High CRMP2 expression in peripheral T lymphocytes is associated with recruitment to the
brain during virus-induced neuroinflammation
Vuaillat C, Varrin-Doyer M, Bernard A, Sagardoy I, Cavagna S, N Chounlamountri, IA, Lafon M *,
Inserm, U842, Lyon, F-69372; Inserm, IFR19, Bron, F-69677 France; Université de Lyon, Lyon, F-
* Institut Pasteur, Viral Neuroimmunology, Paris, F-75015 France
Corresponding author: Pascale Giraudon, U842 Inserm, Rue Guillaume Paradin, Faculté de
Médecine Laënnec, F-69372 Lyon, Cedex, France
firstname.lastname@example.org; Fax 33 4 78 77 86 16
Funds: ARSEP, Internal grants from Institut Pasteur (ML) and INSERM
Collapsin Response Mediator Protein (CRMP)-2 is involved in T-cell polarization and migration. To
address the role of CRMP2 in neuroinflammation, we analyzed its involvement in lymphocyte
recruitment to the central nervous system in mouse infected with neurotropic and non-neurotropic
virus strains (RABV, CDV). A sub-population of early-activated CD69+CD3+ T lymphocytes highly
expressing CRMP2 (CRMP2hi) peaked in the blood, lymph nodes and brain of mice infected with
neurotropic viruses, and correlated with severity of disease. They displayed high migratory properties
reduced by CRMP2 blocking antibody. These data point out the potential use of CRMP2 as a
peripheral indicator of neuroinflammation.
Key words: neuroimmunology, neurovirology, neuroinflammatory disease, T-cell motility, rabies
virus, canine distemper virus
In acute viral encephalitis, the recruitment of immune cells to the central nervous system (CNS) plays
a crucial role in the outcome of this disease. Notably, T-lymphocytes are key players in the initiation of
specific intrathecal immune responses by directly destroying virus-infected cells, increasing the
phagocytotic activity of macrophages and stimulating the production of antibodies by B cells (Fujinami
et al., 2006; Salazar-Mather and Hokeness, 2006; Tishon et al., 2006). Thus, immune cells control the
spread of viruses and participate in viral clearance. However, T-lymphocytes can mediate
inappropriate inflammatory responses, producing multiple cytokines, chemokines, and reactive oxygen
species that contribute to severe neural tissue injury and CNS dysfunction (Carlson et al., 2006)
(Peterson and Fujinami, 2007; Galelli et al., 2000; Giraudon et al., 2004; Lane et al., 2000; Patterson
et al., 2003). Considering the importance of viral encephalitis in humans, the homing behavior and
functional status of immune cells that infiltrate the virus-infected brain need to be clarified.
Characterizing the mechanisms that favor the migration of T-lymphocytes to the CNS in response to
viral infection is of interest both for understanding the physiopathology of virus-induced
neuroinflammation and to propose new therapeutic approaches.
T-lymphocytes constantly move in and out of lymph nodes, circulating through the vasculature.
Probably following activation and as a result of chemotactic stimuli, T-cells acquire a polarized
morphology, leave the vasculature and start to crawl within neural tissue (Norman and Hickey, 2005;
Sanchez-Madrid and del Pozo, 1999; Westermann et al., 2005). Dynamic cytoskeletal remodeling is
fundamental to drive the rapid conversion of lymphocyte cytoarchitecture from a semi-rigid to a highly
deformable state needed for transendothelial migration and crawling within CNS parenchyma
(Vicente-Manzanares and Sanchez-Madrid, 2004). Thus, motile T-cells display a polarized
morphology, with two distinct T-cell compartments. At the advancing front, the leading edge is
enriched in chemokine receptors and acts as a sensor to the microenvironment. At the trailing edge,
the uropod is an adhesive and retractile structure required for T-cell migration (Barreiro et al., 2004).
Rapid reorganization of actin at the leading edge and the retraction of microtubule and vimentin
networks in the uropod during polarization and migration have been well described (Brown et al.,
2001; Mitchison and Cramer, 1996). Such a cytoskeletal reconfiguration requires a complex and still
not completely defined array of intracellular signaling pathways implicating second messengers,
several kinases and Rho family small GTPases (del Pozo et al., 1999; Vicente-Manzanares et al.,
2002). In search of molecules involved in cell motility and shared by the immune and nervous
systems, we recently identified Collapsin Response Mediator Protein-2 (CRMP2) in T-lymphocytes
(Vincent et al., 2005), a phosphoprotein described in neuron growth cone advance (Goshima et al,
1995) and neural cell migration via microtubule organization. We showed that CRMP2 redistributes in
the uropod in motile T-cells, binds cytoskeletal elements and is implicated in T-cell polarization and
migration (Vincent et al., 2005). A correlation between CRMP2 expression levels and cell migratory
rates was demonstrated by over-expression and knock-down experiments in primary T-lymphocytes.
The importance of CRMP2 in the immune response was emphasized by the observation that, in
humans suffering from virus-induced neuroinflammatory disease, peripheral activated immune cells
displayed high migratory rates associated with elevated CRMP2 expression (Vincent et al., 2005).
These observations led us to suspect a role for CRMP2 in immune cell trafficking and recruitment to
the inflamed CNS. In the present study, we evaluated the CRMP2 contribution in brain lymphocyte
recruitment by analyzing CRMP2 expression in peripheral and brain-infiltrating immune cells in mouse
models of infection using neurovirulent and non-neurovirulent strains of rabies virus (RABV) and
Canine distemper virus (CDV). After inoculation of the neurovirulent RABV strain Challenge virus
standard (CVS) in the muscle of the hind limb- to mimic the inoculation of this virus by an animal bite-
the virus enters nerve endings via the neuromuscular junction and then travels from one neuron to
another along the spinal cord before reaching the entire brain, causing severe neuronal dysfunctions
and fatal encephalitis (Lafon, 2004). Infection in the CNS triggers a strong innate immune response in
both neurons and astrocytes, resulting in inflammatory and chemo-attractive responses characterized
by IFN, IL-1, IL6, TNF CCL5, CXCL-10 expression capable of attracting T cells, B cells and
macrophages through the blood brain barrier (Camelo et al., 2001a; Baloul et al., 2004; Wang et al.,
2005; Roy et al., 2007). Inoculation of the Pasteur Virus (PV) RABV strain another neurovirulent RAV
strain causes only spinal cord and the bulb infection, resulting in abortive rabies: mice cleared the
CNS infection and survive despite paralytic sequel (6). By contrast, injection by the same route of the
non-neuronotropic vaccine RABV strain ERA does not cause infection of the CNS and animals remain
healthy. In the mouse model infected with CDV, intracerebral inoculation of a neurovirulent strain
results in acute encephalitis followed, in surviving animals, by the development of motor impairment
and metabolic disease (Bernard et al., 1999; Griffond et al., 2004; Verlaeten et al., 2001). CDV infects
the neurons of defined brain structures (Bernard et al., 1993) and provokes a brain inflammatory
reaction with the infiltration of T lymphocytes, the expression of the cytokines TNF, IL6, IFN and IL-4
associated with the disruption of the tissue inhibitor of the metalloprotease TIMP/ metalloprotease
MMP balance (Benscsik et al., 1996; Khuth et al., 2001) capable of modifying vascular permeability.
We show here that, in contrast of what was observed in mice infected with non-neurovirulent virus
strain, infection of the CNS with neuronotropic CDV and RABV enhanced both CRMP2 expression
and migratory rate of activated T lymphocytes. Interestingly, the presence in blood and lymph nodes of
lymphocytes expressing high CRMP2 level was associated with the intensity of recruitment of immune
cells into the brain of mice suffering from encephalitis and correlated with the severity of clinical score.
2- Material and Methods
Mice, rabies infection and assessment of clinical symptoms
Experiments were performed with 6-week-old female BALB/c mice from Janvier (St. Berthevin,
France). Mice were inoculated intramuscularly in both hind legs, with 1x107 infectious RABV particles
of different strains: Challenge Virus Standard, CVS-11, ADCC 959; ERA strain ADCC 332 (American
Type Cell Collection, ADCC, Rockville); Pasteur virus strain (PV). Disease progression was evaluated
by scoring mobility and mortality as follows: 0=normal mice, 1=ruffled fur, 2=loss of agility, 3=one
paralyzed hind leg, 4=two paralyzed hind legs, 5=total loss of mobility and 6=death. Clinical signs
were expressed as individual clinical sign values or as cumulative clinical signs by adding the
individual clinical signs of the group. At various times after infection, blood was taken from the retro-
orbital plexus and groups of two or three mice were perfused by intracardiac injection of 50 ml 0.1M
phosphate buffer pH 7.4. Spleen, lymph nodes, spinal cords and brains were removed and processed
to isolate mononuclear cells (8 infected mice and 8 controls analyzed). The presence of virus in the
brain was checked during the experiments (not shown).
