Background
Bacterial meningitis is still an important cause of mortality and morbidity despite advances in antimicrobial therapy [
1,
2]. Especially, the exact role of the blood-cerebrospinal fluid (CSF) barrier, which is constituted by epithelial cells of the choroid plexus (CP), in bacterial meningitis is under investigation [
3,
4]. Important functions of the CP are maintaining homeostasis in the CNS, CSF secretion and participation in neurohumoral brain modulation and neuroimmune interaction [
5,
6].
Streptococcus suis (
S. suis) is a swine and emerging human pathogen causing a wide range of infections like meningitis and septicaemia [
7].
S. suis has been suggested to enter the brain via the blood-CSF barrier. In fact, lesions have been observed at the CP in natural and experimentally induced cases of
S. suis meningitis in pigs and mice [
8‐
10]. In an inverted Transwell filter model of primary porcine CP epithelial cells (PCPECs)
S. suis invades PCPECs specifically from the basolateral side in a capsule-dependent manner [
4]. Furthermore, after apical infection of PCPECs with
S. suis, tight junction function, morphology and protein expression is significantly altered [
3,
11].
Inflammatory activation of epithelial and endothelial cells, e.g. after bacterial infection, induces the release of interleukin-8 (IL-8) and other chemokines that recruit polymorphnuclear neutrophils (PMNs), which transmigrate across the cellular barriers and build the first line of defence to subcellular spaces [
12,
13]. For endothelial cells two possible routes for leukocyte transmigration have been described: paracellular and transcellular [
14,
15]. Ling et al. have reported that monocytes traverse epiplexus cells by a process called emperipolesis, whereby monocytes migrate through the epithelial cells [
16]. In contrast, for PMNs only data for paracellular transmigration through epithelia exist so far [
12,
13].
The molecular events of a transcellular pathway involve a rather complicated mechanism wherein a zipperlike model of junctional disruption is easy to envision. Many studies may undervalue the frequency of transcellular events since they appear in very close proximity to the junctions and thus might be mistaken for paracellular migration [
17,
18]. Thus, the visualization of a leukocyte migrating through endothelial cytoplasm very close to, but distinct from, the junctional area requires advanced ultrastructural technical settings.
The molecular mechanisms employed by PMNs to cross endothelia and epithelia have been intensively investigated. So far only a few molecules have been found to be involved in the transmigration across epithelial monolayers; these include the leukocyte α
Mβ
2-integrin CD11b/CD18 and the leukocyte and epithelial integrin-associated glycoprotein CD47 [
12,
19]. In comparison, rolling PMNs adhere firmly to endothelium via leukocyte α
Lβ
2-integrin CD11a/CD18 (leukocyte function-associated antigen-1, LFA-1) and CD11b/CD18, which bind to adhesion molecules such as intercellular adhesion molecule-1 (ICAM-1) [
20].
As the CP epithelium is a potential entry point for PMNs during bacterial meningitis, we here analyzed the transmigration process of PMNs through PCPEC monolayers after infection with S. suis. Analyzing the traversal route of PMN, we found that paracellular migration stops in front of a tight junction. Strikingly, the epithelial cell layer forms funnel-like structures protruding from the apical membrane, which are regularly found adjacent to PMNs. We propose a new mechanism of PMN transmigration in epithelial cells, whereby PMNs transmigrate through these funnels to finally overcome the epithelial barrier via a final transcellular step.
Methods
Bacterial strains and growth conditions
S. suis serotype 2 virulent strain 10 and its isogenic non-encapsulated mutant strain 10cpsΔEF, named strain 10 Δcps in this study, were kindly provided by H. Smith (DLO-Institute for Animal Science and Health, Lelystad, The Netherlands) [
21]. Bacteria were maintained and inoculated as previously described [
4]. Bacteria were washed twice in phosphate-buffered saline (PBS) (pH 7.3) and adjusted to an optical density at 600 nm (OD
600) of 0.65. This stock solution had approximately 2 × 10
8 colony-forming units (CFU)/ml and was further diluted in fresh culture medium without antibiotics for the experiments.
