j immunol-2010-bako-evi--1317-27

7/30/2019 J Immunol-2010-Bako-evi--1317-27

1/12

of July 21, 2012.This information is current as

Tolerance and Immunityduring the Induction of Respiratory

Dendritic Cell Interaction DynamicsT Cell

Reinhold FrsterNadja Bakocevic, Tim Worbs, Ana Davalos-Misslitz and

http://www.jimmunol.org/content/184/3/1317doi: 10.4049/jimmunol.0902277December 2009;

2010; 184:1317-1327; Prepublished online 30J Immunol

MaterialSupplementary

7.DC1.htmlhttp://www.jimmunol.org/content/suppl/2009/12/30/jimmunol.090227

Referenceshttp://www.jimmunol.org/content/184/3/1317.full#ref-list-1

, 15 of which you can access for free at:cites 38 articlesThis article

Subscriptionshttp://jimmunol.org/subscriptions

is online at:The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/ji/copyright.htmlSubmit copyright permission requests at:

Email Alerts http://jimmunol.org/cgi/alerts/etocReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606.Immunologists, Inc. All rights reserved.Copyright 2010 by The American Association of9650 Rockville Pike, Bethesda, MD 20814-3994.The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology
http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/cgi/adclick/?ad=34608&adclick=true&url=http%3A%2F%2Fwww.ebioscience.com%2Fspecials%2Fpromotions%2Ffree-ipad.htm%3Futm_campaign%3DQ2-ipad-promo%26utm_source%3Donline%2520banner


2/12

The Journal of Immunology

T CellDendritic Cell Interaction Dynamics during the

Induction of Respiratory Tolerance and Immunity

Nadja Bakocevic, Tim Worbs, Ana Davalos-Misslitz, and Reinhold Forster

Dendritic cells (DCs) residing in the lung are known to acquire inhaled Ag and, after migration to the draining bronchiallymph node

(brLN), to present it to naive T cells in an either tolerogenic or immunogenic context. To visualize endogenous lung-derived DCs, we

applied fluorescent latex beads (LXs) intratracheally, thereby in vivo labeling the majority of phagocytic cells within the lung. Of

note, LX-bearing cells subsequently arriving in the draining brLN were found to represent lung-derived migratory DCs. Imaging

explanted brLN by two-photon laser-scanning microscopy, we quantitatively analyzed the migration and interaction behavior of

naive CD4+ T cells and endogenous, lung-derived DC presenting airway-delivered Ag under inflammatory or noninflammatory

conditions. Ag-specific naive CD4+ T cells engaged in stable as well as transient contacts with LX-bearing DCs in both situations

and displayed similar overall motility kinetics, including a pronounced decrease in motility at 1620 h after antigenic challenge. In

contrast, the comparative analysis of T cellDC cluster sizes as well as contact durations strongly suggests that lung-derived

migratory DCs and naive CD4+ T cells form more stable, long-lasting contacts under inflammatory conditions favoring the

induction of respiratory immunity. The Journal of Immunology, 2010, 184: 13171327.

The airways are continuously exposed to numerous envi-

ronmental particles, such as dust, smoke, and pollen, as well

as various pathogens of bacterial or viral origin. Repre-

senting a common pathogenic consequence of these complex anti-

genic challenges, allergic airway diseases have been hypothesized to

develop during instances when the immune systemof the lung fails to

correctly discriminate between innocuous and harmful Ags.Over the

last four decades of the twentieth century, the prevalence of allergic

asthma increased substantially in industrialized countries (1),

making a more detailed understanding of the pathogenic mecha-

nisms of this allergic disease highly desirable.

Dendritic cells (DCs) are professional APCs that function as key

players during the induction phase of adaptive immune responses,directing the outcome toward either tolerance or immunity (25).

Airway DCs in particular have been shown to be indispensable for

the induction and maintenance of Th2-dependent airway in-

flammation (6). In the lung, immature DCs are located both within

and right beneath the surface of the respiratory epithelium, thereby

being ideally positioned for the efficient sampling of airborne Ags

(7). Upon capture of innocuous Ag in the absence of so-called

pathogen-derived danger signals, lung-resident DCs are thought

to acquire a semimature phenotype that includes intermediate

expression of costimulatory molecules, as well as high levels of

MHC class II molecules (reviewed in Ref. 4), while migrating

toward the draining bronchial lymph node (brLN). Consequently,

Ag presentation by these semimature DCs is assumed to result in

T cell tolerance, whereas bacterial or viral pathogens induce full

maturation of DCs with subsequent generation of strong effector T

cell responses. Interestingly, however, proliferation of Ag-specific

T cells is occurring in both situations, as confirmed by several

in vitro and in vivo studies (4). Furthermore, during the induction

of both tolerance and immunity, the trafficking of skin DCs toward

the draining lymph nodes (LNs) was shown to crucially require

the upregulation of CCR7 (8). In a recent study, we have dem-

onstrated that, in the absence of inflammatory signals, the in-

duction of respiratory tolerance to harmless aerogenic Ags

depends on the function of CCR7 as well and is mediated bymigratory DCs that transport Ag from the lung to the brLN (9).

Thus, the migratory capacity of DCs is likely to play an important

role in the development of both tolerance and immunity (10).

Employing diverse experimental models, several studies have

compared the kinetics of early T cellDC interactions during the

development of both tolerance and immunity (1113). Up to now, no

consensual model has been established for the interaction behavior

of T cells and DCs during the onset of tolerance and immunity.

Furthermore, T cellDC interactions during the development of

immune responses triggered by airway Ags have not been evaluated.

In this study, we provide a detailedanalysis of the dynamic behavior

of CD4+ T cells within the brLN during the induction of respiratory

tolerance and immunity, using two-photon laser-scanning micros-copy. The intratracheal (i.t.) application of fluorescently labeled latex

beads (LXs) coupled with Ag allowed the simultaneous visualization

of endogenous migratory DCs and the characterization of their in-

teractions with T cells within explanted brLN. We found the overall

migration behavior of Ag-specific CD4+ T cells being activated in the

brLN under tolerance- or immunity-inducing conditions to be highly

comparable regarding general motility parameters such as cellular

velocity or motility coefficients. In both situations, the T cells un-

derwent a phase of largely reduced motility while interacting with

Ag-presenting DCs 1620 h after Ag challenge. Interestingly, how-

ever, the quantitative analysis of the size and stability of T cellDC

clusters revealed that stable, long-lasting contacts were clearly more

predominant under conditions leading to immunity. These data

Institute of Immunology, Hannover Medical School, Hannover, GermanyReceived for publication July 15, 2009. Accepted for publication November 29,2009.

This work was supported by Deutsche Forschungsgemeinschaft Grant SFB587-B3and in part by EXC62 (to R.F.).

Address correspondence and reprint requests to Prof. Reinhold Forster, Institute ofImmunology, Hannover Medical School, Carl-Neuberg Strasse 1, 30625 Hannover,Germany. E-mail address: [email protected]

The online version of this article contains supplemental material.

Abbreviations used in this paper: BAL,bronchoalveolar lavage;brLN, bronchial lymphnode; DC, dendritic cell; FRC, fibroblastic reticular cell; GR, granulocyte; i.t., intratra-cheal; LN, lymph node; LX, latex bead; LYM, lymphocyte; MHCII, MHC class II; MO,monocyte; OVA-ALUM, OVA and aluminum hydroxide; OVA-LX, OVA-loaded latexbead; wt, wild-type.

Copyright 2010by The American Association of Immunologists, Inc. 0022-1767/10/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.0902277

yg

y

,

pj

g
mailto:[email protected]://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/mailto:[email protected]


3/12

therefore suggest that the outcome of a CD4+ T cell response against

airway-derived Ags might actually be represented by the stability of

early T cellDC interaction within the brLN.

