Download Two-photon laser scanning microscopy imaging of intact spinal cord and

Journal of Immunological Methods 352 (2010) 89–100
Contents lists available at ScienceDirect
Journal of Immunological Methods
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j i m
Research paper
Two-photon laser scanning microscopy imaging of intact spinal cord and
cerebral cortex reveals requirement for CXCR6 and neuroinflammation in
immune cell infiltration of cortical injury sites
Jiyun V. Kim a,⁎, Ning Jiang a,1, Carlos E. Tadokoro a,2, Liping Liu b, Richard M. Ransohoff b,
Juan J. Lafaille a, Michael L. Dustin a,⁎
a
Helen L. and Martin S. Kimmel Center for Biology and Medicine of the Skirball Institute of Biomolecular Medicine, New York University School of Medicine,
540 First Avenue, New York, NY 10016, USA
b
Neuroinflammation Research Center, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA
a r t i c l e
i n f o
Article history:
Received 16 July 2009
Received in revised form 22 September 2009
Accepted 24 September 2009
Available online 2 October 2009
Keywords:
Rodent
EAE
Cell trafficking
Microscopy
Chemokines
a b s t r a c t
The mouse spinal cord is an important site for autoimmune and injury models. Skull thinning
surgery provides a minimally invasive window for microscopy of the mouse cerebral cortex,
but there are no parallel methods for the spinal cord. We introduce a novel, facile and
inexpensive method for two-photon laser scanning microscopy of the intact spinal cord in the
mouse by taking advantage of the naturally accessible intervertebral space. These are powerful
methods when combined with gene-targeted mice in which endogenous immune cells are
labeled with green fluorescent protein (GFP). We first demonstrate that generation of the
intervertebral window does not elicit a reaction of GFP+ microglial cells in CX3CR1gfp/+ mice.
We next demonstrate a distinct rostrocaudal migration of GFP+ immune cells in the spinal cord
of CXCR6gfp/+ mice during active experimental autoimmune encephalomyelitis (EAE).
Interestingly, infiltration of the cerebral cortex by GFP+ cells in these mice required three
conditions: EAE induction, cortical injury and expression of CXCR6 on immune cells.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
1.1. Two photon laser scanning microscopy and CNS intravital
surgeries
Two-photon laser scanning microscopy (TPLSM) has revolutionized the study of the nervous system and the immune
⁎ Corresponding authors. Helen L. and Martin S. Kimmel Center for Biology
and Medicine of the Skirball Institute of Biomolecular Medicine, New York
University School of Medicine, Molecular Pathogenesis Program, 540 First
Avenue, SKI-2, New York, NY 10016, USA. Tel.: +1 212 263 3208; fax: +1
212 263 5711.
E-mail addresses: [email protected] (J.V. Kim),
[email protected] (M.L. Dustin).
1
Present address: Department of Bioengineering, James Clark Center,
Stanford University, 318 Campus Drive, Stanford, CA 94305, USA.
2
Present address: Instituto Gulbenkian de Cienza, R. da Quinta Grande 6,
Oeiras, 2780-156, Portugal.
0022-1759/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.jim.2009.09.007
system by providing dynamic views of sub-cellular structures
and cells in a tissue context in vivo (Denk et al., 1994; Miller
et al., 2003). These methods generally require surgical exposure
of the tissue of interest with resulting potential for induction of
inflammation, which may contaminate observations. Imaging of
the intact CNS has been accomplished in vivo using thinned
skull methods, in which imaging is performed through a thin
plate of bone without the need for a craniotomy (Grutzendler
et al., 2002; Xu et al., 2007). Microscopy of the living mouse
spinal cord reported to date requires a laminectomy, the
removal of the vertebral bone, with potential disturbance of
the underlying dura and the surface of the spinal cord
(Engelhardt et al., 2003; Odoardi et al., 2007; Davalos et al.,
2008). The exposed spinal cord in this system is vulnerable to
sterile injury and foreign materials that may contain innate
immune activators such as endotoxin. Such injury has been
reported in intravital microscopy observations through a
craniotomy, but not with the thinned skull method (Xu et al.,
90
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
2007). Induction of the microglial injury response is a sensitive
measure of such damage (Davalos et al., 2005; Kim and Dustin,
2006). The underlying tissue is also potentially more vulnerable
to additional inflammatory artifacts such as acute neutrophil
recruitment (Kim and Dustin, 2006). These shortcomings of the
existing methods prompted us to devise a method for
microscopy of the living mouse spinal cord that does not
require laminectomy, but leaves the dura fully intact by taking
advantage of the natural intervertebral space as an imaging
window.
We applied our intervertebral window to determine the
dynamic behavior of immune cells in experimental autoimmune encephalomyelitis (EAE), a mouse model for inflammatory aspects of multiple sclerosis (MS) (Zamvil et al., 1985;
Zamvil and Steinman, 1990). EAE is characterized by white
matter inflammation with T cell infiltration in the optic nerves,
the spinal cord and the cerebellum (Bettelli et al., 2003; Wensky
et al., 2005) and evaluated based on an ascending paralysis,
starting with tail weakness, progressing to hindlimb weakness
and in most severe cases involving forelimb paralysis. Paralysis
is associated with white matter inflammatory lesions in the
spinal cord that include CD4+ T cells generally without
substantial T cell accumulation in the gray matter. The dynamics
of these cells in the intact spinal cord during EAE is not known.
1.2. Role of CXCR6 in EAE
Three chemokine receptors have been implicated in guidance of T cells into the central nervous system. In the steady
state, a population of memory T cells enters the cerebrospinal
fluid (CSF) via the choroid plexus (Kivisakk et al., 2003, 2004).
These cells are highly enriched for expression of CCR7 and active
PSGL-1, which suggests these molecules are involved in CNS
entry. Recent reports also indicate that the induction of EAE
involves the CCR6 dependent entry of Th17 type CD4+ T cells
through the choroid plexus (Reboldi et al., 2009). These cells
then trigger the CCR6 independent recruitment of additional T
cells and myeloid cells via parenchymal vessels. Antibodies to
CXCL16 (ligand to CXCR6) reduce EAE severity (Fukumoto et al.,
2004). Although CXCL16 also functions as a scavenger receptor
referred to as the scavenger receptor that binds phosphatidylserine and oxidized lipoprotein (SR-PSOX) (Shimaoka et al.,
2000), CXCR6 (also called BONZO, STRL33 and TYM-STR), the
CXCL16 receptor (Matloubian et al., 2000; Shimaoka et al.,
2000) could be important in T cell infiltration in EAE. CXCL16 is
expressed on potential antigen-presenting cells including
dendritic cells (Matloubian et al., 2000), macrophages (Shimaoka et al., 2003) and astrocytes (Ludwig et al., 2005) and, is a
chemokine receptor that is expressed in activated CD4+ T cell
subsets (Loetscher et al., 1997; Matloubian et al., 2000), NKT
cells (Matloubian et al., 2000), and activated CD8+ T cells
(Matloubian et al., 2000; Sato et al., 2005). CXCR6 is known to be
associated with antigen-specific interferon-γ producing cells in
multiple sclerosis (MS) (Calabresi et al., 2002). Therefore, we
have further examined the role of CXCR6 in EAE here.
We used mice in which the major coding exon of CXCR6 has
been replaced with GFP in gene-targeted mice (Unutmaz et al.,
2000) in which CXCR6 expressing cells can be tracked in vivo
(Geissmann et al., 2005). The heterozygous mice (CXCR6gfp/+)
have half the wild type levels of CXCR6 and express cytoplasmic
GFP, while the homozygous mice (CXCR6gfp/gfp or CXCR6−/−)
lack CXCR6 expression, but instead express only cytoplasmic
GFP (Geissmann et al., 2005). We utilized this mouse strain on
the B6 background to study the dynamics of GFP+ cells in the
living mouse spinal cord and brain during EAE. CXCR6+ cells
migrated rapidly in the spinal cord with a rostrocaudal bias in a
CXCR6 independent manner. Unexpectedly, our studies also
addressed the molecular basis of T cell infiltration into cortical
injury sites in the context of EAE.
