A critical role for the sphingosine analog AAL-R in dampening the cytokine response during influenza virus infection

Abstract

Pulmonary tissue damage resulting from influenza virus infection is caused by both the cytolytic activity of the virus and the host immune response. Immune-mediated injury results from T cell-mediated destruction of virus-infected cells and by release of cytokines and chemokines that attract polymorphonuclear leukocytes (PML) and macrophages to the infected site. The cytokines/chemokines potentiate dendritic cell (DC) activation and T cell expansion, which further enhances local damage. Here we report that immune modulation by local administration to the respiratory tract of sphingosine analog AAL-R significantly dampens the release of cytokines and chemokines while maintaining protective neutralizing antibody and cytotoxic T cell responses. As a result there was a marked reduction of infiltrating PML and macrophages into the lung and resultant pulmonary tissue injury. DC maturation was suppressed, which limited proliferation of specific antiviral T cells in the lung and draining lymph nodes. Further, AAL-R was effective in controlling CD8(+) T cell accumulation in the lungs even when given 4 days after initiation of influenza virus infection. These data indicate that sphingosine analogs display useful potential for controlling the immunopathology caused by influenza virus.

Full text

A critical role for the sphingosine analog AAL-R in
dampening the cytokine response during influenza
virus infection
David Marsolais
a,1
, Bumsuk Hahm
b,c,1
, Kevin B. Walsh
b,1
, Kurt H. Edelmann
b
, Dorian McGavern
b
, Yasuko Hatta
d
,
Yoshihiro Kawaoka
d,e
, Hugh Rosen
a,2
, and Michael B. A. Oldstone
b,2
aDepartments of Chemical Physiology and Immunology, The Scripps Research Institute, La Jolla, CA 92037; bDepartment of Immunology and Microbial
Science, The Scripps Research Institute, La Jolla, CA 92037; cDepartments of Surgery, and Molecular Microbiology and Immunology, Center for Cellular
and Molecular Immunology, University of Missouri, Columbia, MO 65212; dDepartment of Pathobiological Sciences, University of Wisconsin-Madison,
Madison, WI 53706; and eDivision of Virology, Department of Microbiology and Immunology, and International Research Center for Infectious
Diseases, Institute of Medical Science, University of Tokyo, Tokyo 108-8639, Japan
Contributed by Michael B. A. Oldstone, December 12, 2008 (sent for review November 6, 2008)
Pulmonary tissue damage resulting from influenza virus infection
is caused by both the cytolytic activity of the virus and the host
immune response. Immune-mediated injury results from T cell-
mediated destruction of virus-infected cells and by release of
cytokines and chemokines that attract polymorphonuclear leuko-
cytes (PML) and macrophages to the infected site. The cytokines/
chemokines potentiate dendritic cell (DC) activation and T cell
expansion, which further enhances local damage. Here we report
that immune modulation by local administration to the respiratory
tract of sphingosine analog AAL-R significantly dampens the re-
lease of cytokines and chemokines while maintaining protective
neutralizing antibody and cytotoxic T cell responses. As a result
there was a marked reduction of infiltrating PML and macrophages
into the lung and resultant pulmonary tissue injury. DC maturation
was suppressed, which limited proliferation of specific antiviral T
cells in the lung and draining lymph nodes. Further, AAL-R was
effective in controlling CD8
ⴙ
T cell accumulation in the lungs even
when given 4 days after initiation of influenza virus infection.
These data indicate that sphingosine analogs display useful potential
for controlling the immunopathology caused by influenza virus.
immunopathology 兩dendritic cells 兩cytokine storm 兩T cells
The antiviral host response evolved to limit the spread of
infection at the cost of causing tissue injury. There is a balance
between the protective and injurious responses that leads either to
the purging of infectious virus and host rec overy or to severe disease
and even death. Thus, strategies to balance the antiviral immune
response in favor of host outcome need to be developed.
T cell response elicited early in the course of infection recognizes,
attacks and lyses virus-infected cells to eliminate potential factories
of progeny viruses (1–5). The protective inf luenza antibody re-
sponse is elicited later and plays a role in controlling re-infection (6,
7). The innate and adaptive immune systems release wide varieties
of cytokines and chemokines that activate and attract inflammator y
cells to the site of infection (2, 8, 9). However, these molecules can
also cause the host harm by a phenomenon known as cytokine
storm. Cytokine storm has been convincingly documented both in
experimental animals infected with the 1918/1919 and H5N1 in-
fluenza viruses (2, 10 –13) as well as in humans (14–18) succumbing
to H5N1 infection. Although antiviral drugs can be used to treat the
virus, a strategy to balance the resultant cytokine release and lung
injury while maintaining benefits of the antiviral protective immune
response is needed (19). Influenza virus replication is most often
limited primarily to the respiratory tract but the systemic signs and
symptoms of disease, e.g., fever, muscle pain and shakes, intestinal
tract involvement, are related to cytokine effects (20, 21).
We reported that intratracheal (i.t.) delivery of the sphingosine
analog AAL-R or its phosphate ester inhibited virus-specific T cell
responses to influenza-virus infection whereas, in contrast, neither
intraperitoneal (i.p.) delivery of AAL-R nor i.t. administration of
the non-phosphorylable stereoisomer AAL-S were effective (22).
Here, we extend those findings by determining the mechanism of
AAL-R down-modulation of the antiviral T cell response. Using a
mouse model of influenza virus infection, three findings are made.
First, local administration of sphingosine analog AAL-R to the
respiratory tract down-modulates the anti-influenza virus T cell
response and inhibits cytokine/chemokine release without altering
the formation, kinetics or titer of anti-inf luenza-neutralizing anti-
bodies. Limiting the navigation signals of cytok ines and chemokines
reduces the infiltration of PML and macrophages into the lung.
