Heteroduplex Typing of Clinical Isolates of Acinetobacter from Intensive Care Units in a Tertiary Health Care Centre and their Antibiogram

Introduction

Acinetobacter species are low virulence pathogens that tend to cause infections in immunocompromised individuals. Their ubiquitous nature and resistance to environmental conditions and antibacterial agents play a major role in their emergence as an important nosocomial pathogen. Last three decades have seen the emergence of Acinetobacter as a nosocomial pathogen of immense significance whose pathogenicity was a matter of debate hitherto.^1-4 Acinetobacter is implicated in various disease conditions like ventilator-associated pneumonia (VAP), skin and wound infections, blood stream infections as well as endocarditis, peritonitis in patients receiving peritoneal dialysis, meningitis and urinary tract infections, to name a few.² Occasionally these organisms may cause conjunctivitis, osteomyelitis, and synovitis.

The infection is acquired through a variety of sources like patient’s own flora or hospital environment. Community acquired Acinetobacter infections are on rise.⁵ Acinetobacter species spread from sources like soil, water, human skin etc.² The ubiquity of these organisms can be attributed to their ability to adapt genetically to new environmental challenges. In addition, they have formidable ability to acquire drug resistance on exposure to antibiotics.^6,7 Prevention of infection requires knowledge about the epidemiology of infection. Therefore, it is essential to identify the source or environmental colonization of these bacteria, especially in the hospitals, to prevent nosocomial infections. Typing of pathogens is critical in source tracing. Various potent molecular typing methods are available to investigate the epidemiology of infections based on the molecular heterogeneity of the isolates. The established methods of typing are expensive, laborious and time consuming.^8-12 Therefore, a less expensive, sufficiently rapid and less laborious method is desirable.

Heteroduplex mobility assay or heteroduplex analysis (HDA) is a novel, less expensive, adequately rapid and robust typing method. HDA combines some of the advantages of allele-specific Polymerase Chain Reaction (PCR) and Restriction Fragment Length Polymorphism (RFLP) methods with the advantage of direct sequence analysis to detect new alleles.¹³ Heteroduplexes are formed when PCR amplicons from reference strain and test strain are denatured and allowed to anneal. When the strands of reference DNA duplex with allele test DNA strands, heteroduplexes are formed. HDA detects the difference in mobility of heteroduplex products on electrophoresis. The number and type of mismatched bases within a given heteroduplex product determines the conformation and mobility of the DNA duplex during electrophoresis. HDA is a very sensitive method. If the fragment length is <500 bp, it can detect even single nucleotide difference in fragments.^14,15 HDA has been extensively used to determine the variation in microbial isolates either within the genus and / or different genera from sources of the isolates.^16-18 HDA being a rapid and less strenuous method, multiple isolates can be screened together.^19,20

Ours is a 1000 bed tertiary health care centre catering to the needs of around 10 lakh people in and around North Karnataka. We have well equipped intensive care units. We often encounter Acinetobacter species from various clinical samples and also from the hospital environment. We believe that it is imperative to know molecular diversity of these isolates which would give us an insight into the source of infection. It also helps us to take appropriate measures to control the contamination of hospital environment.

There is a need to type the Acinetobacter isolates from this hospital to know the molecular diversity among them. This information would help us to monitor and take appropriate measures to control and prevent hospital infections. With this background, we aimed to establish HDA and type the Acinetobacter isolates from clinical samples received from different ICUs of our hospital and understand the distribution of heteroduplex types of Acinetobacter isolates in our ICUs.

Material and Methods

The study was initiated after obtaining the Institutional Ethics Committee clearance.

Isolation and identification of Acinetobacter species from clinical samples

Various clinical samples such as pus, sputum, endotracheal tube (ET) aspirates, central venous catheter (CVC) tips, intra-jugular vein catheter (IJC) tips, lumbar drain tips, urine, wound swabs, blood, other sterile body fluids etc. were received at the Clinical Microbiology laboratory. Acinetobacter species isolated from clinical samples originating from various Intensive care units (ICUs) such as Medical ICU (MICU), Neonatal ICU (NICU), Paediatric ICU (PICU), Surgical ICU (SICU), and ICU Super speciality Hospital (ICUSH) were stored on nutrient agar butts for further study. The isolates were identified using standard established protocols.² Briefly, oxidase negative, catalase positive, non-motile, non-fermenting, Gram negative coccobacilli were presumptively identified as Acinetobacter species. Antibiotic sensitivity of the isolates was performed by VITEK-2 system (bioMérieux, France).

Extraction of DNA from Acinetobacter isolates and quantification of DNA

The isolates were sub cultured on brain heart infusion (BHI) agar plate to check for purity and to obtain fresh cells. Two well characterized, similar looking colonies from BHI agar were inoculated into 1 mL of Luria Bertani (LB) broth (Hi Media, India) and incubated overnight at 37°C. Extraction of DNA was carried out by phenol-chloroform method. DNA samples were quantified in UV/Vis spectrophotometer (Eppendorf BioSpectrometer Kinetic, Germany). All DNA samples were stored at -200C (Celfrost).

