Bacterial Strains And Culture Conditions

Published Date: 02 Nov 2017

First Year Report

Michelle Lister

School of Molecular Medical Sciences, Centre for Biomolecular Sciences, University of Nottingham NG7 2RD, UK

The work presented in this report by the author, except where stated in the text, is her own. Technical assistance with the project has been acknowledged where appropriate.

1. Abstract

2. Introduction

Clostridium difficile is a major nosocomial pathogen that causes gastrointestinal disease in patients who have usually received antibiotic treatment (1). The effects of this pathogen can range from mild diarrhoea to life threatening pseudomembranous colitis (2) placing a huge burden on the heath care service (3). C. difficile spores are ubiquitous in the health care setting enabling them to be passed from patient to patient and facilitating reinfection in the same patient (4). Once spores have been ingested they germinate during their descent through the gastrointestinal tract and establish colonisation within the large intestine (2). At this stage the pathogen has the ability to cause the disease state by producing toxins A and B which affect the epithelial cells of the intestine causing reorganisation of the cell cytoskeleton and resulting in cell death (5).

The pathogenicity locus (PaLoc) is a 19.6kb genetic region in C. difficile that encodes the proteins for both toxin A and B. It consists of five open reading frames; tcdR, tcdA, tcdC, tcdB and tcdE (Fig 1.). tcdA and tcdB encode toxin A and B respectfully. tcdR is believed to be a positive regulator of toxin production, tcdC is believed to be a negative regulator of toxin production and tcdE is believed to be involved with transport of toxins out of the cell (6). There has been a lot of research on the roles of these regulators with conflicting findings (7�14). There is also increasing evidence to show that other genes other than those found in the PaLoc are involved in toxin regulation (15, 16), suggesting that regulation it is even more complex than originally thought.

Fig 1: A schematic of the pathogenicity locus showing the open reading frames and their direction of transcription.

Effective countermeasures against C. difficile infection can be developed once there is a better understanding of the genetic and molecular basis of this pathogen. Recent advances in the genetic manipulation of Clostridium sp. means that it is now possible to create random mutant libraries in Clostridium difficile (17). This enables a relationship of interesting phenotypes to be linked with genetic regions by using inverse polymerase chain reaction (PCR) to amplify the DNA located around the insertion site of the transposon. This DNA can then be sequenced and compared to genomic DNA to identify the insertion site. To ensure that the phenotypic traits seen have not been caused by polar effect the gene can be inactivated by using allelic coupled exchange (ACE) and observing to see if the phenotype has been preserved.

This process has been completed in strain R20291. The mariner-transposon system has been constructed (17) and a reporter strain has been built (Clostridia Resarch Group (CRG), unpublished). This system has not been constructed for strain VPI10463 and will form the first part of this study. Firstly the conjugation plasmid needs to be selected. These are previously published and have been shown to conjugate into C. difficile 630 with a high frequency (18). The four plasmids being trailed all contain a Gram-negative replicon ColEI, the conjugation apparatus traJ, and antibiotic selection marker catP and multiple cloning site lacZ?. The distinguishing feature between the four plasmids is the Gram-positive replicon found in each. These replicons are pBP1 (Clostridium botulinum), pCB102 (Clostridium butyricum), pCD6 (Clostridium difficile) and pIM13 (Bacillus subtilis) (Fig 2).

Fig 2: Schematic of pMTL plasmids, adapted from (17).

Once a plasmid has been selected, the mariner-transposon element can be constructed and inserted into the plasmid. A reporter strain will also be constructed by replacing the tcdB gene with a licB gene originally found in Clostridium thermocellum. This gene encodes lichenase which degrades starch. This degradation can be visualised by adding lichenin to media and using a dye to look for zones of clearance (ZOC) around individual colonies. If toxin B would have been produced there will be a ZOC as none of the other genes i.e toxin promoters will have been affected. The creation of mutant libraries and screening of phenotypes can then occur.