1.3. An inflammation model for initiating cortical MS lesions
MS is also characterized by cortical lesions with inflammation
which may contribute to inducing direct damage to neurons and
neurites (Bo et al., 2003; Geurts and Barkhof, 2008). Cortical
demyelinating lesions can be either leukocortical, involving the
cortex and the adjacent subcortical white matter, or subpial,
involving the first three layers of cortex without associated white
matter demyelination (Moll et al., 2008). In very late stage
chronic progressive MS cortical lesions contain relatively few T
cells. However, recent studies of subpial cortical lesions in
relapsing-remitting MS show abundant T cell and macrophage
inflammation in the affected gray matter (Moll, N., R.M.R., B.
Kelley, J.E. Parisi, and C. Lucchinetti. personal communication).
Gray matter lesions could precede cognitive deficits and increase
the morbidity of MS (Geurts and Barkhof, 2008). Perry and
colleagues developed a rodent model for cortical inflammation
based on stereotactic injection of heat killed bacillus CalmetteGuérin (BCG) in the parenchyma and subcutaneous inoculation
with complete Freund's adjuvant, which contains heat killed
Mycobacterium tuberculosis (MTB) (Matyszak and Perry, 1998).
Either stimulus alone was insufficient to induce demyelination
(Matyszak and Perry, 1995). In mice, one can image sterile
cortical injury through a thinned skull window (Grutzendler
et al., 2002), but this appears to be a local reaction, much like that
to injection of BCG into the parenchyma. Injury in the context of
inflammation elsewhere in the tissue may lower the threshold of
the blood brain barrier through effects of cytokines (Schnell et al.,
1999; Sun et al., 2004).
Therefore, since classical EAE models in the mouse do not
demonstrate cortical lesions, we examined recruitment of T
cells in response to the combination of EAE and laser injury in
layer 1 barrel cortex (Davalos et al., 2005).
2. Materials and methods
2.1. Transgenic mice
CXCR6+/gfp, CXCR6gfp/gfp and CX3CR+/gfp
, and LysMgfp/+ mice
1
were a gift of D.R. Littman (NYU School of Medicine, New York,
NY) (Jung et al., 2000; Geissmann et al., 2005) and Thomas Graf
(AECOM, Bronx, NY) (Faust et al., 2000), respectively. All strains
were backcrossed onto C57BL/6 for at least 12 generations and
housed in specific pathogen-free conditions in accordance with
Institutional Animal Care and Use Committee protocols of New
York University School of Medicine.
2.2. Peptides
The Dana-Farber Cancer Institute Molecular Biology Core
Facility (Boston, MA) synthesized peptides.
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
91
2.3. Quantitative PCR
2.5. Two-photon microscopy
We perfused anesthetized mice with PBS containing 2 mM
EDTA. We isolated brain and spinal cord and snap froze in liquid
nitrogen and stored in −80 °C until further processing. We
isolated RNA from tissue with TRIzol (InVitrogen, Carlsbad, CA),
and treated with DNase I from which we generated cDNA using
Superscript III enzyme system (InVitrogen) and amplified with
ABI 7900 cycler (Applied Biosystems, Foster City, CA). The
primer and probe oligonucleotides used for CXCL16 were,
cxcl16-forward, ggaagccaagaccagtgggt, cxcl16reverse,
tttggtggtgaaaactcttccc and cxcl16 probe, [6-fam]ttgagcgcaaagagtgtggaactggtc[TAMRA]. We used HPRT and G6PD for
internal control housekeeper genes, HPRT-forward, aggttgcaagcttgctggt, HPRT-reverse, tgaagtactcattatagtcaagggc, and
HPRT probe, [5TET]tgttggatacaggccagactttgttggat[TAMRA],
and G6PD-forward, tgaagctccctgatgcctat, G6PD-reverse,
caatcttgtgcagcagtggt, and G6PD probe [5TET]gaagcctggcgtatcttccac[TAMRA]. Cycle temperatures were 50 °C for 2 min,
95 °C for 2 min, 40 cycles of 95 °C for 15 s and 60 °C for 30s. We
performed melting curve analyses at the genomics core facility
of the NYU Cancer Institute.
We tuned a mode locked Ti-Sapphire laser (Spectraphysics) connected to a Bio-Rad Radiance multiphoton microscope to 920 nm to excite GFP and quantum dots (Shakhar
et al., 2005), and acquired stacks of images using step sizes of
1–3 µm to a depth of 200 µm bellow the skull using x40 or
x60 water dipping objectives. We then generated time-lapse
movies with 1- to 1.5-min intervals between 3D stacks.
2.4. Surgery and experimental procedures
2.4.1. EAE
We induced classical EAE with MOG peptide in CFA and
pertussis toxin (Mendel et al., 1995). Briefly, we injected
100 µg MOG (35–55) peptide (MEVGWYRSPFSRVVHLYRNGK) emulsified in CFA with heat killed mycobacterium
tuberculosis (4 mg/ml) at the base of the tail on day 0 and
pertussis toxin (100 ng in sterile saline) into the tail vein on
days 0 and 2. We scored mice for EAE as follows: 0, no
disease; 1, tail paralysis; 2, hind limb weakness; 3, hind limb
paralysis; 4, fore limb weakness; 5, moribund. We hydrated
mice with scores >2 daily and sacrificed for score ≥ 4.
2.4.2. Thinned skull and laser injury
In all the intravital imaging described below, we maintained anesthetized mice on warming plates and the
intravital window through an objective heater to maintain
core temperature at 37 °C. We performed thinned skull
intravital window surgeries and laser ablation as previously
described (Grutzendler et al., 2002; Davalos et al., 2005). The
size of injury induced was typically 15 µm in diameter. We
defined vascular spaces by intravenous injection of 655 nm
emitting quantum dots (R&D Systems, Minneapolis, MN).
2.4.3. Intravital spinal cord (See Fig. 1 and Results (3.1) for
details of the method.)
We anesthetized mice and exposed the thoracic spine by
dissecting the overlying muscle and connective tissue. We bent
a thin stainless steel plate with an open central slot to separate
the inner edges of the plate to engage two adjacent vertebrae in
a funnel clamp and then immobilized the plate by attaching it to
two posts using screws. We then acquired Images through the
thinned intervertebral connective tissue sealed above with 2%
low melting agarose (Sigma, St. Louis, MO) and a cover glass
over the area.
2.6. Data analysis
We used Volocity software (Improvision, Waltham, MA)
to track T cell movements. We obtained cell counts within
three-dimensional volumes to calculate cells/mm3. We
determined average speed by using mean value of instantaneous speeds spanning 20–30 min. Confinement index is the
ratio of displacement to path length with a range of 0 to 1.
Immobile fraction is the percent of cells with average
displacement rate of <1 μm/min over a 30 min interval and
can be visualized as cells whose movements are confined
within a maximum confinement radius of 30 µm (Sims et al.,
2007). One-dimensional motility coefficient is the square of
mean displacement divided by 2t1/2 (Sumen et al., 2004). We
determined statistical significance (p < 0.05) using a Student's t test and a Mann–Whitney rank sum test for
populations with non-Gaussian distributions.
3. Results
3.1. Intervertebral spinal cord microscopy preparation
We first devised an inexpensive apparatus to immobilize
and create an intervertebral window to image the intact spinal
cord in a live anesthetized mouse (Fig. 1). We manipulated
standard double edge razor blades to generate a spinal cord
clamp and mounting system with correct dimensions for an
adult mouse. We assembled the apparatus in three stages: in
the first stage, a small stack of 2–3 cemented blades served as a
spring clamp to hold the vertebral column; in the second
stage, we immobilized this bottom plate by cementing to a
transverse blade that was firmly attached to two posts to a
position resulting in a stabilized mouse; and in the third stage,
a small amount of low melt agarose further stabilized the
intact dura and the underlying spinal cord for high-resolution
imaging. Bending the inner tabs at 45° below the plane
generated a spring clamp to hold the spinal cord (Fig. 1A). A
midline dorsal incision exposed the thoracic muscles over ∼T3
to T7, which are dissected free from the sides of the vertebral
bone in an anesthetized mouse. The bent bottom plate
separated the tabs, which in turn can grip either side of the
vertebral bone T3–T7 vertebrae (Fig. 1B). We then cemented a
second plate composed of two razor blades transversely to the
bottom plate such that the large central opening is aligned
over the intervertebral space. We then secured the top plate to
two magnetic posts on either side of the mouse that are
positioned to partly lift the thoracic region off the underlying
heating plate to reduce effects of breathing movements on the
stability of the spinal cord (Fig. 1C). We can then carefully
peeled the intervertebral fascia leaving the intact dura
(∼50 µm) to protect the spinal cord and further dampen the
92
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
Fig. 1. Intervertebral intravital window. (A) 2–3 dulled razor blades are cemented together to produce a more stable steel plate. Inner wings of the blade are bent to
serve as funnel clamps. (B) This bent plate is then inserted on either side of the thoracic spinal vertebral bone in an anesthetized mouse after cutting the thoracic
muscles on either side of the spine. (C) A second flat plate is glued perpendicularly to the first to immobilize the mouse to the magnetic posts on either side of the
holding plate. (D) The resultant anatomy and intravital window imaged through the objective is illustrated in this schematic. The intervertebral space is dissected
down to leave the dura layer of about 50 µm and 2% agarose in aCSF is overlayed to further dampen any organ movement without inducing pressure. A microscope
coverglass then seals the intravital window for which water or oil immersion objectives can be used. (E) The anesthetized mouse is placed under the objective on a
37 °C surface with low flow oxygenation mask. The objective heater also provides a close temperature control to the intravital window. (F and G) The
intervertebral space with the spinal cord and the posterior spinal vein are translucently visible under the coverglass.
