| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Immunology and Divisions of
* Allergy and Immunology, and
Pulmonary and Critical Care Medicine, Department of Internal Medicine, Washington University School of Medicine, and
Howard Hughes Medical Institute, St. Louis, MO 63110
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although considerable information exists regarding the mechanisms controlling the differentiation and secretory profile of Th2 cells, little is known about the events that lead to the selective recruitment of Th2 cells into the lungs and airways. Our group has shown that, when it is limited to two 30-min aerosol treatments on a single day, airway challenge of naive mice with1% Ag induces recruitment of adoptively transferred resting Th2 cells to the lung’s draining paratracheal lymph nodes (LN),5 but does not induce localization of these cells to the lung or stimulate airway inflammation. In contrast, adoptively transferred Th1 cells do respond to this 1-day dose of aerosolized Ag and localize to the lungs, creating tissue inflammation (7). This inability of resting Th2 cells to initiate a tissue inflammatory response can be overcome by administering repetitive doses of the same concentration of Ag (R. Stephens, unpublished data and Ref. 8); however, studies by other investigators have shown that Th2 cells also enter other Ag-challenged tissues much less readily than do Th1 cells. This has been shown in the skin (9), peritoneal cavity (10), and pancreatic islets (11). In experimental autoimmune gastritis, the data suggest that differential migration of Th1 and Th2 cells may actually play a role in establishing the local cytokine milieu since IL-4-secreting cells are produced by activation in the local LN but are then excluded from the effector site, the gastric mucosa (12). In all of these tissues, Th1 cells are found in substantially higher numbers than Th2 cells. D’Ambrosio et al. (13) have suggested that this means that Th2 cells have a higher threshold for entry into effector sites than do Th1 cells. Th1 cells and Th2 cells manifest differential expression of many adhesion receptors. These include receptors for chemokines (14, 15), integrins (16), and selectin ligands (9). Differential expression of these molecules probably determines the signaling threshold for migration of each cell population into different types of effector sites.
The Th1 and Th2 cell subsets differ importantly in their
effector functions and in some instances appear to be
counterregulatory. For example, although IL-4 is a growth factor for
Th2 cells, it can also inhibit the production of the Th1
cytokine IFN-
(17). Although IFN- activates
macrophages and directs B cells to class switch to the production of
distinct Th1-specific Ab isotypes, it can also block switching to
the Th2 isotypes, IgG1 and IgE, and can directly antagonize
recruitment of eosinophils (18, 19). This
cross-regulation, however, may not lead to exclusively polarized
responses. Many responses are characterized by participation by both
Th1 and Th2 cells and cytokines (e.g., the host response
against malaria (20)), although one Th subset
generally predominates in the response to a given Ag determining the
nature of the inflammatory response and potentially the outcome of
infection.
Our laboratory has previously observed that challenge of OVA-sensitized
mice using an OVA aerosol leads to accumulation of Th1 cells in
the lungs before accumulation of Th2 cells in the airways
(7). Although the dominant immune effector functions
associated with asthmatic inflammation are Th2 in character, we
have observed that when Th2 cells are adoptively transferred into
a naive mouse, airway challenge does not induce recruitment of the
transferred Th2 cells into the airways. This suggests that
Th2 cells are on their own poorly able to initiate a tissue
inflammatory response when Ag is presented at low doses or for short
time periods. This adoptive transfer model clearly showed that fully
differentiated Th cells remain preferential producers of IL-4 or
IFN- in vivo after transfer (Ref. 7 , and see Fig. 6
).
Interestingly, when Th1 cells and Th2 cells are transferred
together into a naive host, airway challenge induces the recruitment of
both Th1 cells and Th2 cells to the airway, resulting in the
accumulation of large numbers of eosinophils (21). Thus,
the Th2 cells that on their own appear poorly competent to
initiate a tissue inflammatory response are fully competent to enter
the tissue and drive the local inflammatory response to one of Th2
character if the environment is prepared, in this case by Th1
cells. These findings substantiate the hypothesis that Th1 and
Th2 responses are not exclusively counterregulatory.