Mice, CDV infection and clinical symptoms
Four-week-old female outbred SV129 mice (Harlan-France) were inoculated intracerebrally (10l) as
already reported (Bernard et al., 1983) with a neonatal mouse brain suspension containing 200 to
1000 PFU of the mice neuroadapted Onderstepoort CDV strain shown to be neuronotropic. Control
(sham-inoculated) mice were inoculated with brain homogenate from non-infected neonatal mice.
Concomitant with active virus replication in selective brain structures, acute meningoencephalitis
occurred in all mice and peaked on day 14 post-inoculation. At various times after virus inoculation,
cardiac perfusion was performed and blood, spleen, lymph nodes, spinal cords and brains were
removed and processed to isolate mononuclear cells on day 3 (n=3), day 7 (n=12), day 11 (n=5) and
day 14 (n=8) as described above (3 different experiments). The presence of virus in the infected brain
was checked during the experiments (not shown). Mice were housed in A2 animal colonies according
to European Economic Community (86/609/EEC) and French (Decree 87-848) animal care
regulations, in a temperature- and light-controlled room.
- Cell preparation
For peripheral blood mononuclear cell (PBMC) analysis, blood was collected at the retro-orbital
plexus. PBMCs were then isolated on Ficoll-Hypaque (Eurobio), washed three times with PBS and re-
suspended in PBS/2%FCS before immunocytochemistry and flow cytometry analysis. For lymph node
mononuclear cell analysis, mononuclear cells were collected from popliteal (for RABV model) or
cervical lymph nodes (for CDV model), following dissociation with a syringe plunger. Lymphocyte
enrichment was performed by a 30 min contact of the cell preparation on a plastic dish to remove
adherent phagocytes. Mononuclear cells infiltrating the CNS were collected using Percoll (Amersham
Biosciences) gradients, as previously described (Camelo et al., 2001b).
- Flow cytometry phenotyping
Mononuclear cells were washed in PBS and then incubated for 15 min at room temperature (RT) with
various combinations of fluorescence-conjugated mAbs (all from Beckman Coulter, unless otherwise
specified) as follows: phycoerythrin (PE)-labeled anti-CD3, PE-labeled anti-CD69 and PE-labeled anti-
CD11c. The stained cells were washed and fixed with 4% paraformaldehyde (PFA- Sigma) in PBS (pH
7.4) for 30 min at 4°C and washed again. Cells were then re-suspended in PBS at 4°C awaiting
analysis (if surface staining alone was required) or processed further for intracellular CRMP2 protein
detection. For the latter, cells were re-suspended in PBS containing 0.5% saponin (permeabilization
buffer -PB) and incubated for 30 min at RT with rabbit anti-peptide4/CRMP2 antibody, prepared
against a CRMP2 (Ricard et al., 2001). The cells were washed once with PBS 0.05% saponin and
then incubated in PBS 0.5% saponin with fluorescein isothiocyanate (FITC)-labeled goat F(ab’)2 anti-
rabbit for 20 min at RT. Finally, the cells were washed with PBS 0.05% saponin, re-suspended in PBS,
and analysed by flow cytometry on a Coulter EPICS XL (Beckman Coulter). Nonspecific staining was
excluded by isotypic control gating.
Analysis of infiltrating T cells was performed on brain slices using two different protocols of fixation. 1)
For Diaminobenzidine staining, mice were perfused with PBS then PBS containing 1% PFA. Removed
brains were then post-fixed in 1% PFA (1h, RT), immersed in 15% sucrose (overnight), frozen in
isopentane cooled to -60°C and stored at -80°C until use. Coronal brain sections (cryo microtome, 14
to 20m thick) from mice sacrificed on day 14 post-infection with CDV were collected on Superfrost
glass slides. Following treatment with reagent to eliminate nonspecific labeling due to endogenous
biotin (Vector kit), slides were incubated for 24h with anti-mouse CD4 monoclonal antibody (clone
H129.19, Pharmingen Int.). Samples were then sequentially washed and incubated with biotinylated
anti-rabbit or anti-mice IgG antibodies (2h, RT), avidin-biotin peroxidase complex (VectaStain, 1h RT)
and 3.3‘-diaminobenzidine. 2) For immunofluorescence analysis, brain were removed and immediately
frozen in isopentane cooled to -60°C and stored at -80°C until coronal sections and use. Fixation in
absolute ethanol (10 min, RT) was required for CD4 and CD8 detection while acetone fixation (10 min,
4°C) was required for CRMP2 and TCR beta chain detection. Following 1h incubation with a blocking
solution (PBS, 0,3% Triton, 1% BSA, 1% normal goat serum) to eliminate nonspecific labeling, slides
were incubated with anti-mouse CD4 and CD8 (clone H129.19 and clone53-6.7, Pharmingen Int.)
monoclonal antibodies or polyclonal anti-peptide4/CRMP2 antibody (Ricard et al., 2001). (2h, 37°C),
then with anti-rat and anti-rabbit IgG Alexa 488 antibodies (2h, RT) (Molecular Probes, Inc.). T-cell
receptor was detected using a FITC anti TCR beta chain (clone H57-597, BD Pharmingen). Omission
of the primary or secondary antibody resulted in no signal.
- Transmigration assay
Transmigration of lymphocytes, isolated from lymph nodes as reported above, was performed both in
micro-Transwell system as described (Vincent et al., 2005) and in organotypic culture of mice brain.
For ex vivo transmigration on mice CNS, hippocampus cultures were prepared as follows. Hippocampi
from postnatal (P7) C57BL6 mouse were dissected and placed immediately in cold Gey’s balanced
solution supplement with glucose (6,5mg/ml). Four hundred micrometer slices were cut
perpendicularly to the septotemporal axis of the hippocampus using a McIllwain tissue chopper. Slices
were carefully trimmed for excess tissue, and 6 slices were placed immediately on 30 mm semi-
permeable membrane inserts (Millicell-CM, Millipore) in a 6-well plate containing 1 ml of culture
medium in each well. The culture medium consisted of 50% Minimum Essential Medium (Gibco), 25%
Hank’s balanced salt solution, 25% heat-inactived horse serum (Gibco), 1% L-glutamine 200Mm
(Gibco) and 6,5 mg/ml D-glucose. Plates were incubated at 37°C and 5% CO2. The culture medium
was exchanged twice a week. Lymphocytes (1.106 cells per slice) stained ex vivo using the vital
fluorochrome carboxyfluoroscein succinimidyl ester/CFSE (1mM, 5 min, 37°C) were spotted closely to
hippocampus slices. Following a 18h contact at 37°C, slices were extensively washed with D-MEM,
fixed with ethanol (10 min, 4°C), incubated with 4’, 6’-diamidino2-phenylindol/DAPI for nucleus
staining. Numbers of infiltrating lymphocytes were counted under fluorescence microscopy.
For ex vivo transmigration in micro-Transwell system (Boyden chamber, Costar, 3 m-diameter pore
size membrane coated with fetal calf serum, in triplicate), lymphocyte preparations (4.105 cells/well)
were added in the upper chambers and chemokines in the lower compartments: CXCL12 (10ng/ml),
CCL5 (100ng/ml), CCL2 and CXCL10 (20ng/ml). Transwell were incubated at 37°C for 1.5-2h. In
blockade experiments, anti-pep4 antibody CRMP2 blocking antibody (3 mg/mL) was added to the
lymphocyte preparation for 30 min at 37°C just before transmigration assay as already reported
(Vincent et al., 2005). Migratory cells in the lower chambers or in hippocampus slices were counted
under the microscope. Data were expressed as total number of migratory T cells.
Statistical significance in comparing two means was tested with the unpaired Student’s t test, p
values<0.05 were considered significant. In the migration test, the number of migratory lymphocytes
was counted by light microscopy (15-20 microscope fields per condition -2 or 3 independent
experiments) and data expressed as mean number of migratory lymphocytes per field.