Preparation and cultivation of PCPECs on inverted Transwell filters
Epithelial cells from porcine CP were obtained by a modified preparation basically as described previously [
22]. For inverted cell cultures the cells were seeded on laminin- (Sigma, Deisenhofen, Germany) coated Transwell filters (pore diameter 3.0 μm, 0.33 cm
2; PET membrane, Falcon, BD, Le Pont De Claix, France), which were flipped over and placed in a medium-flooded 12-well plate as described recently [
4].
Measurement of transepithelial electrical resistance (TEER)
Confluence of PCPEC monolayers and barrier properties were documented by measuring TEER. TEER was measured using an epithelial tissue voltohmmeter (EVOM®, World Precision Instruments, Sarasota, FL, USA) and an STX-2 electrode system. PCPEC inverted cultures were used when TEER values reached more than 180 Ω × cm2. In PMN transmigration experiments, TEER was monitored over a range of 4 h. Resistance values of cells in medium alone were used as negative control values and stayed above 180 Ω × cm2 during all experiments.
Determination of paracellular permeability
As an independent measure of paracellular permeability of CP epithelium monolayers, the passage of Texas Red-labelled dextran (MW 3000; Sigma, Deisenhofen, Deutschland) across cell monolayers in the basolateral-to-apical direction was determined during PMN transmigration experiments. Texas Red-dextran (TR-dextran, 100 μg/ml) was loaded into the upper compartment during the incubation period. At indicated time intervals samples from the lower compartment were collected and fluorescence was measured in duplicates in a Tecan Infinite M200 Multiwell reader (Tecan, Switzerland). TEER and permeability measurements were performed with the same cultures as PMN transmigration.
Isolation of PMNs
For the PMN transmigration assays blood was taken from freshly slaughtered pigs at the abattoir. PMNs were isolated from non-coagulated citrate blood by Percoll density sedimentation according to the manufacturer's instructions (Biochrom, Berlin, Germany). Contaminating erythrocytes were lysed with NH4Cl on ice. PMNs were resuspended in culture medium at a cell density of 1 × 107/ml. For transmigration assays PMNs were loaded with the fluorochrome 2',7'-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein, acetomethyl ester (BCECF-AM; Molecular Probes, Eugene, OR, USA) according to the manufacturer's instructions.
Stimulation of PCPECs and PMNs
For transmigration experiments inverted PCPEC cultures were used when TEER had reached 180 Ω × cm2. PCPECs were stimulated with either TNFα from the apical and basolateral side (10 ng/ml) for 24 h or basolaterally (blood-side) infected with S. suis strain 10 or strain 10 Δcps [multiplicity of infection (MOI): 10] and hereafter incubated for 2 h at 37°C and 5% CO2. After the incubation period penicillin/streptomycin (100 U/ml/100 μg/ml) was added to the upper and lower compartment of the Transwell filter to inhibit further extracellular bacterial growth and therefore to prevent cytotoxic effects. PMN transmigration assays were performed after an additional 22 hours on the following day. In a second set of experiments we pre-incubated PMNs with S. suis strain 10 or strain 10 Δcps (MOI 10) for 1 h and hereafter performed with pre-stimulated PMN transepithelial migration assays in antibiotic-containing medium (to prevent further bacterial growth) as described below.