The second part of this project will look at the role of TcdR in the regulation of toxin production. Two mutant strains of C. difficile R20291 have been created using ACE to firstly delete the tcdR gene from the wild type strain (?tcdR). This mutation was repaired using the same wild type gene but with a nucleotide substitution to not only act as a �watermark� but also to create a restriction site to enable differentiation (tcdR[EcoRI]). The strains will have their growth characteristics and toxin production compared using enzyme linked immuno sorbent assay (ELISA) and a cell cytotoxicity assay. This will firstly show that the mutations have not affected the ability of the strain to replicate and secondly, if the production of toxin if affected in any way.

The clinical aspect of this project will focus on quantitatively measuring the level of toxin production in clinical samples from patients with confirmed Clostridium difficile infection (CDI). These data will then be compared to the level of toxin produced by the strain once isolated from the stool sample. These strains will be ribotyped and compared to strains isolated, if any, from stools collected from the same patient after a four and twelve week period. Clinical information from the patients will be collected. All this information will give some insight on how different ribotypes/hosts/disease severities etc. can affect toxin production. One difficulty that will be faced is to ensure that the best yield of C. difficile is isolated from clinical samples after the patients have received antibiotic treatment for their illness. There has been a number of literature realised comparing different isolation methods (Ref***). The first phase of this aspect of the project will be to find a suitable isolation method.

3. Materials and Methods

3.1 Bacterial Strains and Culture Conditions

Bacterial strains used in this study are presented in Table 1 and plasmids in Table 2. Escherichia coli strains were cultured in aerobic conditions at 37�C on Luria-Bertani (LB) agar or broth shaken at 200rpm. C. difficile strains were routinely cultured in an anaerobic workstation (Don Whitley, Yorkshire, UK) at 37�C with the atmospheric conditions CO2:H2:N2 (80:10:10 vol:vol:vol) on BHIS agar or broth. All cultures were supplemented with antibiotics as appropriate for the selection of plasmids; concentrations are listed in Table 3.

3.2 General molecular biology techniques

E. coli strains were transformed using a Gene-Pulser (BioRad) by electroporation as recommended by the manufacturer. DNA was isolated and purified from agerose gel using GenElute Gel Extraction Kit (Sigma Aldrich). DNA was extracted from cells using Bacterial Genomic Miniprep Kit (Sigma Aldrich) or phenol-chloroform extraction was performed as described previously (19). Before genomic DNA was isolated from cells they were treated with lysozyme (10mg/ml in phosphate buffered saline (PBS)) to break down cell walls and SDS (65�C for 30min) to lyse cells. Enzymes were sourced from New England Biolabs (NEB) unless otherwise stated. PCRs were performed using Phusion High Fidelity polymerase (New England Biolabs) or KOD hot start polymerase (Novagen). All primers used in this study are listed in Table 4. Thermocyling conditions were 95�C for 3 min, followed by 30 cycles of 95�C for 30s, and appropriate temperature for annealing of primers for 30s, followed by 72�C for 1 min for short products and 3 min for long products with a final extension step of 72�C for 10 min. Ligation, restriction digestion and electrophoresis were performed using standard protocols (19). Southern blot assays was performed using a digoxigenin (DIG) High-Prime labelling and detection kit (Roche) as directed by the manufacture.

3.3 Selection of an appropriate Gram-positive plasmid replicon

Plasmids from the pMTL80000 series described previously (18) were conjugated as previously described (20). Briefly, 1mL E. coli containing pMTL80000 plasmids were collected from antibiotic containing media by centrifugation at 4000 x g for 1 minute. Cells were washed once in 1�L PBS and then re-suspended in 200�L of overnight C. difficile recipient culture. The mating mixture is then spotted onto non-selective media and incubated in anaerobic conditions at 37�C for 24h. The mating mixture was harvested into 500�L of PBS. 100�L volumes were re-suspended onto media containing antibiotics to select for the plasmid and against the E. coli donor and incubated in anaerobic conditions at 37�C for 24-48h before enumerating transconjugates. One transconjugant colony was selected to be re-streaked onto identical selective agar to check plasmid transfer. The E. coli donor culture was serially diluted to check cell density and determine the conjugation frequency.