Second, AAL-R acts on pulmonary and draining lymph node
dendritic cells (DC). AAL-R significantly down-regulates major
histocompatibility (MHC) class I and II molecules, as well as
co-stimulatory B7–2 molecule from the surfaces of DC in the lung
and draining lymph nodes, leading to a significant decrease in
accumulation of virus-specific T cells in the lung. Third, AA L-R
therapy is effective when administered 4 days after initiation of
influenza virus infection. Thus, short-term modulation of S1P
receptors in lungs interferes with antigen presentation by DC and
provides an additional therapeutic approach in treating aberrant
cytokine production in influenza and likely other infectious respi-
ratory diseases.
Results
An In Vivo Model for Quantitating and Monitoring Migration of
Influenza Virus-Specific T Cells into the Lung. We established an in
vivo model system to study inf luenza virus-mediated host immune
responses by generating a recombinant influenza virus. A recom-
binant A/WSN/33 (WSN; H1N1) virus (FLU-LCMV) was engi-
neered to bear the immunodominant H-2D
b
CD8 (GP 33–41) and
I-A
b
CD4 (GP 65–77) T cell receptor epitopes of lymphocytic
choriomeningitis virus (LCMV) in its neuraminidase (NA) stalk
using reverse genetics (23) (Fig. 1A). Availability of transgenic (tg)
mice in which the T cell receptor (TCR) is specific for these LCMV
immunodominant epitopes allowed in vivo quantification of T cell
trafficking (24) and visualization of T cell immunobiology (25).
Author contributions: D. Marsolais, B.H., K.B.W., K.H.E., H.R., and M.B.A.O. designed
research; D. Marsolais, B.H., K.B.W., K.H.E., and D. McGavern performed research; Y.H.
and Y.K. contributed new reagents/analytic tools; D. Marsolais, B.H., K.B.W., K.H.E., D.
McGavern, H.R., and M.B.A.O. analyzed data; and D. Marsolais, B.H., K.B.W., H.R., and
M.B.A.O. wrote the paper.
The authors declare no conflict of interest.
1D. Marsolais, B.H., and K.B.W. contributed equally to this work.
2To whom correspondence may be addressed. E-mail: hrosen@scripps.edu or
mbaobo@scripps.edu.
This article contains supporting information online at www.pnas.org/cgi/content/full/
0812689106/DCSupplemental.
© 2009 by The National Academy of Sciences of the USA
1560–1565
兩
PNAS
兩
February 3, 2009
兩
vol. 106
兩
no. 5 www.pnas.org兾cgi兾doi兾10.1073兾pnas.0812689106

Intranasal (i.n.) or i.t. inoculation of 1 ⫻10
5
PFU of FLU-LCMV
resulted in expression of influenza viral antigens throughout the
lungs. As shown in Fig. 1C, at 6 days post-infection (dpi), viral
antigen is observed throughout the lung parenchyma. High power
examination showed viral antigen in bronchioles and epithelial cells
surrounding the airways (data not shown). Mice adoptively trans-
ferred with 2.5 ⫻10
4
Thy1.1
⫹
GP 33–41-restricted CD8
⫹
(GP33/
CD8
⫹
) T cells and 2.5 ⫻10
4
GFP-expressing GP 65–77-restricted
CD4
⫹
(GP65/CD4
⫹
) T cells and subsequently inoculated i.n. with
FLU-LCMV displayed pulmonary infiltration of virus-specific
CD8
⫹
and CD4
⫹
T cells (Fig. 1D) at six dpi. As shown in Fig. 1E
and F, pulmonary accumulation of GFP
⫹
GP33/CD8
⫹
T cells
occurred after 4 dpi, reached a plateau between 6 and 8 dpi, then
dramatically decreased at 10 dpi, as revealed by both fluorescence
imaging on whole lung sections (Fig. 1E) and on single-cell sus-
pension from lung tissue analyzed by f low cytometry (Fig. 1F). This
model system provided an opportunity for quantification of virus-
specific T cells and a mechanistic study of the interplay between
host immunity, inf luenza virus, and their modulation by chemical
probes.
Sphingosine Analog AAL-R Inhibits the Accumulation of Influenza
Virus-Specific T Cells in the Lung but Does Not Interfere with the
Production of Protective Neutralizing Antiiinfluenza Antibodies. Ear-
lier we reported that local administration of a single 0.1-mg/kg dose
of AAL-R by an i.t., but not i.p. route, specifically down-modulated
virus-specific T cell accumulation in the lung, whereas both routes
of delivery induced T cell recirculation from the blood to secondary
lymphoid tissues (22). This inhibition of T cell accumulation in the
lung depended on phosphorylation of AAL-R as the chiral enan-
tiomer molecule AAL-S that cannot be phosphorylated efficiently
in vivo was impotent in restricting T cell infiltration in the lung.
We now assessed the effect of AA L-R treatment on the virus-
specific immune responses. Treatment with 0.1 mg/kg AAL-R i.t.
following FLU-LCMV inoculation inhibited the accumulation of
total CD4
⫹
and CD8
⫹
T cells (Fig. 2A), as well as virus-specific T
cells in the lung (Fig. 2B). However, AAL-R i.t. administration had
no effect on the production of influenza virus-specific total immu-
noglobulins (Ig) (Fig. 2C) and IgM (data not shown) at 6, 10, 15, and
30 days following virus infection. Assays for influenza virus neu-
tralizing antibodies revealed that the amount of immunoglobulins
was not impaired by AAL-R administration when compared to
AAL-S and vehicle-treated groups (Fig. 2D). Administration of
AAL-R did not significantly increase viral burden within the lung,
but slightly delayed viral clearance (Fig. 2 E). By day 10 post-
infection, both vehicle and AAL-R-treated mice were negative for
infectious virus. These results indicate that AAL-R therapy given
locally into the respiratory tract interferes with neither influenza
virus replication nor the generation of protective antibodies but
significantly down-regulated the numbers of virus-specific T cells in
the lung. Although numbers of virus-specific CD8
⫹
T cells were
reduced, influenza virus infection was still controlled.