|
. Similar
results in mice show that infection with one of several viruses can
exacerbate airway inflammation and hyperresponsiveness
(23, 24, 25). These authors and others have proposed various
mechanisms by which the tissue inflammation resulting from viral
infections could lead to decreased lung function in asthmatics, but
none of these accounts for the allergic nature of the asthma
exacerbation (reviewed in Ref. 26). Our previous
observation that Th1 cells can provide help for the recruitment of
Th2 cells and consequently for the development of allergic
inflammation led us to hypothesize that a major mechanism underlying
the allergic phenotype of virally induced exacerbations of asthma could
involve regulation of Th2 cell recruitment. This recruitment
effect could depend on viral Ag-specific Th1 cells. Alternatively,
it could be a manifestation of a generalized change in the airway
environment that allows for nonspecific recruitment of immune
cells. The current experiments were designed to test whether respiratory virus infection could provide a stimulus for the recruitment of Th2 cells to the airway and to define the role of the T cell’s cognate Ag in this recruitment. Using OVA-specific transgenic Th2 cells, we show that respiratory infection with the mouse parainfluenza virus, commonly designated as Sendai virus, can stimulate recruitment of adoptively transferred Th2 cells to the lungs. Additional experiments using adoptive transfer of hen egg lysozyme (HEL)-specific Th1 cells along with OVA-specific Th2 cells demonstrate that recruitment of Th2 cells can be triggered by simultaneous Th1-type inflammation in the absence of the Th2 Ag. These experiments provide a mechanism by which the host response to respiratory virus infection may support the recruitment to the airways and activation of an unrelated population of Th2 cells. Furthermore, they demonstrate the Ag-nonspecific recruitment of Th2 cells to sites of tissue inflammation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Female BALB/cAnHsd mice obtained from Harlan Breeders
(Indianapolis, IN) were used between 6 and 8 wk of age. DO11.10 mice
(27) that are transgenic for a V
8.2 TCR that recognizes
chicken OVA peptide 323–339 in the I-Ad class II
MHC protein were generously provided by K. Murphy (Washington
University, St. Louis, MO). Thy1.1+ DO11.10 mice
were generated as previously described (21). For these
experiments, the DO11.10 strain on the BALB/c background was crossed
with B10.BR mice (The Jackson Laboratory, Bar Harbor, ME) to generate
(DO11.10 x B10.BR)F1 mice. 3A9 mice (on the
B10.BR background) that are transgenic for a V8.2 TCR that
recognizes HEL peptide 48–62 on I-Ak
(28) were a generous gift from E. Unanue (Washington
University). For all of the experiments described here, these mice and
B10.BR nontransgenic mice were crossed with BALB/cAnHsd (Harlan
Breeders) to generate F1 mice. All mice were kept
in microisolator cages in the specific pathogen-free facility of the
Division of Comparative Medicine at the Washington University Medical
Center. Experimental infections with Sendai virus were performed in a
biohazard containment facility. Mice were supplied with sterile chow
and water ad libitum. All animal procedures were approved by the
Washington University Institutional Committee for the humane use of
laboratory animals.
Culture and differentiation of Th cells
T cells were cultured in complete IMDM (T cell medium, all
components from Invitrogen, Carlsbad, CA) as previously described
(21). Single-cell suspensions were prepared from the
spleens of the DO11.10 and 3A9 mice using a disposable mesh (BD Falcon,
San Jose, CA) and were cultured at 2.5–5 x
106 cells/ml with 0.3 µM OVA or 0.1 µM HEL
peptide and the indicated cytokines. For production of Th2 cells,
cultures contained 40 ng/ml recombinant murine IL-4 (R&D Systems,
Minneapolis, MN) and a 1/8 dilution of anti-IFN- hybridoma
supernatant (clone TOSH, provided by E. Unanue, Washington
University). For preparation of Th1 cells, recombinant murine
IL-12 10 ng/ml (BD PharMingen, San Diego, CA) and a 1/8 dilution of
monoclonal anti-IL-4 supernatant (clone 11B11, provided by K.
Murphy, Washington University) were added. Three days after initiating
the cultures (day 3), the cells were split (1:4–1:6) and IL-2 (BD
PharMingen) was added to 40 U/ml. On day 7, cells were split onto fresh
irradiated splenocytes with peptide. On day 10, the cells were split
again and more IL-2 was added. Four days later (day 14), the cells were
transferred to fresh medium and were "resting." At this time they
do not secrete cytokines without further stimulation with Ag or with
PMA plus ionomycin (both from Sigma-Aldrich, St. Louis, MO), as
assessed by intracellular staining and flow cytometry or by ELISA (R&D
Systems) of culture supernatants. Resting but previously activated
cells express intermediate levels of CD25 (29).
Restimulation of the rested polarized cells with Ag and fresh APC led
to expression of cytokines. Assessment of cytokine production by
intracellular staining of monensin-, PMA-. and ionomycin-treated cells
followed by flow cytometry showed that >30% of T cells in the
Th1 cultures made IFN-, whereas <2% made IL-4 and 20–30% of
cells in the Th2 cultures made IL-4, while <2% made IFN-. For
some experiments, cells were labeled with CFSE (Molecular Probes,
Eugene, OR) by washing the cells in PBS without
Ca2+ or Mg2+ (Invitrogen)
and then incubating them at 1 x 107
cells/ml in 5 µM CFSE for 15 min. at 37°C in the dark. The labeling
reaction was stopped by adding an equal volume of FCS and the cells
were then washed thoroughly in PBS before adoptive transfer.
Adoptive transfer of differentiated cells and challenge with Ag
Differentiated, resting DO11.10 and 3A9 Th cells were
transferred to naive mice i.v. via the retro-orbital plexus. The
following day, 1% (w/v) Ag in PBS was administered via aerosol or
intranasally. The aerosol was generated using a clinical nebulizer
(UltraNeb 99; DeVilbiss Healthcare/Sunrise Medical, Sommerset, PA)
which was calibrated to deliver droplets <4 µm in diameter and was
administered to mice in a 15-inch square Plexiglas chamber as
previously described (7). Aerosol treatment was for 
30
min and was repeated once 6–8 h later. For intranasal challenge, 40
µl of 1% Ag was administered twice to mice anesthetized with
methoxyflurane (Metofane; Schering-Plough Animal Health, Union, NJ). In
experiments involving infection with Sendai virus, Th2 cells were
administered on day 5 or day 8 following inoculation with virus.
Aerosol or intranasal challenges were 1 day after T cell transfer and
the mice were analyzed 3 days after Ag challenge. In all cases, aerosol
and intranasal Ag challenges yielded results that were
indistinguishable.
Analysis of airway and lung leukocytes by flow cytometry
Identification of adoptively transferred Th2 cells in
isolated cell populations was performed by flow cytometry using
CyChrome-conjugated clonotypic Ab for the DO11.10 transgene
(KJ1-26-cyc; Caltag Laboratories, Burlingame, CA) (30) and
the 3A9 transgene (1G12-biotin, a generous gift from E. Unanue and D.
