Fibroblastic Reticular Cells Guide T Entry into and Migration within the Splenic Zone

This information is current as Marc Bajénoff, Nicolas Glaichenhaus and Ronald N. of September 24, 2021. Germain J Immunol 2008; 181:3947-3954; ; doi: 10.4049/jimmunol.181.6.3947 http://www.jimmunol.org/content/181/6/3947 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2008/08/29/181.6.3947.DC1 Material

References This article cites 41 articles, 16 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/181/6/3947.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists by guest on September 24, 2021

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Fibroblastic Reticular Cells Guide T Lymphocyte Entry into and Migration within the Splenic T Cell Zone1

Marc Baje´noff,*†‡ Nicolas Glaichenhaus,† and Ronald N. Germain2*

Although a great deal is known about T cell entry into lymph nodes, much less is understood about how T access the splenic (WP). We show in this study that, as recently described for lymph nodes, fibroblastic reticular cells (FRCs) form a network in the T cell zone (periarteriolar lymphoid sheath, PALS) of the WP on which T lymphocytes migrate. This network connects the PALS to the (MZ), which is the initial site of lymphocyte entry from the . T cells do not enter the WP at random locations but instead traffic to that site using the FRC-rich MZ bridging channels (MZBCs). These data reveal that FRCs form a substrate for T cells in the , guiding these lymphocytes from their site of entry in the MZ into the PALS, within which they continue to move on the same network. The Journal of Immunology,

2008, 181: 3947–3954. Downloaded from

he “spleen is quantitatively the most important organ in blood content (8). This critical function is ensured by specialized the lymphoid system, with more lymphocytes passing populations of termed marginal zone macrophages through this organ each day than all the other lymphoid and marginal zone metallophilic macrophages (MMMs), as well as T 3 tissues combined” (1). In lymph nodes (LNs), blood-borne lym- by dendritic cells designated marginal zone dendritic cells (9–11). phocytes enter the paracortex via , spe- Located between the MZ and the WP, the MZ sinus and its sinus- http://www.jimmunol.org/ cialized blood vessels that support the rolling, arrest, and diape- lining cells are believed to be the place where recently incoming desis of T and B lymphocytes across the endothelium into the lymphocytes can access the neighboring WP (4, 8, 12, 13), al- surrounding parenchyma (for reviews, see Refs. 2, 3). The molec- though there is little direct experimental support for this model. ular events involved in this migration from blood to LN are well Based on histological observation of human spleen sections, an characterized and sequentially involve selectins, , and alternative scheme has been proposed in which specialized fibro- integrins. In the spleen, high endothelial venules are absent (4) and blasts guide CD4ϩ T cell entry into the periarteriolar lymphoid although a key role for chemokines such as CCL21 and CXCL13 sheath (PALS) (14). are appreciated in the intrasplenic localization of T (5, 6) and B Recently, we demonstrated that a fibroblastic reticular cell

cells (7), respectively, there is scant evidence for any special role (FRC) network supports and guides T and motility in the T by guest on September 24, 2021 of nonhematopoietic structural elements of this organ in guiding cell area of LNs (15), dictating the apparent random migratory lymphocytes to their sites of accumulation once they have entered behavior of these cells. Lymphocytes adapt their shape to the cell from the vasculature. bodies and processes of these large stellate fibroblastic cells and The spleen has a complex and well-described microanatomy follow the supporting fibers of the FRCs during migration within (see Fig. 1A). The white pulp (WP), where T and B cell popula- the paracortical region (T cell zone) that is itself defined by the tions segregate, is surrounded by the (RP), a loose mesh- extent of this FRC network (15). In the spleen, the only known work of reticular fibers and fibroblasts where blood is filtered and function of the FRCs is their ability to create a conduit system that old erythrocytes removed (4). Localized between the WP and the transports blood-derived material inside the PALS in the same way RP, the marginal zone (MZ) creates a transit area for recently that the comparable FRC-based conduit system can transport immigrating blood lymphocytes as well as a filtering zone for lymph content in the paracortex of the LN (16, 17). In this study, we characterize the exact location and describe ad- ditional functions of the FRC network in the spleen. As anticipated *Lymphocyte Biology Section, Laboratory of Immunology, National Institute of Allergy from our previous work, we show that these stromal cells are located and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892; and †Institut National de la Sante´et de la Recherche Me´dicale and ‡Centre National de la Recherche in the splenic PALS and support T cell motility in this region. Sur- Scientifique, Universite´de Nice-Sophia Antipolis, Valbonne, France prisingly, we also found that FRCs connect the PALS to the MZ only Received for publication October 30, 2007. Accepted for publication July 14, 2008. where the MZ sinus and MZ macrophages rims are interrupted (i.e., The costs of publication of this article were defrayed in part by the payment of page the so-called MZ bridging channels (MZBCs)) (18). Using T cell charges. This article must therefore be hereby marked advertisement in accordance homing experiments, we show that T cells entering the PALS do not with 18 U.S.C. Section 1734 solely to indicate this fact. cross the MZ randomly but only use these FRC-rich bridging chan- 1 This research was supported in part by the Intramural Research Program of National nels. Thus, by their unique location, FRCs not only support T cell Institute of Allergy and Infectious Diseases, National Institutes of Health, Department of Health and Human Services, by the Institut de la Sante´et de la Recherche Me´dicale motility in the PALS but also provide access roads to this area for T (INSERM), and by the Centre National de la Recherche Scientifique (CNRS). cells that have recently immigrated into the spleen. 2 Address correspondence and reprint requests to Dr. Ronald N. Germain, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Building 10, Room 11N-311, 10 Center Drive, MSC 1892, Bethesda, MD 20892. E-mail ad- Materials and Methods dress: [email protected] Mice 3 Abbreviations used in this paper: LN, ; WP, white pulp; RP, red pulp; MZ, marginal zone; MMM, marginal metallophilic ; PALS, periarteriolar C57BL/6 and C57BL/6 ubiquitin-GFP mice (UBI-GFP/BL6, strain 4353) lymphoid sheath; FRC, fibroblastic reticular cell; MZBC, marginal zone bridging were purchased from The Jackson Laboratory and maintained in the channel; 2P, two photon. National Institutes of Health animal facilities. Hu-CD2 GFP mice were www.jimmunol.org 3948 REGULATION OF T CELL MIGRATION IN THE SPLENIC WP originally a gift from D. Kioussis (Mill Hill, London, U.K.). For the gen- eration of chimeras, C57BL/6 ubiquitin-GFP mice were gamma-irradiated with a single dose of 950 rads (or twice with 500 rads) from a cesium source and were reconstituted with 2 ϫ 106 C57BL/6 cells. At 8 wk after reconstitution, mice were tested for chimerism. Chimeras were used for subsequent experiments only if analysis of blood leukocytes showed the presence of less than 2% of CD3ϩ T cells of host origin. All procedures performed on animals in this study have been approved by the Animal Care and Use Committee, National Institute of Allergy and Infec- tious Diseases, National Institutes of Health.