AAL-R Does Not Decrease FLU-LCMV Titers or Cytotoxic Activity of
Remaining Virus-Specific CD8
ⴙ
T Cells. We next determined how
AAL-R treatment would influence viral kinetics and ex vivo
cytotoxic T lymphocyte (CTL) activity following FLU-LCMV
infection. As expected, AAL-R reduced the pulmonary content of
CD8
⫹
T cells having the ability to produce IFN-
␥
in response to
GP33 peptide stimulation in vitro 7 dpi, and this effect was
maintained until 8 dpi (Table 1). Also, the kinetics of viral clearance
in mice treated with AAL-R was not significantly altered compared
to viral clearance observed in mice administered with VEH or
AAL-S (Table 1), suggesting the remaining anti-viral T cell activity
was sufficient to clear virus from the lungs. We further tested this
idea by measuring the cell-based cytotoxic activity of GP33-specific
CD8
⫹
T cells extracted from lung and spleen after treatment with
AAL-R, AAL-S or VEH (Fig. S1). CTL activity of virus-specific
CD8
⫹
T cells in the lung (Fig. S1A) or spleen (Fig. S1B) was not
adversely affected by AAL-R when compared with VEH and
AAL-S treatments. Intracellular cytokine staining revealed a sig-
nificant reduction in the number of virus-specific CD8
⫹
T cells in
AAL-R treated mice (Fig. S1C). Calculation of cytolytic activity on
a per cell basis revealed a significant increase in the CTL activity of
GP33-specific splenocytes in AAL-R-treated mice when compared
with AAL-S or vehicle treatments.
Decreased Accumulation of Virus-Specific T Lymphocytes in the Lung
Following AAL-R Treatment Is Associated with Suppression of Cyto-
kine Synthesis and Alleviation of Immunopathology. Next we assessed
the ability of AAL-R to regulate inf luenza virus-induced cytokines/
Fig. 1. Generation of recombinant influenza virus and pulmonary accumula-
tion of viral antigens and virus-specific T cells in vivo. (A) Reverse genetics strategy
to insert CD4 and CD8 immunodominant epitopes of LCMV glycoprotein (GP),
amino acids (aa) 33–41 (GP33) for CD8⫹T cells, and GP amino acid 65–77 (GP65)
for CD4⫹T cells, into the neuraminidase (NA) stalk of A/WSN/33 virus. (B) Cartoon
to display the administration of fluorochrome-labeled or phenotypically identi-
fiable congenic LCMV-specific GP33/CD8⫹T cells and GP65/CD4⫹T cells from T cell
receptor transgenic mice and instillation of influenza virus. (C) Distribution of
influenza viral antigens (green) in the whole lung at day 6 postinfluenza virus i.n.
administration. Nuclei appear in blue. (D) Simultaneous detection of adoptively
transferred CD4⫹(green) and CD8⫹T cells (red) specific for FLU-LCMV at day 6
post-influenza i.n. inoculation. Nuclei appear in blue. (E) Analysis of virus-specific
GFP⫹TCR transgenic GP33/CD8⫹T cells on day 4 (d4), d6, d8, and d10 after i.n.
FLU-LCMV infection. GFP⫹TCR transgenic GP33/CD8⫹T cells appear in green and
nuclei were stained with DAPI (blue). (F) By comparison to (E), flow cytometric
analysis of GFP⫹TCR transgenic GP33/CD8⫹T cells after i.n. FLU-LCMV infection.
The percentages of GFP⫹TCR transgenic GP33/CD8⫹T cells are depicted on
scattergrams. Eand Fshow representative samples of one mouse from a group of
three.
Marsolais et al. PNAS
兩
February 3, 2009
兩
vol. 106
兩
no. 5
兩
1561
MICROBIOLOGY

chemokines synthesis in vivo. We noted that upon FLU-LCMV
infection there was a significant up-regulation of multiple cytokines/
chemokines, most notably marked increase of interleukin (IL)-1
␣
,
IL-1
␤
, IL-6, IL-10, IL-12, monocyte chemoattractant protein
(MCP-1, CCL2), TNF-
␣
, macrophage inflammatory protein (MIP-
1
␣
, CCL3), granulocyte macrophage colony stimulating factor
(GM-CSF), and RANTES (CCL5) (Fig. 3A). Of interest IL-6 and
MCP-1 levels were strikingly increased in our model, an event also
observed in both H5N1-infected humans (16), as well as in ma-
caques (11) and mice (10) infected with the 1918/1919 recon-
structed influenza virus. Treatment with AAL-R significantly
inhibited the production of IL-1
␣
, IL-1
␤
, IL-6, IL-10, MCP-1,
TNF-
␣
, and GM-CSF 2 dpi, when compared with mice treated with
AAL-S (Fig. 3B). Histological examination of lungs revealed that
AAL-R, but not AAL-S, dramatically decreased the accumulation
of lymphocytes, leukocytes, monocytes and macrophages and in-
jury in the distal airways (Fig. 3C) while modestly enhancing
survival time (Fig. 3D). Thus, delivery of AAL-R i.t. diminished
cytokine-mediated histopathologic injury of the lungs during inf lu-
enza virus infection. Cytokine/chemokine suppression was associ-
ated with a decrease of GR-1
⫹
PML and F4/80
⫹
macrophages (Fig.
3E) into the lung which directly correlated with the diminished
alveolar cytopathology and infiltration in the lung. However, de-
spite AAL-R successfully down-modulating T cell-mediated injury
and cytokine/chemokine release there was no change in lung
A/WSN/33 viral titers even at 2 mg/kg of AAL-R compared with
treatment with AAL-S or vehicle (Fig. 3F). This indicated that
AAL-R neither disrupted viral replication nor the resultant pro-
tective antiviral T cell immune responses but did successfully
downmodulate immunopathology.