Peterson, Washington University). In selected experiments, the
transgenic cells were identified by virtue of their common usage of a
V8.2 TCR chain using anti-V8.2-CyChrome (Caltag
Laboratories). Finally, where indicated, the transgenic Th2 cells
were marked genetically with the allotypic marker Thy1.1 (and
transferred into Thy1.2 BALB/c mice) or fluorescently with CFSE. Other
fluorescent reagents included anti-CD25-PE, anti-CD4-CyChrome,
anti-CD4-allophycocyanin, and streptavidin-allophycocyanin
(BD PharMingen). Analysis was performed using a FACSCalibur with two
lasers (red diode and 488 nm of argon) and CellQuest software (version
3.3; BD Immunocytometry Systems, Mountain View, CA). Intracellular
staining for cytokines was achieved only after activating recovered
cells with PMA (0.01 µg/ml) and ionomycin (1 µM) in the presence of
2 µM monensin (all from Sigma-Aldrich) for 4–6 h. The surface
molecules were then stained as described and washed in 2% FCS. The
cells were permeabilized by washing three times in PBS with 2% FCS
containing 0.1 mg/ml saponin and stained for intracellular cytokines in
this permeabilization buffer. Parallel cultures were stained with
isotype control Abs (all from BD PharMingen).
Respiratory tract infection with Sendai virus
Five hundred egg infectious units of Sendai virus were instilled into the nostrils of mice anesthetized with methoxyflurane. On the indicated day after intranasal infection, Thy 1.1+- or CFSE-labeled DO11.10 Th2 cells were adoptively transferred i.v. Airway Ag challenge was 1 day after the adoptive transfer of Th2 cells by exposing mice to an aerosol of 1% OVA in PBS for 20 min or to 40 µl of 1% OVA in PBS administered intranasally. The mice were sacrificed and analyzed 3 days after Ag challenge.
Analysis of chemokine RNA expression
A total of 1.2 x 108 DO11.10 Th1 cells and/or Th2 cells was transferred i.v. into naive BALB/c recipients. One day later, these mice were challenged with a 30-min exposure to an aerosol of 1% OVA in PBS. Twelve hours later, their lungs were removed, frozen in liquid nitrogen, and total RNA was prepared using a Qiagen (Valencia, CA) miniprep kit according to the manufacturer’s instructions. Levels of chemokine RNA were determined using the Riboquant mCk-2b probe set (BD PharMingen) according to the manufacturer’s instructions.
In vitro proliferation of transgenic T cells
Splenocytes from DO11.10, 3A9, or nontransgenic mice (all on the (BALB/c x B10.BR)F1 background) were cultured in triplicate at 2.5 x 106/ml in 96-well plates. OVA or HEL (10-5 or 10-6 M) was added, followed 5 days later by [3H]thymidine. Twenty-four hours later, the cells were harvested and [3H]thymidine incorporation was determined. Similar results were obtained on day 3, although the DO11.10 T cells showed maximal proliferation to the whole protein Ag later than did the 3A9 T cells. One micromolar OVA or 0.3 µM HEL peptide and medium alone provided positive and negative controls, respectively. In selected experiments, HEL (Sigma-Aldrich) that had been further purified by size exclusion column chromatography to eliminate albumin (kindly provided by P. Allen, Washington University) was used for Ag challenge.
Immunohistochemistry
Paratracheal LN were frozen in OCT (Tissue-Tek, Torrance, CA) and 8-µm sections were fixed for 10 min in acetone and stained with KJ1-26-biotin followed by avidin multimer, Elite (ABC)-alkaline phosphatase and Vector Blue alkaline phosphatase substrate (Vector Laboratories) to observe the distribution of DO11.10 Th2 cells in the LN.
Assessment of Ag-nonspecific IL-5 production
Populations of resting DO11.10 or 3A9 Th1 or Th2 cells
on a (B10.BR x BALB/c)F1 background were
obtained 7 days after the last Ag challenge in vitro.
CD4+ cells were purified from these cultures
using CD4-specific Dynal magnetic beads (Dynal, Lake Success, NY)
according to the manufacturer’s instructions. At this point the
purified Th1 and Th2 cells were negative by intracellular
FACS staining or 72-h ELISA for IFN- or IL-4 unless they were
stimulated again with PMA and ionomycin. A total of 2.5 x
105 purified (B10.BR x
BALB/c)F1 TCR transgenic Th1 or Th2
cells was cultured together with 5 x 106
irradiated splenocytes from (B10.BR x
BALB/c)F1 mice in 1 ml of T cell medium.