Table 1: Bacterial strains and plasmids used in this study.

Strains Description Origin

E. coli

TOP10 Plasmid replication Invitrogen

CA434 (21)

C. difficile

630 Wild type, PCR ribotype 078

630?erm Erythromycin sensitive strain

R20291 Wild type, PCR ribotype 027 (Stoke Mandeville outbreak strain) Jon Brazier (Anaerobe Reference Laboratory, Cardiff, United Kingdom)

VPI10463

B1-7 Transposon reporter strain; R20291 ?tcdB::licB Clostridia Research Group (unpublished)

?? VPI10463 ?tcdB::licB This study

CRG*** R20291 ?tcdR Clostridia Research Group (unpublished)

CRG*** R20291 tcdB(EcoRI) Clostridia Research Group (unpublished)

NCTC 11209 Ribotype control

Table 2: Plasmids used in this study.

Plasmid Description Reference

pMTL82151 (18)

pMTL83151 (18)

pMTL84151 (18)

pMTL85151 (18)

pMTL-SC0 (17)

pMTL-SC1 (17)

Table 3: Antibiotic working concentrations

Antibiotic E. coli C. difficile

Chloramphenicol 25�g/mL

Thiamphenicol 15�g/mL

Kanamycin 50�g/mL

Lyncomycin 20�g/mL

Cycloserine 250�g/mL

Cefoxitin 8�g/mL

Lyncomycin

3.4 Plasmid segregational stability assay

The method for establishing plasmid segregational stability was slightly modified from that published by Bron et al. (22). Briefly, 1mL of BHIS broth was supplemented with 7.5�g/mL thiamphenicol was inoculated with a transconjugant colony and incubated in anaerobic conditions for 12h. A 1% inoculum was made from this culture in the same medium and incubated for 12h. Cells were collected and washed twice in PBS and used to inoculate 1mL of unsupplemented BHIS at a 1% inoculum. This marks the start of the stability experiment. The transconjugate was sub-cultured every 12h in unsupplimented BHIS broth. At 0, 24, 48 and 72 hour time points transconjugates were plated onto both BHIS agar and BHIS agar containing thiamphenicol, colony forming units (CFU) were enumerated to calculate the segregation stability.

Segrigational stability per generation was calculated as , where n is the number or generations of C. difficile that have passed without antibiotic selection by this timepoint and R is the number of cells containing the plasmid at the last timepoint that it could be ascertained. From the 1% inoculum used at each 12h subculture and assuming that maximum cell density is reached it was assumed that the number of generations in each 12h to be 6.64 because 1 x 26.64 = 100.

Table 4: Primers used in this study.

Primer name Sequence (5�-3�) Function Reference

catP-INV-F1 TAA ATC ATT TTT AGC AGA TTA TGA AAG TGA TAC GCA ACG GTA TGG Inverse PCR

catP-INV-R1 TAT TGT ATA GCT TGG TAT CAT CTC ATC ATA TAT CCC CAA TTC ACC Inverse PCR

catP-INV-R2 TAT TTG TGT GAT ATC CAC TTT AAC GGT CAT GCT GTA GGT ACA AGG Sequencing of inverse PCR products

catP-F1 GGC AAG TGT TCA AGA AGT TAT TAA GTC GGG AGT GCA GTC GAA GTG G Southern probe synthesis and PCR screening for transposon based catP gene

catP-R1 TGA AGT TAA CTA TTT ATC AAT TCC TGC AAT TCG TTT ACA AAA CGG Southern probe synthesis and PCR screening for transposon based catP gene

tcdR-Fs1 ? Tcdr primer Clostridia research group (unpublished)

tcdR-Rs1 ? Tcdr primer Clostridia research group (unpublished)