Materials and MethodsMice

CCR72/2 (14) as well as DO11.10 mice were bred and maintained underspecific pathogen-free conditions at the central animal facility of theHannover Medical School and were used at the age of 812 wk. BALB/cmice were purchased from Charles River Laboratories (Sulzfeld, Ger-many). All animal experiments were conducted in accordance with localand institutional guidelines.

i.t. instillation of LXs, OVA-loaded LXs, and OVA-loaded

LXs+LPS

Micewere anesthetized i.p. with 0.2mg ketamineand 0.02 mg xylazine/g bodyweight. Subsequently, a blunt cannula (0.7 3 19 mm; Introcan; B. Braun,Melsungen, Germany) was instilled through the larynx into the trachea. Thetotal injected volume was 60 ml for all i.t. applications. Plain yellow-green0.5mm LXs (Polysciences, Warrington, PA) were instilled at a concentrationof 3.6 3 1011 particles/ml, with each animal receiving 4.3 3 10

9 LXparticles. OVA (OVA, Grade VI; Sigma-Aldrich, St. Louis, MO) was pas-sively adsorbed onto 0.5 mm LXs according to the manufacturers directions

and

4.33

10

9

OVA-loaded LXs (OVA-LX) were applied i.t. where in-dicated. All batches of OVA were tested for the presence of LPS (LAL QCL1000; BioWhittaker-Cambrex, Walkersville, MD), and only batches con-taining ,5 EU/mg were used for adsorption loading. Some animals addi-tionally received 10 mg LPS (LPS; Sigma-Aldrich) i.t. where indicated.

Flow cytometry

For flow cytometric analysis, lungs and brLNs were cut into small pieces anddigestedfor30min(brLN)or60min(lung)at37CinPBScontaining0.5mg/mlcollagenase A and 20 U/ml DNase I (both from Roche Applied Science, In-dianapolis, IN). To obtain single cell suspensions, tissues were subsequentlyminced through a nylon mesh and washed with PBS supplemented with 3%FCS. Leukocytes isolated from lungs were additionally purified using a Lym-pholyte M-gradient (Cedarlane,Hornby, ON, Canada). Cells werestained usingthe following Abs: antiCD4-Cy5 (clone Rm CD4-2), anti-mouse DO11.10TCR(KJ1-26,Caltag), antiCD103-bio(M290), antiCD11c-APC(HL3), anti-

MHC class II (MHCII [I-Ad

])-biotin (AMS-32.1), and antiCD11b-PECy7(M1/70), all purchased from BD Pharmingen (San Diego, CA). BiotinylatedAbs were revealed by Streptavidin-PerCP (BD Biosciences, San Jose, CA).Lung-derived migratory DCs were identified by the expression of CD103 (15).FACS analysis was performed on an LSRII flow cytometer (BD Biosciences).

Histology

Mice were anesthetized and bled to death from the retro-orbital plexus. Thethorax was opened, and the lung was perfused with cold PBS via the rightheart ventricle. The lung wasremoved, filled viathe trachea with OCT Tissue-Tek (Sakura Finetek USA, Torrance, CA) diluted 1:4 in PBS, embedded inOCT, and frozen on dry ice. Next, 8-mm cryosections of lungs and brLNswere prepared, air dried, and fixed for 10 min in ice-cold acetone. Forfluorescence microscopy, cryosections of brLNs were blocked with 5% ratserum and stained with the following Abs: antiCD3-Cy5 (clone 17A2) andantiB220-Cy3 (clone TIB 146), both provided by E. Kremmer (Helmholtz

Zentrum Munchen, Neuherberg, Germany). For brightfield microscopy, lungcryosections were stained with H&E. Images were acquired using anOlympus (Melville, NY) BX61 upright microscope.

Cytospins

For the preparation of cytospins, FACS-sorted cells were centrifuged at 600rpm for 10 min onto HistoBond-coatedglass microscope slides (Marienfeld,Lauda-Konigshofen, Germany), using a Cytospin 4 centrifuge (ThermoShandon, Munchen, Germany). After overnight drying, cytospins werestained using May-Grunwald and Giemsa eosin methylene blue stainingsolutions (Merck, Darmstadt, Germany).

Adoptive transfer of OVA-specific TCR-transgenic T cells

Single-cell suspensions were prepared from spleen and LNs (mesenteric,inguinal, brachial, and axillary) of DO11.10 mice. After E lysis, cells were

labeled with 10 mM 5-(and-6-)-TAMRA SE or 5 mM CFSE (Invitrogen,Carlsbad, CA) for 15 min at 37C. Untouched CD4+ T lymphocytes weresorted using a MACS CD4+ T cell isolation kit applying an AutoMACS

(Miltenyi Biotec, Auburn, CA). he purity after MACS was typically.90%.BALB/c recipient mice (68 wk old) received 5 3 106 CD4+ DO11.10T cells by i.v. injection into the tail vein.

In vivo T cell proliferation assay

Twenty-four hours after adoptive transfer of 5 3 106 CFSE-labeledMACS-purified CD4+ DO11.10 T cells into BALB/c recipients, LXs orOVA-LX 6 LPS (10 mg) were applied i.t. Five days after the initialadoptive transfer, brLNs were analyzed by flow cytometry for the fre-quency and proliferation of OVA-specific CD4+ DO11.10 T cells.

Model of lung inflammation

BALB/c mice received i.t. applications of OVA-LX, OVA-LX+LPS, or plainLXs on days 24 and 21. On day 0, animals that had received OVA-LX orOVA-LX+LPS were immunized i.p. with OVA-ALUM (150 mg OVA gradeVI (Sigma-Aldrich) in 200 ml aluminum hydroxide (ALUM) gel adjuvant(2.0% Alhydrogel; Brenntag Biosector, Frederikssund, Denmark). On days14 and 17 after the i.p. sensitization, the respective recipient mice were re-challenged with i.t. applications of either OVA-LX or OVA-LX+LPS. At thesame time points, LX-treated animals received i.t. applications of LXs.Twenty-four hours after the last i.t. challenge, mice were sacrificed anda bronchoalveolar lavage (BAL) was performed using 3 3 0.8 ml RPMI1640 supplemented with 0.05 mM EDTA. BAL fluid was analyzed by flowcytometry for the presence of infiltrating CD11b+CD11c2 granulocytes,CD11b+CD11clo monocytes, CD11b+CD11c+ DCs, and lymphocytes.

Two-photon laser-scanning microscopy

One hour after adoptive transfer of 5 3 106 TAMRA-labeled MACS-purifiedCD4+ DO11.10 T cells into BALB/c recipients, mice received plain LXs,OVA-LX, or OVA-LX+LPS i.t. Three hours after the adoptive transfer ofCD4+ DO11.10 T cells, mice additionally received 100 mg anti-CD62L mAb(clone MEL-14) i.v. to prevent any further LN homing of lymphocytes. At1620 h as well as 3640 h after i.t. application, the recipient mice weresacrificed and the brLN was explanted. After fixation in a custom-built in-cubation chamber, using tissue glue (Abbott, Abbott Park, IL), the brLN wasconstantly superfused with oxygenated (95% O2 plus 5% CO2) RPMI me-dium containing penicillin/streptomycin (Life Technologies, Rockville, MD;Grand Island, NY). The temperature was monitored directly at the brLN andmaintained at 37C. Two-photon laser scanning microscopy was performedusing an upright Leica Microsystems (Wetzlar, Germany) DM LFSA mi-croscope equipped with a 20 3 0.95 NA water immersion objective(Olympus) and a MaiTai Ti:Sa pulsed infrared laser (Spectra-Physics,Darmstadt, Germany). For simultaneous two-photon excitation of yellow-green LXs and TAMRA, the MaiTai laser was tuned to 858 nm. Green andred fluorescence emission was detected with non-descanned detectors fittedwith 535/50 and 610/75 bandpath filters, respectively. To generate time-lapseseries, stacks of 11 images (with 1.52.53 electronic zoom and 3.9- to 6-mm

z-spacing) were acquired every 15 s, yielding total imaging volumes of 215360 mm in x and y and 4466 mm in z.

Data analysis

Imaris (Bitplane, Zurich, Switzerland) was used for four-dimensionalimageanalysis and automated tracking of cells. The accuracy of the automatedtracking was manually controlled, and only tracks with durations .60 swere included in the analysis. Average cell velocities and arrest co-efficients were calculated in Imaris; mean displacement plots, motilitycoefficients, and turning angle distributions were calculated using Excel

(Microsoft, Redmond, WA) based macros after exporting the x-, y-, and z-coordinates of all individual spot positions. The arrest coefficient repre-sents the percentage of time points of a given track for which the in-stantaneous velocity of the cell is ,2 mm/min (11); for details on the othermotility parameters, see Ref. 16. For the analysis of DCT cell clusterformation, individual DCT cell interactions were defined as stable whenpresent over the entire imaging interval (22.525 min), whereas contactswith a duration of.75 s (more than five time points) were defined astransient. Statistical analysis was performed with GraphPad Prism 4(GraphPad Software, La Jolla, CA). All significant values were determinedusing the unpaired two-tailed t test.