Adoptive transfers T cells were purified from the LNs of wild-type mice with a pan T cell isolation kit while B cells were purified from their with a pan B cell isolation kit (Miltenyi Biotec). Cells were stained with either 5-chlorom- ethyl fluorescein diacetate-2 ␮M, carboxy-fluorescein diacetate, succinimi- dyl ester-2 ␮M, CellTracker red CMTPX (2.5 ␮M), or SNARF-1 (2.5 ␮M) (Invitrogen) at 37°C for 15 min. The indicated numbers of cells were trans- ferred into host mice by i.v. injection. Downloaded from Antibodies ERTR-7 Ab specific for an unknown FRC-secreted molecule and anti- desmin serum were purchased from Acris Abs. RA3–6B2 Ab specific for B220, 17A2 specific for the CD3 complex, and MECA-89 specific for MadCAM-1 were from BD Pharmingen. MOMA-1 Ab specific for MMM was purchased from Cedarlane. A goat polyclonal anti-murine CCL21/

6cKine was purchased from R&D Systems. These Abs were visualized by http://www.jimmunol.org/ direct coupling to biotin; allophycocyanin; Alexa Fluor 488, 568, or 647; or through the use of Alexa Fluor 488, 568, or 647 coupled secondary Abs or streptavidin.

Immunostaining Spleens were harvested and fixed in a 0.05 M phosphate buffer containing

