapenem, caused cell swelling, lysis, or free floating in the medium. Irrespective of the antibiotics, dAST could determine antimicrobial susceptibility for GNBs at 90 min. To examine the inter-observer agreement, two independent observers interpreted the MIC results of the tested drugs for 15 randomly selected isolates based on the morphological changes, as shown in Figure 4. In each run, morphological changes of bacteria were compared with antibiotic-free controls. The EA and CA of the MIC results between two observers was 98.3% and 93.3%, respectively. For quality control, the MICs of the tested drugs for a standard strain, E. coli ATCC 25922, were measured by the dAST monthly during the study period. Thus, the dAST MIC results were regarded as being reproducible and consistent. Quantitative measurements of sequential morphological changes. Quantitative measurements of morphological changes for different GNBs after antibiotic treatment were calculated and shown in Figure 5. Compared with initial cell lengths (0 min), cell lengths of E. coli ATCC 25922 treated by the MIC concentration of cefazolin, ceftazidime, or cefepime increased with time (Figure 5A, P

Dielectrophoresis System for Testing Antimicrobial Susceptibility of

Recommend Documents