ResultsSpecific labeling of endogenous lung-derived DCs in the brLN

after i.t. application of fluorescent LXs

After trafficking toward the draining brLN, lung-derived migratoryDCs are assumed to initiate primary T cell immune responses

1318 T CELL DYNAMICS IN BRONCHIAL LYMPH NODES

yg

y

,

pj

g
http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/http://jimmunol.org/


4/12

directed against inhaled Ags. To visualize these DCs during Ag

presentation in the brLN by in situ imaging, we applied LX i.t., as it

had been previously described that LX-bearing cells, presumably

representing lung-derived DCs, arrive in the brLN following i.t. (9)

or intranasal application of LX (17). To further characterize this

in vivo labeling of endogenous cells, we applied 4.33 109 yellow-

green fluorescent LX (0.5-mm diameter) i.t. and phenotypically

analyzed LX-bearing cells (LX+ cells) of both lung and brLN by

flow cytometry. In the lung, the majority of LX

+

cells were foundto be either CD11c+CD11b2 macrophages (18), CD11b+CD11c2

granulocytes (17), or CD11c+MHCIIhi DCs (Fig. 1A), indicating

that all major leukocyte populations with phagocytic capacity take

up LX within this tissue. Only a minor population of LX-bearing

cells were CD103+CD11c+ DCs. In contrast, LX+ cells in the

brLN were mostly confined to the CD11c+MHCIIhi DC population

(Fig. 1B). In addition, a high proportion of LX+ DCs in the brLN

was CD103+, suggesting that they were actually of lung origin, as

CD103 is a surface marker of a large subpopulation of lung DCs

(15). To confirm this hypothesis, we applied LX i.t. into CCR7-

deficient recipient mice, which are known to possess a profound

defect in the mobilization of DCs from the periphery to the

draining LNs. As shown by flow cytometric analysis, LX+ DCs

were virtually absent from brLNs of CCR72/2 animals, con-firming the notion that LX+ cells present in the brLNs of wild-type

(wt) mice are indeed migratory DCs of lung origin (Fig. 1C).

Furthermore, LX+ cells isolated from the brLN by FACS sorting

displayed a typical DC-like morphology when analyzed by cyto-

spin (Fig. 1D). Thus, the i.t. application of fluorescent LX allows

the subsequent identification of lung-derived DCs in the brLN

under steady-state conditions.

Leukocytic infiltration in the lung and draining brLN after i.t.

application of LX in the absence or presence of LPS

To further characterize the i.t. application of LX as a model system

for the encounter of airway Ag under inflammatory and non-

inflammatory conditions, we analyzed by flow cytometry the

amount of leukocytic infiltration as well as the phenotype of Ag-

bearing cell populations in the lung after i.t. application of LX with

or without LPS. Twenty to twenty-four hours posttreatment, four

main subpopulations of phagocytic cells, distinguished by the

differential expression of CD11b and CD11c, could be detected in

the lungs of recipient mice under both conditions (Fig. 2A).

However, indicative for a situation of acute lung inflammation

(19), we observed a particularly strong recruitment of CD11b+

CD11c2 neutrophils and CD11b+CD11clo monocytes in animals

treated with LX+LPS (Fig. 2A, 2B). As both of these populations

express CD11b, they were further distinguished by their expres-

sion of CX3CR1 (20). CD11b+CD11c2 neutrophils were found to

express intermediate levels of CX3CR1, whereas CD11b+CD11clo

monocytes expressed high levels of CX3CR1 (data not shown).

The increase in these populations under inflammatory conditionswas accompanied by a concomitant reduction of CD11c+CD11b2

alveolar macrophages, whereas the absolute number of DCs in-

creased under these conditions (Fig. 2A, 2B). Similar to the situation

in PBS-treated and untreated mice, only a moderate leukocyte re-

cruitment into the lung was observed in animals treated with plain

FIGURE 1. LX-bearing cells within brLNs are predominantly lung-derived migratory DCs. Yellow-green fluorescent LX were applied i.t. into the

airways of mice. Twenty to twenty-four hours postapplication, leukocytes were isolated from lungs (A) and brLNs (B), and the phenotype of LX-bearing

cells was analyzed by flow cytometry using Abs against CD11c, CD11b, MHCII, and CD103 as indicated. C, Number of LX+ DCs (CD11c+ MHCIIhigh)

present in brLNs of CCR72/2 and wt recipients 24 h after i.t. application of fluorescent LX. D, Cytospin of FACS-sorted CD11c+ LX+ cells from brLNs of

wt BALB/c recipients. Scale bar, 20 mm. Data shown are representative for two to three independent experiments.

The Journal of Immunology 1319

yg

y

,

pj

g


5/12

LX (Fig. 2A, 2B). The phenotype of LX-bearing cells within the

draining brLN was highly similar after treatment with either plain

LX or LX+LPS. In both cases, LX-bearing cells in the brLN were,

to a large degree, confined to the CD11c+MHCIIhiCD103+ pop-

ulation of migratory DCs (Figs. 1B, 2C). Of note, however, the

mean number of DCs migrating into the brLN increased 7-fold in

the LX+LPS-treated mice compared with mice treated with plain

LXs (Fig. 2D), which is in line with a strongly increased DC traf-

ficking that occurs under inflammatory conditions.

In conclusion, the i.t. application of LXs in the presence or

absence of LPS represents an antigenic challenge within the air-

ways under inflammatory or noninflammatory conditions, poten-

tially leading to the induction of respiratory immunity or tolerance,

respectively. The differing patterns of leukocytic infiltrations in the

lung closely mirror the state of inflammation, whereas the phe-

notype of LX+ APCs arriving in the brLN is largely confined to

migratory DCs in both cases.

Kinetics of LX+ lung-derived DC migration into draining brLN

Ag uptake by DCs within peripheral tissues is followed by their

migration into the paracortical T cell zone of draining LNs, where

they are ideally positioned to present Ags to naive T cells. To

determine the kinetics of DC migration from the lung toward the

draining brLN, we have performed both histological and flow

cytometric analyses, addressing the localization as well as the

absolute numbers of LX+ cells present at various time points after

i.t. application of LX. In line with previously published data for

intranasal administration of LX (21), the histological analysis

showed that single LX+ cells can be detected in the subcapsular

sinus as early as 12 h after i.t. administration (data not shown).

Sixteen hours after i.t. application, the first LX+ DCs clearly

populate interfollicular regions of the brLN (Fig. 3A). Flow cy-

tometric analysis of both LX+CD11c+MHCIIhi and LX+CD11c+

CD103+ DC populations confirmed that the first lung-derived

migratory DCs arrived in the brLN 12 h after the instillation of LX

FIGURE 2. Differential leukocyte recruitment

within lung and brLN of mice treated with LX under

inflammatory and noninflammatory conditions. A,

Yellow-green fluorescent LX were i.t. applied into

BALB/c mice in the presence (LX+LPS) or absence

(LX) of LPS as an inflammatory stimulus; 24 h later,

leukocytes isolated from the recipients lungs were

analyzed by flow cytometry for the expression ofCD11b and CD11c. B, Number of CD11b+CD11c2

granulocytes, CD11b+CD11clo monocytes, CD11c+

CD11b+ DCs, and CD11c+CD11b2 macrophages in

the lungs of untreated, PBS, LX-, and LX+LPS-treated

animals 24 h after i.t. application. C, Phenotype of LX+

cells in the brLN of LX+LPS-treated animals 24 h after

i.t. application. D, Number of CD11c+MHCIIhi DCs

present in the brLN of untreated, PBS, LX-, and LX

+LPS-treated animals 24 h after i.t. application. Data

shown are representative of at least two independent

experiments.

FIGURE 3. Lung-derived LX+ DCs localize to in-terfollicular regions of the brLN within 16 h after LX

application. A, At 16 h or 38 h after the application of

LXs, brLNs were isolated, and cryosections of brLNs

were stained as indicated. Scale bars, 200 mm. B, At

various time points after i.t. application of LXs, leu-

kocytes from brLNs of wt recipients were isolated and

analyzed by flow cytometry. The percentage of LX+

cells within CD11c+MHCIIhi and CD11c+CD103+ DCs

was calculated.


yg

y

,

pj

g


6/12

(Fig. 3B). The fact that no LX+ cells were detectable in the brLN

at earlier time points strongly supports the idea of an active car-

riage of LX particles by migrating DCs instead of a passive

transport by the lymph flow. Sixteen hours after LX application,

7.7 6 3.3% of all brLN DCs were found to harbor LX (Fig. 3B).