0.1 M L-lysine (pH 7.4), 2 mg/ml NaIO4, and 10 mg/ml paraformaldehyde (PLP) for 12 h, then washed in phosphate buffer and dehydrated in 30% sucrose in phosphate buffer. Spleens were snap frozen in Tissue-Tek by guest on September 24, 2021 (Sakura Finetek). In brief, 10–30 ␮m frozen sections were cut and then FIGURE 1. MZBCs directly connect the MZ to the PALS. A, Schematic stained with the indicated Abs as previously described (19). For in situ fixation experiments, animals were anesthetized with avertin and given an representation of the cellular elements of a WP unit and its surrounding MZ intracardiac injection of 15 ml PLP fixative. After excision from perfused and RP. B, Two adjacent spleen cryostat sections from HuCD2 GFP animals, spleens were treated using the protocol described above. Immu- (green ϭ PALS) mice were stained for B220 (blue ϭ B cell follicles), nofluorescence confocal microscopy was performed with a Leica SP5 con- Madcam-1 (left panels; red ϭ MZ sinus), MOMA-1 (right panels; red ϭ focal microscope. Separate images were collected for each fluorochrome MMM), and imaged using confocal microscopy. This picture is represen- and overlaid to obtain a multicolor image. Quantitative analysis of T/B cell tative of two different experiments. distribution into different areas of the LN sections was performed using ImageJ software (National Institutes of Health). For each tissue section, the number of cells within manually defined regions of interest was calculated by the image processor. These values were then normalized to the number age stacks are maximum intensity projections and play at 100ϫ or 300ϫ of cells present in 0.2 mm2 of each analyzed region. Final image processing real time.4 was performed with ImageJ software (National Institutes of Health) and Adobe Photoshop. Results FRCs connect the MZ and PALS via the bridging channels Two photon (2P) microscopy In the spleen, the MZ and WP are separated by a rim of ϩ ϩ Freshly isolated T cells were labeled with CellTracker red or SNARF-1 and MOMA-1 MZ macrophages underlying the MadCAM-1 MZ injected i.v. into chimeric recipient mice. Twenty-four hours later (unless sinus (9). However, this rim is incomplete and interrupted in re- otherwise specified), the spleen was removed, fixed on a tissue holder, and gions known as MZBCs. Interestingly, these MZBCs are only sliced into two nonsymmetric pieces using a vibratome (Leica, VT 1000 S) present where PALS, but not B cell follicles, abut the MZ (Fig. in a bath of ice-cold PBS. The holder containing the thickest piece of 1B), suggesting that the PALS may be directly connected to the spleen was then incubated in a tissue chamber (Bioptechs). Splenic tissue was perfused with a 37°C RPMI 1640 medium bubbled with a gas mixture MZ at the site of these special bridging corridors. containing 95% O2 and 5% CO2 while being imaged with a Bio-Rad Ra- Given our previous demonstration that FRCs are the substrate diance 2100 MP system attached to a Nikon 600 FN upright microscope for migration of T cells in the paracortical region (T zone) of LNs fitted with a 20ϫ water immersion lens (NA ϭ 0.95, Olympus). The 2P (15), we characterized the exact location of FRCs in the WP as a laser was a Chameleon XR femtosecond pulsed laser (Coherent) tuned to 880 nm. The bandpass filters used to detect GFP and SNARF were 525/50 first approach to understanding how lymphocytes enter into and nm and 620/100 nm, respectively. Three-dimensional images were col- move within the PALS and the possible relationship of such move- lected in the PALS, a region characterized by the accumulation of dye- ment to this organization of the splenic microanatomy. Immuno- labeled T cells, the presence of a central arteriole, and surrounding FRC staining of spleen sections revealed that the conduit system and its fibers. This volume collection was repeated every 20–30 s to create 4-D data sets that were then processed with Imaris software (Bitplane) and Adobe AfterEffects (Adobe). Supplemental movies created from these im- 4 The online version of this article contains supplemental material. The Journal of Immunology 3949

FIGURE 2. FRCs connect the MZ to the PALS at the MZBC. Spleen cry- ostat sections from HuCD2 GFP (blue) mice were stained for ERTR-7 (green ϭ FRC-derived matrix protein), MOMA-1 (red, A and C) or desmin (B), and imaged using confocal microscopy. C, Downloaded from Inset shows a higher magnification of two MZBCs. Arrowheads point to the location of MZBCs while “B” indicates the pres- ence of B cell follicles. This picture is rep- resentative of three different experiments. See also movie S1. http://www.jimmunol.org/ by guest on September 24, 2021

associated FRCs, whose outlines can be respectively delineated by Confocal microscopy has been successfully used for intravital their reactivity with ERTR-7 and anti-desmin Abs, were present in imaging of the spleen (20, 21), but useful data can only be obtained the PALS and also in the MZBC (Fig. 2, A and B). A closer anal- in the most superficial region of the organ with this technology. ysis indicated that these reticular fibers span across from the MZ to Furthermore, because the WP is usually located deep in the mouse the PALS exactly where the rim of MMMs and the MZ sinus are spleen where thin GFPϩ FRC fibers are difficult to image effec- interrupted, that is, the MZBCs (Fig. 2C, movie S1). tively even by 2P microscopy (Ͼ200–300 ␮m under the thick capsule and surrounded by RP full of RBC that absorb the 2P laser FRCs support T cell motility in the PALS signal), direct intravital or whole spleen explant imaging was not The location of FRCs in the PALS suggested the possibility that adequate for our purpose. Therefore, to analyze the dynamic be- the FRC network might be the substratum for lymphocyte migra- havior of T cells in relation to the green fluorescent FRC network, tion in this region. To determine whether naive T cells migrate we developed a different approach. Wild type naive T cells were along FRC fibers in the PALS, these lymphocytes were visualized labeled with the red fluorescent dye SNARF-1 and injected i.v. in the spleens of mice using 2P laser scanning microscopy. To into chimeric animals. One day later, each recipient spleen was permit simultaneous imaging of the nonhematopoietic stromal cell sliced into two nonsymmetric pieces using a vibratome to allow populations within the spleen, we generated chimeric mice by us- direct imaging access to the deep WPs of the thickest piece. Bi- ing wild-type bone marrow cells to reconstitute irradiated ubiquitin sected spleens were then perfused with warm and oxygenated me- promoter-GFP transgenic animals, as previously reported (15). To dium while being imaged by 2P microscopy as previously reported confirm that GFP-expressing cells within the PALS of chimeric for LN vibratome sections (15). Using this technique, we found animals represented the FRC population, splenic sections from that T cells maintain their characteristic migratory behavior, mov- such animals were stained for desmin and ERTR-7 expression and ing in the PALS at 8.95 Ϯ 0.9 ␮m/minϪ1, a speed comparable to analyzed using confocal microscopy (Fig. 3A). In the PALS, GFPϩ that observed in intact LNs (data not shown, movie S2) (22–24). cells form a three-dimensional network that surrounds the Analysis of 4-D (x, y, z, and time) datasets suggested that migrat- ERTR-7ϩ conduit system and overlaps with desmin staining, in- ing SNARF-1 labeled T cells actively crawled on GFPϩ FRCs, dicating that this network is indeed formed by FRCs. following and morphologically adapting to the paths established 3950 REGULATION OF T CELL MIGRATION IN THE SPLENIC WP