Molecular identification of clinical isolates of Acinetobacter

A genus level PCR was used for molecular identification of clinical Acinetobacter isolates. The primers (Ac696F; Ac1093R), targeting 350 bp hyper variable zone in the rpoB gene specific to Acinetobacter species were used. The PCR was carried out in thermal cycler (QB – 96, Quanta Biotech, UK). The 350 base pair targets with the flanking region gave a band on gel electrophoresis at 397 bp. PCR reaction mixture (50 μl) comprised of 2 μl DNA template, PCR Master-mix (10 X, Sigma Aldrich) and primers (0.2 μM). The amplification protocol suggested by Kulkarni SS et al.²¹ was used with some modification; briefly, initial denaturation at 94°C for 2 minutes, followed by 35 cycles of denaturation at 94°C for 30 seconds, annealing at 53.8°C for 30 seconds, extension at 72°C for 1 minute followed by final elongation at 72°C for 7 minutes. The amplicons were resolved in 1% agarose gel containing ethidium bromide (0.5 µg/mL) in Tris-acetate EDTA buffer at 100 volts for 1 hour. The amplicons were observed and documented using E-box imaging system (Vilber Lourmat, France).

Heteroduplex HD analysis

A total of 570 clinical isolates of Acinetobacter from ICUs were subjected to HDA. The PCR amplicons of clinical isolates of Acinetobacter and that of in-house standard strain were used for HD analysis. The checker board matrix was conceptualized and implemented to select the best laboratory isolate as an in-house standard isolate to perform HD analysis (Table 1). A laboratory isolate giving consistent and reproducible HD bands was identified by this matrix. An isolate from endotracheal aspirate from an adult MICU patient (Isolate No. 229) gave robust and stable bands with other isolates. This isolate was preserved as our in-house standard for HDA

The HDA protocol used was modified from the protocol used by Eugene et al.¹⁹ 10 μl of test amplicon was added in 0.2 mL sterile PCR tube. To this 10 μl of amplicon of inhouse standard strain was added. 2 μl of annealing buffer (1000 mM NaCl, 100 mM Tris-HCl, pH 7.8, 20 mM EDTA) was added to the tube. The reaction mixture was vortexed for five seconds followed by a brief centrifugation at 5000 rpm. Total reaction volume was 22 μl. The tubes were subjected to HD formation in thermocycler (QB-96, Quanta Biotech, UK) using the following protocol. Denaturation at 94°C for 3 minutes, initial annealing at 60°C for 30 minutes followed by final annealing by taking the temperature to 30°C with 0.1°C/ sec negative ramping and holding at 30°C for 5 minutes. The reaction mixture was immediately subjected to native polyacrylamide gel electrophoresis (PAGE) to analyse heteroduplex formation.

Native polyacrylamide gel electrophoresis (PAGE) of heteroduplexes

Heteroduplexes were detected by native polyacrylamide gel electrophoresis (10% Acrylamide: Bis Acrylamide 29:1, 30%). PAGE was performed in Vertical Dual Mini Gel System (Genei Laboratories Pvt Ltd., Bangalore) of gel size 8 cm x 7 cm. Electrophoresis was performed in Tris-borate-EDTA at a constant voltage of 100 volts for 3.5 hours. The wells were loaded with 10 μl HD mixture containing 2 μl gel loading buffer. DNA ladder (Thermo Scientific GeneRuler, 100-1000 bp) was included with each run. In house standard homoduplex DNA samples were also loaded with each run to ensure the detection of homoduplexes on the gel. Gels were stained by ethidium bromide (0.5 μg/mL) for 15 minutes and were observed under gel doc system (Vilber Lourmat, France). The distribution of bands according to their mobility had a few common patterns. These patterns were assigned to types 1 through type 25 by consensus between the investigators.

Results

A total of 570 clinical isolates of Acinetobacter from ICUs were subjected to HDA. After extraction, DNA samples were quantified to know the yield and purity of DNA using Epoch Microplate spectrophotometer (BioTek/ Agilent, India). The quantity of extracted DNA was found to be between 500-900 µg/mL with minimal contamination with protein and other biomolecules. We observed that most of the isolates were multi drug resistant (Table 2) and significant variation in antibiogram was not observed between isolates of same HD types.

Heteroduplex Assay

An identified isolate of Acinetobacter was used as an inhouse standard strain in the study for heteroduplexing. Based on the electrophoretic migration of heteroduplexes, we recognized a total of 25 heteroduplex (HD) types (Table 3). Out of 570 isolates, 212 (37.19%) isolates were assigned to one of the 25 HD types. Remaining 358 (62.81%) isolates did not form heteroduplex with inhouse standard. A total of 10 bands of various sizes were identified from 212 isolates. The band sizes obtained were - 450 bp, 500 bp, 550 bp, 600 bp, 700 bp, 800 bp, 850 bp, 900 bp, 1000 bp and 1100 bp (Figure 1- 5). The number of bands varied from isolate to isolate from one to a maximum of six bands. The permutations and combinations of these bands were used for classifying the isolates into types (Table 3). HD type 6 was the commonest having 94 (44.3%) isolates, followed by type 22 (19; 9.0%) and type 19 (12; 5.7%). For HD types 1, 3, 10, 12, 23 and 25, only one isolate per type was encountered and these isolates were rerun to confirm the results.