Neuronotropic RABV and CDV infections trigger the stimulation of CRMP2+ cell populations in
Mouse models of virus-induced encephalitis, RABV and CDV were developed using two different
strains of mice, C57B6 and SV129. Preliminary analysis indicated that the basal expression of CRMP2
in immune cells in healthy C57B6 and SV129 mice was different. While in healthy BALBC mice, less
than 20 % of PBMC and lymph node mononuclear cells expressed CRMP2, virtually all PBMC and
lymph node mononuclear cells of SV129 mice expressed CRMP2 (98.6±0.7% positive cells). To
investigate how RABV and CDV infections regulate CRMP2 in these two strains of mice, CRMP2
expression was investigated by cytofluorimetry in PBMC and lymph nodes mononuclear cells of
BALB/C and SV129 mice during the course of RABV or CVD infection, respectively.
BALB/C mice were infected with three different types of laboratory RABV strains: CVS, PV and ERA.
These strains are different in respect of neurotropism and issue of the disease. ERA is a vaccine virus
strain, which has lost the property to infect the nervous system after peripheral infection in BALB/c
mice. PV is a neuronotropic virus causing only abortive rabies (Galelli et al., 2000) as mice are
paralyzed but survived. In contrast, CVS is a neuronotropic virus strain causing a fatal myelo-
encephalitis (Baloul et al., 2004). PBMC and lymph nodes were sampled at different times (days 1, 7
and 9) after intramuscular injection with RABV or injection of medium only (non-infected mice). After
CVS inoculation, infection did not significantly modify the % of CD3+ cells among the PBMC of BALB/c
mice during the time period (days 1, 7 and 9) of the infection (Figure 1-A). In contrast, the percentage
of cells expressing CD69, this molecule being taken as a marker of early activation (Hamann et al.,
1993), increased significantly on day 7 before declining thereafter, suggesting that infection with CVS
triggered the transient activation of PBMC (Figure 1-A). In parallel, a drastic increase in CRMP2+ cells
among the PBMC of CVS-infected animals occurred at day 7 (78% of the PBMC in infected animals
expressed CRMP2 compared to less than 20% in non-infected animals) (Figure 1A). This increase
was still high (70%) on the 9th day of infection. The CRMP2+ cells recruited among the CD3+ T
lymphocytes and activated CD69+ cell subsets, as shown by double-staining analysis (80% of
CRMP2+ cells from CVS infected animals at day 7 were CD3+ and 20% were CD69+) (Figure 1-B).
Virtually all CD69+ cells (98.7±1.2%) triggered by the infection at day 7 and most CD3+ T cells
(75.4±2.8%) expressed CRMP2. In addition, the frequency of T lymphocytes expressing CRMP2
increased with time (20% on day 1, 70-80% on days 7 and 9, respectively (Figure 1-C). Similar
patterns were obtained with popliteal lymph nodes (data not shown). After infection with PV, a
neuronotropic RABV strain causing abortive rabies, similar expansion of CRMP2+ cells was observed
in PBMC than those observed after CVS infection (Figure 1D). In contrast, after infection with the non-
neuronotropic RABV strain ERA, the number of cells expressing CRMP2 remained unchanged in the
course of the infection and not different from those observed in non-infected mice. However,
expansion of CD69 positive cells was similar in ERA to those observed in PV infected mice,
suggesting that the early steps of T cell activation have occurred after ERA infection as they have
occurred after PV and CVS infection. This suggests that CRMP2 expression by lymphocytes in
periphery is associated with RABV neurotropism.
Altogether, these data indicated that neurotropic strains of RABV triggered CRMP2 expression in
peripheral T lymphocytes and that most of the early-activated cells (CD69+), which increase in number
through infection, also expressed CRMP2.
CRMP2 expression was then evaluated in a model of mice infected by intracerebral injection of a
mice-adapted neurovirulent CDV strain previously shown to infect neurons (Bernard et al., 1993). In
CDV infected SV129 mice, PBMC and lymph nodes were sampled at different times (day 3, 7, 11 and
14) following injection of neuro-adapted virus (infected-mice) or brain suspension (sham-inoculated
mice). At first, a significant decline was detected in the CD3+ lymphocyte frequency in PBMC on day 7
post virus-inoculation (47.2% versus 63.3% in sham mice), probably due to the immunosuppressive
effect of CDV (Figure 2-A). Nevertheless and similarly to what was observed in RABV infected mice,
the frequency of CD69+ cells among PBMC increased (10.5% versus 5.2% in sham mice on day 7
post-inoculation). When CRMP2 expression was analyzed in terms of CRMP2+ cell frequency (% of
positive cells in the total population), CDV infection did not modify the % of CRMP2 expressing cells
(≈98% positive cells, not shown). In contrast, when the CRMP2 level per cell (mean fluorescence
intensity= MIF) was analyzed in double stained cells (Figure 2-B), CRMP2 expression showed an
increase in MIF in CD3+ and CD69+ cells of CDV-infected mice, reaching significance on day 7 post-
inoculation. Focusing on the CD69+CRMP2+ subpopulation (Figure 2-C), cytofluorimetry analysis
indicated that CDV infection triggered the expansion of a discrete population characterized by a strong
CRMP2 and CD69 expression. This cell subset, termed CRMP2hi, was recruited preferentially among
the activated cell population CD69+, was characterized by an elevated CRMP2 expression (MIF for
CRMP2hi cell population= 26-45 vs 16-20 for non-CRMP2hi cells) and constituted 57±13% of the
CD69+CRMP2+ cell population and so, probably participated to the elevated mean fluorescence
intensity detected in CD69+ activated cells of infected mice. CRMP2hi cells were specifically detected
in PBMC and lymph node of the infected animals, at the peak of encephalitis, namely on day 7 (in 9
out 12 infected mice analyzed) and on day 11 post-inoculation (in 4 out 5 infected mice analyzed)
(Figure 2-D). Thus, even under conditions of CRMP2 constitutive expression, CDV infection increased
CRMP2 level in a subset of activated cells in blood and lymph nodes.
In conclusion, analysis of CRMP2 expression in peripheral immune cells performed in different models
of viral infection indicated that neurotropic strains of CDV and RABV specifically increased the level of
CRMP2 in circulating lymphocytes. Depending on the constitutive level of CRMP2 expression, neural
infection triggered the expansion of CRMP2+ or CRMP2hi cell populations. These CRMP2+ cells
recruited mainly among CD3+ T lymphocytes, notably in CD69+ early-activated cells.
Selective recruitment of CRMP2+CD3 T cells in neurovirulent RABV and CDV infected CNS
Assuming that elevated CRMP2 expression may favor T cell motility and transmigration to inflamed
tissue, we analyzed whether CRMP2+ cells expansion in the periphery after RABV or CDV inoculation
correlated with trafficking to the CNS as the viral infection progressed in this tissue. Mononuclear cells
were isolated from the CNS using Percoll gradients and phenotyped by cytofluorimetry, at different
times after viral inoculation. In the RABV infection model with the neurovirulent strain CVS we focused
our analysis on CD3+ T cells, which selectively migrate into the CNS of infected mice (Figure 3-A). In
the absence of infection and also in the first 5 days of CVS infection, CD3+ T cells could only be
detected in minimal amounts (10-15% of the Percoll-collected cells), whereas later in the infection,
CD3+ T cells constituted 30% (day 8) and up to 60% (day 12) of the migratory cells. Double staining
analysis (CRMP2 and CD3) showed a recruitment of CRMP2+CD3+ T cells in the CNS on day 8 (27%
of the Percoll-collected cells). This percentage was unchanged on day 12 (Figure 3-B). The
combination of the CRMP2+ Percoll collected cell population was further investigated by double
staining of CRMP2 with markers chosen to roughly identify B-lymphocytes (B220), T-lymphocytes
(CD3), monocyte/macrophages and microglia (CD11b), NK and dentritic cells (CD11c), on days 0, 5, 8
and 12 post-infection. At each time point, the total % of CRMP2+ cells was taken as 100 and the
frequencies of B220, CD11C, CD11b and CD3 were calculated (Figure 3-C). Under non-infected
conditions, CRMP2+ cells could be sorted into CD11b (80%) and B220 (20%) in the brain. CVS
infection modified these ratio, inducing a drastic increase in CD3+CRMP2+ cells (38% and 62% are T
lymphocytes on days 8 and 12, respectively), a decrease of B220+CRMP2+ cells and the appearance
of a CD11c+CRMP2+ cell population accounting for a small percentage of the CRMP2+ population
(peak 12% on day 8). CD11b antigen, a marker of both macrophages and microglia, could not
differentiate the resident microglia trapped in the Percoll preparation from infiltrated immune
Cytofluorimetry analysis of Percoll-collected cells was further performed in the mouse model of CDV
infection on days 7 and 14 post-infection. Compared to non-infected conditions, the frequency of
CD3+ T-lymphocytes and early-activated CD69+ cells detected in the brain infiltrate was significantly
enhanced under CDV infection (reaching significance on day 14) (Figure 4-A), indicating that in the
course of CNS infection, inflamed brain attracts immune cells. Co-detection of CRMP2 with CD69 and
CD3 antigens on migratory cells showed that almost all expressed high CRMP2 levels, thus behaving
like CRMP2high cells (illustration of one infected-brain and sham-inoculated brains in Figure 4-B).