Supernatants were recovered after culture for 72 h under 8%
CO2 and analyzed by ELISA (R&D Systems) for the
presence of IL-5 in the supernatant.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To investigate the relationship between respiratory virus
infection and recruitment of Th2 cells to the airway after Ag
challenge, we analyzed the effect of acute parainfluenza virus
infection on the recruitment of adoptively transferred Th2 cells
in otherwise naive mice. Animals infected with 500 egg infectious
doses of Sendai virus (1/10 the LD50)
did not become fatally ill. They did, however, show evidence of
significant inflammation in the airways. There was a large influx of
monocytes, macrophages, and neutrophils observed by Wright’s staining
of the bronchoalveolar lavage (BAL) cells. Analysis by flow cytometry
and intracellular cytokine staining also showed that 50% of the
infiltrating endogenous CD4+ T cells were capable
of producing IFN-. These data indicated that there was a significant
Th1-type inflammatory response to Sendai virus infection. Infected
mice (BALB/c, Thy1.2) were given 107 OVA-specific
DO11.10 Thy1.1+ Th2 cells at the peak of
virus infection (days 5–8). One day later, the infected mice were
exposed to an aerosol of 1% OVA. Three days after this challenge, the
airways were sampled by BAL. The transferred Th2 cells were
identified by surface staining of the BAL cells for the allotypic
marker Thy1.1 and phenotyped by intracellular cytokine staining and
flow cytometry as Th2 on the basis of their production of IL-4 but
not IFN-. Mice infected with Sendai virus showed recruitment of a
larger number of Thy1.1+,
IL-4+, IFN-- cells
compared with uninfected mice (Fig. 1). Thus, infection with a sublethal dose
of Sendai virus enhanced the recruitment of the OVA-specific Th2
cells. Interestingly, a 4-fold increase in the numbers of recruited
eosinophils was also observed in the infected mice challenged with OVA
compared with mice that were challenged with OVA alone (Fig. 1A). Together, these data suggested that the antiviral
(Th1-type) response could occur simultaneously with the allergic
inflammatory (Th2-type) response. In fact, the antiviral response
appeared to enhance Th2 cell recruitment and not to preclude
Th2 effector functions.
|
B). Interestingly, virus infection resulted in a dramatic
increase in the number of OVA-specific Th2 cells that could be
recovered by BAL even without OVA challenge. Infection without OVA
challenge induced airway inflammation equivalent to that seen in mice
infected with virus and then subjected to airway Ag challenge. The
numbers of Th2 cells recruited to the airways in the presence of
an antiviral response approach the numbers obtained when Th1 and
Th2 cells were transferred together before challenge with OVA.
This observation indicated that these Th2 cells could be recruited
to the airways in an Ag-independent fashion.
The Ag-independent recruitment of Th2 cells to the lungs seen in
the context of respiratory viral infection did not represent a global,
systemic mobilization of Th2 cells. Although transgenic
OVA-specific Th2 cells were found in large numbers in BAL
recovered from mice infected with Sendai virus but not challenged with
OVA, paratracheal LN of mice infected with Sendai virus but not
challenged with OVA showed a percentage of DO11.10 Th2 cells
similar to that of nondraining inguinal LN (Fig. 1C). To
determine whether the increase in cell number seen in the lungs was due
to increased recruitment to the airways or to Ag-independent
proliferation of the Th2 cell population, we tested the adoptively
transferred cells for evidence of cell division. To permit analysis of
cell division, the Th2 cells were labeled with CFSE before
adoptive transfer. Proliferation of the adoptively transferred Th2
cells was monitored using flow cytometry by observing the decrease in
intensity of CFSE fluorescence that occurs when labeled cells
proliferate. Evidence of Th2 cell proliferation was detected in
cells that were localized in the paratracheal LN 3 days after OVA
challenge of Sendai virus-infected mice. Only those mice that received
an OVA aerosol challenge showed an increased percentage of Th2
cells in the paratracheal LN. Transgenic cells recruited to this tissue
showed diminished CFSE fluorescence indicating that they had
proliferated. Although the percentage of nonproliferating
KJ1-26+ cells in the LN did not vary greatly
between experimental groups, the percentage of proliferating cells did
increase in the presence of Ag (Fig. 1C). Thus,
proliferation of Th2 cells in the regional node was Ag dependent.
Importantly, however, even when proliferation did not occur in the
local LN, Th2 cells could still localize efficiently to the
airways in response to respiratory virus infection.
Matching Ag specificity is not required for the increase in Th2 recruitment elicited by cooperating Th1 cells
To test the hypothesis that a local Th1 response of any
antigenic specificity might trigger recruitment of circulating allergic
Th2 cells, we substituted transgenic HEL-specific Th1 cells
for the virally induced inflammatory response. In this double
transgenic adoptive Th cell transfer system, the Ag specificity of
the Th1 cells was different from that of the Th2 cells,
allowing the two Th subsets to be activated independently. This
also allowed for investigation of the role of antigenic specificity in
recruitment of T cells. Although it is widely accepted that innate
cells are recruited in an Ag-nonspecific fashion, this has not been
previously established for the Ag-specific cells of the immune system.
DO11.10 Th2 cells recognizing OVA were adoptively transferred into
naive mice along with 3A9 Th1 cells recognizing HEL. Since both
3A9 and DO11.10 T cells express Ag receptors using the V8.2 chain,
the transferred Th2 cells that localized to the lung could be
identified by flow cytometry as surface V8.2+
and intracellular
IL-4+IFN--. The V8.2
gene is expressed by 5–10% of all circulating T cells (depending
on the strain), introducing some background into the determination of
OVA- and HEL-specific cells; however, use of this marker along with
anti-CD4, anti-IL-4, and anti-IFN- Abs permitted us to
measure intracellular cytokines. The cytokine staining allowed us to
identify clearly that the transferred cells were previously activated.