FIGURE 3. T cells crawl on the splenic FRC network. A, Twenty mi- crometer thick spleen cryostat sections from GFP (green) chimeric mice were examined using confocal microscopy after staining for desmin (red) and ERTR-7 (blue). C.A., Central arteriole. M.Z, Marginal zone. B, Five ϫ 106 SNARF-1 labeled T cells were injected i.v into chimeric mice. One day later, spleens were sectioned using a vi- bratome, perfused with warm, oxygen- ated medium, and imaged using 2P Downloaded from microscopy. Data show intravital snap- shots of a single T cell (red) moving over time on FRC fibers (green) in a 12 ␮m thick volume. C, Quantification of T cells showing turns in 4D datasets with respect to their location on or off http://www.jimmunol.org/ GFP-marked stromal fibers in chimeric animals. Data are representative of at least three experiments. by guest on September 24, 2021

by the cell bodies and extended processes of these nonhematopoi- MZ macrophages in this process has been proposed because of etic cells (Fig. 3B, movie S3). their unique location in the MZ as well as their ability to bind to To assess quantitatively whether T cells actively followed the lymphocytes deposited on spleen cryostat sections in a sialoadhe- paths laid out by the FRC network, we used the same approach that sin-dependent manner (25). However, this hypothesized function was applied in our prior study of migration of T cells in LNs (15). is unlikely in light of data showing that lymphocyte homing to the We assumed that if the fibers provided guidance for cell move- PALS is unaltered in mice in which MZ macrophages have been ment, then any directional turns made by a T cell should always be depleted using chlodronate liposomes (26). Because T cells mi- associated with a corresponding turn or branch of a supporting grate on FRCs in the PALS and because FRCs connect the MZ to FRC fiber. Conversely, a lack of correspondence between T cell the PALS via the MZBCs, we considered the possibility that FRCs directionality and fiber pathways would indicate that spontaneous also support T cell migration from MZs to the PALS, as previously turns or physical impediments posed by the many other cells in the suggested based on static imaging analyses using human material densely packed PALS environment accounted for T cell direc- (14). tional changes. Analysis of T cell turns using this previously de- We first attempted to use our vibratome cut method to address scribed method (15) revealed an 89% correlation (155 of 174 cells) this issue but found that using this procedure, the structure of the between changes in T cell direction and the presence of T cell- MZ loosely attached to the WP becomes compromised, inducing ϩ associated GFP FRC fibers running at the corresponding angle lymphocytes to leak out of the MZ over time, a phenomenon we (Fig. 3C). Given that this is likely an underestimate of the corre- never observed in the more rigid and compact WP (data not spondence between T cell movement and FRC organization due to shown). As a consequence, we used an alternative approach. Naive the inability to image the thinnest or dimmest FRC fibers, we con- polyclonal T cells were labeled with 5-chloromethyl fluorescein clude that as observed in LNs, T cells migrate along FRCs in the diacetate-2 ␮M or CFSE and injected i.v. into recipients that were splenic WP. euthanized 10, 20, 30, and 180 min later. Spleens were harvested and used to prepare fixed sections that were stained with B220 to FRCs support T cell entry in the PALS locate B cell follicles, as well as with ERTR-7 to highlight the Blood circulating lymphocytes enter the spleen in the MZ (8, 20, PALS and the MZBC that connect the MZ to the PALS (Fig. 4A). 21). It is thought that lymphocytes and DCs can enter the WP from We reasoned that if T cells are not constrained to enter the WP via the MZ sinus by passing through a layer of sinus-lining cells that FRCs, we should observe T cells initially accessing the PALS not form a barrier between the MZ and the WP (4) (Fig. 1A). A role for only adjacent to the MZBCs but everywhere through the WP. At The Journal of Immunology 3951

FIGURE 4. T cells access the PALS using MZBCs. A, Ten ϫ 106 CFSE la- beled polyclonal T cells (green) were transferred in recipient mice. At 10, 20, 30, and 180 min later, spleen cryostat sections were stained for B220 (blue) and ERTR-7 (red) and imaged using confocal microscopy to identify where T cells enter the WP. Arrow- heads point to the location where MZBC connect the MZ. B, Quantifi- cation of T cells entering the WP in Downloaded from ERTR-7Ϫ and ERTR-7ϩ areas of the WP over time. Data are representative of three different experiments. http://www.jimmunol.org/

