Uncorrected Version. Published on June 3, 2008 as DOI:10.1189/jlb.0108048

Laminin isoforms of lymph nodes and predominant role of ␣5laminin(s) in adhesion and migration of blood lymphocytes Gezahegn Gorfu,* Ismo Virtanen,† Mika Hukkanen,† Veli-Pekka Lehto,‡ Patricia Rousselle,§ Ellinor Kenne,㛳 Lennart Lindbom,㛳 Randall Kramer,¶ Karl Tryggvason,# and Manuel Patarroyo*,1 Departments of *Odontology, 㛳Physiology and Pharmacology, and #Medical Biochemistry and Biophysics, Karolinska Institutet, Stockholm, Sweden; †Institute of Biomedicine/Anatomy and ‡Haartman Institute, University of Helsinki, Finland; §Institut de Biologie et Chimie des Prote´ines, Universite´ Lyon I, Lyon, France; and ¶Department of Cell and Tissue Biology, University of California, San Francisco, California, USA

Abstract: During extravasation and within lymph nodes (LNs), blood lymphocytes interact with laminins (Lms), major components of vascular basement membranes (BMs) and of reticular fibers (RFs), a fibrillar extracellular matrix. However, the identity and role of these Lm isoform(s) are poorly known. By using confocal microscopy examination of human LNs, we show that BMs of high endothelial venules express Lm␣3, -␣4, -␣5, -␤1, -␤2, and -␥1 chains and that the same chains, in addition to ␣2, are found in RFs. In functional studies with Lm isoforms covering all Lm␣ chains, Lm␣5 (Lm-511) was the most adhesion- and migration-promoting isoform for human blood lymphocytes, followed by Lm␣3 (Lm-332) and Lm␣4 (Lm-411), and the lymphocytes used the ␣6␤1 integrin (INT) as the primary receptor for the Lm␣5. Moreover, Lm-511 strongly costimulated T cell proliferation, and blood lymphocytes were able to secrete Lm␣4 and -␣5 following stimulation. The LN cell number in Lm␣4-deficient mice compared with wild-type did not differ significantly. This study demonstrates a predominant role for Lm␣5 in blood lymphocyte biology and identifies LN Lms and their INT receptors in blood lymphocytes. J. Leukoc. Biol. 84: 000 – 000; 2008. Key words: leukocyte 䡠 extracellular matrix 䡠 basement membrane 䡠 chemotaxis

INTRODUCTION Interaction of lymphocytes with extracellular matrix (ECM) components, such as collagens, fibronectin (FN), and laminins (Lms), regulates several lymphocyte activities in vitro, including adhesion, migration, differentiation, and proliferation/survival [1]. In lymph nodes (LNs) and other secondary lymphoid tissues, these ECM components are nonrandomly distributed and constitute barriers and scaffolds, such as vascular basement membranes (BMs) and reticular fibers (RFs) [2]. In addition to these structural functions, these large molecules may facilitate in situ-specific localization and/or migration of lym0741-5400/08/0084-0001 © Society for Leukocyte Biology

phocytes and other immune cells and hence, contribute to the generation of immune responses [3, 4]. During trafficking, blood lymphocytes adhere to and migrate across high endothelial venules (HEVs) to enter LNs, tonsils, Peyer’s patches, and other secondary lymphoid tissues [5, 6]. Whereas the initial steps of this extravasation process (rolling and firm adhesion) are well-characterized, the following transendothelial migration and the interaction with the HEV BM is poorly understood. Moreover, once within the lymphoid tissue, lymphocytes may also interact with the RFs [4, 7]. Lms are a growing family of large, heterotrimeric ␣␤␥ glycoproteins, mainly found in BMs [8]. Five ␣, three ␤, and three ␥ Lm chains constitute, by combination, over 15 Lm isoforms, which are expressed in a cell- and tissue-specific manner [9, 10]. The ␣ chains of these molecules are recognized by at least seven different integrins (INTs) and strongly promote cell adhesion, migration, and differentiation [9]. In secondary lymphoid tissues, Lms are present in the BMs of ordinary blood vessels and HEVs, as well as in RFs [2, 11–13]. The latter fibers, which also contain other BM components such as nidogen, perlecan, and type IV collagen, constitute the reticular network, a conduit system for transport of small, soluble molecules [11–14]. Interactions of lymphocytes with Lm isoforms are poorly understood. As identification of the prototype Lm-111 [Lm-1, Lm ␣1␤1␥1, Engelbreth-Holm-Swarm (EHS) Lm] in mice over 25 years ago, a large number of additional Lm isoforms with different tissue expression and biological properties have been found [8 –10]. Early studies describing Lm expression in lymphoid tissues and subsets of lymphocytes used polyclonal antibodies to Lm-111, which cross-react with most Lm isoforms or anti-Lm mAb of unknown or misassigned chain specificity [15]. Moreover, early functional studies of lymphocytes and Lms examined the biological effect of Lm-111, which is neither synthesized by lymphoid cells nor expressed in lymphoid tissues [9]. Alternatively, the studies used poorly characterized, commercial Lm preparations isolated from human placenta,