spinal cord motions by sealing with a drop of low melt agarose
on the dura and a coverglass (Fig. 1D). The curvature of the
mouse thoracic spin resulted in a very natural posture of the
anesthetized mouse in the apparatus (Fig. 1E). The white
matter to either side of the central vein was visible through
the intervertebral window after removal of the fascia and
application of the agarose and coverslip (Fig. 1F and G). We
verified that microglial cells in CX3CR+/gfp
mice did not show
1
evidence of surface injury (Supplementary Fig. 1) and that
neutrophils from LysMgfp/+ mice did not arrest inside the
vessels or extravasate spontaneously over a 2 hour observation period (data not shown). With this method we isolated
the inflammation of our disease model, EAE, with minimal
artifacts induced by surgical trauma.
3.2. CXCR6 is not required for EAE-associated ascending paralysis
To study the behavior of inflammatory lymphocytes within
the intact spinal cord and cerebral cortex we characterized
reporter mice in which GFP replaces CXCR6. First, we determined the course of EAE in CXCR6gfp/+ and CXCR6gfp/gfp
C57BL/6 mice compared to the wild type (WT) by inducing EAE with MOG (35–55) peptide in CFA. EAE disease
course was similar between all three groups up to day 21
(Fig. 2A), with mean onset of disease at day 9.7 ± 1.25 (WT),
8.9 ± 1.5 days (CXCR6gfp/+) and 9.0 ± 1.6 (CXCR6gfp/gfp) days
(Fig. 2C), mean maximal score of 2.75 ± 0.75 (WT), 2.7 ± 0.6
(CXCR6gfp/+) and 2.7 ± 0.9 (CXCR6gfp/gfp) (Fig. 2D), and
proportion responding of 10/10 (WT), 29/29 (CXCR6gfp/+)
and 20/22 (CXCR6gfp/gfp) (Fig. 2E), respectively. We further
followed EAE disease up to 50 days between CXCR6gfp/+ and
CXCR6gfp/gfp mice and found no statistical difference (data not
shown). Immunohistochemistry and flow cytometry of cells
isolated from the spinal cord indicated that GFP+ T cells were
CD4+ (Supplementary Figs. 2–5) and were pro-inflammatory
since less than 5% expressed FOXP3 (data not shown). Although
2/22 animals in the CXCR6gfp/gfp group remained completely
healthy, this was not statistically significant when compared to
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
93
Fig. 2. CXCR6 and EAE course. (A) CXCR6 did not affect the course of EAE at the onset, active (days 0–15) and remission phases (days 20–52) (scores ± s.d) for
C57BL/6 WT (n = 10) CXCR6gfp/+ (n = 29) and CXCR6gfp/gfp mice (n = 22) aged 9–12 weeks. (B) Day of onset defined by the day on which the score ≥ 1 (day ± s.d.).
(C) The mean maximum score for each genotype (score ± s.d.). (D) The fraction of animals developing EAE, which was 10/10 WT, 29/29 CXCR6gfp/+ and 20/22 of
the CXCR6gfp/gfp (the results here are not statistically significant by Pearson's Chi-square test).
the heterozygous mice. These healthy mice were not included in
the analysis or imaging experiments due to concerns that there
may have been problems with the inoculation given that the
other 20 animals had a very similar disease course to the
CXCR6gfp/+. We concluded that CXCR6 is not required for
ascending paralysis in this model of EAE and thus these mice can
be used to study the migratory behavior of inflammatory T cells
in the spinal cord.
3.3. T cell movement is biased in the spinal cord
The GFP expression by the CXCR6gfp/+ and CXCR6gfp/gfp mice
enabled imaging the behavior of inflammatory T cells in the
spinal cord of living mice with EAE and to examine whether
CXCR6 contributes to migration patterns. We imaged the spinal
cord through the intervertebral space in the thoracic spine area
by TPLSM in CXCR6gfp/+ (n = 10 animals) and CXCR6gfp/gfp
(n = 5) mice with EAE (Fig. 3A and B; Movie 1). GFP+ cells were
mostly confined to the spinal cord white matter based
immunofluorescence histology (Supplementary Fig. 2). Twophoton imaging in perfused thick sections demonstrated that
GFP+ cells also rarely penetrated into the underlying gray
matter, which can be distinguished from white matter based
upon differential autofluorescence (Supplementary Fig. 3).
Furthermore, the GFP+ cells were positive for CD4 but negative
for CD8, NK1.1, CD11b and CD11c (Supplementary Figs. 2, 4 and
5). There were no significant differences in numbers of GFP+
cells or their movement between CXCR6gfp/+ and CXCR6gfp/gfp
mice: the mean speeds were 4.9 µm/min and 4.7 µm/min
(Fig. 3C), the mean confinement indices were 0.53 and 0.53
(Fig. 3D) and the mean arrested fractions were 22.2% and 26.2%
(Fig. 3E), respectively. Arrested cells may represent autoantigen
specific T cells forming immunological synapses with APC. The
CXCR6 expressing T cells include autoreactive T cells in MS
(Calabresi et al., 2002), but will also include many activated T
cells specific for non-CNS antigens (Zamvil and Steinman,
1990). Visual inspection of the movies showed that GFP+ cells
in both genotypes displayed movement that was strikingly
biased in that the cells moved more on the rostrocaudal axis
than the left–right axis. To quantify this bias in GFP+ cell
movement, we aligned the rostrocaudal axis of the spinal cord
with the y-axis of our analysis grid and plotted the x and y
coordinates of the trajectories (Fig. 3F) from 30 minute long
intravital movies. Statistically significant difference between x
(CXCR6gfp/+, 7.5 ±0.9 µm; CXCR6gfp/gfp, 5.9 ±0.8 µm) and y
displacements (CXCR6gfp/+, 13.4 ±1.5 µm; CXCR6gfp/gfp, 13.0 ±
1.4 µm) between 1.5 to 9 min was found with p < 0.001 for both
genotypes (Fig. 3G). This analysis confirmed that movement
was biased along the rostrocaudal axis. GFP+ cells in CXCR6gfp/+
and CXCR6gfp/gfp genotypes exhibited similar rostrocaudal bias
between 1.5 to 9 min (p = 0.06 for x and p = 0.75 for y). Plotting
the y-displacement against the square root of time yielded a
linear plot consistent with a linear random walk, whereas
plotting the x-displacement against the square root of time
showed a plateau, consistent with confinement (Fig. 3H). This
analysis strongly indicated that the GFP+ cell movement is
described by a rostrocaudal biased random walk with confined
lateral movement.
3.4. Gray matter infiltration in EAE requires CXCR6
In contrast to MS, the monophasic mouse EAE model does
not display gray matter lesions (Zamvil and Steinman, 1990).
