Bacterial strains and culture preparation
In these experiments, we used a S. aureus isolate from a dominant mastitis clone. This S. aureus isolate was identified as Publsed Field Gel Electrophoresis (PFGE) type A (A7 and A3) corresponding to clfA subtype Q, identical to the sequenced RF122 isolate [11]. It was used in mammary alveolar cell-T (MAC-T) invasions under normal (incubator with 5% CO2) and reduced oxygen (5% O2) conditions. The reduced oxygen condition was maintained in a Modular Incubator Chamber as explained below. The isolate was stored at -80°C without subjecting to extensive subculturing to avoid genetic changes. Innoculum cultures were prepared according to Bayles et al., (1998) [12] with some modifications. Overnight cultures (grown in the invasion medium, which was the cell culture growth medium without antibiotics and Fetal Bovine Serum (FBS), as defined below) were centrifuged and the pellet was washed once with sterile phosphate-buffered saline (pH 7.2) and resuspended in 10 ml of the invasion medium to give a density of 1010 CFU ml-1. Serial dilutions of this were prepared and 1 ml/well of the 102 dilution was used to inoculate MAC-T cell monolayers at the multiplicity of infections of 100 (MOI 100).
Cell culture
A bovine mammary epithelial cell line designated MAC-T [13] was used. The growth medium contained 44.5% Dulbecco’s Modified Eagle Medium (DMEM)(Gibco BRL), 44.5% Roswell Park Memorial Institute (RPMI) Medium 1640 (1X) with L-glutamine (Sigma-Aldrich Ltd, Oakville, Ontario, Canada), 10% FBS (Invitrogen Canada Inc., Burlington, Ontario, Canada), and 1% of 100x concentrated antibiotic/antimycotic solution (Invitrogen Inc.). It was then sterilized by 0.22 µm filters. Cells were seeded at 1 x 106 cells/well and grown for 2-3 days at 37°C with 5% CO2 before being used for the experiments.
For normal oxygenation, cells were cultured in the incubator with 5% CO2. To ensure adequate amount of oxygen flux into cells, culture plates with wide-surface, flat-bottom, and shallow wells were used that were inoculated with a thin layer of media (1ml) with frequent swirling of plates. These measures have been found necessary, as the oxygen does not readily diffuse into culture media [14]. For reduced oxygenation, the Modular Incubator Chamber model 101 (Billups-Rothenberg, Inc, California, USA) was used to create lower limits of microaerophilic conditions. To achieve low oxygen level, 5% oxygen in the Modular Incubator Chamber (37.5 mm Hg) was used. For each experiment in these chambers, two 6-well tissue culture plates were simultaneously incubated under the same condition; one with co-cultures and the other with free bacteria without host cells to act as control. The experiment was repeated three times (3x) each under normal and reduced conditions. This design aimed to mimick the intra-mammary oxygen levels during mastitis.
Invasion Assay: The assay was carried according to Shompole et al.(2003) [15]. Approximately 16 h prior to experiment the cell growth medium was replaced with 1 ml of the invasion medium. The morning of the experiment, the medium was removed and monolyayers were washed once with the invasion medium. The MAC-T cell plates were then inoculated with S. aureus in the invasion medium at the MOI 100 and incubated at 37°C under normal oxygenation or reduced oxygenation. Plates containing only free bacteria in the invasion medium without host cells were used as the control for both oxygenation conditions After 1 h, supernatants of the cocultures were removed and monolayers washed three times with the invasion medium containing lysostaphin (10 µg ml-1; Sigma) to kill extracellular bacteria [16, 17]. Plates were then re-washed with the invasion medium before incubating in the fresh invasion medium containing 100µg /ml gentamicin (Invitrogen Inc.) for 8h. Supernatants were then removed and discarded. MAC-T cell monolayers were carefully and quickly washed with sterile distilled water, scraped with disposable sterile scrapers (Fisher Scientific, Ottawa, Ontario, Canada), and quickly stored at -80°C in 5-10 volumes of RNAlater as described by the manufacturer (Applied Biosystems, Ambion Inc., Streetsville, Ontario, Canada), before RNA extractions. Extractions were done according to the Ambion RiboPure Bacteria kit (Ambion Inc.) at the PFGRC, J. Craig Venter Institute (JCVI), Maryland, USA.
Whole-genome transcriptional profiling:
We compared gene expression in S. aureus recovered from a MAC-T intracellular environment under normal oxygen tension with gene expression in free-living S. aureus under the same condition. Similarly we compared gene expression in S. aureus recovered from a MAC-T intracellular environment under low oxygen tension with that of free-living S. aureus under the same condition.
The whole-genome qPCR was conducted on twelve samples under four treatments: internalized S. aureus and free S. aureus under normal or reduced oxygen conditions, with three samples for each treatment. Primer pairs covering 5182 open reading frames on three genomes (COL, Newman, and RF122), were designed (JCVI, Maryland, USA). We needed 13.5 of 384-well plates/run to cover 5182 ORFs. Duplicate wells from triplicate samples from 4 experiments = 13.5 x 2 x 3 x 4 = 324 plates. The large number of these open reading frames were for hypothetical proteins based on theoretical information or for ‘unknown’ proteins due to lack of enough information on them. Primers were diluted to 1.25 uM and the final concentration was 0.125 uM. (i) cDNA synthesis and purification: cDNA was synthesize by taking 2 µg of total RNA, mixing with a final concentration of 0.5 µM of dNTP mix and random hexamers using SuperScript III (Invitrogen Inc.). cDNA was purified using Qiagen minElute column (Qiagen), according to the manufacturer protocol. (ii) High-throughput qRT-PCR Purified cDNA was diluted 1:10 in DEPC water. Nine ml of diluted cDNA and 15ml of Roche 2x SYBR Green master were mixed. Eight µl of cDNA/mastermix was aliquoted into 384-well plates. Primers were added to cDNA master mix to a final volume of 10µl. Plates were sealed with qRT-PCR tape and stored in the -80 until use. Plates were run in a Roche LifghtCycler480 (LC480) RealTime PCR system 384-plate format for gene-expression, interfaced with robotics to create an automated high-throughput. qRT-PCR cycles were as follow: Step 1: 5 minutes at 95°C. Step 2: 95°C for 10 secs, 60°C for 10 secs, 72°C for 10 secs. Cycle 65 times. Step 3: melt curve analysis on all wells.
Normalization and data analysis the data from the qRT-PCR was first normalized on average crossing-poing (Cp) values by plates by subtracting the mean Cp value of the plate from each Cp value on the plate. The known S. aureus housekeeping genes (gmk, gyrA, rplD, rpoB and 16s rDNA sequence) were used to normalize data from different plates. Each gene (represented by the open reading frame) was evaluated in duplicate and the duplicates were averaged before normalization. For the purpose of expression profiling for each gene, the difference between the internalized S. aureus and free S. aureus under either normal or reduced oxygen conditions was calculated. The difference was considered significant only if there was 4-fold difference in expression levels.
The normalized expression data generated by the high-throughput LC480 system was run in the Biological Role Query Tool (BioQT) linked to the server at the JCVI. BioQT retrieves and presents annotation information from several databases (CMR Cellular Roles, Gene Ontology (GO) Terms, Protein Families (PFam), TIGRFams, Kyoto Encyclopedia for Genes and Genomes (KEGG) Pathways, Enzyme Commission information (EC) numbers, and KEGG Orthologs for lists of protein accessions or gene locus IDs for all organisms currently entered into JCVI’s Comprehensive Microbial Resource (CMR) or into a JCVI small genome database. BioQT reports summaries of cellular roles and also includes a pie chart that represents the role distribution.