AAL-R Protects by Impairing the Antigen-Presenting Capacity of
Pulmonary DC. Upon pulmonary infection, inf luenza virus-specific
T cells are rapidly induced and proliferate in mediastinal lymph
node (MLN), then migrate to the infected sites, including the lungs
(26, 27). Given i.t., AAL-R significantly reduced the number of
virus-specific CD8
⫹
(Fig. 4A) and CD4
⫹
(Fig. 4B) T cells in the
MLNs 5 and 6 dpi with FLU-LCMV, indicating that AAL-R
inhibited clonal expansion of T cells upon influenza virus infection.
The number of Annexin V-positive virus-specific T cells in both
MLNs and lungs was not significantly altered by local treatment
with AAL-R (data not shown), suggesting that AAL-R altered T
cell stimulation by influencing antigen-presenting DC (28) rather
than deleting T cells.
We then focused on pulmonary DC and found that i.t. delivery
of AAL-R neither reduced the numbers of total DC (data not
shown) nor of specific DC subsets (Fig. 4Cand D). Moreover,
AAL-R did not interfere with the viability of CD11c
⫹
cells (Fig.
4E) in the lungs upon influenza virus infection. However, AAL-R
strongly suppressed influenza virus-induced DC activation in the
lungs as measured by down-regulation of MHC-I, MHC-II and
co-stimulatory molecule B7–2 on the surfaces of CD11c
⫹
cells.
AAL-R impaired maturation of CD11c
⫹
cell subsets
(CD11b
⫹
CD103
-
; CD11b
⫹
CD103
⫹
, CD11b
-
CD103
⫹
, and CD11b
-
CD103
-
) in the lung by reducing surface expression of MHC-I,
MHC-II and B7–2, with suppression of the CD11b
⫹
CD103
-
subset
being affected the most (Fig. 4Fand G) (data not shown). Stimu-
latory capacity of CD11c
⫹
cells was impaired by in vivo AAL-R
treatment, as confirmed by inefficient induction of virus-specific
CD8
⫹
T cell proliferation in vitro (Fig. 4H).
We found that impaired activation of DC in the lungs resulted in
Fig. 2. Effect of sphingosine analog on influenza virus-specific T cells and
antibodies as well as virus titers in the lung. (Aand B) Mice (n⫽4 mice per group)
were infected with 1 ⫻105PFU of FLU-LCMV and administered i.t. with vehicle
(VEH) or 0.1 mg/kg AAL-R, or in B, with VEH, 0.1 mg/kg of AAL-R, or isomer control
AAL-S. At 6 dpi, lungs were processed for flow cytometric analysis to obtain
numbers of CD4⫹and CD8⫹T cells in lungs (A) or percentage of virus-specific TCR
transgenic GP33/CD8⫹T cells among total CD8⫹T cells in lungs (B). (Cand D) Four
to six mice per group were infected with mock or FLU-LCMV and treated with
VEH, AAL-S or AAL-R (0.1 mg/kg) 1 hpi. (C) Influenza virus-specific total immu-
noglobulins in serum were assessed by ELISA at 6, 10, 15, and 30 dpi, mean ⫾SD.
(D) Titers of virus neutralizing antibodies in serum were evaluated 30 dpi, mean ⫾
SEM. (E) PFU/gram of lung tissue in normal, unaltered C57BL/6 mice following i.t.
inoculation of 1 ⫻105PFUs of FLU-LCMV and i.t. delivery of 0.1 mg/kg of AAL-R
or vehicle 1hpi. At day 8 following AAL-R treatment, 2 of 4 mice had detectable
levels of virus at 1 ⫻102PFU/gram of lung. The level of detection is 1 ⫻102PFU/g
tissue. Four mice were used per group per timepoint. P⬎0.23 for days 2, 4, and
6 after infection. ND: not detectable.
Table 1. Effect of AAL-R on antigen-specific CD8
ⴙ
T cells accumulation and viral titers
Treatment n
Day 5 after infection Day 7 after infection Day 8 after infection
GP33-specific cells Titer GP33-specific cells Titer GP33-specific cells Titer
VEH 4 4.1 ⫻10
2
⫾1.1* 14.3 ⫻10
4
⫾5.0
‡
7.6 ⫻10
3
⫾2.0 3.9 ⫻10
2
⫾0.8 3.0 ⫻10
4
⫾1.3 1.3 ⫻10
2
(1/4)
‡
AAL-S 4 3.3 ⫻10
2
⫾0.9 7.9 ⫻10
4
⫾2.0 4.7 ⫻10
3
⫾1.8 3.6 ⫻10
2
⫾0.9 3.8 ⫻10
4
⫾0.8 0.2 ⫻10
2
⫾0.03 (3/4)
AAL-R 4 1.5 ⫻10
2
⫾0.5 6.9 ⫻10
4
⫾1.6 1.8 ⫻10
3
⫾0.2* 26.0 ⫻10
2
⫾12.0 2.3 ⫻10
4
⫾0.8 10.1 ⫻10
2
⫾6.8 (2/4)
Mice were infected with 1 ⫻105PFU of FLU-LCMV and administered i.t. with VEH and or 0.1 mg/kg AAL-S or AAL-R 1 hpi. *,Pⱕ0.05 when compared to
vehicle-treated mice.
*Values are presented as average number ⫾SEM.
†Values are presented as average PFU/g of tissue ⫾SEM.
‡Virus-positive mice/n.
1562
兩
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0812689106 Marsolais et al.

subsequent inhibition of influenza virus-induced elevation of
CD11c
⫹
cell numbers (both CD11b
⫹
and CD8
␣
⫹
) in the MLNs
(Fig. S2A), as well as impaired activation of CD8
␣
⫹
DCs present
in the lymph nodes, which are efficient at priming CD8
⫹
T cells
(29). Intratracheal delivery of A AL-R also down-regulated MHC-I,
MHC-II and B7–2 co-stimulatory molecules on DC in MLN (Fig.