later times, T cells would eventually gather in the PALS. Alter- T cells, B cells were observed entering the WP via the MZBC over natively, if T cells enter the WP using the FRCs located in the time. Interestingly, B cells seemed to preferentially use the exter- by guest on September 24, 2021 MZBCs, they should be observed entering the WP only where nal part of the corridor corresponding to the T/B interface (the these structures connect to the PALS. equivalent of the cortical ridge present in the LN (27)) to gain Using this sequential static imaging approach, T cells were access to the follicles. Upon closer examination of B cells within found throughout the MZ surrounding the WP 10 min after the this region using a previously described fixative-perfusion tech- transfer. Starting 20–30 min after the transfer, T cells started to nique (15), we found that the vast majority (48 of 55) of elongated enter the WP. Strikingly, T cells only entered the WP at the and polarized B cells were touching an ERTR-7ϩ desminϩ FRC MZBCs (Fig. 4A). Three hours after the transfer, T cells were fiber (Fig. 5C). These data suggest that B cells migrate on the FRC localized in the PALS. To quantify this phenomenon, we counted network as they transition between PALS and B follicle regions of the numbers of T cells present at 10, 20, and 30 min (i.e., when the WP. It is important to point out that a small number of B cells they were in the process of entering the WP) in both ERTR-7Ϫ and were observed in the follicles at the earliest time point examined. ERTR-7ϩ regions of the WP and normalized these numbers to the Elucidating whether these cells directly crossed the MZ sinus or surface areas of the respective regions. The analysis revealed that alternatively, rapidly homed to the follicles after entering the T cell entry into the WP almost exclusively occurred in MZBC PALS, is difficult but this observation may indicate that some B regions where FRCs connect the PALS to the MZ (Fig. 4B). cells may follow a path other than through the MZBCs to enter the WP. B cells move across MZBCs and on FRCs outside the follicles B cells reside in WP follicles that lack an FRC network, leaving Discussion open the question of how this lymphocyte subset accesses this In this study, we present evidence that FRCs support T cell access region. Forster et al. (6) observed that before accessing the adja- to the splenic WP by creating physical roads for T cell migration. cent follicles, B cells are transiently retained in the PALS in a These FRC pathways connect the MZ to the PALS via breaks in CCR7-dependent manner, indicating that B cells do not directly the shell of MZ macrophages and the MZ sinus, regions called cross the MZ/WP border to access the follicles but rather follow a MZBCs. After accessing the PALS along these guides, the T cells path similar to T cells. To assess whether B cells enter the WP migrate on the FRC fibers within the T zone of the spleen, as using the MZBC and to quantify this phenomenon, we adoptively previously observed in the paracortical region of LNs (15). Inter- transferred CFSE-labeled B cells into WT recipients and counted estingly, like their LN counterparts, splenic FRCs secrete and are the numbers of labeled B cells present at 30 min, 1, 2, 3, and 8 h tightly associated with CCL21 (Ref. 28 and Fig. 6A), a in three different regions of the WP: the B cell follicles (B220ϩ that is an important regulator of T cell motility and positioning in ERTR-7Ϫ), the T cell zone (B220Ϫ ERTR-7ϩ), and the T/B in- the PALS and that has been shown to provide an important che- terface (B220ϩ ERTR-7ϩ). These numbers were then normalized mokinetic stimulus to T cells in the LN paracortex (29–31). In to the surface areas of the respective regions (Fig. 5, A and B). Like Plt/Plt mice that lack expression of CCL19/CCL21 (ELC/SLC), 3952 REGULATION OF T CELL MIGRATION IN THE SPLENIC WP

FIGURE 5. B cells access the PALS using MZBCs. Twenty ϫ 106 CFSE la- beled polyclonal B cells (red) were transferred i.v. into recipient mice. At 0.5, 1, 2, 3, and 8 h later, spleens were harvested and sectioned using a cryo- stat. A, Sections were stained for ERTR-7 (green) and B220 (blue) to identify by confocal microscopy the MZBCs, the B cell zone (B220ϩ ERTR-7Ϫ) and the T cell zone sepa- rated into the deep T zone (B220Ϫ ϩ

ERTR-7 ), and the superficial T zone Downloaded from (or T/B interface) where T and B cells intermingle (B220ϩ ERTR-7ϩ). At t ϭ 1h,theinset provides a magnified view of an MZBC where B cells gain access to the WP. B, Quantification of B cells entering the WP in the B zone, deep T zone, and superficial T zone of the WP http://www.jimmunol.org/ over time. Data are representative of two different experiments. C, Three hours after B cell adoptive transfer, mice were perfused with a fixative so- lution and spleens were harvested, sectioned using a cryostat, and stained for desmin (red) and ERTR-7 (blue). Three examples of CFSE-la- beled B cell (green) interactions with by guest on September 24, 2021 FRC fibers are shown. Left panels, FRC fiber staining alone. Right pan- els, Overlay with B cells. Scale bar, 10 ␮m. Data are representative of two different experiments.