In other experiments, we have shown that adoptively transferred
Th1 cells do not switch phenotype under these conditions in vivo
(7, 21). Since genetic background can affect studies of
Th1 and Th2 cells in some systems (31) and since
the DO11.10 and 3A9 TCR transgenes required that the experiments be
performed using the H-2d x
H-2k ((BALB/c x
B10.BR)F1) background, we first tested whether
this mixed genetic system showed cooperation for recruitment between
Th1 and Th2 cells as we had seen in our earlier studies
(7, 21). Using either DO11.10 or 3A9 Th2 cells, we
found that few Th2 cells were recruited into the lungs of
BALB/c x B10.BR recipients unless Ag-stimulated Th1 cells
were also recruited to the tissue (Fig. 2, A and B). Thus,
the (BALB/c x B10.BR)F1 system yielded
similar results compared with the BALB/c system. Interestingly,
although we confirmed that Th1 cells are recruited to the airway
in response to aerosolized Ag when adoptively transferred without
Th2 cells, we found that the numbers of Th1 cells that were
recruited was increased when Th2 cells were also present. This
suggests that perhaps in reciprocal fashion Th2 cells can provide
signals supporting enhanced Th1 cell recruitment.
|
C). Having demonstrated
that both DO11.10 and 3A9 Th1 and Th2 cells could cooperate
in the (BALB/c x B10.BR)F1 genetic
background, we then tested whether Th1 and Th2 cells of
different antigenic specificities could affect each other’s
recruitment in the same way that virus infection or OVA-specific
Th1 cells influenced OVA-specific Th2 cell accumulation (Fig. 3). Th1 and Th2 cells derived
from transgenic mice encoding two different antigenic specificities
were transferred into unprimed syngeneic mice and various combinations
of the Th1 and the Th2 Ags were administered via aerosol. In
this experiment, the OVA-specific DO11.10 Th cells were identified
in the BAL using anti-CD4 plus the clonotypic Ab KJ1-26 and the
HEL-specific 3A9 Th cells were identified using anti-CD4 plus
the 1G12 mAb. Although this allowed us to identify each population of
transgenic cells specifically, it did not permit measurement of
intracellular cytokines using a four-color FACS machine. In previous
experiments, however, we showed that Th1 and Th2 cells retain
their starting phenotype over the course of experiments such as these.
When HEL and OVA were administered together to recipients of a mixture
of DO11.10 Th2 cells and 3A9 Th1 cells, substantial numbers
of Th2 cells were recruited (Fig. 3). Thus, the ability of
Th1 cells to provide stimuli for the recruitment of Th2 cells
is not dependent on both cell types recognizing the same Ag.
|
, left panel and
right panel, respectively). Although in the experiment shown
here the number of eosinophils recruited to the BAL was higher when
recruited by 3A9 Th2 cells, this was an inconstant finding that
was not reproduced in replicate experiments. Recruitment of Th2
cells to the airways in response to an Ag that was distinct from their
own cognate Ag suggested that the challenge Ag or the Th1 response
it elicited enhanced, in a nonspecific fashion, the recruitment
functions within the tissue for lymphocytes in general. To investigate
this further, we tested the effects of Ag challenge and Th1 cell
recruitment on expression of chemokine genes in the lung. Recruitment of Th1 cells to the lungs is associated with up-regulation of chemokine expression
Our observation that Th2 cell recruitment to the lungs and
airways can occur without the Th2 Ag is consistent with a model in
which the primary stimulus for Th2 cell recruitment is the
triggering by locally produced inflammatory mediators of adhesive and
chemotactic signals that render the tissue receptive for recruitment of
leukocytes circulating in the bloodstream. Because local recruitment of
Th1 cells can potentiate Th2 cell recruitment, we next
investigated the nature of the signals that aerosolized Ag and
recruited Th1 cells elicited in the tissue that might contribute
to subsequent homing of Th2 cells to the airway. DO11.10 Th1
and/or Th2 cells were transferred into naive BALB/c mice and the
mice were subsequently challenged with aerosol OVA. Twenty-four hours
(Fig. 4) or 36 h (data not shown)
after challenge, the lungs were collected and total lung RNA was
analyzed for the presence of chemokine transcripts using an RNase
protection assay. Twenty-four hours after OVA challenge of naive mice
that received no adoptively transferred cells, there was only low,
baseline expression of RANTES and very low expression of eotaxin-1,
macrophage-inflammatory protein (MIP) 1, MIP-1
, and MIP-2.
These levels of expression were indistinguishable from those of mice
receiving no Ag challenge (data not shown). Mice that had received only
OVA-specific Th2 cells showed only small increases in
Ag-stimulated chemokine RNA levels. In contrast, mice that received
OVA-specific Th1 cells or Th1 cells plus Th2 cells
showed a dramatic increase in expression of the tested chemokines.
Several of the chemokines up-regulated in association with Th1
cell recruitment, including eotaxin (32), RANTES, and
MIP-1 (33), have been shown to support the development
of eosinophilic airway inflammation in a coordinated sequential
fashion. In the course of investigating the expression of other
molecules that participate in leukocyte recruitment, we have previously
reported that VCAM-1 is up regulated in association with Th1 cell
recruitment and that administration of blocking Abs to VCAM-1 limits
the ability of Th1 cells to cooperate in the recruitment of
Th2 cells and eosinophils (21). In this study, we
report that when Th1 cells are recruited to the airways, there is
increased expression of several chemokines that may contribute to
establishing a milieu that is permissive for subsequent Ag-independent
recruitment of Th2 cells to the lungs.