S2B). Thus, these results indicate that upon influenza virus infec-
tion, AAL-R locally administered to the respiratory tract disrupts
antigen-presenting DC network by blocking DC-mediated signal
transmission from the infection site to the MLNs, leading to a
dramatic decrease in T cell expansion.
AAL-R Therapy Administered at a Later Time Point During Influenza
Virus Infection Is Still Effective in Modulating the CD8
ⴙ
T Cell Response
to the Virus. The last experiment determined whether administering
0.1 mg/kg of AAL-R locally into the respiratory tract at 4 dpi was
effective in inhibiting T cell accumulation in the lung. As shown in
Fig. 5, six and eight days after giving 1 ⫻10
5
PFU of FLU-LCMV
the number of virus-specific CD8
⫹
T cells was significantly reduced
in the lung. In contrast, providing vehicle had no therapeutic effect.
Discussion
Here we make three main points. First, a single low-dose admin-
istration of the sphingosine analog AAL-R locally into the respi-
ratory airway, when provided at the time of inf luenza virus infec-
tion or after 4 days of infection, is able to effectively alter one of the
two major arms of the antiviral immune response. Virus-specific T
cell expansion in MLNs and accumulation in the lungs are signif-
icantly down-regulated whereas the production, kinetics and
amounts of protective neutralizing antibodies are not altered. In
addition, CD8
⫹
T cell cytotoxic activity is not impaired. Second,
AAL-R delivery in the respiratory tract can significantly decrease
the release of a variety of cytokines and chemokines known to
contribute to the cytokine storm effect, especially inhibition of IL-6
and MCP-1. The diminished synthesis of cytokines/chemokines
following AAL-R treatment correlates directly with a decreased
infiltration of PML and macrophages in the lung, resulting in less
cytopathology of alveolar cells and inflammatory driven congestion
of air spaces. Third, mechanistically, AA L-R acts on DC from both
draining lymph nodes and the lung. The drug inhibits DC accu-
mulation in the draining lymph nodes and down-regulates MHC
class I and II molecules as well as co-stimulatory molecule B7–2.
The decreased accumulation of DCs in the draining lymph nodes
combined with the decreased expression of stimulatory and co-
stimulatory molecules likely accounts for the diminished prolifer-
ation and expansion of T cells in draining lymph node and subse-
quent accumulation in the lung.
Injury to the lung accompanying influenza virus infection de-
pends on both the virus and action of the host’s immune response.
The virus being cytopathic lyses cells. In addition, infected cells may
apoptose thus removing the nidus of infection. Similarly, the
cytotoxic function of antiviral T cells is to kill virus-infected cells,
especially early in the infectious process, removing cells that make
progeny virus. However, by virus or immune-induced lysis of cells,
factors are released that attract scavenger PMLs and macrophages
to the infected site. These cells, while removing cellular debris, also
contribute to airway disease. Most important, the antiviral T cells
release cytokines/chemokines at the site of infection via their
specific recognition of infected targets. These factors then further
act to provide navigation signals to attract the PML and macro-
phages thereby potentiating the pulmonary injury. In addition, the
released cytokines/chemokines play the dominant role in causing
the generalized systemic manifestations of inf luenza virus infection.
Administration of sphingosine analog AAL-R strikingly im-
paired accumulation of DC in the draining lymph nodes, as well as
maturation of DC in the lung and MLN. There was a down-
regulation of MHC and co-stimulatory molecules required to
engage specific antiviral T cells. This led to diminished proliferation
and clonal expansion of T cells in the MLN. These effects could also
be shown with the phosphorylated S1P receptor agonist AFD-R
(22), suggesting that targeting S1P receptors locally in the lung and
on DC was sufficient to inhibit proinflammatory cytokine release
and virus-specific T cells expansion. Others have reported that
systemic delivery of sphingosine analog FTY720 can sequester
normal expanding lymphocytes in lymphoid organs in various
models (30–32) including during virus infection (33). However, this
was unlikely a mechanism in our studies for the following reasons.
First, elsewhere we documented that delivery of a single dose of the
sphingosine analog AA L-R (0.1 mg/kg) in the airways did not cause
sequestration of lymphocytes in draining lymphoid organs at 6 days
Fig. 3. AAL-R impairs cytokine release and leukocyte infiltration in the lungs
resulting in diminished alveolar inflammation and congestion but without alter-
ing infectious influenza virus titer. (Aand B) Multiplex ELISA was performed on
lung homogenates obtained at 2 dpi from uninfected mice (n⫽4) or mice
infected with 1 ⫻105PFU FLU-LCMV i.t. (n⫽12). Virus-infected mice were treated
with VEH or AAL-R (0.1 mg/kg) (n⫽6). Cytokines detected at levels of more than
40 pg/ml in supernatant of lung homogenates and significanlty increased by at
least two-fold by viral infection over uninfected mice are shown. (A) Fold increase
of cytokines/chemokines from FLU-LCMV-infected mice over that of uninfected
mice. (B) Percentage of inhibition caused by AAL-R (black bars) and AAL-S
treatment (white bars) relatively to vehicle-treated mice. Mean ⫾SEM. *,Pⱕ
0.05. (C) Naive C57BL/6 mice were (i) left untreated or (ii,iii,iv) infected i.t. with
1⫻105PFU of FLU-LCMV. One hpi mice were either administered i.t. with VEH (ii),
AAL-S (iii: 0.1 mg/kg) or AAL-R (iv: 0.1 mg/kg), and euthanized 6 days later. Lungs
were processed for hematoxylin and eosin staining. Representative fields are
shown. (Scale bars, 50
␮
m.) Results were equivalent in more than 20 fields
analyzed per group (⬎5 fields/section, 2 sections/mouse and 2 mice/group). (D)
Percent survival was observed with 13 mice per group from two combined
experiments. Mice were infected i.t. with 5 ⫻103PFU of A/WSN/33 virus. One hour
later, mice were administered i.t. with vehicle or AAL-R (0.1 mg/kg) and moni-
tored daily for mortality. (E) Decrease in PML (GR-1⫹) and macrophages (F4/80⫹)
infiltrates in the lung. Mice were infected with 1 ⫻105PFU of FLU-LCMV and then
received i.t. either vehicle or AAL-R (0.1 mg/kg). Two dpi, Macrophages (F4/80)
and PML (GR1) numbers were evaluated by flow cytometry. Pool of 2 experi-
ments, n⫽7–8, mean ⫾SEM; *significantly different from vehicle, P⬍0.05. (F)
Titers of A/WSN/33 in lung three days after i.n. infection with 1 ⫻105 PFU of
A/WSN/33 and i.t. administration of VEH, or 2 mg/kg of AAL-S and AAL-R. Virus
plaques were quantitated on MDCK cells. Mean ⫾SEM; *P⬍0.05.