Discussion

This work was initiated to study the possibility of using heteroduplex analysis (HDA) for typing clinical isolates of Acinetobacter. Acinetobacter species are important nosocomial pathogens and are capable of producing outbreaks in hospitals. These organisms get adapted to new environments and show great genetic diversity.^22,23 A sequence from beta subunit of rpoB gene, consistently found in all Acinetobacter species was used as a target.¹⁵

Typing of clinical isolates is of cardinal importance in source tracing or epidemiological study of hospital pathogens. The commonly used typing methods are cumbersome, expensive, time consuming and less sensitive. With a frightening increase in drug resistant hospital infections, increasing patient awareness and ever-increasing enquiries by the insurance companies, there is a definite need for a more specific, reliable, reproducible, quicker and cost-effective typing method. The precision of molecular methods over other typing methods has been proved time and again.^8-12 Molecular methods can be accurately standardized, the protocols can be precisely defined, can be easily adopted by other laboratories and we may say, molecular methods have mathematical precision.

A total of 570 Acinetobacter isolates from clinical samples originating in various ICUs of this hospital were included in the study. All the isolates were phenotypically and genotypically identified as explained above. The amplicons from the PCR of these isolates were subjected to HDA. Of the 570 isolates, 212 isolates could be typed into 25 HD types while 358 isolates could not be typed or differentiated. These 358 isolates belonged to a homologous HD-Type. This could be a strain prevalent in our hospital environment. Environmental sampling was not a part of this study and therefore, we were not able to comment on presence of this HD-Type in our hospital environment. The environmental sampling was dropped from the original study design because of the suggestion from the funding agency.

Most of our isolates belonged to HD-Type-6 (94/212; 44.3%). They were isolated from samples from all the ICUs. This, therefore, may be the next most common HD-type prevalent in our hospital. The large number of isolates belonging to HD-Type-6 suggests that it could be a hospital pathogen. Table 2 shows the HD types found in this study. Following the group of 358 untypable isolates, HD-Type-6 (94, 44.5%) HD-Type-22 (19/8.96%) and HD-Type-19 (12/5.66%) were the commonest HD types observed. The occurrence of Type-6 and Type-22, when compared with untypable isolates was found to be statistically significant with a p value of <0.001 (z test).

Of the 570 isolates, 358 isolates (62.81%) did not form HD. The inhouse standard strain we had selected for typing the isolates was recovered from a clinical sample. This strain is probably belonging to the most prevalent Acinetobacter type in our set up. This is, therefore, showing homology with the 358 isolates.

There is a need to continue this work using environmental sampling. Environmental sampling would be important to identify if any of the currently found HD-Types are replacing the existing types and getting established in the hospital environment. Different molecular methods have been utilized to demonstrate genetic diversity in Acinetobacter species.²² HDA is a comparatively simpler, sensitive and rapid method capable of detecting even single bp variation.^14,15 We compared the antibiograms of isolates belonging to same HD-Type and noted that there was no significant variation in the antibiotic resistance pattern of the isolates. The isolates may vary in other characters such as ability to form biofilm, adhesion and other attributes of pathogenicity; however, this was not investigated. HD-Typing will be an excellent tool for laboratories not affording sequencing. Sequencing studies, however, would be the final answer to identify the source as it has unmatched precision.

Through this study, we developed a rapid, inexpensive and reliable HDA for clinical isolates of Acinetobacter. Such an analysis is essential to understand the epidemiology of any infection including nosocomial infections. This will help in contact tracing, identification of source and to undertake control measures in the outbreak situations. HDA can be applied to other hospital pathogens. Evaluation of this method at different centers on various pathogens will refine this technique. There is certainly a strong potential for HDA to evolve into a potent inhouse typing method for tertiary health care facilities.

Limitation of the study

Sequencing of various bands generated was not performed.

It is likely that as this gene is highly conserved, it is not showing a great variation. This may limit the typeability of the technique

Conclusion

Heteroduplex analysis provides a simple and accurate method to analyse the strain diversity of Acinetobacter isolates. Epidemiological typing and source tracing of the pathogens cab be done using HD typing. The technique has potential to be applied for other organisms. Various gene targets in the same pathogen may be explored for better precision.

Financial support

Research Grant by Advanced Research Wing, RGUHS, Bengaluru, Karnataka.

Conflict of interest

None

Acknowledgement

We would like to acknowledge the Research Grant given to the corresponding author by Advanced Research Wing, RGUHS, Bengaluru, Karnataka. We acknowledge the help of technical staff of Department of Microbiology, SDM College of Medical Sciences & Hospital, Shri Dharmasthala Manjunatheshwara University, Dharwad – 580009 (Karnataka), India.