Thus, 16 h also seemed to represent an adequate time point for the

dynamic imaging of T cell interactions with early arriving DCs by

two-photon microscopy. After 24 h, already 20.8 6 6.1% of all

brLN DCs were LX

+

(Fig. 3B). Finally, the most pronounced ac-cumulation of LX+ DCs in the T cell zone of brLNs could be ob-

served at 40 h after i.t. application (Fig. 3A, 3B). As LX+ DCs

were not present earlier than 16 h in the paracortical T cell zone of

the brLN in sufficient quantities to be easily observed by two-

photon microscopy, we started analyzing the interaction dynamics

of T cells and early arriving DCs at 1620 h after i.t. application of

LX. As we further observed a pronounced, almost 15-fold increase

in the number of migratory DCs present in the brLN within the

following day (not shown), we decided to additionally study the

cellular dynamics of Ag-dependent contacts between T cells and

migratory DCs at 3640 h after i.t. application of the particulate Ag.

OVA-LXbearing DCs are potent APCs

To visualize the Ag presentation of endogenous lung-derived DCswithin the brLN, we aimed to combine the in vivo labeling ap-

proach using fluorescent LX with the targeted delivery of protein

Ag. Therefore, OVA was passively adsorbed to fluorescent LX prior

to their i.t. application. To test whether DCs bearing OVA-LXs

efficiently present OVA-derived peptides to naive T cells, we an-

alyzed the in vivo proliferation of OVA-specific transgenic

DO11.10 CD4+ T cells in the brLN following i.t. application of

OVA-LX. For this purpose, we adoptively transferred MACS-

purified CFSE-labeled CD4+ T cells isolated from DO11.10 donor

mice, which carry a transgenic TCR specific for an MHC class II-

restricted OVA peptide prior to the i.t. application of OVA-LX,

OVA-LX+LPS, or plain LXs. The expansion of DO11.10 CD4+

T cells in the brLN was evaluated 4 d after the i.t. challenge. Asexpected, the i.t. application of OVA-LX as well as OVA-LX+LPS,

but not of plain LX, induced a massive proliferation of DO11.10

CD4+ T cells within the brLN (Fig. 4). Together, these data

demonstrate that the i.t. application of fluorescent LX loaded with

OVA results in an efficient activation of Ag-specific T cells in the

draining brLN under both inflammatory and potentially tolerizing

conditions.

Induction of Ag-specific tolerance and immunity after i.t.

application of OVA-LX or OVA-LX+LPS

The previous experiment had shown that i.t. application of both

OVA-LX and OVA-LX+LPS leads to a strong proliferation of Ag-

specific T cells in the draining brLN. To further characterize the

nature of the ensuing immune response with regard to the lung asa target organ, mice received two i.t. applications of either OVA-LX

or OVA-LX+LPS prior to a systemic OVA-ALUM sensitization by

i.p. injection. Two weeks postsensitization, the same animals were

rechallenged i.t. with either OVA-LX or OVA-LX+LPS (Fig. 5A).

Twenty-four hours later, the number and phenotype of infiltrating

cells present in the BAL, including granulocytes, monocytes, DCs,

and lymphocytes, were determined by flow cytometry, and the

lungs were examined histologically. As a negative control, one

group of animals received only plain LX by i.t. application at the

corresponding time points. All animals treated i.t. with OVA-LX

+LPS displayed a strong inflammatory immune response in the

lung, leading to the massive recruitment of infiltrating cells as

evidenced by FACS (Fig. 5B, green columns) and histologic ex-

amination (Fig. 5E, arrow). In contrast, the second i.t. challengedid not result in the development of overt lung inflammation in

OVA-LX-treated mice (Fig. 5B, red columns; Fig. 5D), despite the

previous systemic sensitization with OVA-ALUM. Whereas the

number of lymphocytes within the lungs is slightly higher in these

animals compared with those receiving plain LX (Fig. 5B, black

columns), the numbers of all inflammatory cell populations are not

significantly increased (Fig. 5B). Furthermore, whereas OVA-LX

+LPS-treated mice display massive peribronchiolar infiltrations

(Fig. 5E), the lung tissues of OVA-LXtreated, and LX-treated

animals were virtually indistinguishable by histological analysis

(Fig. 5C, 5D). Taken together, these data indicate that i.t. appli-

cation of OVA-LX induces a state of respiratory tolerance re-

garding rechallenge with the same Ag, whereas coadministrationof LPS induces respiratory immunity.

Dynamics of CD4+

T cell migration in brLNs during induction

of tolerance and immunity to pulmonary particulate Ags

To study by two-photon microscopy thedynamic behavior of T cells

and DCs during induction of tolerance and immunity to lung-de-

livered Ags, we adoptively transferred TAMRA-labeled CD4+

DO11.10 T cells prior to i.t. application of either plain LX in the

presence or absence of LPS, of OVA-LX, or of OVA-LX+LPS. As

previously described (22), we injected antiL-selectin mAb i.v. 3 h

after initial transfer of DO11.10 T cells to prevent further lym-

phocyte homing into LNs, thereby allowing the analysis of LN-

resident T cells with synchronized dwell times (Fig. 6A). Between16 and 36 h after application of LXs, brLNs of recipient animals

were harvested and placed in a custom-built incubation chamber

for two-photon imaging. Explanted brLNs were kept at a constant

temperature of 37C and were continually superfused with O2-

enriched RPMI medium.

Adoptively transferred TAMRA-labeled T cells, as well as LX-

bearing migratory DCs, were observed to primarily localize to the

paracortical T cell zone of brLNs (Fig. 3A). Consequently, further

imaging analysis was particularly focused on this region 200

240 mm below the subcapsular sinus of explanted brLNs. De-

pending on the presence or absence of OVA adsorbed to LXs, the

intranodal migration behavior of the OVA-specific DO11.10

T cells was highly different: Whereas TAMRA+ DO11.10 T cells

exhibited a highly motile migration covering large areas of theT cell zone after application of plain LXs (Fig. 6B, upper panel),

FIGURE 4. LX-bearing DCs induce strong Ag-specific T cell pro-

liferation within brLNs under inflammatory and noninflammatory con-

ditions. A total of 5 3 106 CFSE-labeled MACS-purified CD4+ DO11.10

T cells were adoptively transferred into wt recipient mice. One day later,

mice received i.t. either plain LX or OVA passively adsorbed to LX (OVA-

LX) in the presence or absence of LPS. Five days after the adoptive

transfer of CD4+ DO11.10 cells, brLNs were analyzed by flow cytometry.

Four animals per group were analyzed.


yg

y

,

pj

g


7/12

the i.t. application of LXs coupled with Ag resulted in a slower,

much more confined movement, being largely restricted to the

vicinity of individual LX+ Ag-loaded DCs (Fig. 6B, middle and

lower panels). At 16 h after the application of plain LXs, T cells

displayed average velocities of 11.7 mm/min (Fig. 6C, Supple-

mental Video 1), whereas T cells of OVA-LX or OVA-LX+LPS

treated animals exhibited median cellular velocities of 6.5 mm/min

and 6.2 mm/min, respectively (Fig. 6C, Supplemental Videos 2, 3).

At 36 h after i.t. application, DO11.10 T cells within the brLNs ofOVA-LX and OVA-LX+LPStreated animals showed an 57%

and 71% increase in T cell velocities compared with the 16-h

time point: OVA-LX, 10.2 mm/min; OVA-LX+LPS, 10.6 mm/min

(Fig. 6C, Supplemental Videos 4, 5). Thus, during this later stage

of their interactions with Ag-presenting DCs, DO11.10 T cells had

already substantially regained motility. Finally, the mean velocity

of DO11.10 T cells migrating in the brLNs of untreated mice (data

not shown) was 12 mm/min, equaling the value reported for skin-

draining peripheral LNs (23). Taken together, analysis of the

migration velocity of DO11.10 T cells within brLNs after i.t.application of Ags revealed that neither at early nor at late time

points were significant differences in the cellular velocity of

DO11.10 T cells observed between conditions leading to induction

of tolerance and those leading to immunity. Therefore, to compare

in more detail the character and directionality of migration, we

additionally analyzed the following motility parameters: mean

displacement, motility coefficient, arrest coefficient, and distri-

bution of turning angles.