Marsolais et al. PNAS
兩
February 3, 2009
兩
vol. 106
兩
no. 5
兩
1563
MICROBIOLOGY

after AAL-R treatment, but a decreased accumulation of virus-
specific T cells does occur in the lung (22). The latter was not
observed with systemic delivery of A AL-R. Further, i.p. delivery of
AAL-R, the route used by others to show sequestering of lympho-
cytes in lymphoid organs, was not associated with inhibition of DC
accumulation in draining lymph nodes (not shown). The fact that
MLN DC numbers were significantly altered by i.t. but not i.p.
treatment with AAL-R further indicates the likelihood of local
effects of the AAL-R analog on DC maturation and function
rather than changes of DC or lymphocyte migration from the
blood to MLN (34, 35). Thus, the effects of local administration
coupled with the failure of systemic dosing to induce suppression
of DC molecules indicate that lymphopenia alone could not
explain the immunomodulation we observed in the influenza
model. Further, we noted a dramatic decrease in numbers of
both CD11b
⫹
and CD8
␣
⫹
DC in MLN. Others have shown that
CD8
␣
⫹
DC are potent in stimulating T cells and do not originate
from the lung after inf luenza virus infection (26). Our results
thus suggest that the immunosuppressive effect of AAL-R is
caused by modulation of DC function rather than by altering T
cell trafficking.
Sphingosine analog did not inhibit the production of protective
antibody (Fig. 2) even in the occurrence of significant T cell
suppression. Generation of antibodies upon viral infection depends
on multiple factors including infection route, duration of viral
components’ presence in specific locations, contribution of various
anatomically distributed germinal centers (36), and dependency on
Fig. 4. Sphingosine analog AAL-R inhibits
both induction of influenza virus-specific T
cells in MLN and activation of DC in the lung.
(A) Number (left) and percentage (right) of
Thy1.1⫹TCR transgenic GP33/CD8⫹T cells in
MLN 5 dpi with mock or FLU-LCMV and i.t.
administration of VEH or AAL-R (0.3 mg/kg) 1
hpi. Representative of two different experi-
ments, both showed similar results, n⫽3–4
mice per group, mean ⫾SEM; *P⬍0.05. (B)
Number (left) and percentage (right) of
Thy1.1⫹TCR transgenic GP65/CD4⫹T cells in
MLN 6 dpi with mock or FLU-LCMV and i.t.
instillation with VEH or AAL-R (0.1 mg/kg) 1
hpi. n⫽4 mice per group, mean ⫾SEM; *P⫽
0.07. Number of (C) CD11c⫹CD11b⫹CD103-
cells and (D) CD11c⫹CD11b-CD103⫹cells in
the lungs at 2 dpi with mock or FLU-LCMV and
i.t. treatment with VEH or 0.1 mg/kg AAL-R 1
hpi. (E) Number of annexin-V⫹CD11c⫹cells in
lungs 12 h after infection with mock or FLU-
LCMV, and i.t. treatment with VEH or 0.1
mg/kg AAL-R 1 hpi. Representative of two
different experiments, all showed similar re-
sults, n⫽4 mice per group; *P⬍0.05. (Fand
G) Mean fluorescence intensity (MFI) of
MHC-I (H-2Kb), MHC-II (I-Ab) and B7–2 on (F)
CD11b⫹DC and (G) CD103⫹DC was analyzed.
Cells were isolated from lungs 2 dpi with
FLU-LCMV and i.t. treatment with VEH or
AAL-R (0.1 mg/kg) 1 hpi. n⫽4 mice per group;
mean ⫾SEM. *,P⬍0.05. (H) Proliferation of
T cells cultured alone (top); with DC from
FLU-LCMV ⫹VEH treated mice (center); with
DC from FLU-LCMV ⫹AAL-R-treated mice
(bottom). See SI Materials and Methods for
details. Representative of 3 different experi-
ments, all showed similar results.
Fig. 5. Sphingosine analog AAL-R administered i.t. 4 days after FLU-LCMV
infection is effective in limiting the accumulation of virus-specific CD8⫹T cells
in the lungs. (A) Cartoon of the experimental protocol. Mice, 4 per group, were
infected with mock or 1 ⫻105PFU of FLU-LCMV i.t. followed 4 days later by i.t.
instillation of VEH or AAL-R (0.1 mg/kg). (B) Number of GP33/CD8⫹T cells in
lungs is shown. Mean ⫾SEM; *,P⫽0.09; **,P⫽0.07.
1564
兩
www.pnas.org兾cgi兾doi兾10.1073兾pnas.0812689106 Marsolais et al.