PMN transepithelial migration assay
For transepithelial migration assays BCECF-AM-loaded PMNs were added to the upper Transwell filter compartment (blood-side) of control, TNFα or
S. suis stimulated cells in a PMNs:PCPECs ratio of 10:1. As chemoattractant IL-8 (10 ng/ml) was used in indicated samples and added to the lower Transwell filter compartment (CSF-side) 30 min before starting the transmigration experiments. After 4 hours of transmigration the Transwell filter inserts were removed and the 24-well plates were centrifuged (5 min, 300 × g) to ensure that all PMNs are attached to the bottom of the wells. The supernatants were collected for permeability measurements. The PMNs were washed once with HBSS with Ca
2+/Mg
2+ and again centrifuged (5 min, 300 × g). Transmigrated PMNs were lysed by 1% Triton X-100 in PBS and quantified by fluorescence measurement with a Tecan 200 M Infinite Multiwell reader (Tecan, Switzerland) in relation to an internal standard. For antibody blocking experiments BCECF-AM loaded PMNs or PCPECs were pre-incubated with antibodies as indicated for 30 min at room temperature or at 37°C/5% CO
2, respectively. Hereafter, a transmigration assay was performed as described above. Antibodies specific to porcine epitopes (Table
1) were selected referring to the analyses of the "Third International Workshop on Swine Leukocyte Differentiation Antigens" [
23,
24]. Cross reactivity of anti-CD47 has been described by Shahein et al. [
25].
Table 1
Antibodies used for inhibition studies, flow cytometry and immunofluorescence.
IgG
1
, κ Isotype | MOPC-31C | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | BD Pharmingen (Heidelberg, Germany) |
IgG
2b
, κ Isotype | MPC-11 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | BD Pharmingen (Heidelberg, Germany) |
CD11a, IgG2b
| BL2F1 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC)l | BD Pharmingen (Heidelberg, Germany) |
CD11R3, IgG1
= CD11b
| 2F4/11 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | Serotec (Oxford, UK) |
CD18, IgG1
| PNK-I | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | BD Pharmingen (Heidelberg, Germany) or Serotec (Oxford, UK) |
CD47, IgG2b
| BRIC126 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | Serotec (Oxford, UK) |
CD49d, IgG1
| HP2/1 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | Serotec (Oxford, UK) |
CD49e, IgG1
| VC5 | Mouse | 20 μg/ml (inhibition) 1:10, 10 μl/test (FC) | BD Pharmingen (Heidelberg, Germany) |
Occludin
| | Rabbit | 1:250 (IF) 1 μg/ml | Zymed Laboratories (South San Franscisco, CA, USA) |
ZO-1
| | Rabbit | 1:250 (IF) 1 μg/ml | Zymed Laboratories (South San Franscisco, CA, USA) |
SWC3a-FITC
| 74-22-15 | Mouse | 1:10 (IF) 10 μl/Test | Southern Biotech (Birmingham, AL, USA) |
Secondary antibodies
|
Alexa fluor
®
594 goat anti chicken
| | Goat | 1:1000 (IF) | Molecular Probes (Eugene, OR, USA) |
Goat anti mouse-PE
| | Goat | 1:10, 10 μl/test (FC) | Southern Biotech (Birmingham, AL, USA) |
Flow Cytometry
Flow cytometric analyses was performed to test the specificity of antibodies, which were used in inhibition studies, and to investigate the expression patterns of PMN surface markers before and after transmigration. For this purpose, PMNs were stained with various antibodies according to manufacturer's instructions (Table
1). The PMN populations showed high and homogeneous integrin expression with regard to all tested integrins antibodies (data not shown). Integrin regulation before and after PMN transmigration was measured by analyzing the mean fluorescence intensities. To study the specific effect of IL-8, TNFα and
S. suis and the transmigration itself on the integrin expression, two different additional analyses were performed. In the first, PMNs were analyzed that had transmigrated towards an IL-8 gradient (10 ng/ml in the lower compartment) through control or
S. suis (MOI 10) or TNFα (10 ng/ml) stimulated PCPEC. In the second, PMNs were used that had not yet transmigrated, but which were stimulated in non-tissue culture-treated 24-well plates (employing 0, 6.7 or 10 ng/ml IL-8). After 4 h of transmigration or incubation with
S. suis, TNFα and IL-8, the PMNs were collected and divided into tubes. After centrifugation (5 min, 300 × g) the PMNs were incubated with the primary antibodies for 12 min. Then the cells were washed twice with cell culture medium and incubated with the secondary antibody (goat anti mouse-PE) for 7 min followed by two washing steps. Flow cytometry was performed using a FACScan (Becton Dickinson, USA) with appropriately set light scatter gates.