P3 CTG GGG TGA AGT CGT AAC AAG G Ribotyping

P5 GCG CCC TTT GTA GCT TGA CC Ribotyping

3.5 Construction of the mariner-transposon element

3.6 Construction of the VPI10463 reporter strain

3.7 Isolation of transposon mutants

The plasmids containing the mariner transposon where transferred into R20291 and VPI10463 via conjugation and selected for on BHIS agar containing cycloserine, cefoxitin and thiamphenicol. After which transconjugant colonies were picked and restreaked onto TY agar containing cycloserine, cefoxitin and lincomycin to enhance transposon delivery by driving the tcdB promoter. To select for the transposon based catP marker, harvested cells were plated onto BHIS supplemented with cycloserine, cefoxitin and thiamphenicol. All growth was harvested after 72 h into BHIS with 10% glycerol for storage. Serial dilutions of this growth were made, plated BHIS agar containing cycloserine, cefoxitin and thiamphenicol and incubated for 24 h before being enumerated. The most appropriate dilution factor was harvested and stored and this process repeated three times to create passage 1 � 3 (P1 � 3).

3.8 Inverse PCR and sequencing

To ensure the diversity of the mutant pool inverse PCR and sequencing were performed. At each passage genomic DNA from 10 individual transposon mutants was subjected to HindIII digestion overnight at a concentration of 200 ng/�L. Inactivation of the HindIII enzyme (65�C for 30 min) was followed by ligation with T4 ligase after digests were diluted to 5 ng/�L to favour self-ligation of restriction fragments. Ligation reactions were left at ambient temperature for 1 h and then heat inactivated (65�C for 30 min). Inverse PCRs were performed using 50 �L volumes using 100 ng of ligated DNA, KOD Hot Start DNA polymerase and FailSafe Master Mix kit (Novagen) and primers catP-INV-F1 and catP-INV-R1 (Table 4). The primers face out from the catP sequence found in the transposon therefore the unknown DNA sequence around the insertion site can be amplified. Inverse PCR products were run out on a 0.8% (wt/vol) agarose gel.

3.9 Conformation of transposon inserts and clean deletions

3.10 Identification of a suitable culture method for the isolation C. difficile from stool samples

To identify the most suitable method for the isolation of C. difficile from stool samples two separate experiments were conducted. The first experiment was to identify which broth enrichment media gave the best recovery of C. difficile. Two broths were compared; cycloserine, cefoxatin, fructose broth (CCFB) and cycloserine, cefoxatin, manatol (CCMB) broth with added lysozyme. The broths differ only in the carbon source and CCMB has lysozyme (Table 5). Faeces from hamsters known to be infected with C. difficile were added to 1mL PBS, homogenised and heat shocked (10 min at 80�C). 100�L supernatant was added to 5mL broth and incubated anaerobically at 37�C for 24 � 96h. A positive broth was indicated by both turbidity and a pH change altering the colour of the broth from red to yellow. 50�L of broth once positive was cultured onto cycloserine, cefoxatin, fructose agar (CCFA) using a four quadrant streak method. Plates were further incubated for 24 � 48h. Recovery of C. difficile was enumerated semi-quantitatively (0 = no growth. 1 = 1st quadrent, 2 = 2nd quadrent, 3 = 3rd quadrent, 4 = 4th quadrent).

For the second experiment the broth that showed the best recovery of C. difficile was chosen for the enrichment step. Spore stocks of C. difficile 630?erm were produced by preparing an overnight culture of C. difficile in BHIS broth. 100�L volumes were plated onto BHIS plates and incubated for 5 days anaerobically at 37�C. Cells from three plates were harvested into 1ml dH2O and incubated overnight at 4�C. The spore/cell suspension was washed 10x in ice cold water 12,000 xg for 2 min. Spores were then enumerated using phase contrast microscopy. 1mL volumes of mouse faeces and PBS was inoculated with 104 spores and diluted, to give equal volumes, to 10 spores per sample and stored at -20�C. Faecal/spore mixtures were heat shocked as before. For each dilution 50�L of heat shocked mixture was directly plated, using a four quadrant streak, onto CCFA, Braziers agar (LabM, Heywood) supplemented with 4% egg yolk emultion (Oxoid, ?), cefoxatin (8�g/mL) and cyclocerine (250�g/mL) and chromID C. difficile (bioM�rieux). 100�L of heat shocked mixture was added to enrichment broth and incubated anaerobically at 37�C for 24 � 96h. 50�L of broth once positive was cultured onto CCFA, Braziers agar, ChromID and tryptone soy agar with 5% sheep�s blood. Semi-quantification was performed as before.