Immunocytochemistry performed on brain slices from mice examined on day 14 post-inoculation
confirmed that CRMP2+ cells mainly infiltrated the infected neural tissue (Figure 5-A) and showed the
accumulation of CD4+ T lymphocytes in the parenchyma of several brain structures previously
demonstrated to replicate CDV, including hippocampus and thalamus (Bernard et al., 1993) (Figure 5-
B). T lymphocytes, in particular, were present in brain parenchyma, at the proximity of structures
involved in blood-CNS exchange: the velum interpositum of meninges, the choroid plexuses in
ventricle. Immunofluorescence assay detected CD4+ and CD8+ cells in ethanol fixed neural tissue,
confirming the presence of T lymphocytes in the brain infiltrate of CDV infected mice (Figure 5-C). Co-
detection of CRMP2 and CD4 or CD8 antigens on brain slices could not be performed (different tissue
fixation requirement). However, CRMP2 and TCR beta chain positive cells were co-detected in same
brain areas (acetone fixation) (Figure 5C) confirming the presence of CRMP2+ lymphocytes in
Altogether, these observations supported the idea that infection of the CNS with a neurotropic virus
modifies CRMP2 expression in activated immune cells in the periphery, and favored the recruitment,
mainly of CRMP2+ T-lymphocytes, to the infected brain.
Enhanced migratory rates for peripheral immune cells displaying elevated CRMP2 expression
in CDV-infected mice
As brain infiltration may result from the elevated migratory rate and/or enhanced survival of activated
immune cells within neural tissue, the effect of CNS viral infection on the motility of peripheral immune
cells was investigated ex vivo on T-lymphocytes isolated from lymph nodes on day 7 post inoculation
(3 infected- and 3 non-infected mice tested). Immune cells from lymph nodes were first analyzed for
CRMP2 expression. Cytofluorimetry detected a clear increase in CRMP2 expression level (Figure 6-A)
and the presence of CRMP2hi cells (38.3 to 77.2% of the total CD69+CRMP2+ cells) in CDV-infected
mice. The migratory rate was further analyzed, as previously reported (Vincent et al., 2005), in a
Transwell chamber system allowing the determination of the number of cells migrating towards a
gradient of chemokines known to be present in virus-infected inflamed brain (CXCL12, CCL5,
CXCL10) (Patterson et al., 2003). As shown in Figure 6-B, lymphocytes from CDV-infected mice
displayed a greater migratory rate compared to those isolated from non-infected mice (89,1±4 versus
43,7±17 migrating cells). In blockade experiment using an anti-CRMP2 antibody previously shown to
lower T cell migration, the migratory rate was reduced by 30% and 10% for lymphocytes of CDV- and
sham-inoculated mice, respectively. The ability for lymphocyte to migrate to CNS tissue was further
examined ex vivo using mouse hippocampus maintained in organotypic culture (Figure 6-C). Brain
slices were either non-infected or infected with CDV for three days before migration assay.
Lymphocytes from lymph nodes of CDV-infected and sham-inoculated mice were stained with vital dye
CFSE (1M) then spotted close to CDV- and non-infected brain slices. Following 18h incubation,
migrating lymphocytes were detected on brain slices (Figure 6-C). Compared to lymphocytes of sham-
inoculated mice, lymphocytes of CDV-infected mice greatly invaded the CDV-infected brain slices
(Figure 6-C and graph in 6-D). Following treatment with anti-CRMP2 antibody, lymphocytes of CDV-
and sham-inoculated mice displayed similar migratory rates, demonstrating the role of CRMP2 in the
elevated ability to traffic to neural tissue displayed by CDV-infected mice lymphocytes.
All these experiments indicated that virus-induced elevated CRMP2 expression conferred to
lymphocytes a greater ability to migrate into neural tissue, thus increasing the potential of T cells to
target infected/inflamed brain.
Presence of peripheral CRMP2hi cells correlates with the severity of clinical scores in mice
suffering from RABV encephalitis
In contrast to mice infected with CDV, which develop abrupt signs of encephalitis, RABV-infected mice
displayed progressive and well-scored clinical signs (Camelo et al., 2001b), allowing a correlative
study between clinical score and CRMP2 expression levels in the periphery. After intra-muscular
injection with the highly neurotropic strain of RABV, CVS-1, BALB/c mice develop fatal acute
encephalitis. The first symptoms (ruffled fur, muscle stiffness) detected at day 6 post-infection (pi)
were quickly followed by paralysis of the hind limbs and then forelimbs, hunch back, severe loss of
weight and prostration followed by death (day 13 pi). A representative evolution of the cumulative
scores (addition of individual scores of the 8 animals of the group) after RABV injection is presented in
Figure 7-A. To analyze the time course of CRMP2 expression in the peripheral immune cells, together
with the clinical score, RABV infected mice were bled and sacrificed at different times post infection
and individual clinical scores at the time of sacrifice were recorded. CRMP2 expression levels in
PBMC and lymph node cells were then evaluated by cytofluorimetry. Similar to the CDV infection
model, CRMP2hi cells were detected in infected mice, as shown by the high CRMP2 expression in the
CD3+ T lymphocyte population of infected compared to non-infected mice (Figure 7-B). When
CRMP2hi cells were plotted against the individual clinical score displayed by encephalitic mice (Figure
7-C), they appeared preferentially in sick mice displaying high clinical scores (>3-5). Indeed, a
significant correlation between the cumulative clinical scores displayed by RABV-infected mice and the
frequency of CRMP2+ cells in blood (R2= 0.69) was detected using a logarithmic representation of
non-linear regression analysis (Figure 7-D). After infection with the neuronotropic RABV strain, PV,
causing abortive RABV and sustained infiltration of T cells into the CNS (Galelli et al., 2000), the
correlation index was 0,78. This strongly suggested the participation of CRMP2hi cells in the
deleterious neuroinflammatory reaction induced in the infected CNS, and indicated CRMP2 expression
levels in PBMC as a potential peripheral marker of virus-induced neuroinflammation.
Collectively, our data revealed an association between clinical features and the presence, both in the
periphery and in the brain of mice suffering from virus-induced encephalitis, of activated T-
lymphocytes displaying high CRMP2 levels.
In the present study, we showed that brain infection by neurovirulent strains of two viruses, either
RABV or CDV, causing fatal encephalitis, triggers CRMP2 expression in activated T lymphocytes.
These cells of increased mobility find their way to the infected brain. CRMP2 expression by T cells
seems to be a hallmark of infection/inflammation of the brain, since it was not observed after infection
with non-neurovirulent RABV infection. Interestingly, the kinetics of CRMP2 expression in the
periphery reflected the severity of the brain disease.
Following intramuscular (IM) inoculation, neurovirulent strains of RABV induced local (popliteal lymph
node) and systemic (spleen) proliferative and cytotoxic responses, with notably cytokine secreting T
cells and specific RABV antibodies (Irwin et al., 1999). We showed here that, as early as 5 days post
infection, the T cells of lymph nodes and blood were activated (CD69 expression) and expressed
CRMP2. After intracerebral (IC) injection of CDV, a similar immune response was obtained,
suggesting that viral inoculation by the IM and IC routes both activated the systemic immune
response. This can be surprising as the nervous system is an immuno specialized organ where
directly injected viruses evade an immune response, probably through the installation of an
immunosuppressive milieu (Galea et al., 2007). However, this is not the case when antigens or virus
particles are delivered by intrathecal immunization (Cserr et al., 1992; Cserr and Knopf, 1992;
Harling-Berg et al., 1991; Panda et al., 1965; Stevenson et al., 1997; Shin et al., 2006). It is likely that
after IC inoculation of the CDV, virus induced leakage of neural cells and redirected viral antigens to
the cerebrospinal fluid, the meninges and finally to the cervical lymph nodes (Hatterer et al., 2006).