Immunofluorescence
Confluent PCPECs were grown on inverted Transwell filters, stimulated with
S. suis or TNFα and co-cultured with PMNs as described above. After 4 h of transmigration towards a gradient of IL-8, the cells were washed, fixed and permeabilized as described previously [
3]. Subsequently, the cells were washed with PBS and incubated overnight at 4°C with the primary antibodies (Table
1) to stain the TJ proteins. On the following day the cells were washed again, incubated for 60 min with the secondary antibody (Alexa fluor
® 594 goat anti-chicken), with Phalloidin Alexa fluor
® 660 for staining the actin cytoskeleton and with 4'-6-diamidino-2-phenylindole dihydrochloride (DAPI) (1:25.000) for staining nuclei. PMNs were labelled with the granulocyte-monocyte marker SWC3a-FITC (1:10 in PBS) (Southern Biotech, Birmingham, AL, USA) for 30 min. After washing the cells three times with PBS the filters were embedded in ProLongAntifadeReagent (Invitrogen, Karlsruhe, Germany). Images were acquired with Zeiss Apotome
® and Axiovision software (Carl Zeiss, Jena, Germany) using a 63×/1.4 objective lens. The image acquisition was carried out using the Zeiss scanning software Axiovison 4.6 and Axiovison module Inside 4D. Assays were performed in triplicates for each value and repeated at least four times.
Transmission electron microscopy (TEM)
PCPECs were grown on inverted Transwell filters, stimulated and co-cultured with PMNs as described above. After a 4-h coincubation period, the cells were washed once with culture medium, twice with Dulbecco's-PBS and hereafter fixed with 2% glutaraldehyde (Polyscience, Warrington, PA, USA) in D-PBS, pH 7.4, for 24 h at 4°C. The Transwell filter membranes were cut out of the insert and washed three times with D-PBS and post-fixed with 1% osmiumtetroxide. After another three washes with D-PBS the samples were dehydrated by a graded ethanol series (30%, 50%, 70%, 90%, 96% for 15 min each, 2 × 99% for 30 min each) and two washes with propylenoxide. During the 70% ethanol step of the graded ethanol series, the specimens were incubated in saturated uranyl acetate. After completion of dehydration, the preparations were embedded in Araldite 502 (Sigma-Aldrich) at 60°C for 48 h. Ultrathin sections were prepared on a Leica FCR Ultracut ultramicrotome and stained with lead citrate. Sections were examined using a Zeiss EM 10 electron microscope.
Scanning electron microscopy
Samples were fixed with 2.5% glutaraldehyde in cacodylate buffer, postfixed with 1% osmium tetroxide in phosphate-buffered saline, dehydrated in a graded series of ethanol and critical-point-dried using CO2. Finally, the samples were sputter-coated with a layer of 7 nm gold/palladium (Bal-Tec MED 010) and examined at 20 kV accelerating voltage in a Hitachi S-800 field emission scanning electron microscope.
Quantitative real-time PCR
After treatment with
S. suis [strain 10 and strain 10cpsΔEF (short: strain 10 Δcps)] or TNFα (10 ng/ml) for 2 and 4 h inverted PCPEC monolayers were washed with PBS and total cellular RNA was extracted using the RNeasy mini kit (Qiagen, Hilden, Germany) modified for inverted cell cultures. For every stimulus, three Transwell filters were used. Briefly, Transwell filter membranes were cut out of the insert and transferred into 48 wells. The cells on the filter membrane were lysed by resuspending them in 100 μl of RLT (ready to load) buffer. Lysed cells of every triplicate were collected and filled to 350 μl with RLT buffer and homogenized by QIAshredder (Qiagen, Hilden, Germany) followed by RNA extraction. Contaminating DNA was digested with the RNase-free DNase I (Roche, Mannheim, Germany) for 60 min at room temperature. DNase was inactivated by 1 min at 72°C. After spectrophotometrical determination of the RNA concentration using a NanoDrop (Thermo Scientific, Wilmington, USA) 145 - 300 ng of total RNA was reverse-transcribed with the SuperScript™ III First-Strand Synthesis System (Invitrogen, Karlruhe, Germany) for real-time PCR according to the manufacturer's instructions. Quantitative real-time PCR was performed with 2× Quanti Tect SYBR Green PCR Master Mix (Qiagen, Hilden, Germany) and the 7900HT Fast Real-Time PCR System (Applied Biosystems, Darmstadt, Germany). cDNA quantities were measured as critical threshold (C
T) values, which were then normalized using simultaneously measured β-actin levels (ΔC
T). Final ΔΔC
T values were obtained by comparing bacteria-stimulated or TNFα-treated cells with unstimulated cells using PCPECs from three different preparations. Primers used were previously described [
26] (Table
2). The specificity of PCR products was checked by agarose gel electrophoresis (data not shown).