Therefore, it is not surprising that GFP+ cells were not detected
in the cortical gray matter while imaging through thinned skull
windows of CXCR6gfp/+ or CXCR6gfp/gfp mice with EAE. One
reason for the lack of subpial cortical lesions in EAE could be the
relatively shorter time frame of the EAE process compared to
that of MS. We hypothesized that in humans the neuroinflammation associated with MS may be combined in a long lifetime of small, sterile injuries in the cerebral cortex, which may
in combination serve as a nidus for cortical lesions in MS. In
order to model the effects of small sterile injury in the context of
EAE, we applied a previously established injury model in which
we induced acute gray matter injury using a brief exposure to
radiation from the femtosecond pulsed laser (Davalos et al.,
2005). We reported that such microscopic injuries do not
disrupt the blood–brain barrier or induce leukocyte infiltration
in healthy mice (Kim and Dustin, 2006). Consistent to this
previous finding, there was no T cell infiltration to gray matter
parenchymal injury focus in healthy mice based on the absence
of GFP+ cells detected through thinned skull windows by TPLSM
in injured CXCR6gfp/+ mice over 7 days observation period (data
not shown). We chose a stable period of EAE (days 16–20) to
monitor the injured brain intravitally through the thinned skull
window using TPLSM in anesthetized mice after inducing a single
sterile focal laser injury (∼15 µm in diameter). We found that in
mice with EAE, injury caused a significant infiltration by GFP+
cells around the injury site in the cortex of CXCR6gfp/+ mice, but
not, in CXCR6gfp/gfp mice. 3D rendering of such an injury site
94
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
revealed subpial infiltration of GFP+ cells around the laser
injury (arrow) (Fig. 4A). We excluded that meningeal inflammation induced by repeated thin skull procedures contributed
to the GFP+ cell infiltration by quantifying cells that are farther
than 60 μm from the skull (i.e. not in the meninges) (Fig. 4B,D)
(Kim and Dustin, 2006). GFP+ cell recruitment began as early as
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
95
Fig. 4. CXCR6 dependent recruitment of T cell into injured brain in EAE. The images are all 3D reconstructions that include the thinned skull (blue) the vasculature
(red) and the GFP+ T cells (green). (A) Area around a laser injury in a CXCR6gfp/+ mouse with laser injury at day 17 of EAE and imaged on day 27 (10 days post
injury). (B) Thinned skull control in CXCR6gfp/+ mouse with EAE showing an example of some GFP+ cells in the meninges, but few in the parenchyma. (C)
CXCR6gfp/gfp mouse with injury showing rare GFP+ cells in meninges and perivascular spaces 10 days after injury. Representative images of at least n = 3. Grid
scale for A–C, 30 µm. Arrows in A and C show the depth level of laser injury. (D) Number of GFP+ cells recruited in the brain parenchyma ≥ 60 µm below the bone
undersurface (Kim and Dustin, 2006) as a function of time in CXCR6gfp/+ and CXCR6gfp/gfp mice. Each curve represents a mouse. (E) Quantitative PCR of CXCL16
mRNA from whole brain tissue at various days after EAE induction and injury in CXCR6gfp/+ and CXCR6gfp/gfp mice (n = 3 mice per bar graph). Controls are mice
without EAE on days 15 and 26.
24 h after injury, peaked between days 4 and 10 in 4 different
CXCR6gfp/+ mice with EAE score 1–2, and then waned to basal
levels by day 21 after injury (Fig. 4D). In contrast, the same type
of injury did not recruit GFP+ T cells into the parenchyma of
CXCR6gfp/gfp mice with a similar EAE disease score to the
CXCR6gfp/+ mice (Fig. 4C,D). The increased infiltration of GFP+
cells into cortical injury sites in mice with EAE compared to
healthy mice may be due to increased abundance of GFP+
effector cells in circulation in addition to increased permeability
of the blood brain barrier induced by circulating cytokines from
the vaccination site and draining lymph nodes. To test this
possibility, we vaccinated CXCR6gfp/+ mice (n = 5) with
ovalbumin in CFA with pertussis toxin (on days 0 and 2). We
then induced laser injury into the barrel cortex on day 7 after
vaccination, the peak of this response, and found no infiltration
of GFP+ cells to the injured focus under these circumstances.
Therefore, the accumulation of GFP+ cells in the gray matter
after laser injury requires both CXCR6 and CNS inflammation.
This result is consistent with work by Perry and colleagues
(Matyszak and Perry, 1995, 1966a,b, 1998), in which BCG
implanted in the CNS did not elicit a T cell response unless mice
were vaccinated with CFA in the periphery.
3.5. EAE induces CXCL16 expression in the brain
Although the small size of the subpial laser lesion precluded
characterization of the GFP+ cells in the injury focus by
histology, global CXCL16 expression in the brain was evaluated
by qPCR (Fig. 4E). CXCL16 signal was undetectable in the brains
of healthy mice. However, the levels of CXCL16 mRNA increased
Fig. 3. Spinal cord T cell dynamics in EAE determined by TPSLM. (A and B) Representative intravital spinal cord 2-photon images from both sides of the posterior
spinal vein of a CXCR6gfp/+ mouse at day 14 of EAE. GFP+ cells are shown in white and the vasculature, in red. Tracks of the GFP+ cells are outlined from a timelapse movie spanning 30 min (see the corresponding Movie 1). Scale bar (A + B) = 50 µm. Average speed (C), confinement indices (D) and immobile fractions (E)
from CXCR6gfp/+ (n = 10 animals) and CXCR6gfp/gfp (n = 5 animals) mice with stable active disease at EAE days 11–18. (F) Tracks of CXCR6gfp/+ (black) and
CXCR6gfp/gfp (red) movement over 30 min show apparent rostrocaudal migrational bias (each panel represents one animal). The + direction in y is toward the
head. (G) Rostrocaudal (y) vs. left–right (x) displacements over 9 min for each track in n ≥ 4 animals and n ≥ 80 cells for each category. Scales for x and y values are
in µm. (H) Rostrocaudal (y) vs. left–right (x) displacements vs. square root of time for CXCR6gfp/+ (black) and CXCR6gfp/gfp (red) cell movement. Open and closed
symbols represent x and y-displacement, respectively (n ≥ 4 mice in each category). The asterisks signify statistical significance.
96
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
dramatically during EAE in both CXCR6gfp/+ and CXCR6gfp/gfp
mice with or without laser injury. Therefore, CXCL16 mRNA is
increased in EAE brain tissue, but was not sufficient on its own
to induce CXCR6+ immune cell infiltration. T cell accumulation
in the cortex required an additional event, provided in our
study by the sterile laser injury.
3.6. CXCR6+ cells decelerate and become confined close to the
injury focus
The dynamics of the accumulated GFP+ cells in the gray
matter was analyzed by 3D tracking of cell movements by TPLSM
through the thinned skull. Inspection of the movies revealed that
the GFP+ cells formed a highly dynamic swarm around the injury
site with cells farther from the injury moving faster than cells
close to the injury, which were sometimes arrested (Fig. 5). The
average speed of the distant cells was statistically indistinguishable from that of the cells in the spinal cord at 4.7±2.5 (day 4
post injury) and 5.4±1.6 (day 10 post injury) μm/min (mean±
s.d.) (Fig. 5C) but with less confinement than in the spinal cord
with a confinement index of 0.69±0.21 (day 4 post injury) and
0.67±0.26 (day 10 post injury) (Fig. 5D). In contrast, the cells
within 50 μm of the injury site decelerated to 2.2±1.1 (day 4
post injury) and 3.1±1.9 (day 10 post injury) μm/min (Fig. 5C)
and were more restricted in movement than both the cells
outside the 50 μm radius of injury and those in the spinal cord
with a confinement index of 0.42±0.26 (day 4 post injury) and
0.38±0.28 (Fig. 5D).
4. Discussion
We report a new intravital apparatus and methodology,
both simple and inexpensive, for TPLSM of the intact mouse
spinal cord. We then used this apparatus and the related
thinned skull methods for intact cerebral cortex to examine
the role of CXCR6 in the MOG (35–55) EAE-B6 mouse model.
There was no difference in disease course or intravital GFP+
cell dynamics in the spinal cord between CXCR6gfp/+ and
CXCR6gfp/gfp mice. However, we report for the first time a
preference for rostrocaudal movement by the infiltrating T
cells within the living spinal cord in both genotypes.