naive T cells do not enter the PALS efficiently (Ref. 5 and Fig. 6B). these corridors govern cell exchanges between the RP/MZ and the Similarly, in mice deficient for CCR7, the cognate receptor for WP in both directions. these chemokines, naive T cells cannot enter the PALS, indicating Although our present data together with past results point to a the crucial role of this molecular interaction in regulating naive T clear role for CCL21 and FRC fibers in guiding T cell migration cell access to this region (6). Interestingly, in the MZ, a subset of into and within the PALS, the mechanisms underlying T cell DCs called MZ DCs is localized at the border of MZBC (10). migration in the MZ itself remain unclear. In the MZ, MMM Upon LPS activation, these DCs migrate to the PALS and this location is regulated by CCL19/CCL21 because Plt/Plt mice migration is correlated with increased expression of CCR7 by the show a reduction in MMMs colonizing the MZ (36). Because, DCs (10, 32, 33). It is therefore likely that both MZ DCs and naive as this result indicates, these chemokines are present and func- T cells access the PALS by using CCL19/CCL21 present on FRCs tional in the MZ, T cell migration in this region of the spleen to guide their migration. In addition, large heavily carbon-laden might also be regulated by CCL19/CCL21 until the presence of macrophages located only in the RP at 30 min after carbon injec- FRCs at the MZBCs provides a presumably more attractive path tion start to appear along MZBCs 1–6 h later (34). Taken together, for the T cells to follow. Alternatively, because the these observations suggest that FRC-rich MZBCs are the entry meshwork discharges blood in the MZ, this specific area is con- door to the PALS for several different cell types. MZBCs have tinuously perfused (20, 21). In the MZ, a dense network of previously been suggested to be the place where lymphocytes exit reticular fibers creates a maze that may slow down lymphocytes (but do not enter) the WP (18, 35). Therefore, it is possible that (Fig. 2). As a consequence, it is possible that, while flowing in The Journal of Immunology 3953

FIGURE 6. Splenic FRCs secrete and are covered with CCL21. A, Twenty mi- crometer thick splenic cryostat sections from chimeric (green) mice were exam- ined using confocal microscopy after staining for CD3 (blue) and CCL21 (red). 6 C.A., Central arteriole. B, Ten ϫ 10 Downloaded from CFSE labeled T cells were injected i.v into wild type and plt/plt mice. Three hours later, splenic cryostat sections were exam- ined using confocal microscopy after staining with ERTR-7 (red) and B220 (blue). T cells enter the PALS in wild type but not plt/plt mice, as previously http://www.jimmunol.org/ described (5). Scale bar, 60 ␮m. Data are representative of three different experiments. by guest on September 24, 2021

the MZ, lymphocytes that can “catch” an FRC present at the substratum that provides a sufficient grip. This adhesiveness MZ/PALS border in the MZBC will begin their journey to the may be regulated by integrins or other adhesions molecules that underlying PALS, while those that do not successfully interact remain to be identified but it is also possible that nonspecific with these FRCs will continue their movement and either enter physical interactions such as those mediated by charges or Van a downstream PALS or remain in the RP. der Waals forces may be involved. In addition, our previous A central question remains: which molecules beside CCR7 observation that lymphocytes maintain numerous microvilli ligands, if any, govern T cell locomotion on FRCs in the spleen while moving in the LN environment raises the possibility that and LNs? So far, CCR7 ligands are the only molecules that have T cells are using these protrusions to exert the mandatory trac- been shown experimentally to have a direct role in controlling tion forces required for their migration (15). the naive T cell movement in LNs in vivo in the steady state Overall, the data we present in this study reinforce the conclu- (29–31), but eliminating expression of this chemokine or its sion we reached from our prior study of lymphocyte migration in receptor only fractionally reduces the velocity of the T cells; it the LN (15). Although at the gross level in the absence of inflam- does not prevent their movement. Recently, Woolf et al. (37) mation, T cell migration appears random in secondary lymphoid demonstrated that, in a shear-free environment, immobilized tissues (40, 41), when examined closely, it is clear that the move- but not soluble CCR7 ligands are chemokinetic agents for T ment of these lymphocytes occurs along preformed paths. In the cells and that neither LFA-1 nor VLA-4 are required for T cell LN, these paths are studded with DCs, thus facilitating the inter- motility in LNs. FRCs wrap around collagen fibers, preventing action of these two key cell types. As noted in this study, activated lymphocytes from directly contacting these ECM molecules DCs traffic use similar chemokine signals as the T cells to move (17, 38). However, in absence of chemokines, lymphocytes into the PALS and it is likely they occupy places on the same FRC move in vitro on collagen lattices but fail to do so on 2-D network trafficked by these lymphocytes. Thus, it seems that collagen coated surfaces (37, 39). This observation raises the throughout lymphoid tissues, initiation of adaptive immune re- possibility that T cells can crawl on any 3-D (but not 2-D) sponses is left less to chance than might be imagined. 3954 REGULATION OF T CELL MIGRATION IN THE SPLENIC WP