Table 2
Primers used for quantitative real-time PCR
ICAM-1 | forward | CACAGGCCGCCACTAACAA |
ICAM-1 | reverse | GGTTCCATTGATCCAGGTCTT |
VCAM-1 | forward | GCACGAGCTTCCTGAGCACTT |
VCAM-1 | reverse | CTGTGTGACGAGGAAACAATG |
β-actin | forward | TCCAGAGGCGCTCTTCCA |
β-actin | reverse | CGCACTTCATGATCGAGTTGA |
Measurement of cell viability
Cell vitality was measured using the Live/Dead® Viability/Cytotoxicity Kit for mammalian cells (Molecular Probes, Göttingen, Germany) according to the manufacturer's instructions. The results were photodocumented by fluorescence microscopy. In all experiments no cytotoxicity was observed (data not shown).
Statistical analysis
All data are expressed as means ± standard deviation (SD) as indicated. TEER, dextran-flux, FACS analysis and real-time PCR data were analyzed by Student's t test respectively, PMN transmigration data by an analysis of variance (ANOVA) with repeated measurements and the method of compound symmetry was performed. P-values for post-hoc test were adjusted by the method of Tukey-Kramer. The procedure of MIXED of SAS was used. A P value of < 0.05 was considered significant. All assays were repeated at least three times. Texas Red-dextran flux data were expressed as percent of tracer in the basolateral compartment.
Discussion
One crucial step in the pathogenesis of bacterial meningitis after bacterial invasion into the CNS is the excessive infiltration of leukocytes into the cerebrospinal fluid leading to massive inflammation. To study PMN transmigration at the blood-CSF barrier, we used a recently established inverted Transwell filter model of PCPECs that enables investigation of leukocyte transmigration in the physiologically relevant basolateral-to-apical direction. [
4]. This model displayed robust barrier properties, which were not compromised by PMNs alone or the chemoatractant IL-8. The chemokine IL-8 (CXCL8) is known to interact with its cognate receptors CXCR1 and CXCR2. CXCR2 is the main receptor involved in neutrophil chemotaxis, leading to cell migration into the brain during injury, infection or disease [
30]. Interestingly, although bacterial infection alone did not influence barrier function, TEER and paracellular TR-dextran flux were significantly affected during the process of PMN traversal under bacterial stimulation, an effect that was already observed after treatment with TNFα alone. Nevertheless, the increased TR-dextran flux alone might not lead to paracellular transmigration of PMNs, since the observed compromise of tight junction function is still much lower than that of our own previously published results in the standard Transwell system after
S. suis infection [
3,
11]. Conflicting data exist as to whether PMN transmigration itself can lead to significant barrier disruption, and this seems to depend on the cells and the stimulus used [
27,
31‐
33].