Surprisingly, a reversible recruitment of CXCR6gfp/+, but not
CXCR6gfp/gfp T cells, was induced in response to a focal sterile
injury inside the gray matter of the brain parenchyma during
EAE. Finally, we demonstrate that CXCL16 expression is
upregulated in the brain during EAE, but that this signal is not
sufficient to induce gray matter infiltration by GFP+ immune
cells in CXCR6gfp/+ mice without additional conditions
established by a small sterile injury.
4.1. Spinal cord intravital window
Our surgical strategy was practical for visualizing the intact
spinal cord. The thoracic spine curvature is naturally amenable
for maximizing the size of the intervertebral space. This method
is both accessible and adoptable in any laboratory; it is
inexpensive and does not require prolonged surgical training.
A minor disadvantage compared to laminectomy is that the
small field of view that limits the tracking distance (maximum
∼ 300 µm). However, we have successfully extended the
rostrocaudal field of view on either end of the intervertebral
space for an additional ∼300 µm by thinning adjacent laminar
bones in lieu of laminectomy to produce an intravital window
through the bone similar to the thinned skull procedure
(Grutzendler et al., 2002) (JVK, unpublished findings). The
quality of imaging is similar through the thinned bone and the
Fig. 5. Dynamics of GFP+ T cells in the gray matter. (A and B) TPLSM 40 µm z-stack through thin skull revealed a laser injury of about 15 μm in diameter at 50 µm
below the brain surface induced on day 17 of EAE in a CXCR6gfp/+ mouse. GFP+ cells are recruited to the gray matter injury focus on day 10 after injury; GFP+ cells
in green, vasculature in red, bone in blue, and the center of the autofluorescent injury site marked with a yellow asterisk in A. A and B are the same images with
tracks superimposed for injury proximal and distal cells (corresponds to Fig. 4A and Movie 2). C, D — Plots of cell speed and confinement coefficient as a function of
distance from the injury site (n = 3 mice, each category). Close and far indicate within and greater than 50 µm from injury focus.
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
97
intervertebral space. A second limitation of this method is that
due to the extensive dissection of most of the dorsal muscles to
expose the sides of the vertebral bone we can only recommend
it as a terminal procedure not amenable to longitudinal studies.
An apparatus that can hold the same points in the vertebral
bone, but not require extensive dissection of the muscle tissue
could allow longitudinal studies in the mouse spinal cord. A
third limitation is that the intravital window examines the
spinal cord under thoracic rather than lumbar spine, where the
most highly disease-correlated lesions are found (Mendel et al.,
1995). Therefore, our findings of T cell dynamics may not reflect
T cells in the disease-correlated EAE foci in the lumbar spine.
However, on staining for myelin basic protein (MBP) in the EAE
afflicted mice used in our studies, we found diminished MBP
staining in the white matter extending into the thoracic spinal
cord regions (data not shown). Thus, our imaging does reflect
areas that are infiltrated to an extent that leads to demyelination, although not as severe as in the lumbar region. By
imaging thoracic rather than lumbar areas we may observe T
cell dynamics at the advancing edge of lesions in EAE rather
than the core of demyelinated foci. A fourth limitation is that
the gray matter of the spinal cord cannot be imaged as they lie
deeper than 1 mm below the surface.
changes from the procedure at longer times that may increase T
cell arrest. Third, we may be visualizing this axial motion more
readily because the thoracic spinal cord is infiltrated by T cells,
but suffers less severe damage than the lumbar region, in which
the rostrocaudal tracks may be destroyed by advanced
demyelination.
Another possibility for the axial bias in T cell migration may be
due to newly formed extracellular matrix (ECM) structures and
cells in inflamed spinal cord. For example, during LCMV
meningitis ER-TR7+ cells on pial surface are extensively infected
and also appear to serve as a substrate for T cell migration (Kim
et al., 2009). In other examples, in both human MS tissue
(Mohan et al., 2008) and mouse CNS toxoplasmosis brain
(Wilson et al., 2009), T cells adhere to inflammation-induced
ECM. Such ECM and cellular networks induced in inflamed CNS
tissue could affect the biased T cell migration we observed in this
study and constitute an interesting body of future research to
explore in terms of examining sequestrations of specific adhesion
and chemokine molecules to these structures (Bajenoff et al.,
2006).
4.2. Rostrocaudal migrational bias
We have shown that CXCR6 deficient mice were equally
susceptible to EAE as CXCR6 positive mice and displayed normal
levels of T cell infiltration into the spinal cord. In this study we
have not examined maintenance of EAE in relapsing and
remitting mouse models of EAE and therefore these should be
examined for future studies. In contrast to our results with
CXCR6, anti-CXCL16 antibodies reduce EAE severity (Fukumoto
et al., 2004). Taken together, this suggests that the effects of
anti-CXCL16 are not mediated by blocking interaction with
CXCR6. Therefore, the effect of CXCL16 may be mediated by its
other function as a scavenger receptor for phosphatidylserine
on the surface of apoptotic cells, as well as, for oxidized lipids,
which are increased in injured tissue (Shimaoka et al., 2000).
Antigen presenting cells may utilize CXCL16 in uptake and
processing of myelin-associated antigens for MHC class IImediated presentation, which may be critical steps for disease
progression and epitope spreading (Vanderlugt et al., 1998).
CXCL16 expression is 10-fold higher in spinal cord than brain
during EAE. This may be attributed to accumulation of inflammatory cells in the spinal cord, including CD11b+ macrophages, which are known to have high levels of CXCL16
(Matloubian et al., 2000; Fukumoto et al., 2004; Ludwig et al.,
2005). However, inflamed endogenous cells such as astrocytes
(Ludwig et al., 2005) and endothelial cells (Hofnagel et al.,
2002) may also contribute to this difference. Under inflammatory conditions, astrocytes can to a lesser extent than microglia
transform into active antigen presenting cells in vitro (Aloisi
et al., 2000) and are certainly capable of innate immune
activation (Suh et al., 2007). The elevated CXCL16 levels in the
EAE may help to recruit CXCR6+ cells across the blood brain
barrier into injured gray matter foci but additional unknown
signals provided by injury are required for gray matter
infiltration. Our imaging study provides an interesting cellular
homing behavior to new injury foci in EAE. Based on our
findings and the current body of research literature we have
discussed many possibilities of future studies to explore the
exact mechanism for both the infiltration and the slower and
We report for the first time that the effector T cells migrate
preferentially along the rostrocaudal spinal cord axis during EAE.
There may be several mechanisms for this phenomenon. One
possibility is that T cells may be migrating along myelinated
axons or their extracellular matrix sheaths, which are oriented
along the rostrocaudal axis and this anatomical environment of
the spinal cord may play a similar role in guiding T cell migration
in the spinal cord as the stromal cell network in lymph nodes
(Bajenoff et al., 2006). Alternatively, rather than be guided,
barriers to lateral movement in the white matter may confine T
cells. It is difficult to distinguish these possibilities in the absence
of direct visualization of guiding or confining elements. The
stromal elements in the spinal cord white matter are oriented
with the fiber tracts paralleling the predominantly rostrocaudal
migration pattern we observe for GFP+ cells (Peters et al., 1991;
Huang et al., 2007). T cells may then preferentially encounter
APCs that are positioned along these tracks. This biased migration
guided or confined by oriented fiber tracks provides a mechanism for spread of inflammation within the spinal cord during
EAE (Peters et al., 1991; Huang et al., 2007). We note that
previous intravital TPLSM publications using laminectomy to
expose the rat spinal cord did not report this striking T cell
behavior thus far (Kawakami et al., 2005; Odoardi et al., 2007).
There are at least three possibilities for this difference. First, we
are not using a T cell transgenic system and the observed
infiltrating T cells in the spinal cord are presumably polyclonal
with a spectrum of affinities for locally expressed mMHC.
Therefore, we may be seeing a higher percentage of lower
affinity T cells and bystander T cells, which do not arrest significantly on antigen presenting cells and thus emphasize
scanning behavior. Second, the laminectomy performed in the
rat model may disrupt this pattern of movement. Consistent
with this, Engelhardt and colleagues reported that their laminectomized intravital spinal cord preparation was stable for only
∼30 min (Engelhardt et al., 2003), suggesting inflammatory
4.3. CXCR6 is not required for EAE
98
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
confined T cell dynamics adjacent to the gray matter injury in
EAE mice.