Thus, despite the tropism of neurovirulent RABV and CDV to the CNS, initial viral entry into the body
by the IM and IC routes triggered a strong immune response in the periphery and the induction of
CRMP2 positive cells in secondary lymphoid organs that circulate further into the blood. After IM
infection with a non-neurovirulent RABV strain, CRMP2 expansion was no observed. This does not
result from an absence of immune activation since expansion of CD69+ cells was similar to those
observed after neurovirulent RABV injection, nor from the route of injection since both groups of mice
were injected IM. It may result of the capacity of the non neurovirulent RABV strain ERA to infect bone
marrow macrophages (Ray et al., 1995) a property which is not shared by neurovirulent RABV strains
which are strictly neuronotropic. This point would require further investigation.
How could CRMP2 integrate the multiple pathways involved in the homing and accumulation of T
lymphocytes in the infected CNS? Access to the brain is strictly controlled by blood-brain interfaces,
blood vessels and the choroid plexus, and accumulation in neural tissue may result from the enhanced
entry/reduced exit rates of T cells from the brain (Kleine and Benes, 2006; Bechmann et al., 2007;
Engelhardt and Ransohoff, 2005). The activation process has a dramatic effect on this pathway, in
particular by modulating the expression of the adhesion molecules, metalloproteases and chemokine
receptors required for lymphocyte transendothelial migration and crawling in inflamed tissue (Barreiro
et al., 2002; del Pozo et al., 1995; Giraudon et al., 1997; Lane et al., 2000; Strazielle et al., 2003). As a
result, more resting T-cells are found in lymphoid organs while activated T-cells mainly target non-
lymphoid organs, including brain (Westermann et al., 2005). In RABV- and CDV-infected mice, the
presence of early-activated cells and T lymphocytes displaying high CRMP2 levels, associated with
high migratory rates for T lymphocytes, suggests that the activation process may modulate T cell
motility through CRMP2 modulation. Our recent observations support this hypothesis. An association
between CRMP2 expression levels, T-cell activation status and in vitro cell migratory rates was
observed in patients suffering from virus-induced neuroinflammatory disease (Vincent et al., 2005). In
addition, T-cell activation through CD3-CD28 stimulation clearly enhanced CRMP2 levels per cell
(unpublished data). How CRMP2 operates in activated T-lymphocytes to drive their migration to the
CNS is currently under investigation. The role of CRMP2 in the CNS provides some clues to
understanding its function in the immune system. By its ability to associate with cytoskeletal elements,
CRMP2 participates in the cytoskeletal reorganization required for neuronal cell reshaping and
migration under semaphorin signals through the promotion of microtubule extensions (Fukata et al.,
2002), a process inactivated by CRMP2 phosphorylation (Yoshimura et al., 2005; Cole et al., 2006;
Uchida et al., 2005). In immune cells, we have shown the ability of CRMP2 to bind vimentin and
tubulin (Vincent et al., 2005) thus to participate in lymphocyte reshaping. Inflammatory molecules,
notably chemokines, could act as modulators of CRMP2 function, promoting T cell reshaping and
motility in neuroinflammatory situations. This idea is supported by the role and distribution of
ckemokines in the inflamed/infected brain. Indeed, chemokines are required for lymphocyte
chemotaxis through actomyosin/ microtubule organization (Vicente-Manzanares and Sanchez-Madrid,
2004; Rey et al., 2002). Chemokines regulate cell entry into the brain microvasculature, direct
navigation across the Virchow-Robin perivascular space and facilitate T cell crawling in brain
parenchyma (Kleine and Benes, 2006; Lane et al., 2000; Lee et al., 2004; Liu et al., 2000; Ransohoff
et al., 2003). Chemokines are expressed at high levels in the virus infected CNS, as it is the case after
RABV infection (Baloul and Lafon, 2003; Roy et al., 2007; Wang et al., 2005).
In several infection models, the infiltration of activated T-lymphocytes into the brain is a well-
characterized parameter linked to the development of encephalitis and neurodegeneration (Irani and
Griffin, 1991; Khuth et al., 2001; Marten et al., 2003). A priming process initiated by a low number of
infiltrating encephalitogenic T-lymphocytes could set the stage for a second greater wave of invading
effector cells, as observed in multiple sclerosis patients and associated animal models (Engelhardt
and Ransohoff, 2005; Norman and Hickey, 2005). The subsequent induction of a shift towards a pro-
inflammatory CNS microenvironment may have beneficial/unfavorable consequences, leading to viral
clearance but promoting immune-related damage to neural cells (Fujinami et al., 2006; Griffond et al.,
2004). Whether or not T cells contribute to immunopathology in RABV is still controversial and seems
to depend upon strain pathogenicity (Camelo et al., 2001b; Galelli et al., 2000; Hooper, 2005; Lafon,
2005). In CDV-infected mice, clinical and histological changes in the CNS do not develop immediately
upon viral inoculation but only after a latency period, coinciding with T lymphocyte infiltration (Khuth et
al., 2001). This suggests that infiltrating immune cells are important effectors of the intrathecal
deleterious process translated in clinical manifestations.
The association we observed between CRMP2 expression in peripheral immune cells, in particular the
presence of CRMP2hi cells, and the clinical scores displayed by encephalitic infected mice and the
fact this is restricted to the infection of brain (not seen after the non neurovirulent RABV infection)
suggest that CRMP2 could be used as a cell marker to isolate and profile the cells with the potential to
infiltrate the infected CNS. This could help in decrypting the association of migrating lymphocytes with
the complex process leading to lethal encephalitis or recovery. Bearing early-activated T-lymphocyte
and dendritic cell markers, CRMP2hi cells could comprise active cytolytic CD8+ T-lymphocytes,
regulatory CD4+ T cells and antigen-presenting cells, each displaying their own role. They could
therefore be crucial either in controlling the immune response to pathogens and virus clearance or in
preventing reactivity to self-antigens, thus reducing the autoimmune reaction within the CNS (Serafini
et al., 2004; Serafini et al., 2006; Esiri, 2007; Lauterbach et al., 2006; Ramakrishna et al., 2006).
Detailed analysis of selected CRMP2hi cells along the course of viral infection and disease would
bring information on the nature and function of cells sequentially recruited to the inflamed CNS.
To conclude, the present work supports the idea that viral infection of the CNS enhances the ability of
activated T-lymphocytes to switch on the intracellular machinery that recruits the cytoskeletal
reorganizer CRMP2 and triggers cell migration. This also identifies CRMP2 molecule as a potential
marker of activated and motile T-lymphocytes generated during viral infection and potentially recruited
in the inflamed CNS.
Baloul, L., Camelo, S., Lafon, M., 2004. Up-regulation of Fas ligand (FasL) in the central nervous
system: a mechanism of immune evasion by rabies virus. J Neurovirol 10, 372-382.
Baloul, L., Lafon, M., 2003. Apoptosis and rabies virus neuroinvasion. Biochimie 85, 777-788.
Barreiro, O., Vicente-Manzanares, M., Urzainqui, A., Yanez-Mo, M., Sanchez-Madrid, F., 2004.
Interactive protrusive structures during leukocyte adhesion and transendothelial migration.
Front Biosci 9, 1849-1863.
Barreiro, O., Yanez-Mo, M., Serrador, J.M., Montoya, M.C., Vicente-Manzanares, M., Tejedor, R.,
Furthmayr, H., Sanchez-Madrid, F., 2002. Dynamic interaction of VCAM-1 and ICAM-1 with
moesin and ezrin in a novel endothelial docking structure for adherent leukocytes. J Cell Biol
Bechmann, I., Galea, I., Perry, V.H., 2007. What is the blood-brain barrier (not)? Trends Immunol 28,
Bencsik, A., Malcus, C., Akaoka, H., Giraudon, P., Belin, M.F., Bernard, A., 1996. Selective induction
of cytokines in mouse brain infected with canine distemper virus: structural, cellular and
temporal expression. J Neuroimmunol 65, 1-9.
Bernard, A., Cohen, R., Khuth, S.T., Vedrine, B., Verlaeten, O., Akaoka, H., Giraudon, P., Belin, M.F.,
1999. Alteration of the leptin network in late morbid obesity induced in mice by brain infection
with canine distemper virus. J Virol 73, 7317-7327.
Bernard, A., Fevre-Montange, M., Bencsik, A., Giraudon, P., Wild, T.F., Confavreux, C., Belin, M.F.,
1993. Brain structures selectively targeted by canine distemper virus in a mouse model
infection. J Neuropathol Exp Neurol 52, 471-480.