The alteration of barrier function after PMN transmigration following infection with
S. suis or by TNFα stimulation was associated with an altered actin cytoskeleton and TJ morphology. Independently of the presence of PMNs, both stimulations led to stress fiber formation and an actin rearrangement within the cells, and a fuzzy and blurred distribution pattern of ZO-1 and occludin. In contrast, TEER of PCPECs was not significantly affected by
S. suis infection alone. Corresponding changes in PCPECs in immunofluorescence analyses of tight junction and cytoskeletal actin morphology were not as strong as described earlier by our group and Zeni and co-workers in the standard Transwell system, which might explain the lower effect on barrier function [
3,
26]. Interestingly, in the present study the TJ structure was specifically altered in areas where PMNs crossed the PCPEC monolayer between cells (as described in Figures
5,
6,
7,
8). Therefore, local changes in PCPEC morphology may correspond to minor or no TEER decrease. Changes in actin rearrangement may also not necessarily go along with a dramatic decrease of TEER. To maintain barrier function during paracellular transepithelial migration, close cell-cell contact and highly regulated mechanisms are necessary for opening and closing of the TJ [
34,
35].
Infection with the well-encapsulated
S. suis wild-type strain 10 and stimulation with the proinflammatory cytokine TNFα led to an augmented PMN traversal in both the presence and absence of IL-8. This goes in line with results that TNFα stimulation of HBMEC or renal proximal tubular cells leads to increased PMN transmigration [
27,
36]. Interestingly, the stimulation of PCPECs with
S. suis strain 10 Δcps caused no increase of PMN transmigration.
S. suis strain 10 Δcps may have induced cell signalling in PCPECs or PMNs that prevented PMN transmigration. Previous results of Chabot-Roy et al. have shown an increased PMN phagocytosis rate of
S. suis strain 10 Δcps compared to the wild-type, which may also influence the transmigration capacity of PMN [
37]. Further experiments are necessary to clarify these questions.
For transendothelial migration a paracellular route between adjacent cells has been postulated for a long time, but in the meanwhile the transcellular route directly through the endothelial cell body has been well documented [
14,
15]. In contrast, there is to date no evidence that PMNs take the transcellular route through epithelial cells [
12,
13]. The concert of experimental techniques applied in this study consistently suggests a general preference of neutrophils for transcellular transmigration pathways across PCPECs. By means of immunofluorescence and transmission and scanning electron microscopy analyzes we were indeed able to provide evidence for migratory events through one single cell. However, in determining the transmigration route of PMNs through PCPECs, we were confronted with the problem that neither our immunofluorescence nor electron microscopic images were absolutely unequivocal. If, for example, a leukocyte appears to lie inside a plexus epithelial cell, this could be interpreted simply as transcellular migration. However, the cell migrating along the intercellular cleft could be sectioned in a way that would allow us to interpret the position of the cell falsely as an intracellular one. Intracellular location in this context is defined as a PMN that is in a vacuolar structure and not in the cytoplasm. By definition, guest cell and host cell form a double membrane. However, the difficulty consists in determining whether the guest cell is only surrounded by the host cell in the section plane, while outside the section plane the cell could have contact with the extracellular space resembling an incomplete phagocytosis. We therefore compared orthogradely sectioned filters with filters sectioned nearly parallel through the apical surface to allow better interpretation of cellular details during transmigration. In every case the maintenance of the tight junctions was definitely observed by electron microscopy. Whether or not tight junctions in this system can be opened cannot be answered for principal reasons: we have currently no means to identify an opened junction. However, both
S. suis or TNFα stimulation led to a fuzzy and blurred distribution pattern of ZO-1 as well as stress fiber formation and actin rearrangement within the cells as described earlier by our group and Zeni and co-workers in the standard Transwell system that may still promote paracellular PMN transmigration [
3,
26]. In scanning electron microscopic images PMNs were regularly found directly on the apical surface of the monolayer and at a distance from the intercellular clefts. Our data are in line with very recent findings by von Wedel-Parlow et al. [
38], who demonstrated, in an in vitro model of the blood-brain barrier under inflammatory conditions, that PMNs preferentially migrate across primary cultured porcine brain capillary endothelial cells via the transcellular route.