Acknowledgments 20. Grayson, M. H., D. D. Chaplin, I. E. Karl, and R. S. Hotchkiss. 2001. Confocal fluorescent intravital microscopy of the murine spleen. J. Immunol. Methods 256: We thank Hai Qi and Jackson G. Egen for suggestions on the manuscript. 55–63. 21. Grayson, M. H., R. S. Hotchkiss, I. E. Karl, M. J. Holtzman, and D. D. Chaplin. 2003. Intravital microscopy comparing T lymphocyte trafficking to the spleen and Disclosures the mesenteric lymph node. Am. J. Physiol. 284: H2213. ϩ The authors have no financial conflict of interest. 22. Bousso, P., and E. Robey. 2003. Dynamics of CD8 T cell priming by dendritic cells in intact lymph nodes. Nat. Immunol. 4: 579–585. 23. Mempel, T. R., S. E. Henrickson, and U. H. Von Andrian. 2004. T-cell priming by dendritic cells in lymph nodes occurs in three distinct phases. Nature 427: References 154–159. 1. Pabst, R., and R. M. Binns. 1989. Heterogeneity of lymphocyte homing physi- 24. Miller, M. J., S. H. Wei, I. Parker, and M. D. Cahalan. 2002. Two-photon im- ology: several mechanisms operate in the control of migration to lymphoid and aging of lymphocyte motility and response in intact lymph node. Science non-lymphoid organs in vivo. Immunol. Rev. 108: 83–109. 296: 1869–1873. 2. Miyasaka, M., and T. Tanaka. 2004. Lymphocyte trafficking across high endo- 25. van den Berg, T. K., J. J. Breve, J. G. Damoiseaux, E. A. Dopp, S. Kelm, thelial venules: dogmas and enigmas. Nat. Rev. Immunol. 4: 360–370. P. R. Crocker, C. D. Dijkstra, and G. Kraal. 1992. Sialoadhesin on macrophages: 3. Warnock, R. A., S. Askari, E. C. Butcher, and U. H. von Andrian. 1998. Mo- its identification as a lymphocyte adhesion molecule. J. Exp. Med. 176: 647–655. lecular mechanisms of lymphocyte homing to peripheral lymph nodes. J. Exp. 26. Claassen, E., and N. van Rooijen. 1985. Evidence that macrophages in the mar- Med. 187: 205–216. ginal zone have no role in the migration of lymphocytes into the periarteriolar 4. Mebius, R. E., and G. Kraal. 2005. Structure and function of the spleen. Nat. Rev. lymphocyte sheaths (PALS). Immunology 56: 689–694. Immunol. 5: 606–616. 27. Katakai, T., T. Hara, J. H. Lee, H. Gonda, M. Sugai, and A. Shimizu. 2004. A 5. Nakano, H., S. Mori, H. Yonekawa, H. Nariuchi, A. Matsuzawa, and T. Kakiuchi. novel reticular stromal structure in lymph node cortex: an immuno-platform for 1998. A novel mutant gene involved in T-lymphocyte-specific homing into pe- interactions among dendritic cells, T cells and B cells. Int. Immunol. 16: ripheral lymphoid organs on mouse chromosome 4. Blood 91: 2886–2895. 1133–1142. 6. Forster, R., A. Schubel, D. Breitfeld, E. Kremmer, I. Renner-Muller, E. Wolf, and 28. Luther, S. A., H. L. Tang, P. L. Hyman, A. G. Farr, and J. G. Cyster. 2000. M. Lipp. 1999. CCR7 coordinates the primary immune response by establishing Coexpression of the chemokines ELC and SLC by T zone stromal cells and Downloaded from functional microenvironments in secondary lymphoid organs. Cell 99: 23–33. deletion of the ELC gene in the plt/plt mouse. Proc. Natl. Acad. Sci. USA 97: 7. Ansel, K. M., V. N. Ngo, P. L. Hyman, S. A. Luther, R. Forster, J. D. Sedgwick, 12694–12699. J. L. Browning, M. Lipp, and J. G. Cyster. 2000. A chemokine-driven positive 29. Asperti-Boursin, F., E. Real, G. Bismuth, A. Trautmann, and E. Donnadieu. 2007. feedback loop organizes lymphoid follicles. Nature 406: 309–314. CCR7 ligands control basal T cell motility within lymph node slices in a phos- 8. Lyons, A. B., and C. R. Parish. 1995. Are murine marginal-zone macrophages the phoinositide 3-kinase-independent manner. J. Exp. Med. 204: 1167–1179. splenic white pulp analog of high endothelial venules? Eur. J. Immunol. 25: 30. Okada, T., and J. G. Cyster. 2007. CC chemokine receptor 7 contributes to Gi- 3165–3172. dependent T cell motility in the lymph node. J. Immunol. 178: 2973–2978.