1 Correspondence: Department of Odontology, Karolinska Institutet at Huddinge, SE-141 04 Stockholm, Sweden. E-mail: [email protected] Received January 18, 2008; revised May 5, 2008; accepted May 5, 2008. doi: 10.1189/jlb.0108048

Journal of Leukocyte Biology Volume 84, September 2008

Copyright 2008 by The Society for Leukocyte Biology.

1

which frequently consisted of large Lm fragments, a mixture of Lm isoforms, and/or contaminating proteins [16]. More recently, mAb of known Lm chain specificity and fully characterized, natural and recombinant (r)Lms have become available, as well as viable Lm␣2 and -␣4 chain knockout (KO) mice [12, 15, 17–24]. Studies with these tools have demonstrated a characteristic distribution of Lm isoforms in thymus and participation of Lm␣2, -␣3, and -␣5 in adhesion, migration, proliferation, survival, and/or development of thymocytes and T cell malignancies [20, 25–31]. Moreover, adhesion and migration of neoplastic lymphocytes have been found to be promoted by Lm␣4 and -␣5 [32, 33]. When compared with Lm␣1, which in adult life is highly restricted to a few epithelia, the latter Lms have a much wider tissue distribution and constitute the major vascular endothelial Lm isoforms [9, 34]. Although a few studies have described expression of Lm chains in LNs [12, 13, 35, 36], a systematic analysis of HEV Lms has not been reported nor has the effect of Lm isoforms covering all Lm ␣ chains, including Lm␣5, in the biology of normal blood lymphocytes. We recently described strong migration-promoting activities of Lm␣4 (Lm-411, ␣4␤1␥1, Lm-8) on blood lymphocytes and neutrophils and impaired extravasation of neutrophils during peritonitis in Lm␣4 KO mice [24, 37]. As an attempt to determine the role of Lm isoforms in the extravasation and migration of blood lymphocytes, we have analyzed in the present study the expression of Lms in LNs, particularly in HEVs and RFs, by using a panel of mAb to all Lm chains, except Lm␥3 and confocal microscopy. With molecularly characterized, natural and rLms covering the five ␣ chains, we have determined the adhesion-, migration-, and costimulation-promoting activities of Lm isoforms on blood lymphocytes and the participation of Lm-binding INTs. Cell number of LNs from Lm␣4 KO mice and synthesis and secretion of Lm␣5 by blood lymphocytes were also investigated.

MATERIALS AND METHODS Cells and purified Lms PBMCs were obtained from healthy donors after Ficoll-Hypaque gradient centrifugation (Amersham Pharmacia AB, Uppsala, Sweden) with approval of the local ethical committee and informed consent of the participating subjects. Lymphocytes were subsequently isolated by removing tissue-culture, flaskadherent cells or by gradients of 60%, 47.5%, and 34% Percoll (Amersham Pharmacia AB) in complete medium (10% FCS in RPMI) with modifications of a method described previously [38]. CD4⫹ T cells were obtained from PBMCs by positive selection using magnetic beads coated with anti-CD4 antibody according to instructions from the manufacturer (Dynal AS, Oslo, Norway). Mouse Lm-111 (mLm-111; ␣1␤1␥1, Lm-1), isolated from the EHS tumor, and human merosin (Lm-211, ␣2␤1␥1, Lm-2) from placenta were obtained from Chemicon International (Temecula, CA, USA). Human Lm-332 (hLm332; ␣3␤3␥2, Lm-5) was immunopurified from the conditioned medium of SCC25 cells [21]. rhLm-211 (␣2␤1␥1, Lm-2), rhLm-411 (␣4␤1␥1, Lm-8), and rhLm-511 (␣5␤1␥1, Lm-10) were produced in a mammalian expression system and purified and characterized as described previously [17, 18, 22]. Another Lm␣5 preparation, Lm-511, was purchased from Sigma-Aldrich (St. Louis, MO, USA) as hLm. This preparation, isolated from placenta by mild pepsin digestion and purified by immunoaffinity chromatography, was molecularly characterized and used only when reactive with a mAb 4C7 to Lm␣5 chain globular domain [16]. Human serum albumin (HSA) and plasma FN (pFN) were also obtained from Sigma-Aldrich.