3.11 Toxin production and cell growth assays

To measure cell growth and toxin production, from pure culture, strains were cultured in tryptose-yeast (TY) broth (3% [wt/vol] Bacto tryptose, 2% [wt/vol] yeast extract, and 0.1% [wt/vol] thioglycolate, adjusted to pH 7.4) (23). Cell growth and toxin A and B production were recorded over a 72 hour time period, starting from an exponential growth phase. This was achieved by culturing -80�C strain stocks on to TY agar supplemented with sodium taurocholate at 0.1% [w/v]. Three colonies where picked, after 16h incubation, into 1mL TY broth and incubated for 8h. In fresh TY broth, serial dilutions (10-1 to 10-8) were made and incubated for 16h. The most dilute inoculum, with visible growth, was then used for the start of the assay and diluted 1 in 100. Optical density (OD600) was then measured at 0, 3, 6, 9, 12, 24, 48 and 72 h to measure cell growth. Supernatants were collected at the same time points (centrifugation at 12,000 g for 1 min), filter sterilised (0.2�m pore size) and tested for toxin A and B using C. difficile TOX A/B II ELISA (TechLab), Vero cell cytotoxicity and neutralisation assays.

3.12 Toxin ELISA assays

Toxin ELISA assays were performed as directed by the manufactures instructions. Toxin supernatants were diluted, if required, 10 fold in TY broth until a measurable OD was obtained. Pure toxin A & B (List Biological Laboratories) were used as standards at a starting concentration of 125ng/mL and serial dilutions (10-1 to 10-7) were made in PBS. Spectrophotometry was performed using *****

3.13 Vero cell cytotoxicity and neutralisation assays

Using Vero (African green monkey kidney) cell monolayers, toxin B cytotoxicity was measured by titration. Each well of a 96-well microtiter plate was seeded with 100�L of cell suspension at a density 2 x 105 cells/mL to create a cell monolayer. The cell culture media, Dulbecco�s modified Eagles medium (DMEM), contained 1% streptomycin/penicillin (vol/vol) and 10% fetal calf serum (vol/vol). To allow the creation of monolayers, cells were incubated at 37�C 5% CO2 for 48h. 20�L of toxin supernatant was added to monolayers, if required serial dilutions were made in PBS. Cultures were then incubated for 24 h (37�C 5% CO2) before examination by phase contrast microscopy (Nickon Eclipse TS100). End point titre was expressed as �1/toxin endpoint titre� due to the inverse relationship between the endpoint titre and the amount of toxin in a sample and was defined as the first dilution in the succession where cell morphology could not be differentiated from the negative controls. Toxin neutralisation assays were performed on identical 96-well microtiter plates using ****. Cells showing no toxic effect should be identical to the negative controls. Pure toxin B and the control provided in the ***** kit were used as positive and/or negative controls depending on which assay was being performed.

3.14 PCR ribotyping

4. Results

4.1 Construction of a mariner-based transposon system for C. difficile VPI10463

Conjugation frequencies and stability assays of pMTL80000 series plasmids into C. difficile strain VPI10463 were assessed to identify an appropriate vector for the mariner-based transposon system. Each of the four plasmids containing different gram positive replicons were conjugated into VPI10463 along with a 630 control to ensure that the process had worked. Transconjugates were assessed as Thiamphenicol resistant (Tmr) colonies which took between 24 � 48 h to grow. Of the four vectors trailed two were unable to be conjugated into VPI10463 and one was unable to be conjugated into 630 (Table 6).