Bretin, S., Reibel, S., Charrier, E., Maus-Moatti, M., Auvergnon, N., Thevenoux, A., Glowinski, J.,
Rogemond, V., Premont, J., Honnorat, J., Gauchy, C., 2005. Differential expression of
CRMP1, CRMP2A, CRMP2B, and CRMP5 in axons or dendrites of distinct neurons in the
mouse brain. J Comp Neurol 486, 1-17.
Brown, M.J., Hallam, J.A., Colucci-Guyon, E., Shaw, S., 2001. Rigidity of circulating lymphocytes is
primarily conferred by vimentin intermediate filaments. J Immunol 166, 6640-6646.
Camelo, S., Castellanos, J., Lafage, M., Lafon, M., 2001a. Rabies virus ocular disease: T-cell-
dependent protection is under the control of signaling by the p55 tumor necrosis factor alpha
receptor, p55TNFR. J Virol 75, 3427-3434.
Camelo, S., Lafage, M., Galelli, A., Lafon, M., 2001b. Selective role for the p55 Kd TNF-alpha receptor
in immune unresponsiveness induced by an acute viral encephalitis. J Neuroimmunol 113, 95-
Carlson, N.G., Hill, K.E., Tsunoda, I., Fujinami, R.S., Rose, J.W., 2006. The pathologic role for COX-2
in apoptotic oligodendrocytes in virus induced demyelinating disease: implications for multiple
sclerosis. J Neuroimmunol 174, 21-31.
Charrier, E., Reibel, S., Rogemond, V., Aguera, M., Thomasset, N., Honnorat, J., 2003. Collapsin
response mediator proteins (CRMPs): involvement in nervous system development and adult
neurodegenerative disorders. Mol Neurobiol 28, 51-64.
Cole, A.R., Causeret, F., Yadirgi, G., Hastie, C.J., McLauchlan, H., McManus, E.J., Hernandez, F.,
Eickholt, B.J., Nikolic, M., Sutherland, C., 2006. Distinct priming kinases contribute to
differential regulation of collapsin response mediator proteins by glycogen synthase kinase-3
in vivo. J Biol Chem 281, 16591-16598.
Cserr, H.F., Harling-Berg, C.J., Knopf, P.M., 1992. Drainage of brain extracellular fluid into blood and
deep cervical lymph and its immunological significance. Brain Pathol 2, 269-276.
Cserr, H.F., Knopf, P.M., 1992. Cervical lymphatics, the blood-brain barrier and the immunoreactivity
of the brain: a new view. Immunol Today 13, 507-512.
del Pozo, M.A., Sanchez-Mateos, P., Nieto, M., Sanchez-Madrid, F., 1995. Chemokines regulate
cellular polarization and adhesion receptor redistribution during lymphocyte interaction with
endothelium and extracellular matrix. Involvement of cAMP signaling pathway. J Cell Biol 131,
del Pozo, M.A., Vicente-Manzanares, M., Tejedor, R., Serrador, J.M., Sanchez-Madrid, F., 1999. Rho
GTPases control migration and polarization of adhesion molecules and cytoskeletal ERM
components in T lymphocytes. Eur J Immunol 29, 3609-3620.
Engelhardt, B., Ransohoff, R.M., 2005. The ins and outs of T-lymphocyte trafficking to the CNS:
anatomical sites and molecular mechanisms. Trends Immunol 26, 485-495.
Esiri, M.M., 2007. The interplay between inflammation and neurodegeneration in CNS disease. J
Neuroimmunol 184, 4-16.
Fujinami, R.S., von Herrath, M.G., Christen, U., Whitton, J.L., 2006. Molecular mimicry, bystander
activation, or viral persistence: infections and autoimmune disease. Clin Microbiol Rev 19, 80-
Fukata, Y., Itoh, T.J., Kimura, T., Menager, C., Nishimura, T., Shiromizu, T., Watanabe, H., Inagaki,
N., Iwamatsu, A., Hotani, H., Kaibuchi, K., 2002. CRMP-2 binds to tubulin heterodimers to
promote microtubule assembly. Nat Cell Biol 4, 583-591.
Galea, I., Bechmann, I., Perry, V.H., 2007. What is immune privilege (not)? Trends Immunol 28, 12-18.
Galelli, A., Baloul, L., Lafon, M., 2000. Abortive rabies virus central nervous infection is controlled by T
lymphocyte local recruitment and induction of apoptosis. J Neurovirol 6, 359-372.
Giraudon, P., Buart, S., Bernard, A., Belin, M.F., 1997. Cytokines secreted by glial cells infected with
HTLV-I modulate the expression of matrix metalloproteinases (MMPs) and their natural
inhibitor (TIMPs): possible involvement in neurodegenerative processes. Mol Psychiatry 2,
Giraudon, P., Vincent, P., Vuaillat, C., Verlaeten, O., Cartier, L., Marie-Cardine, A., Mutin, M.,
Bensussan, A., Belin, M.F., Boumsell, L., 2004. Semaphorin CD100 from activated T
lymphocytes induces process extension collapse in oligodendrocytes and death of immature
neural cells. J Immunol 172, 1246-1255.
Griffond, B., Verlaeten, O., Belin, M.F., Risold, P.Y., Bernard, A., 2004. Specific alteration of the
expression of selected hypothalamic neuropeptides during acute and late mouse brain
infection using a morbillivirus: relevance to the late-onset obesity? Brain Res 1022, 173-181.
Hamann, J., Fiebig, H., Strauss, M., 1993. Expression cloning of the early activation antigen CD69, a
type II integral membrane protein with a C-type lectin domain. J Immunol 150, 4920-4927.
Harling-Berg, C.J., Knopf, P.M., Cserr, H.F., 1991. Myelin basic protein infused into cerebrospinal fluid
suppresses experimental autoimmune encephalomyelitis. J Neuroimmunol 35, 45-51.
Hatterer, E., Davoust, N., Didier-Bazes, M., Vuaillat, C., Malcus, C., Belin, M.F., Nataf, S., 2006. How
to drain without lymphatics? Dendritic cells migrate from the cerebrospinal fluid to the B-cell
follicles of cervical lymph nodes. Blood 107, 806-812.
Hooper, D.C., 2005. The role of immune responses in the pathogenesis of rabies. J Neurovirol 11, 88-
Irani, D.N., Griffin, D.E., 1991. Isolation of brain parenchymal lymphocytes for flow cytometric analysis.
Application to acute viral encephalitis. J Immunol Methods 139, 223-231.
Irwin, D.J., Wunner, W.H., Ertl, H.C., Jackson, A.C., 1999. Basis of rabies virus neurovirulence in
mice: expression of major histocompatibility complex class I and class II mRNAs. J Neurovirol
Khuth, S.T., Akaoka, H., Pagenstecher, A., Verlaeten, O., Belin, M.F., Giraudon, P., Bernard, A., 2001.
Morbillivirus infection of the mouse central nervous system induces region-specific
upregulation of MMPs and TIMPs correlated to inflammatory cytokine expression. J Virol 75,
Kleine, T.O., Benes, L., 2006. Immune surveillance of the human central nervous system (CNS):
different migration pathways of immune cells through the blood-brain barrier and blood-
cerebrospinal fluid barrier in healthy persons. Cytometry A 69, 147-151.
Lafon, M., 2004. Subversive neuroinvasive strategy of rabies virus. Arch Virol Suppl, 149-159.
Lafon, M., 2005. Modulation of the immune response in the nervous system by rabies virus. Curr Top
Microbiol Immunol 289, 239-258.
Lane, T.E., Liu, M.T., Chen, B.P., Asensio, V.C., Samawi, R.M., Paoletti, A.D., Campbell, I.L., Kunkel,
S.L., Fox, H.S., Buchmeier, M.J., 2000. A central role for CD4(+) T cells and RANTES in virus-
induced central nervous system inflammation and demyelination. J Virol 74, 1415-1424.
Lauterbach, H., Zuniga, E.I., Truong, P., Oldstone, M.B., McGavern, D.B., 2006. Adoptive
immunotherapy induces CNS dendritic cell recruitment and antigen presentation during
clearance of a persistent viral infection. J Exp Med.
Lee, B.C., Lee, T.H., Avraham, S., Avraham, H.K., 2004. Involvement of the chemokine receptor
CXCR4 and its ligand stromal cell-derived factor 1alpha in breast cancer cell migration
through human brain microvascular endothelial cells. Mol Cancer Res 2, 327-338.
Liu, M.T., Chen, B.P., Oertel, P., Buchmeier, M.J., Armstrong, D., Hamilton, T.A., Lane, T.E., 2000.