By plotting for all conditions the mean displacement of all

tracked cells against the square root of time (Fig. 6D; Ref. 13), we

found a clearly more confined motility of Ag-specific T cells at

the early compared with late time point, which was again in-

dependent of the presence or absence of LPS (Fig. 6D, solid

lines). Analysis of the motility coefficient not only provides in-formation about the propensity of a cell to displace from its

starting position but also integrates information about cell ve-

locity (24). Consistent with the observation of intense T cellDC

interactions with early arriving LX+ DCs 16 h after i.t. application

of OVA-LX or OVA-LX+LPS, the mean motility coefficients

dramatically decreased under both tolerance- and immunity-

inducing conditions (8.1 mm2/min and 7.8 mm2/min, respectively).

Correspondingly, DO11.10 T cells regained higher motility co-

efficient values at later time points investigated, indicating

a change of the intranodal migration behavior toward higher

displacement under both conditions (Fig. 6E). To identify pauses

in the migration path of individual T cells, we calculated so-called

arrest coefficients for each condition analyzed. This parameterexpresses the percentage of time a cell is considered to be non-

motile (i.e., displaying an instantaneous velocity ,2 mm/min;

compare Ref. 11). As the initiation of long-lasting stable contacts

between T cells and DCs is known to result in a relative immo-

bility of the participating T cell, a low arrest coefficient indicates

that pauses in the T cell tracks and therefore potential stable

contacts are rare, whereas high values imply the presence of

stable T cellDC contacts. Compared with animals treated with

plain LXs, OVA-LXtreated animals displayed a 2.2-fold increase

in the arrest coefficient, whereas OVA-LX+LPStreated animals

showed even a 2.8-fold increase 16 h after Ag challenge (Fig.

6F). Thus, DO11.10 T cells within the brLNs of OVA-LX+LPS

treated animals show the highest propensity to remain immotileduring the early phase of T cellDC interactions, suggesting the

predominance of stable T cellDC contacts. Consistent with the

increase in cell velocities and motility coefficients, the arrest co-

efficient decreased during later stages of T cellDC interactions

(Fig. 6F). Finally, analysis of the turning angle distribution con-

firmed the existence of subtle differences in T cellDC interaction

dynamics in brLNs during conditions leading to tolerance or im-

munity. At early time points, DO11.10 T cells displayed higher

turning angles under conditions that were observed to induce

strong airway inflammation compared with conditions of re-

spiratory tolerance induction (median turning angle value of 76

for OVA-LX+LPS and 67 for OVA-LX; Fig. 6G), again in-

dicating a more confined migration under inflammatory con-

ditions. At late time points after i.t. challenge, the observedturning angle distribution reverted to lower values (Fig. 6H).

FIGURE 5. Repeated i.t. application of OVA-LX+LPS leads to the

formation of inflammatory lung infiltrates. A, Experimental setup of OVA-

LX sensitization. BALB/c mice received two i.t. applications of either

OVA-LX or OVA-LX+LPS (1 challenge). Both groups were subsequently

immunized systemically by i.p. injection of OVA-ALUM. Two weeks later,

all animals received two further i.t. applications (2 challenge) of the same

preparation that was used in the 1 challenge. Control animals received

plain LXs at the time points of 1 and 2 challenge. B, The number of total

cells, granulocytes (GR), monocytes (MO), DCs, and lymphocytes (LYM)

present in the BAL was determined by FACS 24 h after the last i.t. ap-

plication. CE, Lungs of LX-, OVA-LX, and OVA-LX+LPStreated mice

were harvested after the 2 challenge and lung cryosections stained with

H&E. Scale bars, 200 mm. pp , 0.05; ppp , 0.01; pppp , 0.001. Data

shown are pooled from two independent experiments.


yg

y

,

pj

g


8/12

To address the question of whether inflammation per se could

influence the migration behavior of naive T cells, we also analyzed

motility parameters of DO11.10 T cells after i.t. applicationof plain

LX together with LPS. Compared with the situation after appli-

cation of plain LX alone, the additional application of LPS resultedin a moderate butsubstantial decrease of T cell velocities (11.7mm/

min versus 8.5 mm/min; Fig. 6C). The observed lower slope of the

mean displacement plot (Fig. 6D) and, consequently, the reduced

motility coefficient (Fig. 6E) indicated a more confined migration

pattern of T cells during inflammation, which is, as well, reflected

in changes of the arrest coefficient (Fig. 6F) and the turning angledistribution (median turning angle of 67 for LX+LPS; data not

FIGURE 6. Analysis of CD4+ T cell motility during the induction of tolerance and immunity. A, 5 3 106 TAMRA-labeled CD4+ DO11.10 T cells were

adoptively transferred into wt recipient mice. One hour later, mice received either LX alone or together with LPS or OVA passively adsorbed to LX in the

presence (OVA-LX+LPS) or absence (OVA-LX) of LPS. Three hours after the initial transfer of DO11.10 T cells, mice were injected i.v. with 100 mg Mel-

14 (antiL-selectin) mAb to prevent further LN homing of lymphocytes. Between 16 and 40 h after i.t. challenge, brLNs were imaged by two-photonmicroscopy. B, Migration and interaction behavior of TAMRA-labeled CD4+ DO11.10 T cells (red) within the brLNs of LX-, OVA-LX, or OVA-LX

+LPStreated animals 16 h after i.t. challenge. LX+ DCs are depicted in green. Color-coded tracks (blue, start of imaging; white, end of imaging) illustrate

the paths of individual DO11.10 T cells (image data left out for clarity in the right panel). Grid, 20 mm. C, Average velocities of DO11.10 T cells. Each dot

represents an individual cell; bars indicate median values. Data shown are representative from one of at least three independent experiments. D, Mean

displacement plot calculated for DO11.10 T cells migrating within the brLNs 16 h (solid line) and 36 h (dashed line) after the indicated i.t. challenge. E,

Motility coefficient. F, Arrest coefficient. Cells with instantaneous three-dimensional velocities of,2 mm/min were considered immotile. Turning angle

distribution of DO11.10 T cells within brLNs of LX-, LX+LPS-, OVA-LX, or OVA-LX+LPStreated animals at 16 h (G) and 36 h (H) after i.t. challenge.

pp , 0.05; ppp , 0.01; pppp , 0.001; ns, nonsignificant. Data in DH are pooled from at least three independent experiments.


yg

y

,

pj

g


9/12

shown). Taken together, inflammatory conditions induced by LPS

are obviously able to modulate the basal motility level of naive

T cells in the brLN also in the absence of cognate Ag. Importantly,

however, the T cell motility reduction observed under these con-

ditions is clearly less pronounced compared with the effect of

cognate Ag on the migration behavior of naive T cells (Fig. 6).

These results extend circumstantial findings of a decreased T cell

motility in the presence of LPS (22) not observed in a different

experimental setup (11).In summary, the migration behavior of OVA-specific DO11.10

T cells within brLNs appeared to be largely comparable after i.t.

application of OVA-LXs in the presence or absence of LPS. In both

cases, Ag-specific T cells frequently engaged in contacts with LX+

DCs early (16 h) after Ag challenge, displaying a strongly reduced

overall motility and displacement, while clearly regaining cellular

motility at 3640 h after i.t. application. Arrest coefficient and

turning angle distribution suggested a slightly higher confinement

of DO11.10 T cell movement during the early phase of Ag rec-

ognition under immunizing (OVA-LX+LPS) compared with tol-

erizing (OVA-LX) conditions, warranting further detailed

analysis. In addition, our data strongly suggest that inflammatory

processes can affect the intranodal migration behavior of naive

T cells even in the absence of cognate Ag.