Table 6: Transfer frequencies in C. difficile VPI10463 and 630

Vectora Gram positive replicon Organism Conjugation frequency in 630b Conjugation frequency in VPI10463b

pMTL82151 pBP1 C. botulinum 1.86 x 10-5 1.11 x 10-7

pMTL83151 pCB102 C. butyricum -c

pMTL84151 pCD6 C. difficile 8.69 x 10-6 2.50 x 10-8

pMTL85151 pIM13 B. subtilis -d -d

a Each vector has an identical context except the Gram positive replicon as described in the Introduction (18).

b Conjugation frequency calculated as number of transconjugates per CFU of E. coli donor.

c The pCB102 containing vector could not be transferred into C. difficile VPI10463.

d The pIM13 containing vector could not be transferred into C. difficile 630 or VPI10463.

Stability assay*******

4.2 The role of tcdR in toxin production.

Strains R20291 (wild type), CRG*** (R20291 ?tcdR) and CRG*** (R20291 tcdR[EcoRI]) were monitored for cell growth and toxin production in TY medium over a 48 h period. The OD600 readings showed no difference between R20292 and CRG*** (?tcdR), there was however a slight increase in OD600 for strain CRG*** (tcdR[EcoRI]) (Figure 3A). Interestingly, the toxin production in strain CRG*** (?tcdR) was between 100,000 and 10,000 times less than in the wildtype strain (R20291) and the complemented strain CRG*** (tcdR[EcoRI]) (Figure 3B). But the deletion of the tcdR gene has not stopped toxin production altogether.

Fig 3: Cell growth and toxin production in strains R20291, CRG*** (?tcdR) and CRG*** (tcdR[EcoR1]) in TY medium. (A) Optical density was obtained over a 48 h period. (B) Toxin production was measured using a toxin A and B ELISA, samples were diluted as required and absorbance (A450) was measured against a standard curve.

To further check the level of toxin production, cell culture using Vero cells (detection limit 25pg/mL, toxin B) were subject to titrated toxin supernatants taken at the 48 h time point. 1/Endpoint titre for both R20291 and CRG*** (R20291 tcdR[EcoRI]) was determined to be 1 x 109 with 1/Endpoint titre for CRG*** (?tcdR) being 1 x 103. To ensure that the cytotoxic effect on the Vero cells was caused by toxin B and not another artefact, a toxin B neutralisation assay was performed using a duplicate of the cytotoxicity assay. This duplication safeguards against experimental error and proves that the effects seen are caused by toxin B in the supernatant. There were no cytotoxic effects on cells witnessed in the neutralisation assay except for on the positive controls (wells containing no antitoxin).

To prove that the deletion and the complementation had taken place in these strains primers tcdR-Fs1 and tcdR-Rs1 were used to amplify the tcdR gene. The product for both R20291 and CRG*** (R20291 tcdR[EcoRI]) produced a band approximately 1500bp and in the case of CRG*** (?tcdR) a band approximately 1000bp (Figure 4A&B). This shows the deletion of 537bp has occurred. To demonstrate that complemented strain was indeed different to the wild type strain the PCR products were subjected to an EcoRI digestion and then visualised. The products for R20291 and and CRG*** (R20291 tcdR[EcoRI]) were not digested. The product for CRG*** (?tcdR) did show the restriction site that is absent in R20291 (Figure 4C).

Fig 4: PCR analysis of C. difficile R20291 strains. (A) Schematic of the tcdR locus in R20291. The triangle denotes the location of the deletion made in CRG*** (?tcdR) and the approximate location of the EcoR1 site in CRG*** (tcdR[EcoR1]) is shown above the open reading frame. Half arrows show the annealing positions of the primers. (B) PCR analysis using primers tcdR-Fs1 and tcdR-Rs1. (C)

4.3 Isolation of C. difficile from stool samples.