DNA-Microarray-based Genotyping of Clostridium ...

DNA-Microarray-based Genotyping of Clostridium difficile Darius Gawlik 1,2 ,Peter Slickers 3,5, Ines Engelmann3,5, Elke Müller3,5, Christian Lück1, Anette Friedrichs 4, Ralf Ehricht3,5, Stefan Monecke 1,3,5 1) Institute for Medical Microbiology and Hygiene, Technische University of Dresden, Dresden, Germany 2) Hamm-Lippstedt University, Hamm, Germany 3) Alere Technologies GmbH, Jena, Germany 4) Institute for Clinical Molecular Biology, Christian-Albrechts-Universität zu Kiel, Kiel, Germany 5) INFECTOGNOSTICS Research Campus Consortium Jena

Introduction: Clostridium difficile (C. difficile) is a gram-positive, spore-forming bacterium. It is a component of the human intestinal flora but its disturbance by antibiotic therapy can result in a proliferation of C. difficile. Resulting conditions are diarrhoea and pseudomembranous colitis. However, transmissions within hospital settings are common, given that spores are able to survive in a clinical environment and are resistant to many (alcoholic) disinfectants. This warrants molecular typing of C. difficile isolates. For that purpose, a microarray was developed. Material and Methods: 249 clinical C. difficile isolates were tested. They were inoculated on Schaedler haemin cysteine blood agar that was previously pre-reduced under anaerobic conditions, and incubated at 37 °C for 48 hours. For DNA isolation, the QIAamp® DNA Mini Kit (250) was usedaccording to the manufacturer's instructions. For validation and optimisation of the protocol, sequenced strains 630 (GenBank AM180355), BI9 (FN668944), CF5 ( FN665652), M120 (FN665653), CD196 (FN538970) and R20291 (FN545816) were used. All targets were amplified in a linear amplification step incoporating 16-dUTP-linked biotin. Amplicons were hybridised to the array. This was followed by washing and addition of horseradish-peroxidase-streptavidin conjugate and a precipitating dye. An image of the array was recorded and analysed using a Arraymate reader (Alere Technologies). Normalised intensities of the spots were calculated based their average intensities and on the local background. For representative isolates, MLST (Griffith 2010) and ribotyping (Fawley 2012) were performed. Results: Based on slpA alleles and overall hybridisation profiles (excluding possibly mobile or variable elements such as toxin genes and resistance markers), tested isolates clustered into 40 distinct patterns (Table 1). Three additional patterns were predicted from published genome sequences, although they were not found experimentally. If several isolates with identical hybridisation patterns were subjected to MLST, they yielded identical or related sequence types. Occasionally, several ribotypes were observed within one cluster and some ribotypes were present in different, although similar or related clusters. There was a good correlation of array patterns in general with toxin alleles as well as with MLST derived clades as previously defined (Dingle, 2011). Hybridisation pattern HP-01 HP-02 HP-03 HP-04 HP-05 HP-06 HP-07 HP-08 HP-09 HP-10 HP-11 HP-12 HP-13 HP-14 HP-15 HP-16 HP-17 HP-18 HP-19 HP-20 HP-21 HP-22 HP-23 HP-24 HP-25 HP-26 HP-27 HP-28 HP-29 HP-30 HP-31 HP-32 HP-33 HP-34

Fully sequenced reference strains BI-9 (FN668944) Strain 630 (AM180355) CD196 (FN538970)

GenBank entries , that were analysed Tested in silico only isolates ABHD* 76 ABKJ* 5 4 5 2 1 1 10 3 17 7 AHJJ* 2 2 AGAC* 24 1 1 2 2 7 4 9 2 1 1 1 ADEJ* 18 7 2 2 ADEH* 2 4 4 FN668941*, ABKK*, ABHE*, ABHF* 2

HP-35

R20291 (FN545816)

FN665654*, AAML*, ABHG*, ABFD*

HP-36 HP-37

-

HP-38 HP-39 HP-40 HP-41 HP-42 HP-43

MLST Clade

Associated sequence types

slpA allele

Associated ribotypes

I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I II

ST-03 ST45, ST-46* ST-58 ST-04 ST-55 ST-08 ST-03 ST42* ST-17 ST-08 ST-54 ST-54 ST-35 ST-58/related (incomplete)* ST-09 ST-01*

slpABI9 slpABI9 slpA6407 slpA630 slpADJNS0578 slpA negative slpAR12884 trunc. slpAR13711 slpAR13711 slpAR13541 slpA23m63 slpA23m63 slpAJND08162 slpAR12885 slpAKohn slpAKohn slpA79685 slpAJND09041 slpAMRY060211 slpAJ9952 trunc. slpAR12884 slpAR13711 slpAR13711 slpAR13711 slpAR13711 slpA630 slpAJND08037 slpA1446 slpAR13700 slpA6407* slpA6407 slpA6503 trunc. slpA negative slpAR20291

RT-001, RT-015, RT-072 RT-001, RT-013, RT-087 RT-011, RT-049, RT-056 RT-137, RT-150 RT-163 Unidentified pattern RT-054 RT-057, RT-070, RT-094 RT-002, RT-159 RT-009 RT-103 RT-014, RT-049 RT-015 RT-029 RT-064 RT-005 RT-013, RT-087 RT-005, RT-045, RT-054 RT-031 RT-012 RT-012 RT-046 RT-039 RT-010 RT-071 RT-029, RT-081, RT-094 RT-081 RT-027

-

II

ST-01*

slpAR20291*

RT-027

-

2 1

III III

-

slpAR12884 slpAR12884 trunc.

RT-023 -

CF5 (FN665652)

AGAA*, AGAB*, FN668375*

1

IV

ST-37*, ST-86*

slpACF5

RT-017

Strain M120 (FN665653) -

ADVM*, ADNX* ADDE* ABKL* ADEI*

1 8 2 -

IV V V V “VI”

ST11 ST-11* (1 mismatch) ST-11 *(1 mismatch) ST127 dlv*

slpA79685 slpAR13540 slpAR13540* slpA23m63 slpA6503*

RT-017 RT-078 -

Table 1: Hybridisation patterns and their assignment to sequence types, MLST clades and ribotypes. (* = in silico prediction only)

Figure 1: SplitsTree graph based on hybridisation profiles, showing the clustering of profiles into different clades as defined by MLST. Note that tcd-negatives appear as separate clade because of the number of tcd-specific probes that override other, cladespecific differences.

Discussion: A microarray was developed and tested as a rapid, reproducible and convenient method for the molecular typing of C. difficile cultures. Other typing methods (PCR-ribotyping, MLST and pulsed-field gel electrophoresis) need more time for obtaining comparable results. A major advantage for the array-based approach is that typing as well as toxin gene detection can be performed within a single experiment by a single amplification. Acknowledgments: The authors thank K. Hochauf, K. Lück, F. Gunzer (Dresden), C. Seybold (FLI, Jena) and E. Glocker (Freiburg) for collecting and providing strains and isolates, as well as W. Rudolph (Dresden) for help with MLST sequencing. We acknowledge L. v. Müller and colleagues at the National Reference Center for C. difficile, Saarland University, for help with ribotyping, confirmatory toxin PCRs and their hospitality. We thank A. Ruppelt (Dresden) J. Sachtschal and G. Rößler (Jena) for excellent technical assistance as well as Professor E. Jacobs (Dresden), E. Ermantraut (Alere Jena) and the INFECTOGNOSTICS Research Campus Consortium Jena for their support. Contact: [email protected]; Download: http://alere-technologies.com/fileadmin/Media/Paper/Poster/DGHM-2014-Cdiff.pdf