2

Journal of Leukocyte Biology Volume 84, September 2008

Immunofluorescence and confocal laserscanning microscopy (CLSM) For immunofluorescence studies, four normal LN specimens were retrieved from the files of the Institute of Biomedicine/Anatomy and cut at 5 ␮m. The following mAb, all mIgG, against Lm chains were used in frozen sections: 161EB7 (Lm␣1), 5H2 (Lm␣2), BM-2 (Lm␣3), 168FC10 (Lm␣4), 4C7 (Lm␣5), 114DG10 (Lm␤1), S5F11 (Lm␤2), 6F12 (Lm␤3), 113BC7 (Lm␥1), and D4B5 (Lm␥2 chain). These antibodies were detected with Alexa Fluor 594 goat anti-mIgG antibodies (Molecular Probes, Eugene, OR, USA). mAb M3F7 (collagen type IV ␣1/␣2) and A9 (nidogen-1) were also tested. The source of the mAb was described recently [39]. For double-immunolabeling, mAb 3G4 against type III collagen (mIgM, Life Technologies, Gaithersburg, MD, USA) or mAb MECA-79 against sulfation-dependent, HEV-associated carbohydrate (rat IgM, BD Biosciences, Stockholm, Sweden) was used, together with Alexa Fluor 488 goat anti-mIgM (␮ chain) antibody or Alexa Fluor 488 goat anti-rat IgM (␮ chain), respectively. Tetramethylrhodamine-isothiocyanate-coupled Ulex Europaeus I-lectin (Vector Laboratories, Burlingame, CA, USA) was also used in double-immunostainings to identify blood vessel endothelia. Doubleimmunolabeling and CLSM were carried out as described recently [39].

Immunofluorescence flow cytometry Lymphocytes, gated from isolated PBMCs by forward- and side-scatter, were analyzed for cell surface (nonpermeabilized) and intracellular (permeabilized) expression of Lm chains as well as for cell surface expression of INTs and differentiation markers by immunofluorescence flow cytometry in a FACScan flow cytometer (Beckton Dickinson, Mountain View, CA, USA), as described previously [40]. For INT expression, mAb 6S6 (INT␤1; Chemicon), FB12 (INT␣1; Chemicon), P1E6 (INT␣2; Chemicon), P1B5 (INT␣3; Chemicon), BQ16 (INT␣6; Ancell Corp., Bayport, MN, USA), 9.1 (INT␣7) [41], 69.6.5 (INT␣V; Beckman Coulter, Bromma, Sweden), IB4 (INT␤2) [24] (generous gift from Prof. Samuel Wright, Merck Research Labs, Rahway, NJ, USA), H12 (INT␣L; kindly provided by Hans Wigzell, Karolinska Institutet, Stockholm, Sweden), 2LPM (INT␣M; Dakopatts, Copenhagen, Denmark), 3.9 (INT␣X; Harlan SeraLab, Loughborough, UK), SZ.21 (INT␤3; Beckman Coulter), SZ.22 (INTIIb; Beckman Coulter), LM609 (INT␣V␤3; Chemicon), ASC-9 (INT␤4; Chemicon), and P1F6 (INT␣V␤5; Chemicon) were used. For Lm chain expression, mAb 15H5 (Lm␣5) [42] (kindly provided by Prof. Kiyotoshi Sekiguchi, Osaka University, Japan), LN-26 (Lm␤1; Takara Shuzo, Kyoto, Japan) [15], C4 (Lm␤2; Neomarkers, Fremont CA, USA), and LN-82 (Lm␥1; Takara Shuzo) [15] were used. mIgG and mAb 9.6 to CD2, a lymphocyte marker, were also included as negative and positive controls, respectively. Over 90% of the gated cells (lymphocytes) expressed CD2. INT, Lm, and CD2 antibodies, all mIgG, were detected with FITC-conjugated F(ab⬘)2 fragments of rabbit antimIgs (Dakopatts), whereas lymphocyte populations were identified by using PerCP-labeled anti-CD4 (Th cells), APC-labeled anti-CD8 (cytotoxic T cells), and PE-labeled anti-CD19 (B cells) antibodies (BD Biosciences, San Diego, CA, USA). INT expression in different lymphocyte populations was investigated by multicolor immunofluorescence.