Another striking observation in our experiments is that funnel-like structures were regularly found adjacent to PMNs protruding from the apical PCPEC membrane. We observed these funnel-like structures in stimulated as well as unstimulated PCPECs only in the presence of PMNs. The present funnel-like structures are apparently used by transmigratory PMNs, independent of their stimulatory status. However, we could not confirm the possibility that funnel-like structures are a
conditio sine qua non for the transmigration of PMNs. Although we cannot unequovically exclude a completely paracellular process, our data imply that tight junctions induce a stop of paracellular migration and lead to a deviation of the PMN travelling route to involve a transcellular step via the indented apical compartment of the epithelial cell. This mechanism is to some extent different from the suggested paracellular transmigration of PMNs across the CP after brain injury [
39]. In a mouse model it has been described that neutrophils of wild-type mice migrate predominantly paracellularly whereas CD11b/CD18-deficient mice use predominantly the transcellular pathway [
40]. The reason for this difference is related to PMNs crawling to cell junctions that are absent in CD11b/CD18-deficient PMNs. The increase in permeability was nearly identical for both forms of migration, because the endothelium forms a dome over the migrating neutrophil, which sealed the emerged gap and thereby limited alterations in vascular permeability independent of the migration route [
18]. It should be noted that the two ways of transmigration - paracellular and transcellular - are not mutually exclusive.
The exact routes and molecular mechanisms of PMN transmigration through epithelium are still not fully understood. For T84 intestinal epithelial cells important roles of CD11b/CD18, but also CD11b/CD18-independent mechanisms, have been described under specific inflammatory conditions [
31,
41]. Along these lines we observed that PMN transmigration through PCPECs is also CD11b/CD18-dependent. Interestingly, we found no evidence for a role of the integrin-associated glycoprotein CD47 during PMN migration through choroid plexus epithelium. In contrast, in intestinal epithelial cells CD47 plays an essential role in PMN transepithelial migration and in signalling events involving the signal regulatory protein α (SIRPα) [
12]. Other potential binding partners of PMNs on the surfaces of PCPECs are ICAM-1 for the β
2-integrins and VCAM-1 for the β
1-integrins. In cultured intestinal epithelial monolayers ICAM-1 is not involved in PMN transmigration and is not a counter receptor for CD11b/CD18 on the basolateral cell side of the cells [
42]. Instead, CD11b/CD18 promotes adherence of PMNs at the apical cell surface and, rather, binding to fucosylated glycoproteins at the basolateral cell side [
43,
44]. In contrast, the transendothelial migration of most leukocytes has been shown to be ICAM-1-, VCAM-1-, CD11a/CD18- and CD11b/CD18-dependent [
15]. Ultrastructural studies in an experimental autoimmune encephalomyelitis model revealed polar localization of ICAM-1, VCAM-1, and MAdCAM-1 on the apical surface of choroid plexus epithelial cells and their complete absence on the fenestrated endothelial cells within the choroid plexus parenchyma [
6].
Since contrary data exist about the role of ICAM-1 and VCAM-1 in PMN adhesion and transmigration as described above we were interested in their role on PCPECs after infection with
S. suis. We analyzed mRNA levels for these adhesion molecules in PCPECs under bacterial infection with
S. suis strain 10 and its isogenic mutant strain 10 Δcps and after TNFα stimulation, using quantitative real time-PCR. The mRNA expressions of ICAM-1 and VCAM-1 were considerably upregulated with VCAM-1 to a greater extent than ICAM-1 and with TNFα as the strongest stimulus. ICAM-1 upregulation was significantly more enhanced after infection with
S. suis strain 10 Δcps. Thus, we speculate that this effect may be due to the higher invasion capacity of this strain as recently demonstrated [
4]. These data indicate a certain involvement of the adhesion molecules on PCPECs after
S. suis infection. Further studies are necessary to determine the exact localization of these CAMs on PCPECs and their binding partners.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
TT and CW conceived and coordinated the study, and drafted the manuscript. CW, AS, UQ performed transmigration, immunofluorescence, qReal Time-PCR and FACS experiments. HW, LG, MAS, JB, UQ performed the electron microscopic studies. HJG, CS, HS have co-conceived the study and have been involved in drafting the manuscript. All authors have read and approved the final version of this manuscript.