4.4. CXCR6 is required for T cell entry into injured the gray
matter during EAE
A limitation of most mouse EAE models is that gray matter
lesions are not observed. In MS, gray matter involvement
correlates with more severe disease. MRI and pathology
findings demonstrate that gray matter infiltration can be
significant in patients with relapsing-remitting disease (Audoin
et al., 2007; Lassmann et al., 2007) and dendritic cells are located
in the gray matter in chronic active disease (Cudrici et al., 2007).
Cortical pathology in early relapsing MS has been demonstrated
both by neuropathological evaluation and by magnetic resonance imaging studies showing gray matter atrophy at this
stage of disease (Bo et al., 2006; Summers et al., 2008). Recent
results demonstrate that subpial cortical lesions display marked
T cell accumulation early in MS (Moll, N., R.M.R., B. Kelley, J.E.
Parisi, and C. Lucchinetti, personal communication) — this is
often carried out by comparing human MS patients MRI atrophy
levels that follow previously infiltrated areas (Rudick et al.,
2009).
To determine if sterile injury can occur to promote gray
matter infiltration, we have extended the EAE mouse model by
inducing small focal injury followed by monitoring of CXCR6GFP+ cells. Our study suggests that CXCR6 could have a
potential role in gray matter infiltration in the context of MS.
CXCR6 was required for the accumulation of activated T
cells in the injured cerebral cortex in the presence of EAE. It is
not known how T cells enter these injury sites since we did not
directly observe the entry process. However, although it is also
possible that the T cells migrated from the nearby meninges or
over a greater distance from the underlying white matter, it is
most likely that recruitment was through parenchymal
microvessels (A. Flügel, N. Kawakami, personal communications). CXCR6-dependent accumulation may arise from
increased entry, decreased egress, or could also arise from
increased survival as previously shown for NKT cells (Geissmann et al., 2005). This response in the cortex was not
sustained and resolved over a period of weeks. This suggests
that the immune cell infiltration at the cortical injury site does
trigger a sustainable cycle of tissue injury and recruitment.
An ongoing controversy surrounds the role of trauma and
its relationship to triggering or worsening of EAE and MS
(Phillips et al., 1995; Jung et al., 2000; Marcondes et al., 2005;
Ling et al., 2006). Weller and colleagues reported exacerbated
disease following cortical cryolesions. However, these large
injuries appear to disrupt the blood brain barrier since
activated CD8+ cells are capable of entering cryolesioned
brain parenchyma independent of local antigen presence
(Ling et al., 2006). It is established at least in two rodent
models that injury trigger disease. In a MBP based mouse EAE
model, spinal cord injury initiates disease (Marcondes et al.,
2005). Antigen-specific T cell infiltration occurs along
Wallerian degeneration sites where microglia express class
II molecule I-A (Konno et al., 1990; Molleston et al., 1993). In
the latter model, the blood brain barrier is not affected by the
injury similarly to our model. Although injury has less effect
in provoking inflammation in the brain than in the spinal cord
and the brain (Schnell et al., 1999), systemic inflammation
can alter immunological barrier and therefore the effect of
CNS injury in part mediated by elaborations of in situ cytokine
productions (Sun et al., 2004; Teeling and Perry, 2009). The
small laser injury in the layer 1 barrel cortex in our hands is
the smallest injury reported to date in EAE rodent models and
demonstrated recruitment of CXCR6+ cells to the gray matter
of mice afflicted with EAE, although not surprisingly given the
size of injury, with no apparent side effects in the mouse or
statistically significant changes in EAE scores (data not
shown). We note, however that the injury site in the barrel
cortex might lead to sensory changes near the whiskers,
which is difficult to assess.
5. Conclusions
Our new intravital spinal cord method revealed a biased T
cell migration in EAE spinal cord. The mechanisms remain to
be further examined. Our results on CXCR6 suggest a two hit
model for vulnerability of the cerebral cortex to T cell
infiltration. CXCL16 expression pattern in the CNS suggests
that spinal cord infiltration spills over into the brain. In fact,
early T cell recruitment takes place through the choroid
plexus, which supplies the CSF to the brain and spinal cord
(Reboldi et al., 2009). CXCL16 expression in the EAE brain is
not itself sufficient to mediate T cell infiltration. Similarly,
laser injury alone, which sets up a cascade of astrocyte and
microglial responses (Davalos et al., 2005; Kim and Dustin,
2006) is also insufficient to recruit T cells into the parenchyma. However, when both of these hits are present, robust
CXCR6 dependent T cell infiltration is observed. This may be
due to a lowered threshold for traversing the blood brain
barrier. The prevalence of CXCR6 and CXCL16 expression in
human clinical specimens should be further evaluated as a
first step to assess the potential for anti-CXCR6 based
therapies in treating and/or preventing gray matter involvement in MS and promoting recovery of function by reducing
inflammatory T cell infiltration (De Jager and Hafler, 2007).
Laser injury is but a simple maneuver that was convenient in
our mouse model, but we emphasize that its relationship to
cortical lesions in MS is not established and therefore the
clinical significance of our results in terms of therapeutic
targeting of CXCR6 in treating or preventing cortical inflammation in MS will need further investigation.
Acknowledgments
We thank the following investigators: N. Kawakami and A.
Flügel, for demonstrating spinal cord imaging procedures in the
rat that contributed to methods development for this study; D.
Levy, for his advice on Fig. 1; S. Shen, for help with T cell
analysis; C.F. Brosnan (A.E.C.O.M., Bronx, NY), for discussions
about cortical expression of EAE antigen, myelin basic protein;
V.H. Perry ( Southampton University, U.K.), for discussions
about cryptic antigens and demyelination; W. Gan, for
continuing assistance with intravital TPLSM;. K. Hosiawa and
M. Matloubian (U.C.S.F., San Francisco, CA), for the CXCL16
qPCR protocol; and J. Zavadil of the NYUCI Genomics Core, for
assistance with qPCR analysis. This work was funded by NIH
grants R01 AI055037 (M.L.D.), P30 CA016087 (Core facilities), a
Dana Foundation Research Grant (M.L.D. and W. Gan),
Collaborative Multiple Sclerosis Research Center Award
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
CA1050-A-12 and New York State Spinal Cord Injury Research
grant (J.V.K.). The authors declare no conflicts of interest.
Author contribution
JVK and MLD designed and analyzed most of the
experiments and JVK performed most of the experiments.
Blinded scoring in all EAE mice for Fig. 2 was performed by NJ.
JVK designed the intervertebral space surgery. CET designed
the spinal cord clamp and performed immunohistochemistry
experiments in Supplementary Figs. 2, 4 and 5. RMR, JJL, LL
and MLD advised, analyzed and helped to write the
manuscript. JVK and MLD wrote the manuscript.
Appendix A. Supplementary data
Supplementary data and movies associated with this
article can be found, in the online version, at doi:10.1016/j.
jim.2009.09.007.
References
Aloisi, F., Serafini, B., Adorini, L., 2000. Glia-T cell dialogue. J. Neuroimmunol.
107, 111.
Audoin, B., Davies, G., Rashid, W., Fisniku, L., Thompson, A.J., Miller, D.H.,
2007. Voxel-based analysis of grey matter magnetization transfer ratio
maps in early relapsing remitting multiple sclerosis. Mult. Scler. 13, 483.
Bajenoff, M., Egen, J.G., Koo, L.Y., Laugier, J.P., Brau, F., Glaichenhaus, N.,
Germain, R.N., 2006. Stromal cell networks regulate lymphocyte entry,
migration, and territoriality in lymph nodes. Immunity 25, 989.
Bettelli, E., Pagany, M., Weiner, H.L., Linington, C., Sobel, R.A., Kuchroo, V.K.,
2003. Myelin oligodendrocyte glycoprotein-specific T cell receptor
transgenic mice develop spontaneous autoimmune optic neuritis. J.
Exp. Med. 197, 1073.
Bo, L., Vedeler, C.A., Nyland, H., Trapp, B.D., Mork, S.J., 2003. Intracortical
multiple sclerosis lesions are not associated with increased lymphocyte
infiltration. Mult. Scler. 9, 323.
Bo, L., Geurts, J.J., Mork, S.J., van der Valk, P., 2006. Grey matter pathology in
multiple sclerosis. Acta Neurol. Scand. Suppl. 183, 48.