|
To begin to test whether trafficking of Ag specific Th2 cells
through the LN is needed for these cells to acquire the potential to
migrate to peripheral tissues, we investigated whether recruitment of
Th2 cells to the lungs and airways was associated with
localization to the paratracheal LN. The paratracheal LN from mice that
had received DO11.10 Th2 cells and/or 3A9 Th1 cells before
aerosol Ag challenge with combinations of HEL and/or OVA were analyzed
immunohistochemically for the presence of DO11.10 Th2 cells using
the KJ1-26 Ab (Fig. 5). We observed first
that there were many Ag-specific Th2 cells in these draining LN
after Ag challenge, whether or not this challenge was sufficient to
drive Th2 cell recruitment to the lungs. For example, when DO11.10
Th2 cells were transferred along with 3A9 Th1 cells and
challenge was with OVA alone, there were many Th2 cells in the
local LN (Fig. 5C) even though under these conditions there
was little recruitment of either Th1 or Th2 cells to the
lungs (Fig. 3). The pattern of Th2 cell accumulation in the
regional LN was similar in mice that were challenged with both OVA and
HEL (Fig. 5A), conditions under which there was robust
recruitment of Th2 cells to the airways (Fig. 3). As in mice
infected with Sendai virus but not receiving aerosol Ag challenge (Fig. 1C), recruitment of Th2 cells to the airways did not
appear to be dependent on accumulation of Th2 cells in the
draining LN. This recruitment of Th2 cells to the airway without
accumulation in the draining LN is consistent with the recent data of
Gajewska et al. (34) suggesting that Th2 predominant
experimental allergic airway inflammation can develop in mice with
congenital absence of peripheral LN.
|
Ag-nonspecific recruitment of Th2 cells to the lung may have
very little impact on the tissue if the recruited cells are not
activated. In contrast, Ag-nonspecific Th2 cell recruitment may
elicit the full gamut of Th2-driven pathologic changes if the
recruited Th2 cells manifest an activated phenotype. In this
regard, it is of interest that in mice that received HEL-specific
Th1 cells and OVA-specific Th2 cells followed by challenge
with the Th1 Ag alone there was recruitment of not only the
OVA-specific Th2 cells, but also substantial numbers of
eosinophils (Fig. 4). This was despite the fact that the adoptively
transferred Th cells had reverted to a resting phenotype before
their infusion in vivo. Given that the Th2 cytokine IL-5 seems to
be required for mobilization of eosinophils in murine eosinophilic
airway inflammation (35, 36), our findings suggest that
recruitment of Th2 cells to the lung, even in the absence of the
Th2 cell Ag, may result in activation of the cells to make IL-5
that may then support the mobilization and survival of eosinophils.
These mice also showed increased mucus production as detected by
periodic acid-Schiff staining (data not shown), representing an
additional Th2 effector function (8, 37, 38).
The level of activation of cells that had been recruited to the airways
was examined further by flow cytometry of the BAL lymphocytes for the
level of expression of CD25. In vitro expression of CD25 is elevated on
primary Th cells from day 2 or 3 after challenge with Ag and
remains high until day 5 or 6 when it is reduced to an intermediate
level of surface expression. Ag-experienced cells remain
CD25int for weeks but are induced again for
several days following restimulation, up to two orders of magnitude in
relative fluorescence units, to the high levels seen following the
initial activation (39, 40). Surprisingly, in the BAL from
Ag-challenged mice, similar numbers of Th2 cells expressed high
levels of CD25 whether the airways were challenged with the cognate
Th2 Ag or with a nonspecific Ag (Fig. 6). Similar CD25 expression was observed
on Th2 cells recruited in mice that had received DO11.10 Th2
cells (Fig. 6) or 3A9 transgenic cells (data not shown). Because
resting, CD25int Th2 cells do not enter the
BAL in this system in the absence of Th1 cells and since in the
presence of Th1 cells they acquire an activated phenotype, it is
not possible to recover CD25int, previously
activated but not presently activated, Th2 cells in the BAL as a
control; however, the levels of CD25 on Th2 cells recovered from
BAL of mice that had been treated with either a combination of Th1
and Th2 Ags or with the Th1 Ag alone are substantially higher
than those seen on resting cells in vitro before adoptive transfer.
Taken together, the findings that Th2 recruitment was accompanied
by recruitment of eosinophils and also by increased expression of CD25
on Th2 cells recovered from the BAL indicate that recruitment of
Th2 cells mediated by Th1 cells and Th1 Ag was
accompanied by at least partial activation of the recruited Th2
cells.
Th2 cells can produce IL-5 in the absence of specific Ag
Our observation that Ag-nonspecific recruitment of Th2 cells
to the airway is accompanied by recruitment of eosinophils suggests
that the Th2 cells recruited in this way are at least partially
activated to secrete factors that support eosinophil recruitment. The
Th2 cytokine IL-5 is generally essential for eosinophil
mobilization and survival (35, 36). Thus, although
additional factors may be involved, IL-5 secretion by recruited
Th2 cells may be a crucial parameter for the development of the
type of inflammation detected here. We, therefore, tested the potential
for Th2 cells to produce IL-5 in an Ag-independent fashion. Since
it was unclear from our initial data whether Th1 cells might be
indirect participants in the production of IL-5, we examined both
Th1 and Th2 cells. Polarized cultures of 3A9 and DO11.10
Th cells were maintained for 7 days after stimulation with Ag. At
this time, the cells have a resting phenotype, with no IL-5 detectable
in the culture supernatants of either the Th1 or Th2 cell
cultures. To test whether these cells could be induced to IL-5
production without restimulation by Ag, we repurified these cells based
on their expression of CD4, washed them, and transferred them to new
cultures with irradiated syngeneic splenocytes, but without
addition of Ag. Thirty-six or 72 h later, we tested whether
secreted IL-5 could be detected (Table I). Our data showed that culture of
Th2 cells in the presence of fresh APC resulted in the secretion
of IL-5 regardless of the presence of the Th2 cell Ag or Th1
Ag. Both DO11.10 and 3A9 Th2 cells (Table I) showed this
Ag-independent capability to produce an average of 97 pg/ml IL-5.