Cell adhesion, migration, and proliferation assays Lymphocyte adhesion to Lm isoforms was performed as described previously [37] with some modifications. Briefly, lymphocytes were isolated from PBMCs by Percoll density gradients and thereafter, resuspended in RPMI 1640 (2⫻106 cells/ml), and 96-well flat-bottom polystyrene plates (BD Biosciences) were coated with 50 ␮l/well PBS or 20 ␮g/ml HSA, Lm-111, Lm-411, or Lm-511 in PBS for 3 h at 37°C. After rinsing with PBS, free sites were blocked with 2% polyvinylpyrrolidone (PVP; molecular weight 360 kDa; Sigma-Aldrich) for 1 h at room temperature, and after further washing, 2 ⫻ 105 cells/well (100 ␮l) were incubated in the presence of 200 nM tetradecanoyl phorbol acetate (TPA; also known as PMA, Sigma-Aldrich) at 37°C for 1 h. Following five washes, adherent cells were fixed and stained overnight. The plate was then washed, and adherent cells were quantified in a microplate reader. Transmigration of lymphocytes through protein-coated filters was measured microscopically and by flow cytometry as described before [37]. Briefly, human PBMCs (5⫻105 containing ⬃70% lymphocytes) were added to the top chamber of 3 ␮m pore polycarbonate Transwell culture inserts (Costar, Cambridge, MA, USA) and incubated at 37°C for 3 h in the presence (chemokinestimulated migration) of 500 ng/ml stromal cell-derived factor 1␣ (SDF-1␣;

http://www.jleukbio.org

R&D Systems, Abingdon, UK). Thereafter, the number of cells in the lower chamber was measured microscopically, and the percentage of lymphocytes among the transmigrated was established by flow cytometry, according to forward- and side-scatter. Lymphocyte populations CD4⫹ (Th) cells, CD8⫹ (cytotoxic T) cells, and CD19⫹ (B) cells, before and after transmigration, were identified by immunofluorescence flow cytometry. The filters had been coated initially with proteins and blocked with PVP, as in the cell adhesion assay. To identify INT receptors participating in cell adhesion and migration assays, isolated PBMCs were resuspended at 5 ⫻ 106 cells/ml in RPMI 1640 and incubated separately with 20 ␮g/ml mIgG (negative control) or functionblocking mAb IB4 (INT␤2), 60.1 (INT␣M) [24] (kindly provided by Prof. Patrick Beatty, University of Utah, Salk Lake City, UT, USA), 6S6 (INT␤1; Chemicon) 13 (INT␤1; BD Biosciences), P1B5 (INT␣3; Chemicon), or GoH3 (INT␣6; BD Biosciences) at room temperature for 20 min. Cell migration in the presence of control IgG was defined as 100% migration. All mAb, gift or commercial ones against INTs used in the functional assays, were protein Aor protein G-affinity chromatography-purified IgG. Proliferation of isolated CD4⫹ T cells (⬃99% purity) was measured as 3H-thymidine incorporation after 4 days stimulation, as described previously [37]. Immobilized mAb UCHT1 to CD3 (BD Biosciences) was tested in the absence or presence of HSA or various Lm isoforms. For statistical analysis, Student’s t-test, mean, and SD values were calculated, and the level of significance (*, P⬍0.05; **, P⬍0.01; ***, P⬍0.001) was determined by comparing Lm isoforms with HSA or mAb with mIgG.

Flow cytometric determination of LN cell number in Lm␣4-deficient mice Lm␣4 KO mice were generated by gene targeting in embryonic stem cells, as described previously [23]. Wild-type (WT; ⫹/⫹) and KO (–/–) adult male mice were used to determine the cell number in peripheral LN. Animals were killed by cervical dislocation, and cells were isolated from inguinal LNs by using a mesh iron screen and a 70-␮m cell strainer. Total isolated cells resuspended in 0.3 ml PBS were quantified by flow cytometry using a FACSort (BD Biosciences, San Jose, CA, USA) and identified according to forward- and side-scatter characteristics. All experiments were approved by the regional ethical committee for animal experimentation.