Size and stability of DCT cell interaction clusters during Ag

presentation in brLNs under tolerance- and immunity-inducing

conditions

As differences in the interaction dynamics of T lymphocytes and

DCs can easily be underestimated when analyzing the migration

behavior of Ag-specific T cells solely on the population level, we

next addressed the quality of T cellDC interactions occurring in the

brLN 16 h after i.t. challenge on the level of individual cellular

contacts. To this end, contacts between TAMRA+ DO11.10 T cells

and LX+ DCs were defined as stable when detectable throughout

a complete imaging period (usually lasting 22.525 min). All other

T cellDC interactions that were detected for at least five successivetime points (contact duration .75sec) were designated as transient

contacts. Using this classification system, we found that under

conditions leading to the induction of immunity, LX+ DCs pre-

dominantly formed large interaction clusters, stably interacting with

three or more Ag-specific T cells at the same time (Fig. 7A). By

contrast, under tolerance-inducing conditions, LX+ DCs were more

frequently found to maintain stable contacts with only one or two

T cells at the same time (Fig. 7A). An additional analysis of the

duration of individual transient T cellDC interactions showed that,

on average, transient contacts lasted substantially longer after i.t.

application of OVA-LX with LPS (median contact duration, 6.5

min; Fig. 7B) than without (median contact duration, 4.5 min; Fig.

7B). Collectively, these data indicate that, compared with tolerance-

inducing conditions, lung-derived migratory DCs within the brLN

form more stable, long-lasting interactions with T cells during

presentation of particulate airborne Ag under conditions leading to

the induction of immunity.

DiscussionPrevious studies (10) as well as our own observations (9) had

revealed that tolerance induction to inhaled environmental Ags

relies on the migratory capacity of DCs actively transporting Ag

from the lung to the draining brLN. Although soluble Ags that

reach the brLN via afferent lymph and the conduit system are

readily taken up by LN-resident DCs, these cells fail to activate

T cells even when carrying large amounts of Ag (8). Thus, at least

under tolerance-inducing conditions, Ag presentation and sub-

sequent activation of T cells within the lung-draining brLN is

obviously restricted to those DCs that have actually taken up Ag in

the lung, subsequently migrating to the draining LN (9). In the

current study, we therefore aimed to visualize by two-photon

microscopy how specifically these lung-derived migratory DCsinteract with T cells during the presentation and recognition of

airborne Ag, comparing conditions leading to the induction of

either tolerance or immunity. Applying Ag-coated LX i.t., we

were able to effectively combine the in situ labeling of endoge-

nous lung-resident DCs with the delivery of the model Ag OVA.

The detailed quantitative analysis of the migration and interaction

dynamics of CD4+ T cells and Ag-loaded LX+ DCs in the brLN

revealed characteristic differences between the induction of tol-

erance and immunity toward airway-derived Ags: T cells dis-

played a slightly more confined migration behavior taking sharper

turns while forming more stable long-lasting interaction clusters

with Ag-presenting DCs under immunity-inducing conditions.

With a diameter of 0.5 mm, the LX used for Ag loading weresignificantly below the size range common to pollen of many trees

known to induce allergic airway diseases (25). However, the

allergen-carrying particles that finally enter the small airways and

therefore represent the actual targets for phagocytosis by lung-

resident immune cells have been identified as much smaller,

ranging from far below 0.5 to 4.5 mm (2628). We therefore

conclude that the model Ag used in this study is actually well

suited to model the uptake and transport of particulate Ags po-

tentially involved in allergic sensitization events.

FIGURE 7. Analysis of cluster size and contact duration during Ag-specific T cellDC interactions in the brLN. T cellDC contacts lasting for the

complete imaging period were defined as stable; otherwise they are designated as transient. A, Size distribution of stable interaction clusters between

DO11.10 T cells and LX+ DCs 16 h after i.t. application of OVA-LX or OVA-LX+LPS. Clusters were classified based on the number of T cells stably

interacting with an individual DC. Only DCs stably interacting with at least one T cell were included in the analysis B, Duration of transient contacts

between DO11.10 T cells and LX+ DCs 16 h after i.t. application of OVA-LX or OVA-LX+LPS. Transient contacts of individual T cells were grouped into

four categories with regard to contact duration. Data are derived from three mice per group analyzed.


yg

y

,

pj

g


10/12

Following i.t. application of LX, we found that several subsets of

phagocytic cells within the lung were able to take them up, thus

becoming stably labeled by the ingested fluorescent particles.

Importantly, however, only DCs, but neither neutrophils nor

macrophages, were found to subsequently migrate to the lung-

draining brLN. Inside the brLN, LX-bearing cells were observed to

localize almost exclusively within the paracortical T cell zone, as

expected for immigrating tissue-derived DCs. Thus, i.t. application

of OVA-coated fluorescent LX allowed for specific visualization ofendogenous lung-derived migratory DCs presenting the Ag of

interest within the brLN.

We hypothesized that the initial i.t. application of LXs coated

with OVA would induce tolerance, whereas the addition of LPS

would induce immunity against this model Ag. Indeed, results

presented in this study strongly support this hypothesis. Compared

with the same treatment without LPS, a massive infiltration of

inflammatory cells was observed in mice that received OVA-LX in

the presence of LPS prior to i.p. sensitization (Fig. 5).

Recent two-photon studies have analyzed the T cell migration

and interaction dynamics in skin-draining as well as mesenteric

LNs during the onset of tolerance and immunity (1113). Hugues

et al. (11) have imaged OT-I CD8+ T cells within skin-draining

LNs of wt recipients after systemically applying the model AgOVA coupled to an antiDEC205-mAb, thereby targeting it to LN-

resident DCs. They reported the transgenic T cells, at all time points

analyzed, to primarily form serial brief encounters with CD11c+

DCs under noninflammatory conditions, whereas such T cellDC

interactions were found to be stable and long lasting 1520 h

after Ag challenge under inflammatory conditions. In contrast to

these findings reported for CD8+ T cells, no substantial differences

regarding temporal changes in the intranodal migration dynamics

of OT-II CD4+ T cells had been observed in another study com-

paring T cellDC interactions within skin-draining LNs under

tolerance- and immunity-inducing conditions (12). Similar to the

previously described study, OVA coupled to antiDEC205 mAb

was injected i.v. for systemic Ag delivery to DCs, with or withoutthe coadministration of antiCD40-mAb as an adjuvant to induce

priming conditions. In both situations, the authors observed

a phase of markedly reduced T cell motility due to formation of

stable, long-lasting contacts with Ag-presenting DCs. Conse-

quently, the authors conclude that early Ag-dependent T cell arrest

on DCs seems to be a shared feature of tolerance and priming

associated with activation and proliferation (12). Of note, how-

ever, an increased proportion of arrested T cells at 1117 h after

Ag challenge, as well as a slower return to the rapid migration

phase, was observed after coadministration of the antiCD40-

mAb, suggesting the formation of relatively more stable initial

T cellDC contacts during priming (12). Analyzing the intranodal

motility of adoptively transferred DO11.10 CD4

+

T cells withinperipheral and mesenteric LNs, Zinselmeyer et al. (13) found

a largely similar movement behavior of these transgenic CD4+

T cells 20 h after oral application of OVA in the presence or

absence of the adjuvant cholera toxin. Interestingly, this group as

well reported characteristic differences in T cell clustering be-

havior: Although fewer clusters were found under immunity-

inducing conditions, such clusters were larger and more stable.

Our imaging of explanted brLNs by two-photon microscopy

revealed that T cells interacting with lung-derived migratory DCs

under inflammatory conditions exhibit a slightly more confined

migration behavior compared with tolerance-inducing conditions,

potentially resulting from a higher stability of established T cell

DC interactions. Indeed, we found that, under immunity-inducing

conditions, the majority of Ag-presenting lung-derived DCs formmore stable contacts than under tolerance-inducing conditions, on

average recruiting a higher number of Ag-specific T cells into

individual interaction clusters. Correspondingly, the formation of

transient contacts of shorter duration was more prominent under

tolerance-inducing conditions, and the average duration of these

transient T cellDC contacts was also shorter compared with

immunity-inducing conditions. Thus, similar to previous reports

on the interaction dynamics of T cells and DCs within skin-

draining and mesenteric LNs (1113), data presented in this paper

indicate that also in the lung-draining brLN, Ag-dependent con-tacts between T cells and DCs tend to be more stable under

conditions leading to immunity, whereas temporal changes in the

T cell migration behavior due to interactions with DCs seem to be

largely comparable during the induction of respiratory tolerance

and immunity.