Although the concentration of IL-5 detected in cultures without added
Ag was almost 10-fold lower than that produced when Ag was added, it
was detected consistently and was 10-fold higher than the lower limit
of detection of this sensitive commercial assay. IL-5 production by the
resting Th2 cells was always below the level of detection. The
half-life of peptides in solution and on APC is generally <3 days.
This, along with the fact that the T cells were rested and washed
extensively before this assay, suggests that this production of IL-5 is
indeed Ag independent. We are not, however, able absolutely to exclude
the formal possibility that small quantities of Ag may have been
carried over into the coculture. Collins et al.
(40) have calculated that only 18.3 pmol/kg (0.122
pmol/ml) of circulating IL-5 is sufficient to mobilize eosinophils. In
this context, the low level of IL-5 (2.86 nmol/ml) we observed here
is expected to be well within the range that can induce eosinophilia.
Thus, theoretically, the Ag-independent airway eosinophilia that we
observed may have been the consequence of production of IL-5 by the
recruited Th2 cells, either following their encounter with APC, or
as a consequence of some other stimulating event that occurred during
their migration from the circulation into the tissue. Regardless of the
mechanism operating in vivo, these studies have shown that Th2
cells can produce IL-5 in the absence of added cognate Ag.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In general, the immune response to virus is Th1-like in its effector mechanisms. These mechanisms include, in addition to the antiviral Th1 cells of the adaptive response, Ag-nonspecific elements recruited from the innate immune response. There is a large body of data showing cross-interference of Th1-type and Th2-type responses. This can be observed both at the level of phenotype differentiation and at the level of expression of effector functions. Many studies have suggested that soluble mediators produced in the course of Th1-type and CD8-type responses could inhibit effector functions of coexisting Th2 cells. Although cross-interference of Th1-type and Th2-type responses is broadly accepted, there are also examples, particularly in models of experimental infection, of both types of responses being launched in a single host against a single pathogen, resulting in a composite response of mixed phenotype.
In this study, we have extended our previous findings (7, 21) that Ag-specific Th1 cells can potentiate Th2 cell recruitment and the resulting Th2-like tissue inflammation. We have used respiratory tract infection with Sendai virus as an experimental model in mice for the respiratory tract infections that are associated with exacerbations of asthma in human patients. In this model, the antigenic specificity of the antiviral response is unlikely to be related to the antigenic specificity of the pathologic Th2 cells that ultimately lead to the allergic inflammation. In a parallel system, we have used HEL-specific transgenic Th1 cells as another way of providing a Th1-like inflammatory signal in the context of an unrelated Th2 response (here directed against the model Ag, OVA). In both systems, we have shown that Th1-like responses, irrespective of the Ag that elicited them, can synergize with Th2 cells to promote eosinophilic airway inflammation. This synergy is expressed largely at the level of facilitating recruitment of the Th2 cells to the tissue.
Because recruitment of adoptively transferred Th1 cells leads to
an increase in both airway inflammation and in recruitment of Th2
cells into the lung, we considered that Th1-specific cytokines
might be the primary signals supporting the recruitment of Th2
cells; however, efforts to block signals delivered uniquely by Th1
cells by administration of a neutralizing anti-IFN- Ab, a
neutralizing lymphotoxin receptor-Ig fusion protein, or a
neutralizing anti-IL-2R Ab did not inhibit recruitment of
Th2 cells in mice that had received cotransfer of Th1 and
Th2 cells (Ref. 21 and our unpublished data).
Thus, not one of these Th1 cell products was dominantly
responsible for the cooperation observed between Th1 and Th2
cells, suggesting that this cooperation is in fact not a feature unique
to Th1 cell-driven inflammation.
Our observation that either Th1 cells or infection with Sendai virus could enhance Th2 cell recruitment in the absence of the Th2 Ag led us to investigate in related experiments the role of nonspecific inflammatory signals in this recruitment. We have found that induction of nonspecific airway inflammation by treatment with intranasal LPS also can lead to recruitment of Th2 cells to the lung in the absence of the Th2 Ag (45). Together these data are consistent with a model in which a respiratory virus contributes to Th2 cell recruitment both by induction of an antiviral Th1 cell response and also by activation of local innate inflammatory elements. This does not necessarily imply that Th1 cells can uniquely provide this stimulus. We have also observed that a variety of different inflammatory agents can provide the stimulus for recruitment of Th2 cells. Taken as a whole, our data suggest that recruitment of Th2 cells to a peripheral tissue is initiated by local proinflammatory signals that lead to the generation of an inflammatory milieu. This inflammatory milieu, independent of the Th2 Ag, focuses the accumulation of Th2 cells at the inflammatory site where they can then contribute to the local response depending on what other activating signals are present. Although the molecular mechanisms governing Ag-independent Th2 cell recruitment remain undefined, they probably include enhanced local expression of endothelial cell adhesion molecules and production of chemoattractant molecules. Given our finding that the Th2 Ag is not required for recruitment of these cells to the airways, it is a logical extension that neither proliferation nor an extended stay in a regional LN is required for Th2 cells to be recruited to a peripheral site of inflammation. Our finding that activation within a LN is not required for recruitment to a peripheral tissue is consistent with the recent observations of Gajewska et al. (34) that allergic airway inflammation can be elicited in mice with congenital absence of LN.