The T cell chemoattractant IFN-inducible protein 10 is essential in host defense against viral-
induced neurologic disease. J Immunol 165, 2327-2330.
Marten, N.W., Stohlman, S.A., Zhou, J., Bergmann, C.C., 2003. Kinetics of virus-specific CD8+ -T-cell
expansion and trafficking following central nervous system infection. J Virol 77, 2775-2778.
Mitchison, T.J., Cramer, L.P., 1996. Actin-based cell motility and cell locomotion. Cell 84, 371-379.
Norman, M.U., Hickey, M.J., 2005. Mechanisms of lymphocyte migration in autoimmune disease.
Tissue Antigens 66, 163-172.
Panda, J.N., Dale, H.E., Loan, R.W., Davis, L.E., 1965. Immunologic Response to Subarachnoid and
Intracerebral Injection of Antigens. J Immunol 94, 760-764.
Patterson, C.E., Daley, J.K., Echols, L.A., Lane, T.E., Rall, G.F., 2003. Measles virus infection induces
chemokine synthesis by neurons. J Immunol 171, 3102-3109.
Peterson, L.K., Fujinami, R.S., 2007. Inflammation, demyelination, neurodegeneration and
neuroprotection in the pathogenesis of multiple sclerosis. J Neuroimmunol 184, 37-44.
Ramakrishna, C., Atkinson, R.A., Stohlman, S.A., Bergmann, C.C., 2006. Vaccine-induced memory
CD8+ T cells cannot prevent central nervous system virus reactivation. J Immunol 176, 3062-
Ransohoff, R.M., Kivisakk, P., Kidd, G., 2003. Three or more routes for leukocyte migration into the
central nervous system. Nat Rev Immunol 3, 569-581.
Rey, M., Vicente-Manzanares, M., Viedma, F., Yanez-Mo, M., Urzainqui, A., Barreiro, O., Vazquez, J.,
Sanchez-Madrid, F., 2002. Cutting edge: association of the motor protein nonmuscle myosin
heavy chain-IIA with the C terminus of the chemokine receptor CXCR4 in T lymphocytes. J
Immunol 169, 5410-5414.
Ricard, D., Rogemond, V., Charrier, E., Aguera, M., Bagnard, D., Belin, M.F., Thomasset, N.,
Honnorat, J., 2001. Isolation and expression pattern of human Unc-33-like phosphoprotein
6/collapsin response mediator protein 5 (Ulip6/CRMP5): coexistence with Ulip2/CRMP2 in
Sema3a- sensitive oligodendrocytes. J Neurosci 21, 7203-7214.
Ray NB, Ewalt LC, Lodmell Rabies virus replication in primary murine bone marrow macrophages and
in human and murine macrophage-like cell lines: implications for viral persistence. J Virol.
Roy, A., Phares, T.W., Koprowski, H., Hooper, D.C., 2007. Failure to open the blood-brain barrier and
deliver immune effectors to central nervous system tissues leads to the lethal outcome of
silver-haired bat rabies virus infection. J Virol 81, 1110-1118.
Salazar-Mather, T.P., Hokeness, K.L., 2006. Cytokine and chemokine networks: pathways to antiviral
defense. Curr Top Microbiol Immunol 303, 29-46.
Sanchez-Madrid, F., del Pozo, M.A., 1999. Leukocyte polarization in cell migration and immune
interactions. Embo J 18, 501-511.
Serafini, B., Rosicarelli, B., Magliozzi, R., Stigliano, E., Aloisi, F., 2004. Detection of ectopic B-cell
follicles with germinal centers in the meninges of patients with secondary progressive multiple
sclerosis. Brain Pathol 14, 164-174.
Serafini, B., Rosicarelli, B., Magliozzi, R., Stigliano, E., Capello, E., Mancardi, G.L., Aloisi, F., 2006.
Dendritic cells in multiple sclerosis lesions: maturation stage, myelin uptake, and interaction
with proliferating T cells. J Neuropathol Exp Neurol 65, 124-141.
Shin, J.H., Sakoda, Y., Kim, J.H., Tanaka, T., Kida, H., Kimura, T., Ochiai, K., Umemura, T., 2006.
Efficacy of intracerebral immunization against pseudorabies virus in mice. Microbiol Immunol
Stevenson, P.G., Bangham, C.R., Hawke, S., 1997. Recruitment, activation and proliferation of CD8+
memory T cells in an immunoprivileged site. Eur J Immunol 27, 3259-3268.
Strazielle, N., Khuth, S.T., Murat, A., Chalon, A., Giraudon, P., Belin, M.F., Ghersi-Egea, J.F., 2003.
Pro-inflammatory cytokines modulate matrix metalloproteinase secretion and organic anion
transport at the blood-cerebrospinal fluid barrier. J Neuropathol Exp Neurol 62, 1254-1264.
Tishon, A., Lewicki, H., Andaya, A., McGavern, D., Martin, L., Oldstone, M.B., 2006. CD4 T cell control
primary measles virus infection of the CNS: regulation is dependent on combined activity with
either CD8 T cells or with B cells: CD4, CD8 or B cells alone are ineffective. Virology 347,
Uchida, Y., Ohshima, T., Sasaki, Y., Suzuki, H., Yanai, S., Yamashita, N., Nakamura, F., Takei, K.,
Ihara, Y., Mikoshiba, K., Kolattukudy, P., Honnorat, J., Goshima, Y., 2005. Semaphorin3A
signalling is mediated via sequential Cdk5 and GSK3beta phosphorylation of CRMP2:
implication of common phosphorylating mechanism underlying axon guidance and Alzheimer's
disease. Genes Cells 10, 165-179.
Verlaeten, O., Griffond, B., Khuth, S.T., Giraudon, P., Akaoka, H., Belin, M.F., Fellmann, D., Bernard,
A., 2001. Down regulation of melanin concentrating hormone in virally induced obesity. Mol
Cell Endocrinol 181, 207-219.
Veyrac, A., Giannetti, N., Charrier, E., Reymond-Marron, I., Aguera, M., Rogemond, V., Honnorat, J.,
Jourdan, F., 2005. Expression of collapsin response mediator proteins 1, 2 and 5 is
differentially regulated in newly generated and mature neurons of the adult olfactory system.
Eur J Neurosci 21, 2635-2648.
Vicente-Manzanares, M., Cabrero, J.R., Rey, M., Perez-Martinez, M., Ursa, A., Itoh, K., Sanchez-
Madrid, F., 2002. A role for the Rho-p160 Rho coiled-coil kinase axis in the chemokine stromal
cell-derived factor-1alpha-induced lymphocyte actomyosin and microtubular organization and
chemotaxis. J Immunol 168, 400-410.
Vicente-Manzanares, M., Sanchez-Madrid, F., 2004. Role of the cytoskeleton during leukocyte
responses. Nat Rev Immunol 4, 110-122.
Vincent, P., Collette, Y., Marignier, R., Vuaillat, C., Rogemond, V., Davoust, N., Malcus, C., Cavagna,
S., Gessain, A., Machuca-Gayet, I., Belin, M.F., Quach, T., Giraudon, P., 2005. A role for the
neuronal protein collapsin response mediator protein 2 in T lymphocyte polarization and
migration. J Immunol 175, 7650-7660.
Wang, Z.W., Sarmento, L., Wang, Y., Li, X.Q., Dhingra, V., Tseggai, T., Jiang, B., Fu, Z.F., 2005.
Attenuated rabies virus activates, while pathogenic rabies virus evades, the host innate
immune responses in the central nervous system. J Virol 79, 12554-12565.
Westermann, J., Bode, U., Sahle, A., Speck, U., Karin, N., Bell, E.B., Kalies, K., Gebert, A., 2005.
Naive, effector, and memory T lymphocytes efficiently scan dendritic cells in vivo: contact
frequency in T cell zones of secondary lymphoid organs does not depend on LFA-1
expression and facilitates survival of effector T cells. J Immunol 174, 2517-2524.
Yoshimura, T., Kawano, Y., Arimura, N., Kawabata, S., Kikuchi, A., Kaibuchi, K., 2005. GSK-3beta reg
Figure 1: CRMP2 expression in peripheral blood mononuclear cells of RABV infected mice. A-
After infection in the periphery with the neurovirulent RABV strain CVS (A,B,C) or with the
neurovirulent and partially attenuated RABV strain PV or non neurovirulent RABV strain ERA (D),
PBMC were collected from infected (black or grey histogram) and sham-infected (white histograms)
mice 1, 7 and 9 days post infection (pi). Kinetics of CD3, CD69 and CRMP2 frequency among total
viable PBMC (A and D), % of CD3 or CD69 positive cells among CRMP2 positive cells (B) and % of
CRMP2+ cells among CD3 and CD69 positive cells (C) was followed by cytofluorimetry analysis using
appropriate pairs of antibodies. A, B, C- infection with a neuronotropic CVS RABV strain causing fatal
rabies. CRMP2+ cells recruit in CD3+ T lymphocytes and early activated immune cell populations.