The use of OVA-LX in the current study allowed for an un-

ambiguous identification of those DCs that had actually carried the

Ag of interest from the lung toward the draining LN. Thus, the type

of APC addressed by our approach differs decisively from that of

previous studies, which had analyzed the dynamics of T cellDC

interactions within skin-draining LNs by i.v. injecting OVA-con-

jugated antiDEC205-mAb into either transgenic CD11c-EYFP

mice (12) or wt recipients in which endogenous LN-resident DCs

had been labeled in vivo by Alexa488-conjugated antiCD11c-mAb (11). In both cases, delivery of the model Ag OVA as well as

distribution of the fluorescent labeling probably affected all

(DEC205+ or CD11c+, respectively) DC subpopulations present

within the LN under scrutiny, independent of their individual LN

dwell time and specific origin as tissue-derived migratory DCs or

progeny of blood-borne monocyte progenitors. Importantly, tissue-

derived migratory DCs not only differ from monocyte-derived

LN-resident DCs regarding their surface marker expression and

their capability of Ag presentation and costimulation (29) but have

also been shown to display an intranodal migration behavior quite

different from that of resident DCs (30). Thus, it seems likely that

T cells engaged in Ag-specific contacts with LN-resident DCs

might display motility patterns different from those interactingwith tissue-derived migratory DCs. Furthermore, the LN dwell

time distribution of different DC subpopulations might critically

influence the observed kinetics of T cell motility changes.

The observation that short-lasting transient and long-lasting

stable T cellDC contacts predominate at different time points

after delivery of the cognate Ag has led to a three-phase kinetic

model of T cell priming as initially proposed by Mempel et al.

(13). However, recent studies clearly demonstrated that additional

factors, such as Ag dose, peptideMHC potency, as well as T cell

and DC frequencies, essentially contribute to the stability of early

T cellDC interactions (3134). It is therefore not surprising that

studies addressing T cellDC interactions under tolerizing con-

ditions have reported different kinetics of DCT cell interaction:Hugues et al. (11) found a three-phase kinetic of T cellDC in-

teractions, whereas Zinselmeyer et al. (13) describe the existence

of an early phase, when T cells form transient contacts at 8 h after

Ag delivery, and a late phase characterized by stable contact

formation 20 h after Ag delivery. Stable contacts of DCs and

T cells 16 h after Ag delivery were also reported by Shakhar

et al. (12). Obviously, the experimental models employed by

different groups to study T cellDC interaction dynamics during

the induction of tolerance and immunity have varied with respect

to several of the above-mentioned influencing factors. Considering

this, it seems even more likely that the shared observation of an at

least slightly higher stability of T cellDC contacts under priming

conditions might indeed represent a general phenomenon associ-

ated with the induction of immunity, regardless of the type of LNunder investigation.


yg

y

,

pj

g


11/12

Several factors might actually have contributed to the charac-

teristic differences in T cellDC interaction dynamics observed

during the induction phase of respiratory immune reactions within

brLN. First, when being exposed to inflammatory mediators, DCs

have been shown to undergo massive morphological changes,

resulting in the up- and downregulation of each of several thou-

sand genes (35). Consequently, a more efficient presentation of

exogenous Ags resulting from changes in the metabolism of MHC

class II molecules, as well as a higher surface expression ofcostimulatory molecules on DCs migrating to the brLN under

inflammatory conditions, might have contributed to the observed

higher interaction stability. Second, it seems possible that under

inflammatory conditions in the lung, the composition of migratory

APC arriving in the brLN might be different compared with

steady-state conditions. However, flow cytometric analysis of LX-

bearing cells in the brLN (Figs. 1B, 2C) indicated that lung-

derived DCs are clearly the dominant LX+ cell population after i.t.

application of OVA-LX in the presence as well as absence of LPS,

although they probably differ in their maturation status (see

above). Third, changes in the behavior of the LN-resident DCs,

which are specifically not visualized in our setup, might have

accounted for differences in the movement behavior of Ag-

specific T cells under inflammatory versus noninflammatoryconditions. If so, this hypothetical influence of LN-resident DCs

most probably did not relate to actual Ag presentation, as we

observed clusters of TAMRA+ DO11.10 T cells exhibiting a con-

fined migration behavior typical of Ag-specific interactions (22,

23) only in the direct vicinity of LX+ (lung-derived) migratory

DCs. Fourth, besides Ag-dependent and -independent interactions

between T cells and different DC populations, stroma cells of the

T cell zone have been demonstrated to decisively affect the in-

tranodal migration of T cells (36). These fibroblastic reticular cells

(FRCs) present key guidance cues for T cells and form a structural

network that naive T lymphocytes use during their apparent

random walk migration within the paracortical T cell zone (36).

As FRC undergo substantial remodeling under inflammatoryconditions (37), it seems possible that these cells might also shape

the intranodal migration behavior of T cells differentially during

the event of priming or tolerization. However, such a general

change in the migration-modulating influence of the nonlabeled

FRC under steady-state versus inflammatory conditions should

have been observable throughout the imaged T cell zone, in-

dependent of the localization of a T cell relative to the positions of

the LX+ migratory DCs. In contrast, most general parameters of T

cell movement were largely similar during the induction of tol-

erance and immunity, and the most pronounced differences in T

cell behavior were clearly related to the interaction dynamics of

direct cell-to-cell contacts with LX+ DCs, making a critical role of

the FRC network rather unlikely.Finally, the observed differences in cellular dynamics during Ag

presentation are in line with the previously proposed concept that

long-lasting stable contacts to Ag-presenting DCs actually increase

the probability that a T cell will receive those activating signals that

lead to the development of immunity (38). In contrast, either the

relatively lower stability of T cellDC interactions observed under

conditions leading to the induction of respiratory tolerance may

indicate a different quality of DC signaling, or the lower duration

of individual T cellDC contacts itself might be the factor de-

termining the tolerant fate of this specific T cell.

In addition to cognate Ag, the current study also reveals that

ligands for Toll-like receptors can profoundly affect the migration

behavior of naive T cells independent of the presence of cognate

peptide ligands. The application of plain LX together with LPSclearly reduced the velocity, mean displacement, and motility

coefficient of DO11.10 cells, although not to the same extent as the

delivery of OVA. The underlying mechanisms are currently un-

known, but it seems likely that changes in the LN environment

induced by LPS, such as upregulation of adhesion molecules

(potentially affecting stromal cells and/or resident DCs), might

contribute to the effects observed.

In summary, imaging the intranodal migration of DO11.10

T cells within brLNs at different time points after i.t. application of

OVA, we find a largely similar kinetic of changes in the intranodalmigratory behavior of Ag-specific T cells under conditions leading

to tolerance and immunity. However, analyzing the size and sta-

bility of T cellDC interaction clusters, we find large clusters of

long-lasting stable contacts to be most prominent under in-

flammatory conditions, whereas shorter transient contacts prevail

under steady-state conditions. Therefore, paralleling the situation

within skin-draining LNs of the systemic compartment, the in-

duction of respiratory tolerance and immunity toward particulate

airborne Ags might actually be influenced on the level of cellular

dynamics of T cell migration and interaction with migratory DCs

within brLNs.

AcknowledgmentsWe thank Andreas Krueger for critically reading this manuscript.

DisclosuresThe authors have no financial conflicts of interest.

References1. Anderson, H. R. 2005. Prevalence of asthma. BMJ 330: 10371038.2. Steinman, R. M., D. Hawiger, and M. C. Nussenzweig. 2003. Tolerogenic

dendritic cells. Annu. Rev. Immunol. 21: 685711.3. Steinman, R. M. 2008. Dendritic cells in vivo: a key target for a new vaccine

science. Immunity 29: 319324.4. Lambrecht, B. N., and H. Hammad. 2003. Taking our breath away: dendritic

cells in the pathogenesis of asthma. Nat. Rev. Immunol. 3: 9941003.5. Lutz, M. B., and G. Schuler. 2002. Immature, semi-mature and fully mature

dendritic cells: which signals induce tolerance or immunity? Trends Immunol.23: 445449.

6. van Rijt, L. S., S. Jung, A. Kleinjan, N. Vos, M. Willart, C. Duez,H. C. Hoogsteden, and B. N. Lambrecht. 2005. In vivo depletion of lung CD11c+dendritic cells during allergen challenge abrogates the characteristic features ofasthma. J. Exp. Med. 201: 981991.

7. Jahnsen, F. L., D. H. Strickland, J. A. Thomas, I. T. Tobagus, S. Napoli,G. R. Zosky, D. J. Turner, P. D. Sly, P. A. Stumbles, and P. G. Holt. 2006. Ac-celerated antigen sampling and transport by airway mucosal dendritic cellsfollowing inhalation of a bacterial stimulus. J. Immunol. 177: 58615867.

