Increased spread and replication efficiency of Listeria monocytogenes in organotypic brain-slices is related to multilocus variable number of tandem repeat analysis (MLVA) complex

Bacterial strains

Forty-seven L. monocytogenes strains were investigated in brain-slice cultures and cell lines (Additional file
1). The MLVA-type of 44 strains had been obtained in a previous study 31] and the MLVA-type of three other ruminant isolates (JF4971, JF5052 and JF4978; Additional
file 1) were determined during this study by analysis of tandem repeat numbers at eight
loci according to Sperry et al.23]. A minimal spanning tree was created in the BioNumerics software (Version 6.6, Applied
Maths Inc., Austin, Texas, USA) in order to define the MLVA complex of the strains
31]. Twenty-one strains isolated from ruminant rhombencephalitis cases were selected
based on the following criteria: 1) differences at the 8 MLVA loci and 2) similar
numerical representation of the two large MLVA complexes, to which most of the ruminant
rhombencephalitis strains belong (MLVA complex A: n?=?12; MLVA complex C: n?=?9). The ruminant rhombencephalitis strains were compared to a similar number of
L. monocytogenes strains from other sources (n?=?26) available in our strain collection (Additional file 1). The latter included strains from ruminant non-encephalitic cases including gastroenteritis,
mastitis and abortion (n?=?7), human clinical cases (n?=?9), food and environmental (n?=?10) and mainly belonged to MLVA complex C (n?=?18). Four strains belonged to MLVA complex A, one strain was a single locus variant
associated with MLVA complex A and three strains belonged to MLVA complex B. Non-invasive
Listeria innocua type strain CCUG15531
^T
(Culture Collection University of Göteborg) was used as negative control.

Organotypic brain-slice cultures

Hippocampal brain samples from calves under 6 months of age were obtained from the
slaughterhouse. A vibratome (Leica VT1000S) was used to cut 350 ?m brain-slices and
were cultured on membrane inserts (Vitaris, No. 3450 or 3460) as previously described
42].

Infection assays in ruminant organotypic brain-slice cultures

Brain-slices were infected at day 7 in culture. Penicillin and streptomycin were removed
from the organotypic brain-slice cultures 1 h prior to inoculation in the first set
of experiments and 4 days prior to inoculation in the later experiments due to batch
variation of the antibiotics. Medium was changed to a serum-free formula 1 h prior
to inoculation. Bacteria were plated on trypticase soy agar (TSA), incubated at 37 °C
for 15 h and subsequently diluted using the McFarland optical density standard. One
hundred CFU in 0.1 ?l NaCl (determined by plating on TSA plates) were focally injected
into the dentate gyrus of the hippocampus using a 0.5 ?l syringe (Hamilton, 7402 Bonaduz,
Switzerland. Model 7000.5 KHOC) and a micromanipulator (made in-house). Brain-slices
were incubated with the bacteria for 3 h and subsequently the inoculation medium was
substituted with gentamicin-containing medium (final concentration 0.01 mg/ml). All
experiments were carried out at least in triplicate and included L. innocua (negative control) and an internal control strain (L104, from bovine rhombencephalitis,
MLVA complex A) for normalization. For analysis of bacterial replication CFU’s were
determined 48 h post infection by lysing infected brain-slices in 1 ml PBS containing
55 ?l Isolator (Wampole, Oxoid) and plating serial dilutions on TSA plates. For analysis
of bacterial spread, brain-slices were fixed in 4 % (w/v) paraformaldehyde at 48 h
post infection. Following overnight fixation, brain-slices were incubated in 18 %
(w/v) sucrose (Sigma, S0389) for 12 h, cut with a cryotome into 4.5 ?m-thick sections
and stored at ?20 °C until further use. Immunofluorescence was performed using the
following primary antibodies: anti-Listeria O serotypes 1 and 4 (polyclonal rabbit
antibody, No. 223021, Difco, Sparks, MD, USA), neurofilament (monoclonal mouse antibody,
No M0762, DAKO, Glostrup, Denmark), GFAP (monoclonal mouse antibody, No. Ab4648, Abcam,
Cambridge UK) and CD68 (monoclonal mouse antibody, clone EBM11, DAKO, Glostrup, Denmark).
Alexa Fluor 488 and 544 were used as secondary antibodies (No. A21428, Invitrogen,
Carlsbad, CA, USA,) 42]. Nuclei were stained with TOTO-3 (no T3604, Invitrogen, Carlsbad, CA USA), and 10x
images were acquired on an Olympus FV1000 confocal microscope. The total area covered
by L. monocytogenes was measured on the immunofluorescence labeled cryosections of brain-slices using
the Olympus FV10-ASW Version 03.01.01.09 software and expressed in ?m
²
.

Plaque assay in bovine macrophages and CaCo-2 cells

Plaque forming assays were performed according to a previous study 31] in an immortalized bovine macrophage cell line (BoMac, kindly provided by D. Dobbelaere,
Department of Clinical Research and Veterinary Public Health, Vetsuisse Faculty Bern)
and the human enterocyte-like CaCo-2 cell line (ATCC No. HTB37). Both cell lines were
grown in DMEM (Gibco 61965–026) supplemented with penicillin/streptomycin (Gibco,
15140–122, used 1:100) and 10 % fetal calf serum (FCS) (BoMac cells) or 20 % FCS (CaCo-2
cells), respectively. Cells were grown to confluence in a 24-well plate overnight
at 37 °C, washed with warm PBS and inoculated with 10
³
 CFU?L. monocytogenes (MOI 0.01) in antibiotic free medium supplemented with 2 % FCS. Following 1 h incubation
the medium was removed, cells were washed with PBS and overlaid with medium containing
0.7 % agarose and 0.01 mg/ml gentamicin. The size of five randomly chosen plaques
per well were measured 72 h post infection. Experiments were carried out in duplicate
and the internal control strain L104 was included in all experiments.

Statistical analysis

All results were normalized to the internal control strain L104. Statistical analysis
was performed with the Prism Software (Version 5.03, Graph Pad Software Inc.). The
Mann–Whitney test was used to determine the p-values where two groups were compared.
For comparison of multiple groups, the Kruskal-Wallis test was used with Dunn’s multiple
comparisons as post-test. As only three strains belonged to complex B, they were excluded
from the statistical analysis comparing complexes.