RABV infection triggers CRMP2 expression in CD3 T lymphocytes; virtually all CD69+ cells express
CRMP2. D- Infection with a neuronotropic RABV strain, PV, causing abortive rabies (dark gray
histograms, or infection with a non neuronotropic RABV strain ERA causing no CNS disease (light
gray histograms). In contrast to the neuronotropic PV strain, ERA strain does not trigger the expansion
of CRMP2+ in PBMC of infected mice. These data are representative of at least three independent
Figure 2: CRMP2 expression in peripheral blood mononuclear cells of CDV infected mice. After
infection of brain with CDV (gray histograms) or sham-infection (white histograms), PBMC were
collected and analyzed by cytofluorimetry as decribed above. A- Transient reduction of CD3+ T cells
and elevation of early-activated immune cells CD69+ frequency induced by brain infection; B-
Elevated CRMP2 expression in CD3+ T cells and early-activated CD69+ cells, as shown by increased
Mean Fluorescence Intensity (MFI); C- Detection in infected mice of early-activated CD69+ cells with
elevated CRMP2 expression (MFI≥twice MFI observed in sham-infected mice) called CRMP2hi; D-
Expansion of CRMP2hi cells increased progressively as infection and disease progressed. These
analyses are representative of three independent experiments (±SD).
Figure 3: Immune cells infiltration in the CNS of neurovirulent RABV CVS strain infected mice
analyzed by cytofluorimetry. Cells infiltrating the CNS of CVS infected (gray histograms) and sham-
infected (white histograms) were isolated by Percoll gradients, stained with appropriate pairs of
antibodies and analyzed by cytofluorimetry. Kinetics of CD3+ T cell frequency (A) and double stained
CD3 and CRMP2 cells (B) were followed 1, 5, 8 and 12 days after infection in infected (black
histograms) and non-infected (white histograms) mice. C- The distribution of CRMP2 expressing cells
among CD3, CD11c, CD11b and B220 positive cells from the CNS was analyzed in the course of
RABV infection (5, 8 and 12 days pi) and compared to the non-infected control (day 0). T lymphocytes
constitute the majority of infiltrated immune cells on day 12 post inoculation, while CD11c+ cells
(corresponding roughly to dendritic cells) peaked on day 8. Each point of the curve includes two to
three mice. Data were representative of three different experiments (±SD).
Figure 4: Immune cell infiltration in the CNS of CDV infected mice analyzed by cytofluorimetry.
Cells infiltrating the CNS were collected (7 and 14 dpi) and analyzed by cytofluorimetry as described
above. A- Elevated recruitment of early activated CD69+ and CD3+ T cells under viral infection (black
histograms) compared to sham-infected (white histograms); B- Population of CD3+ cells and
CRMP2hi (CD69+ CRMP2+) cells (dashed circle) is expanded in CDV infected brain (right panels)
compared to sham-inoculated ( left panels) (representative data of three different experiments).
Figure 5: Analysis of the brain infiltrates in CDV-infected mice. Immunodetection of brain
infiltrating immune cells was performed by immunofluorescence (A and C, brain removed and
immediately frozen) and by DAB staining (B, mice perfused with PFA) on coronal brain sections using
anti-CD4, CD8, CRMP2 and TCR beta chain antibodies. A- CRMP2+ infiltrating cells are more
abundant in the brain of CDV-infected than sham-inoculated mice (microscope field x400). B- CD4+ T
lymphocytes infiltrated in CDV-infected brain structures hippocampus, thalamus (Khuth et al., 2001)
and at proximity of the blood-CNS (velum interpositum) and CSF-CNS (third ventricle) interfaces
(avidiin-biotin peroxidase system) (microscope field x100 and x400). C- Identification in CDV infected
brain of CD4+ and CD8+ lymphocytes (microscope field x200) and CRMP2 expression by T
lymphocytes (TCR beta chain positive cells) (microscope field x400).
Figure 6: Association of elevated CRMP2 expression with high migratory rate of lymphocytes
collected in CDV infected mice: CRMP2 expression was evaluated by cytofluorimetry in lymph node
cells of CDV-infected compared to non-infected mice prior to migration tests were performed in
Transwell system (B) and on hippocampus organotypic culture (C). A- Elevated CRMP2 expression
level (MFI) in lymphocytes of CVD-infected mice. B- Elevated migration of lymph node cells of CDV-
infected mice (3 non-infected versus 3 CDV-infected mice) in a Transwell system (3m pore
membrane) in the presence of a chemokine gradient (CXCL12 -10ng/ml, CCL5 -100ng/ml, CCL2 and
CXCL10 -20ng/ml, 37°C, 2h) (±SD). C- Migration analysis of lymph node cells stained with CFSE then
spotted close to CDV-and non-infected hippocampus slices (18h contact then cell counting): elevated
migratory rate of lymphocytes of CDV-infected versus sham-inoculated mice. Blockade experiment:
treatment of lymphocytes with anti-CRMP2 antibody prior migration assay impeded tissue infiltration
both on CDV-infected and non-infected hippocampus slices (open marks in graph ).
Figure 7: Peripheral expansion of CRMP2hi cells in RABV infected mice correlates with clinical
score: A- After infection, RABV infected animals develop clinical symptom which aggravate as the
virus infectionprogresses into the NS. Progression was evaluated by scoring mobility and mortality as
follows: 0=normal mice, 1=ruffled fur, 2=loss of agility, 3=one paralyzed hind leg, 4=two paralyzed hind
legs, 5=total loss of mobility and 6=death. A representative evolution of the cumulative clinical scores
(addition of individual scores of 8 animals per group); B- CD3+ T cells expressing strong CRMP2
fluorescence (CRMP2 high) can be observed in lymph node cells of infected animals (right panel
showing CD3/CRMP2 expression in lymph nodes of 7 day RABV infected mice). These cells were
absent in the CD3+ T cells of non-infected mice (middle panel) (representative data). C- CRMP2hi
cells can be detected not only in lymph nodes but also in PBMC as the disease (individual clinical
score on x axis) progresses (4 mice represented); D- When individual clinical scores were plotted
against % of CRMP2+CD3+ T cells in lymph node, a significant correlation was observed (R= 0.69)
between CRMP2 cells and individual clinical scores (logarithmic representation of non-linear
cell frequency in %
d1 d7 d9 d1 d7 d9 d1 d7 d9 pi
d1 d7 d9 d1 d7 d9 pi
cell frequency in %
among CD3+ among CD69+
d1 d7 d9 d1 d7 d9 pi
cell frequency in %
Figure-1, Vuaillat et al
d3 d7 d11 d14 d3 d7 d11 d14 pi
cell frequency in %
Fluorescence in log
Sham inoc. mice
CRMP2 in PBMC: MFI =13
CDV infected mice
CRMP2 in PBMC: MFI= 22,5
Fluorescence in log
MFI = 19,5
70% MFI= 26
d3 d7 d11 d13 pi
Figure-2 , Vuaillat et al
among CD3+among CD69+
d7 d13 d7 d13 pi
15812 days pi
cell frequency in %
5 8 12 days pi
cell frequency in %
CRMP2+ cells in %
5812 days pi0
Figure-3 , Vuaillat et al
Sham Inoculated CDV
d7 d14 d7 d14 pi
cell frequency in %
Figure-4 , Vuaillat et al
CDV infected Sham inoculated
TCR beta chainCRMP2
Figure-5 , Vuaillat et al
Number of migrating cells
Lympho- CDV mice
Lympho- Sham mice
+ anti-CRMP2 Ab
Lympho- CDV miceLympho- CDV mice
Nb of migrating cells
lymph node cells
CDV-i Sham-i CDV-i
+ CRMP2 Ab
Figure-6 , Vuaillat et al
A Download full-text
678912 13 dpi
Cumulative clinical score
CVS-infected mice Non-infected miceControl
Individual clinical score
CRMP2hi cells in %
2 3 4 5
3 4 5
Individual clinical score
0,5 1 1,5 2 2,5 33,5 4
CRMP2+ cells in %
Figure-7 , Vuaillat et al