8. Ohl, L., M. Mohaupt, N. Czeloth, G. Hintzen, Z. Kiafard, J. Zwirner,T. Blankenstein, G. Henning, and R. Forster. 2004. CCR7 governs skin dendriticcell migration under inflammatory and steady-state conditions. Immunity 21:279288.

9. Hintzen, G., L. Ohl, M. L. del Rio, J. I. Rodriguez-Barbosa, O. Pabst,J. R. Kocks, J. Krege, S. Hardtke, and R. Forster. 2006. Induction of tolerance toinnocuous inhaled antigen relies on a CCR7-dependent dendritic cell-mediated

antigen transport to the bronchial lymph node. J. Immunol. 177: 73467354.10. Hammad, H., and B. N. Lambrecht. 2007. Lung dendritic cell migration. Adv.

Immunol. 93: 265278.11. Hugues, S., L. Fetler, L. Bonifaz, J. Helft, F. Amblard, and S. Amigorena. 2004.

Distinct T cell dynamics in lymph nodes during the induction of tolerance andimmunity. Nat. Immunol. 5: 12351242.

12. Shakhar, G., R. L. Lindquist, D. Skokos, D. Dudziak, J. H. Huang,M. C. Nussenzweig, and M. L. Dustin. 2005. Stable T cell-dendritic cell inter-actions precede the development of both tolerance and immunity in vivo. Nat.

Immunol. 6: 707714.13. Zinselmeyer, B. H., J. Dempster, A. M. Gurney, D. Wokosin, M. Miller, H. Ho,

O. R. Millington, K. M. Smith, C. M. Rush, I. Parker, et al. 2005. In situcharacterization of CD4+ T cell behavior in mucosal and systemic lymphoidtissues during the induction of oral priming and tolerance. J. Exp. Med. 201:18151823.

14. Forster, R., A. Schubel, D. Breitfeld, E. Kremmer, I. Renner-Muller, E. Wolf,and M. Lipp. 1999. CCR7 coordinates the primary immune response by estab-lishing functional microenvironments in secondary lymphoid organs. Cell 99:

2333.15. Sung, S. S., S. M. Fu, C. E. Rose, Jr., F. Gaskin, S. T. Ju, and S. R. Beaty. 2006. Amajor lung CD103 (alphaE)-beta7 integrin-positive epithelial dendritic cell


yg

y

,

pj

g


12/12

population expressing Langerin and tight junction proteins. J. Immunol. 176:21612172.

16. Worbs, T., T. R. Mempel, J. Bolter, U. H. von Andrian, and R. Forster. 2007.CCR7 ligands stimulate the intranodal motility of T lymphocytes in vivo. J. Exp.

Med. 204: 489495.17. Jakubzick, C., F. Tacke, J. Llodra, N. van Rooijen, and G. J. Randolph. 2006.

Modulation of dendritic cell trafficking to and from the airways. J. Immunol.176: 35783584.

18. Gonzalez-Juarrero, M., T. S. Shim, A. Kipnis, A. P. Junqueira-Kipnis, andI. M. Orme. 2003. Dynamics of macrophage cell populations during murinepulmonary tuberculosis. J. Immunol. 171: 31283135.

19. Ulich, T. R., L. R. Watson, S. M. Yin, K. Z. Guo, P. Wang, H. Thang, and J. delCastillo. 1991. The intratracheal administration of endotoxin and cytokines. I.Characterization of LPS-induced IL-1 and TNF mRNA expression and the LPS-,IL-1-, and TNF-induced inflammatory infiltrate. Am. J. Pathol. 138: 14851496.

20. Lin, K. L., Y. Suzuki, H. Nakano, E. Ramsburg, and M. D. Gunn. 2008. CCR2+monocyte-derived dendritic cells and exudate macrophages produce influenza-induced pulmonary immune pathology and mortality. J. Immunol. 180: 25622572.

21. Byersdorfer, C. A., and D. D. Chaplin. 2001. Visualization of early APC/T cellinteractions in the mouse lung following intranasal challenge. J. Immunol. 167:67566764.

22. Mempel, T. R., S. E. Henrickson, and U. H. Von Andrian. 2004. T-cell primingby dendritic cells in lymph nodes occurs in three distinct phases. Nature 427:154159.

23. Miller, M. J., O. Safrina, I. Parker, and M. D. Cahalan. 2004. Imaging the singlecell dynamics of CD4+ T cell activation by dendritic cells in lymph nodes. J.Exp. Med. 200: 847856.

24. Sumen, C., T. R. Mempel, I. B. Mazo, and U. H. von Andrian. 2004. Intravitalmicroscopy: visualizing immunity in context. Immunity 21: 315329.

25. Burge, H. A., and C. A. Rogers. 2000. Outdoor allergens. Environ. HealthPerspect. 108(Suppl 4): 653659.

26. Taylor, P. E., R. Flagan, A. G. Miguel, R. Valenta, and M. M. Glovsky. 2003.Identification of birch pollen respirable particles. Chest 123(Suppl 3): 433S.

27. DAmato, G., G. Liccardi, and G. Frenguelli. 2007. Thunderstorm-asthma andpollen allergy. Allergy 62: 1116.

28. Bacsi, A., B. K. Choudhury, N. Dharajiya, S. Sur, and I. Boldogh. 2006. Sub-pollen particles: carriers of allergenic proteins and oxidases. J. Allergy Clin.

Immunol. 118: 844850.29. Itano, A. A., S. J. McSorley, R. L. Reinhardt, B. D. Ehst, E. Ingulli,

A. Y. Rudensky, and M. K. Jenkins. 2003. Distinct dendritic cell populationssequentially present antigen to CD4 T cells and stimulate different aspects ofcell-mediated immunity. Immunity 19: 4757.

30. Lindquist, R. L., G. Shakhar, D. Dudziak, H. Wardemann, T. Eisenreich,M. L. Dustin, and M. C. Nussenzweig. 2004. Visualizing dendritic cell networksin vivo. Nat. Immunol. 5: 12431250.

31. Henrickson, S. E., T. R. Mempel, I. B. Mazo, B. Liu, M. N. Artyomov, H. Zheng,A. Peixoto, M. P. Flynn, B. Senman, T. Junt, et al. 2008. T cell sensing of antigendose governs interactive behavior with dendritic cells and sets a threshold for Tcell activation. Nat. Immunol. 9: 282291.

32. Skokos, D., G. Shakhar, R. Varma, J. C. Waite, T. O. Cameron, R. L. Lindquist,T. Schwickert, M. C. Nussenzweig, and M. L. Dustin. 2007. Peptide-MHC po-tency governs dynamic interactions between T cells and dendritic cells in lymphnodes. Nat. Immunol. 8: 835844.

33. Varani, L., A. J. Bankovich, C. W. Liu, L. A. Colf, L. L. Jones, D. M. Kranz,J. D. Puglisi, and K. C. Garcia. 2007. Solution mapping of T cell receptor dockingfootprints on peptide-MHC. Proc. Natl. Acad. Sci. USA 104: 1308013085.

34. Celli, S., F. Lematre, and P. Bousso. 2007. Real-time manipulation of T cell-dendritic cell interactions in vivo reveals the importance of prolonged contactsfor CD4+ T cell activation. Immunity 27: 625634.

35. Huang, Q., D. Liu, P. Majewski, L. C. Schulte, J. M. Korn, R. A. Young,E. S. Lander, and N. Hacohen. 2001. The plasticity of dendritic cell responses topathogens and their components. Science 294: 870875.

36. Bajenoff, M., J. G. Egen, L. Y. Koo, J. P. Laugier, F. Brau, N. Glaichenhaus, andR. N. Germain. 2006. Stromal cell networks regulate lymphocyte entry, migra-tion, and territoriality in lymph nodes. Immunity 25: 9891001.

37. Katakai, T., T. Hara, M. Sugai, H. Gonda, and A. Shimizu. 2004. Lymph nodefibroblastic reticular cells construct the stromal reticulum via contact withlymphocytes. J. Exp. Med. 200: 783795.

38. Hugues, S., A. Boissonnas, S. Amigorena, and L. Fetler. 2006. The dynamics ofdendritic cell-T cell interactions in priming and tolerance. Curr. Opin. Immunol.18: 491495.


yg

y

,

pj

g

j immunol-2010-bako-evi--1317-27

Documents