T helper cells. Currently, the effect of systemic delivery of sphin-
gosine analog on B cell response is controversial (33, 37), although
no data are available for antibody response affected by local
treatment of the analog. It is possible that local AAL-R-mediated
split of B and T cell responses was caused by partial blockade of host
immunity to inf luenza virus as: 1) the sphingosine analog was
shown not to completely inhibit the release of cytokines upon
influenza virus infection (Fig. 3B) (22); 2) remaining virus-specific
CD4
⫹
T cells (Fig. 4B) might be sufficient to reach the threshold for
the generation of antibodies dependent on T helper cells; 3) there
was sufficient effector T cell activity to balance inf luenza virus
replication in the lung as infectious virus titers in the lung were
equivalent with AAL-R, A AL-S or vehicle treatment (Fig. 2E,Fig.
3F, and Table 1). Recently, other investigators showed that CD11c
hi
DC depletion upon influenza virus infection caused inhibition of T
cell expansion but did not affect antibody production (38), which is
consistent with our data. We are currently investigating local
modulation of S1P receptors and the mechanism for separation of
B cell response from T cell immunity.
Lastly, the observation that with AAL-R therapy virus titers in
the lung were not significantly increased despite the down-
modulation of numbers of antiviral-specific T cells, along with
diminished cytokine/chemokine release and decreased injur y to the
lung suggest that combinatory therapy using antiviral drugs in
concert with S1P analogs would be a preferred therapy. We are
currently evaluating this possibility in mice models of vir ulent H5N1
influenza virus infection. Further, DC are known to contain at least
four of the five S1P receptors. Dissection of the pulmonary DC S1P
receptor(s) involved could suggest the design of more effective
sphingosine analogs for treatment of lung inflammatory disorders.
Materials and Methods
Mice. Mice were bred and maintained in a closed breeding facility at The Scripps
Research Institute. C57BL/6, C57BL/6 Thy1.1⫹DbGP33–41 TCR-tg, C57BL/6
GFP⫹DbGP33–41 TCR-tg, C57BL/6 GFP⫹I-AbGP61– 80 TCR-tg, and C57BL/6
Thy1.1⫹I-AbGP61–80 TCR-tg were used in this study. The handling of all mice
conformed to the requirements of the National Institutes of Health and The
Scripps Research Institute’s Institutional Animal Care and Use Committee.
Cell Transfer. Virus-specific CD8⫹or CD4⫹T cells were enriched by negative
selection using Stemsep kit (Stemcell technologies), from the spleen of transgenic
mice. Cells (5 ⫻104) were then injected in the tail vein of congenic mice 1 day
before infection. When both CD8⫹and CD4⫹cells were transferred, 2.5 ⫻104of
each was used. To exclude any effect of transferred cells on the immunopatho-
logical response triggered by influenza virus infection, Figs. 2 C–E,3A–F, and 4
C–Hwere conducted without prior adoptive transfer of lymphocytes.
Viruses. To generate a recombinant WSN mutant virus (FLU-LCMV), we inserted
the LCMV immunodominant T cell-specific GP33 and GP65 tandem sequence
AAGGCTGTCTACAATTTTGCCACCTGTGGGGGACGCACAAUGGGTCTTAAGGG-
ACCCGACATTTACAAAGGAGTTTACCAATTTAAGTCAGTGGAGTTTGAT between
nucleotides 145 and 146 of WSN NA gene. Insertion of up to 28 aa into the NA
stalk does not impair NA function but insertion of more than 12 aa attenuates
the virus. A/WSN/33 (WSN; H1N1) and FLU-LCMV were generated by using
plasmid-based reverse genetics (23). Viruses were amplified and plaqued on
Madin-Darby Canine Kidney (MDCK) cells.
Statistical Analysis. Unless otherwise stated, bars represent means ⫾SEM and
averages were compared using a bidirectional unpaired Student’s ttest with a 5%
significance level. Star (*) was used to mark significant differences between 2
groups unless otherwise stated.
For specific information regarding the analytic assays performed in this study,
please consult SI Text.
ACKNOWLEDGMENTS. This is Publication Number 19725 from the Department
of Immunology and Microbial Science, Infectology and Chemical Physiology and
Immunology; and The Scripps Research Institute Molecular Screening Center, The
Scripps Research Institute (TSRI). This work was supported in part by USPHS grants
AI074564 (MBAO, HR, YK, BH, DM, KW), AI009484 (MBAO), AI05509 (HR),
AI069274 (YK), and NIMH-074404 (HR). YK and YH are also supported by Grants-
in-Aid from the Ministries of Education, Culture, Sports, Science and Technology
and of Health, Labor, and Welfare of Japan; HR is supported in part by a grant
from Kyorin Pharmaceutical Company. DM is supported by Le Fonds de la Re-
cherche en Sante du Quebec, Canada. The authors would like to acknowledge the
technical assistance of Megan Welch and Nora Leaf. The authors thank the TSRI
Flow Cytometry Core Facility; and Dusko Trajkovic for histological analyses.
1. Whitton JL, Oldstone MB (2001) in Fields virology, 4th ed., eds Knipe DM, Howley PM,
Griffin DE, Lamb RE, Martin MA, Rolzman B, Straus SE (Lippincott Williams & Wilkins,
Philadelphia), pp 285–320.
2. La Gruta NL, Kedzierska K, Stambas J, Doherty PC (2007) A question of self-preservation:
immunopathology in influenza virus infection. Immunol Cell Biol 85:85–92.
3. Graham MB, Braciale TJ (1997) Resistance to and recovery from lethal influenza virus
infection in B lymphocyte-deficient mice. J Exp Med 186:2063–2068.
4. Zinkernagel RM, Althage A (1977) Antiviral protection by virus-immune cytotoxic T
cells: infected target cells are lysed before infectious virus progeny is assembled. J Exp
Med 145:644– 651.
5. Moskophidis D, Kioussis D (1998) Contribution of virus-specific CD8⫹cytotoxic T cells
to virus clearance or pathologic manifestations of influenza virus infection in a T cell
receptor transgenic mouse model. J Exp Med 188:223–232.
6. Goulding J, et al. (2007) Respiratory infections: do we ever recover? Proc Am Thorac Soc
4:618– 625.