9. Kraal, G., and M. Janse. 1986. Marginal metallophilic cells of the mouse spleen 31. Worbs, T., T. R. Mempel, J. Bolter, U. H. von Andrian, and R. Forster. 2007. http://www.jimmunol.org/ identified by a monoclonal antibody. Immunology 58: 665–669. CCR7 ligands stimulate the intranodal motility of T lymphocytes in vivo. J. Exp. 10. Leenen, P. J., K. Radosevic, J. S. Voerman, B. Salomon, N. van Rooijen, Med. 204: 489–495. D. Klatzmann, and W. van Ewijk. 1998. Heterogeneity of mouse spleen dendritic 32. Sallusto, F., P. Schaerli, P. Loetscher, C. Schaniel, D. Lenig, C. R. Mackay, cells: in vivo phagocytic activity, expression of macrophage markers, and sub- S. Qin, and A. Lanzavecchia. 1998. Rapid and coordinated switch in chemokine population turnover. J. Immunol. 160: 2166–2173. receptor expression during maturation. Eur. J. Immunol. 28: 11. Dijkstra, C. D., E. Van Vliet, E. A. Dopp, A. A. van der Lelij, and G. Kraal. 1985. 2760–2769. Marginal zone macrophages identified by a monoclonal antibody: characteriza- 33. De Smedt, T., B. Pajak, E. Muraille, L. Lespagnard, E. Heinen, P. De Baetselier, tion of immuno- and enzyme-histochemical properties and functional capacities. J. Urbain, O. Leo, and M. Moser. 1996. Regulation of dendritic cell numbers and J. Exp. Med. Immunology 55: 23–30. maturation by lipopolysaccharide in vivo. 184: 1413–1424. 34. Kotani, M., K. Matsuno, and T. Ezaki. 1986. Marginal zone bridging channels as 12. Goldschneider, I., and D. D. McGregor. 1968. Migration of lymphocytes and a pathway for migrating macrophages from the red towards the white pulp in the thymocytes in the rat. I. The route of migration from blood to spleen and lymph rat spleen. Acta Anat. 126: 193–198. by guest on September 24, 2021 nodes. J. Exp. Med. 127: 155–168. 35. Khanna, K. M., J. T. McNamara, and L. Lefrancois. 2007. In situ imaging of the 13. Weiss, L. 1965. The structure of the normal spleen. Semin. Hematol. 39: endogenous CD8 T cell response to infection. Science 318: 116–120. 205–228. 36. Ato, M., H. Nakano, T. Kakiuchi, and P. M. Kaye. 2004. Localization of marginal 14. Steiniger, B., P. Barth, and A. Hellinger. 2001. The perifollicular and marginal zone macrophages is regulated by C-C chemokine ligands 21/19. J. Immunol. zones of the human splenic white pulp: do fibroblasts guide lymphocyte immi- 173: 4815–4820. gration? Am. J. Pathol. 159: 501–512. 37. Woolf, E., I. Grigorova, A. Sagiv, V. Grabovsky, S. W. Feigelson, Z. Shulman, 15. Bajenoff, M., J. G. Egen, L. Y. Koo, J. P. Laugier, F. Brau, N. Glaichenhaus, and T. Hartmann, M. Sixt, J. G. Cyster, and R. Alon. 2007. Lymph node chemokines R. N. Germain. 2006. Stromal cell networks regulate lymphocyte entry, migra- promote sustained T lymphocyte motility without triggering stable integrin ad- tion, and territoriality in lymph nodes. Immunity 25: 989–1001. hesiveness in the absence of shear forces. Nat. Immunol. 8: 1076–1085. 16. Nolte, M. A., J. A. Belien, I. Schadee-Eestermans, W. Jansen, W. W. Unger, 38. Hayakawa, M., M. Kobayashi, and T. Hoshino. 1988. Direct contact between N. van Rooijen, G. Kraal, and R. E. Mebius. 2003. A conduit system distributes reticular fibers and migratory cells in the paracortex of mouse lymph nodes: a chemokines and small blood-borne molecules through the splenic white pulp. morphological and quantitative study. Arch. Histol. Cytol. 51: 233–240. J. Exp. Med. 198: 505–512. 39. Friedl, P., F. Entschladen, C. Conrad, B. Niggemann, and K. S. Zanker. 1998. 17. Sixt, M., N. Kanazawa, M. Selg, T. Samson, G. Roos, D. P. Reinhardt, R. Pabst, CD4ϩ T lymphocytes migrating in three-dimensional collagen lattices lack focal M. B. Lutz, and L. Sorokin. 2005. The conduit system transports soluble adhesions and utilize ␤1 integrin-independent strategies for polarization, inter- from the afferent lymph to resident dendritic cells in the T cell area of the lymph action with collagen fibers and locomotion. Eur. J. Immunol. 28: 2331–2343. node. Immunity 22: 19–29. 40. Miller, M. J., A. S. Hejazi, S. H. Wei, M. D. Cahalan, and I. Parker. 2004. T cell 18. Mitchell, J. 1973. Lymphocyte circulation in the spleen. Marginal zone bridging repertoire scanning is promoted by dynamic dendritic cell behavior and random channels and their possible role in cell traffic. Immunology 24: 93–107. T cell motility in the lymph node. Proc. Natl. Acad. Sci. USA 101: 998–1003. 19. Bajenoff, M., S. Granjeaud, and S. Guerder. 2003. The strategy of T cell antigen- 41. Wei, S. H., I. Parker, M. J. Miller, and M. D. Cahalan. 2003. A stochastic view presenting cell encounter in antigen-draining lymph nodes revealed by imaging of of lymphocyte motility and trafficking within the lymph node. Immunol. Rev. 195: initial T cell activation. J. Exp. Med. 198: 715–724. 136–159.