Calabresi, P.A., Yun, S.H., Allie, R., Whartenby, K.A., 2002. Chemokine receptor
expression on MBP-reactive T cells: CXCR6 is a marker of IFNgammaproducing effector cells. J. Neuroimmunol. 127, 96.
Cudrici, C., Ito, T., Zafranskaia, E., Niculescu, F., Mullen, K.M., Vlaicu, S., Judge, S.I.,
Calabresi, P.A., Rus, H., 2007. Dendritic cells are abundant in non-lesional
gray matter in multiple sclerosis. Exp. Mol. Pathol. 83, 198.
Davalos, D., Grutzendler, J., Yang, G., Kim, J.V., Zuo, Y., Jung, S., Littman, D.R.,
Dustin, M.L., Gan, W.B., 2005. ATP mediates rapid microglial response to
local brain injury in vivo. Nat. Neurosci. 8, 752.
Davalos, D., Lee, J.K., Smith, W.B., Brinkman, B., Ellisman, M.H., Zheng, B.,
Akassoglou, K., 2008. Stable in vivo imaging of densely populated glia,
axons and blood vessels in the mouse spinal cord using two-photon
microscopy. J. Neurosci. Methods 169, 1.
De Jager, P.L., Hafler, D.A., 2007. New therapeutic approaches for multiple
sclerosis. Annu. Rev. Med. 58, 417.
Denk, W., Delaney, K.R., Gelperin, A., Kleinfeld, D., Strowbridge, B.W., Tank, D.W.,
Yuste, R., 1994. Anatomical and functional imaging of neurons using
2-photon laser scanning microscopy. J. Neurosci. Methods 54, 151.
Engelhardt, B., Vajkoczy, P., Laschinger, M., 2003. Detection of endothelial/
lymphocyte interaction in spinal cord microvasculature by intravital
videomicroscopy. Methods Mol. Med. 89, 83.
Faust, N., Varas, F., Kelly, L.M., Heck, S., Graf, T., 2000. Insertion of enhanced
green fluorescent protein into the lysozyme gene creates mice with
green fluorescent granulocytes and macrophages. Blood 96, 719.
Fukumoto, N., Shimaoka, T., Fujimura, H., Sakoda, S., Tanaka, M., Kita, T.,
Yonehara, S., 2004. Critical roles of CXC chemokine ligand 16/scavenger
receptor that binds phosphatidylserine and oxidized lipoprotein in the
pathogenesis of both acute and adoptive transfer experimental autoimmune encephalomyelitis. J. Immunol. 173, 1620.
Geissmann, F., Cameron, T.O., Sidobre, S., Manlongat, N., Kronenberg, M.,
Briskin, M.J., Dustin, M.L., Littman, D.R., 2005. Intravascular immune
surveillance by CXCR6+ NKT cells patrolling liver sinusoids. PLoS Biol. 3, e113.
Geurts, J.J., Barkhof, F., 2008. Grey matter pathology in multiple sclerosis.
Lancet Neurol. 7, 841.
Grutzendler, J., Kasthuri, N., Gan, W.B., 2002. Long-term dendritic spine
stability in the adult cortex. Nature 420, 812.
99
Hofnagel, O., Luechtenborg, B., Plenz, G., Robenek, H., 2002. Expression of
the novel scavenger receptor SR-PSOX in cultured aortic smooth
muscle cells and umbilical endothelial cells. Arterioscler. Thromb. Vasc.
Biol. 22, 710.
Huang, J.H., Cardenas-Navia, L.I., Caldwell, C.C., Plumb, T.J., Radu, C.G., Rocha, P.N.,
Wilder, T., Bromberg, J.S., Cronstein, B.N., Sitkovsky, M., Dewhirst, M.W.,
Dustin, M.L., 2007. Requirements for T lymphocyte migration in explanted
lymph nodes. J. Immunol. 178, 7747.
Jung, S., Aliberti, J., Graemmel, P., Sunshine, M.J., Kreutzberg, G.W., Sher, A.,
Littman, D.R., 2000. Analysis of fractalkine receptor CX(3)CR1 function
by targeted deletion and green fluorescent protein reporter gene
insertion. Mol. Cell. Biol. 20, 4106.
Kawakami, N., Nagerl, U.V., Odoardi, F., Bonhoeffer, T., Wekerle, H., Flugel, A.,
2005. Live imaging of effector cell trafficking and autoantigen recognition within the unfolding autoimmune encephalomyelitis lesion. J. Exp.
Med. 201, 1805.
Kim, J.V., Dustin, M.L., 2006. Innate response to focal necrotic injury inside
the blood–brain barrier. J. Immunol. 177, 5269.
Kim, J.V., Kang, S.S., Dustin, M.L., McGavern, D.B., 2009. Myelomonocytic cell
recruitment causes fatal CNS vascular injury during acute viral meningitis.
Nature 457, 191.
Kivisakk, P., Mahad, D.J., Callahan, M.K., Trebst, C., Tucky, B., Wei, T., Wu, L.,
Baekkevold, E.S., Lassmann, H., Staugaitis, S.M., Campbell, J.J., Ransohoff, R.M.,
2003. Human cerebrospinal fluid central memory CD4+ T cells: evidence for
trafficking through choroid plexus and meninges via P-selectin. Proc. Natl.
Acad. Sci. U. S. A. 100, 8389.
Kivisakk, P., Mahad, D.J., Callahan, M.K., Sikora, K., Trebst, C., Tucky, B., Wujek, J.,
Ravid, R., Staugaitis, S.M., Lassmann, H., Ransohoff, R.M., 2004. Expression of
CCR7 in multiple sclerosis: implications for CNS immunity. Ann. Neurol. 55,
627.
Konno, H., Yamamoto, T., Suzuki, H., Yamamoto, H., Iwasaki, Y., Ohara, Y.,
Terunuma, H., Harata, N., 1990. Targeting of adoptively transferred
experimental allergic encephalitis lesion at the sites of wallerian degeneration.
Acta Neuropathol. 80, 521.
Lassmann, H., Bruck, W., Lucchinetti, C.F., 2007. The immunopathology of
multiple sclerosis: an overview. Brain Pathol. 17, 210.
Ling, C., Sandor, M., Suresh, M., Fabry, Z., 2006. Traumatic injury and the presence
of antigen differentially contribute to T-cell recruitment in the CNS. J.
Neurosci. 26, 731.
Loetscher, M., Amara, A., Oberlin, E., Brass, N., Legler, D., Loetscher, P., D'Apuzzo,
M., Meese, E., Rousset, D., Virelizier, J.L., Baggiolini, M., Arenzana-Seisdedos,
F., Moser, B., 1997. TYMSTR, a putative chemokine receptor selectively
expressed in activated T cells, exhibits HIV-1 coreceptor function. Curr. Biol.
7, 652.
Ludwig, A., Schulte, A., Schnack, C., Hundhausen, C., Reiss, K., Brodway, N.,
Held-Feindt, J., Mentlein, R., 2005. Enhanced expression and shedding of
the transmembrane chemokine CXCL16 by reactive astrocytes and
glioma cells. J. Neurochem. 93, 1293.
Marcondes, M.C., Furtado, G.C., Wensky, A., Curotto de Lafaille, M.A., Fox, H.S.,
Lafaille, J.J., 2005. Immune regulatory mechanisms influence early
pathology in spinal cord injury and in spontaneous autoimmune
encephalomyelitis. Am. J. Pathol. 166, 1749.
Matloubian, M., David, A., Engel, S., Ryan, J.E., Cyster, J.G., 2000. A transmembrane
CXC chemokine is a ligand for HIV-coreceptor Bonzo. Nat. Immunol. 1, 298.
Matyszak, M.K., Perry, V.H., 1995. Demyelination in the central nervous system
following a delayed-type hypersensitivity response to bacillus CalmetteGuerin. Neuroscience 64, 967.
Matyszak, M.K., Perry, V.H., 1996a. A comparison of leucocyte responses to
heat-killed bacillus Calmette-Guerin in different CNS compartments.
Neuropathol. Appl. Neurobiol. 22, 44.
Matyszak, M.K., Perry, V.H., 1996b. Delayed-type hypersensitivity lesions in
the central nervous system are prevented by inhibitors of matrix
metalloproteinases. J. Neuroimmunol. 69, 141.