7. Gerhard W (2001) The role of the antibody response in influenza virus infection. Curr
Top Microbiol Immunol 260:171–190.
8. Biron CA, Sen GC (2007) in Fields Virology, 5th ed., eds Knipe DM, Howley PM, Griffin
DE, Lamb RE, Martin MA, Rolzman B, Straus SE (Lippincott Williams & Wilkins, Phila-
delphia), pp 249–278.
9. Braciale TJ, Hahn YS, Burton DR (2007) in Fields Virology, 5th ed., eds Knipe DM,
Howley PM, Griffin DE, Lamb RE, Martin MA, Rolzman B, Straus SE (Lippincott Williams
& Wilkins, Philadelphia), pp 279–326.
10. Kash JC, et al. (2006) Genomic analysis of increased host immune and cell death
responses induced by 1918 influenza virus. Nature 443:578–581.
11. Kobasa D, et al. (2007) Aberrant innate immune response in lethal infection of
macaques with the 1918 influenza virus. Nature 445:319–323.
12. Cheung CY, et al. (2002) Induction of proinflammatory cytokines in human macro-
phages by influenza A (H5N1) viruses: a mechanism for the unusual severity of human
disease? Lancet 360:1831–1837.
13. Tumpey TM, et al. (2005) Characterization of the reconstructed 1918 Spanish influenza
pandemic virus. Science 310:77–80.
14. Beigel JH, et al. (2005) Avian influenza A (H5N1) infection in humans. N Engl J Med
353:1374–1385.
15. Uiprasertkul M, et al. (2007) Apoptosis and Pathogenesis of Avian Influenza A (H5N1)
Virus in Humans. Emerg Infect Dis 13:708–712.
16. de Jong MD, et al. (2006) Fatal outcome of human influenza A (H5N1) is associated with
high viral load and hypercytokinemia. Nat Med 12:1203–1207.
17. To KF, et al. (2001) Pathology of fatal human infection associated with avian influenza
A H5N1 virus. J Med Virol 63:242–246.
18. Peiris JS, et al. (2004) Re-emergence of fatal human influenza A subtype H5N1 disease.
Lancet 363:617–619.
19. Pearson H (2006) Controlling immune response may cut bird flu death rate. Nat Med
12:1105.
20. Hilleman MR (2002) Realities and enigmas of human viral influenza: pathogenesis,
epidemiology and control. Vaccine 20:3068–3087.
21. Eccles R (2005) Understanding the symptoms of the common cold and influenza. Lancet
Infect Dis 5:718–725.
22. Marsolais D, et al. (2008) Local not systemic modulation of dendritic cell S1P receptors in
lung blunts virus-specific immune responses to influenza. Mol Pharmacol 74:896–903.
23. Neumann G, et al. (1999) Generation of influenza A viruses entirely from cloned cDNAs.
Proc Natl Acad Sci USA 96:9345–9350.
24. Christen U, et al. (2004) A viral epitope that mimics a self antigen can accelerate but not
initiate autoimmune diabetes. J Clin Invest 114:1290–1298.
25. McGavern DB, Christen U, Oldstone MB (2002) Molecular anatomy of antigen-specific
CD8(⫹) T cell engagement and synapse formation in vivo. Nat Immunol 3:918–925.
26. Belz GT, et al. (2004) Distinct migrating and nonmigrating dendritic cell populations
are involved in MHC class I-restricted antigen presentation after lung infection with
virus. Proc Natl Acad Sci USA 101:8670– 8675.
27. Baumgarth N, Kelso A (1996) Functionally distinct T cells in three compartments of the
respiratory tract after influenza virus infection. Eur J Immunol 26:2189–2197.
28. Steinman RM, Hemmi H (2006) Dendritic cells: translating innate to adaptive immunity.
Curr Top Microbiol Immunol 311:17–58.
29. Allan RS, et al. (2006) Migratory dendritic cells transfer antigen to a lymph node-
resident dendritic cell population for efficient CTL priming. Immunity 25:153–162.
30. Budde K, et al. (2006) FTY720 (fingolimod) in renal transplantation. Clin Transplant 20
Suppl 17:17–24.
31. Mandala S, et al. (2002) Alteration of lymphocyte trafficking by sphingosine-1-
phosphate receptor agonists. Science 296:346–349.
32. Sanna MG, et al. (2004) Sphingosine 1-phosphate (S1P) receptor subtypes S1P1 and
S1P3, respectively, regulate lymphocyte recirculation and heart rate. J Biol Chem
279:13839–13848.
33. Pinschewer DD, et al. (2000) FTY720 immunosuppression impairs effector T cell pe-
ripheral homing without affecting induction, expansion, and memory. J Immunol
164:5761–5770.
34. Lan YY, et al. (2005) The sphingosine-1-phosphate receptor agonist FTY720 modulates
dendritic cell trafficking in vivo. Am J Transplant 5:2649–2659.
35. Maeda Y, et al. (2007) Migration of CD4 T cells and dendritic cells toward sphingosine
1-phosphate (S1P) is mediated by different receptor subtypes: S1P regulates the functions
of murine mature dendritic cells via S1P receptor type 3. J Immunol 178:3437–3446.
36. Moyron-Quiroz JE, et al. (2006) Persistence and responsiveness of immunologic mem-
ory in the absence of secondary lymphoid organs. Immunity 25:643–654.
37. Han S, et al. (2004) FTY720 suppresses humoral immunity by inhibiting germinal center
reaction. Blood 104:4129– 4133.
38. GeurtsvanKessel CH, et al. (2008) Clearance of influenza virus from the lung depends
on migratory langerin⫹CD11b- but not plasmacytoid dendritic cells. J Exp Med
205:1621–1634.
Marsolais et al. PNAS
兩
February 3, 2009
兩
vol. 106
兩
no. 5
兩
1565
MICROBIOLOGY