Unexpected in our studies was the observation that Th2 cells that had been adoptively transferred in a resting state (7 days after exposure to Ag in vitro) and recruited to the lung in an Ag-independent fashion were able to express effector functions typical of activated cells. This activated phenotype was manifested by the recruitment of eosinophils to the airways and by the induction on the Th2 cells of increased levels of CD25 in mice that had received HEL-specific Th1 cells along with OVA-specific Th2 cells before challenge with HEL alone. Given the requirement for IL-5 in the recruitment of eosinophils to the airway (41), our data suggested that recruitment of Th2 cells to the airway was accompanied by their activation for cytokine production and that this activation occurred in an Ag-independent fashion. Interestingly, activation of Th2 cells was not always observed in association with their Ag-independent recruitment. For example, Ag-independent Th2 cell recruitment initiated by innate stimuli such as LPS was not associated with Th2 cell activation (45).
We have modeled the Ag-independent activation of Th2 cells in
vitro by culture of resting Th2 cells with syngeneic irradiated
APC. This leads to up-regulation of IL-5 expression as measured by
ELISA of culture supernatants (Table I). Future studies will
investigate the nature of the activating signal that is provided by the
APC. Several previous studies have reported Ag-independent activation
of human CD4+ T cells in vitro. Treatment with a
combination of the cytokines IL-2, IL-6, and TNF has been shown to
stimulate human T cells to proliferate and secrete IL-2
(42). Furthermore, culture of differentiated Th2
cells with IL-10 has been shown to increase transcription of IL-4,
IL-9, and IL-13 (43), and culture of Th1 cells with a
mixture of IL-12 and IL-18 has been shown to induce production of
IFN- without stimulation of the TCR (44). Additional
studies will be required to determine whether any of these signals
contributes to the Ag-independent activation of Th2 cells that we
have observed here. Regardless of the mechanism by which coculture with
APC induces IL-5 expression, we suggest that perhaps similar signals
can be delivered in vivo as the recruited Th2 cells encounter
tissue APC or cells with similar activating potential. The
Ag-independent activation that occurs in vivo may be limited to
cytokine secretion and may represent partial activation as originally
described by Evavold et al. (44). Alternatively, it may
include activation of all effector responses. In experiments using
infection with Sendai virus to mobilize Ag-independent Th2
cell recruitment, we observed a moderate level of tissue eosinophilia
but with no evidence of proliferation of Th2 cells, suggesting
that in this case the Ag-independent Th2 cell activation was
indeed incomplete.
The type of Ag-nonspecific local activation of T cells that we have observed here would have grave implications for patients with large numbers of circulating allergen-specific Th2 cells. Such nonspecific activation would place an atopic individual at risk for a flare of allergic inflammation whenever a local tissue injury occurred. We suspect that this underlies the reported coincidence of respiratory tract infections and clinically significant asthma exacerbations.
Our data demonstrate that the T cell Ag does not play a defining role in T cell recruitment to the airways; however, T cell Ag does lead to proliferation in the LN. We suspect that this action of Ag is a major regulatory one in the majority of T cell-dependent tissue responses in vivo. Under normal conditions, the numbers of Ag-specific T cells are carefully regulated so that large numbers of Ag-specific T cells are present only when a brisk local or systemic response is required. Thus, in the normal setting when the numbers of Ag-experienced cells are low, nonspecific recruitment of T cells to a site of tissue inflammation might result in only limited inflammatory potential. Ag, driving the recruitment of specifically reactive T cells to the regional nodes to permit clonal expansion under the influence of cooperating APC and regulatory cells, would in this setting be required to permit the development of an effective protective response. In contrast, when the numbers of Ag-experienced cells are high, either during the early phases of an active response or in a pathologic condition of chronic tissue inflammation, Ag-independent recruitment of T cells to peripheral tissues and Ag-independent cell activation in these tissues may become clinically relevant, leading to the expression of pathologic inflammatory responses after exposure to otherwise innocuous inflammatory stimuli.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Current address: Division of Parasitology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, U.K.
3 Current address: Department of Pediatrics, University of California, San Francisco, School of Medicine, San Francisco, CA 94143.
4 Address correspondence and reprint requests to Dr. David D. Chaplin, Department of Microbiology, University of Alabama, 845 19th Street South, BBRB 276/11, Birmingham, AL 35294-2170. E-mail address: dchaplin{at}uab.edu
5 Abbreviations used in this paper: LN, lymph node; HEL, hen egg lysozyme; BAL, bronchoalveolar lavage; MIP, macrophage-inflammatory protein; int, intermediate.
Received for publication November 30, 2001. Accepted for publication September 13, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
1 integrin-dependent adhesion in human Th2 T lymphocytes. J. Immunol. 164:3292. regulate allergic airway inflammation and mucus production. J. Exp. Med. 190:1309. and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 236:944.
), DO11.10 (
), or
nontransgenic (
) (B10.BR x BALB/c)F1 splenocytes
were plated in triplicate at 2.5 x 106 cells/ml in T
cell medium with either 10-6M or 10-5M HEL or
OVA protein as indicated. On day 4 of culture,
[3H]thymidine (1 µCi) was added and 24 h later
cells were harvested and [3H]thymidine incorporation was
determined. Data shown are the mean ± SD of triplicate wells.
Similar results (although less absolute proliferation) were obtained if
the cultures were harvested 3 days after stimulation. In addition,
similar data were obtained if stimulation was with synthetic HEL
(48–62) and OVA (323–339) peptides or using repurified OVA and HEL as
described in Materials and Methods. This experiment was
repeated twice.