RT-PCR RT-PCR was performed as described previously [37] with a few modifications. Total RNA was extracted from blood lymphocytes by using the Ribopure TM-WBC kit (Ambion Europe Ltd., Huntingdon, UK). RT and PCR conditions and primers for Lm␣5, Lm␤1, Lm␤2, and Lm␥1 have been described [37, 40].

Gel electrophoresis, immunoblotting, ECL, and purification of Lm from cell lysate by immunoaffinity chromatography and induction of lymphocyte secretion Cell lysate from 1 ⫻ 109 isolated blood lymphocytes was obtained as described previously [37], and Lm was purified from it by immunoaffinity chromatography with mAb 4C7 (Lm␣5; kindly provided by Eva Engvall, Burham Institute, La Jolla, CA, USA), which had been coupled to cyanogen bromide-activated Sepharose 4B (Amersham Pharmacia AB). Following extensive washing, the protein bound to the column was eluted using high pH, and the samples were collected in neutralizing buffer, as reported previously [24, 37]. To study Lm secretion, isolated lymphocytes were resuspended at 5 ⫻ 107 cells/ml in plain RPMI 1640. After preincubation for 15 min, the cells were stimulated with 200 nM TPA for 20 min at 37°C. Following addition of protease inhibitors and centrifugation of the cells, supernatant was collected and analyzed for secreted Lm. After concentration, the purified and secreted materials were separated by SDS-PAGE, and blotted filters were blocked with 0.1% Tween 20/5% dry, skimmed milk in PBS. mAb 6C3 (Lm␣4) [24], 15H5 (Lm␣5), DG10 (Lm␤1), C4 (Lm␤2), and 22 (Lm␥1; BD Biosciences) were used as immunoblotting mAb. mIgG (BD Biosciences) was included as a negative control. Peroxidaseconjugated anti-mIg (Dakopatts) was added as secondary antibody, and ECL (Amersham Pharmacia AB) was used as a developer.

RESULTS Expression of Lms in HEVs and RFs We studied first the localization of Lm chains in normal human LNs by CLSM in HEVs (Fig. 1). Lm␥1 chain immunoreactivity (B) was broad in the basal aspect of HEVs, visualized with mAb MECA-79 (A), as demonstrated by a variable yellow-red color in the merged figure (C). Immunoreactivity for Lm␥1 chain was also found in BM of ordinary blood vessels (B) and in fibril-like structures outside of the vessels. Merged figures for HEV and Lm␣1 (D) and Lm␣2 (E) chains, respectively, showed that these components were not found in BM of HEVs or other vessels but that the Lm␣2 chain was present in stromal fibers. In contrast, immunoreactivity for Lm␣3 (F), Lm␣4 (G), Lm␣5 (H), Lm␤1 (I), and Lm␤2 (J) chains located variably to the BM region of practically all HEVs and other blood vessels and to stromal fibers. Instead, immunoreactivities for Lm␤3 (K) and Lm␥2 (L) chains were confined solely to capillaries and fibers in the germinal center area. Under the present experimental conditions, neither apical staining nor double-layered BM could be observed in HEVs with the Lm antibodies. Immunoreactivity of Lm chains in ordinary blood vessels could be confirmed by double-immunoreaction with Ulex europaeus I lectin-conjugate (data not shown). We then studied the localization of Lm chains in RFs of the LN by using double-immunoreaction for Lm chains and type III collagen, a major component of these stromal fibers. Figure 2 shows that immunoreactivity for type III collagen was prominent in the walls of vessels and in fiber-like structures outside of them (A). Immunoreactivity for the Lm␥1 chain showed a rather similar localization (B), and the merged figure (C) showed an extensive colocalization for these proteins in the fiber-like structures. Immunoreactitivity of the Lm␣1 chain was not found in the LNs (D), and immunoreactivities for Lm␣2 (E), Lm␣3 (F), Lm␣4 (G), Lm␣5 (H), Lm␤1 (I), and Lm␤2 (J) chains showed an extensive colocalization with that of type III collagen. Immunoreactivities for Lm␤3 (K) and Lm␥2 (L) chains were found in colocalization with type III collagen fibers only in the germinal center area. Lm expression in HEVs and RFs of tonsils was similar to the one of LNs (data not shown). Collagen type IV ␣1/␣2 and nidogen-1 were located in all BMs, including those of HEVs, and were also present in RFs (data not shown).