Matyszak, M.K., Perry, V.H., 1998. Bacillus Calmette-Guerin sequestered in
the brain parenchyma escapes immune recognition. J. Neuroimmunol.
82, 73.
Mendel, I., Kerlero de Rosbo, N., Ben-Nun, A., 1995. A myelin oligodendrocyte
glycoprotein peptide induces typical chronic experimental autoimmune
encephalomyelitis in H-2b mice: fine specificity and T cell receptor V
beta expression of encephalitogenic T cells. Eur. J. Immunol. 25, 1951.
Miller, M.J., Wei, S.H., Cahalan, M.D., Parker, I., 2003. Autonomous T cell trafficking
examined in vivo with intravital two-photon microscopy. Proc. Natl. Acad. Sci.
U. S. A. 100, 2604.
Mohan, H., Eisele, S., Krumbholz, M., Sixt, M., Lassman, H., Wekerle, H., Hohlfield, R.,
Meinle, E., 2008. Extracellular matrix in MS lesions: induction of fibrillar
collagens, biglycans and decorin in association with infiltrating immune cells.
J. Neuroimmunol. 203, 131.
Moll, N.M., Rietsch, A.M., Ransohoff, A.J., Cossoy, M.B., Huang, D., Eichler, F.S.,
Trapp, B.D., Ransohoff, R.M., 2008. Cortical demyelination in PML and
MS: similarities and differences. Neurology 70, 336.
100
J.V. Kim et al. / Journal of Immunological Methods 352 (2010) 89–100
Molleston, M.C., Thomas, M.L., Hickey, W.F., 1993. Novel major histocompatibility complex expression by microglia and site-specific experimental allergic encephalomyelitis lesions in the rat central nervous system
after optic nerve transection. Adv. Neurol. 59, 337.
Odoardi, F., Kawakami, N., Klinkert, W.E., Wekerle, H., Flugel, A., 2007. Bloodborne soluble protein antigen intensifies T cell activation in autoimmune
CNS lesions and exacerbates clinical disease. Proc. Natl. Acad. Sci. U. S. A.
104, 18625.
Peters, A., Palay, S.L., Webster, H., 1991. The Fine Structure of the Nervous System.
Oxford University Press.
Phillips, M.J., Weller, R.O., Kida, S., Iannotti, F., 1995. Focal brain damage
enhances experimental allergic encephalomyelitis in brain and spinal
cord. Neuropathol. Appl. Neurobiol. 21, 189.
Reboldi, A., Coisne, C., Baumjohann, D., Benvenuto, F., Bottinelli, D., Lira, S.,
Uccelli, A., Lanzavecchia, A., Engelhardt, B., Sallusto, F., 2009. C–C
chemokine receptor 6-regulated entry of TH-17 cells into the CNS
through the choroid plexus is required for the initiation of EAE. Nat.
Immunol. 10, 514.
Rudick, R.A., Lee, J.C., Nakamura, K., Fisher, E., 2009. Gray matter atrophy
correlates with MS disability progression measured with MSFC but not
EDSS. J. Neurol. Sci. 282, 106.
Sato, T., Thorlacius, H., Johnston, B., Staton, T.L., Xiang, W., Littman, D.R.,
Butcher, E.C., 2005. Role for CXCR6 in recruitment of activated CD8+
lymphocytes to inflamed liver. J. Immunol. 174, 277.
Schnell, L., Fearn, S., Klassen, H., Schwab, M.E., Perry, V.H., 1999. Acute
inflammatory responses to mechanical lesions in the CNS: differences
between brain and spinal cord. Eur. J. NeuroSci. 11, 3648.
Shakhar, G., Lindquist, R.L., Skokos, D., Dudziak, D., Huang, J.H., Nussenzweig, M.C.,
Dustin, M.L., 2005. Stable T cell-dendritic cell interactions precede the
development of both tolerance and immunity in vivo. Nat. Immunol. 6, 707.
Shimaoka, T., Kume, N., Minami, M., Hayashida, K., Kataoka, H., Kita, T.,
Yonehara, S., 2000. Molecular cloning of a novel scavenger receptor for
oxidized low density lipoprotein, SR-PSOX, on macrophages. J. Biol.
Chem. 275, 40663.
Shimaoka, T., Nakayama, T., Kume, N., Takahashi, S., Yamaguchi, J., Minami, M.,
Hayashida, K., Kita, T., Ohsumi, J., Yoshie, O., Yonehara, S., 2003. Cutting edge:
SR-PSOX/CXC chemokine ligand 16 mediates bacterial phagocytosis by
APCs through its chemokine domain. J. Immunol. 171, 1647.
Sims, T.N., Soos, T.J., Xenias, H.S., Dubin-Thaler, B., Hofman, J.M., Waite, J.C.,
Cameron, T.O., Thomas, V.K., Varma, R., Wiggins, C.H., Sheetz, M.P.,
Littman, D.R., Dustin, M.L., 2007. Opposing effects of PKCtheta and WASp
on symmetry breaking and relocation of the immunological synapse. Cell
129, 773.
Suh, H.S., Zhao, M.L., Rivieccio, M., Choi, S., Connolly, E., Zhao, Y., Takikawa, O.,
Brosnan, C.F., Lee, S.C., 2007. Astrocyte indoleamine 2, 3-dioxygenase is
induced by the TLR3 ligand poly(I:C): mechanism of induction and role
in antiviral response. J. Virol. 81, 9838.
Sumen, C., Mempel, T.R., Mazo, I.B., von Andrian, U.H., 2004. Intravital microscopy:
visualizing immunity in context. Immunity 21, 315.
Summers, M., Swanton, J., Fernando, K., Dalton, C., Miller, D.H., Cipolotti, L.,
Ron, M.A., 2008. Cognitive impairment in multiple sclerosis can be
predicted by imaging early in the disease. J. Neurol. Neurosurg.
Psychiatry 79, 955.
Sun, D., Newman, T.A., Perry, V.H., Weller, R.O., 2004. Cytokine-induced
enhancement of autoimmune inflammation in the brain and spinal cord:
implications for multiple sclerosis. Neuropathol. Appl. Neurobiol. 30, 374.
Teeling, J.L., Perry, V.H., 2009. Systemic infection and inflammation in acute
CNS injury and chronic neurodegeneration: underlying mechanisms.
Neuroscience 158, 1062.
Unutmaz, D., Xiang, W., Sunshine, M.J., Campbell, J., Butcher, E., Littman, D.R.,
2000. The primate lentiviral receptor Bonzo/STRL33 is coordinately
regulated with CCR5 and its expression pattern is conserved between
human and mouse. J. Immunol. 165, 3284.
Vanderlugt, C.L., Begolka, W.S., Neville, K.L., Katz-Levy, Y., Howard, L.M., Eagar, T.
N., Bluestone, J.A., Miller, S.D., 1998. The functional significance of epitope
spreading and its regulation by co-stimulatory molecules. Immunol. Rev.
164, 63.
Wensky, A.K., Furtado, G.C., Marcondes, M.C., Chen, S., Manfra, D., Lira, S.A.,
Zagzag, D., Lafaille, J.J., 2005. IFN-gamma determines distinct clinical
outcomes in autoimmune encephalomyelitis. J. Immunol. 174, 1416.
Wilson, E.H., Harris, T.H., Mrass, P., John, B., Tait, E.D., Wu, G.F., Pepper, M.,
Wherry, E.J., Dzierzinski, F., Roos, D., Haydon, P.G., Laufer, T.M.,
Weninger, W., Hunter, C.A., 2009. Behavior of parasite-specific effector
CD8+ T cells in the brain and visualization of a kinesis-associated
system of reticular fibers. Immunity 30, 300.
Xu, H.T., Pan, F., Yang, G., Gan, W.B., 2007. Choice of cranial window type for
in vivo imaging affects dendritic spine turnover in the cortex. Nat.
Neurosci. 10, 549.
Zamvil, S.S., Steinman, L., 1990. The T lymphocyte in experimental allergic
encephalomyelitis. Annu. Rev. Immunol. 8, 579.
Zamvil, S., Nelson, P., Trotter, J., Mitchell, D., Knobler, R., Fritz, R., Steinman, L.,
1985. T-cell clones specific for myelin basic protein induce chronic
relapsing paralysis and demyelination. Nature 317, 355.