Expression of Lm-binding INTs in blood lymphocytes Among the INT family, at least seven members are known to recognize Lm isoforms [9]. Expression of Lm-binding INTs in the lymphocytes was studied by immunofluorescence flow cytometry of purified PBMCs (Fig. 3). Lymphocytes, gated by forward- and side-scatter, displayed marked immunoreactivity for INT␣6, -␤1, -␣L, -␣M, and -␤2 chains. A low expression of INT␣1, -␣3, and -␣V␤3 was also noted. INT␣2, -␣7, -IIb, -␤4, and -␣V␤5 was minimally expressed or not detected. By using multicolor immunofluorescence flow cytometry, expression of the “classical” Lm-binding INTs ␣6␤1 and ␣3␤1 by the blood lymphocytes was further investigated. INT␣6 was found to be expressed by the majority of CD4⫹ (Th) cells (77.7%) and Gorfu et al. Laminin isoforms in lymphocyte biology

3

Fig. 1. Expression of Lm chains in LN HEVs (green), defined by MECA-79-positive reactivity, and Lm chains (red), examined with the corresponding mAb, was analyzed in LNs by confocal microscopy. Following merging, colocalization was revealed by yellow staining. (A) mAb MECA-79 alone; (B) mAb 113BC7 to Lm␥1 alone; (C) merge of A plus B; and (D–L) Merge of MECA-79 with mAb to different Lm chains: (D) mAb 161EB7 to Lm␣1; (E) mAb 5H2 to Lm␣2; (F) mAb BM-2 to Lm␣3; (G) mAb 168FC10 to Lm␣4; (H) mAb 4C7 to Lm␣5; (I) mAb 114DG10 to Lm␤1; (J) mAb S5F11 to Lm␤2; (K) mAb 6F12 to Lm␤3; and (L) mAb D4B5 to Lm␥2. HEVs expressed Lm␣3, -␣4, -␣5, -␤1, -␤2, and -␥1 chains and lacked reactivity with antibodies to Lm␣1, -␣2, -␤3, and -␥2 chains. In contrast to HEVs, germinal center vessels expressed Lm␥2 staining (L). Original scale bar (100 ␮m) in L applies to all pictures.

Fig. 2. Expression of Lm chains in LN RFs (green), defined by type III collagen-positive immunoreactivity, and Lm chains (red), examined with the corresponding mAb, was analyzed in LNs by confocal microscopy. Following merging, colocalization was revealed by yellow staining. (A) mAb 3G4 against type III collagen alone; (B) mAb 113BC7 to Lm␥1 alone; (C) Merge of A plus B; and (D–L) Merge of mAb 3G4 with mAb to different Lm chains: (D) mAb 161EB7 to Lm␣1; (E) mAb 5H2 to Lm␣2; (F) mAb BM-2 to Lm␣3; (G) mAb 168FC10 to Lm␣4; (H) mAb 4C7 to Lm␣5; (I) mAb 114DG10 to Lm␤1; (J) mAb S5F11 to Lm␤2; (K) mAb 6F12 to Lm␤3; and (L) mAb D4B5 to Lm␥2. RFs expressed Lm␣2, -␣3, -␣4, -␣5, -␤1, -␤2, and -␥1 chains and lacked reactivity with antibodies to Lm␣1, -␤3, and -␥2 chains. Staining of germinal center vessels with mAb to Lm␤3 and -␥2 chains was also observed in K and L, respectively. Original scale bar (100 ␮m) in L applies to all pictures.

4

Journal of Leukocyte Biology Volume 84, September 2008

http://www.jleukbio.org

Fig. 3. Expression of Lm-binding INTs in blood lymphocytes. Flow cytometry analysis of INT expression on lymphocytes: mIgG (open peak) and mAb (shaded peak) 6S6 (INT␤1), FB12 (INT␣1), P1E6 (INT␣2), P1B5 (INT␣3), BQ16 (INT␣6), 9.1 (INT␣7), 69.6.5 (INT␣V), IB4 (INT␤2), H12 (INT␣L), 2LPM (INT␣M), 3.9 (INT␣X), SZ.21 (INT␤3), SZ.22 (INTIIb), LM609 (INT␣V␤3), ASC-9 (INT␤4), and P1F6 (INT␣V␤5). The vertical axis shows the cell number, and the horizontal axis shows the log fluorescence intensity [fluorescence 1-height (FL1-H)]. Mean value of mean fluorescence intensity and percentage of positive cells (%) of three experiments were as follows: mIgG: 4.3 (0%), INT␤1: 60.4 (86.9%), INT␣1: 8.6 (28.0%), INT␣2: 5.4 (3.9%), INT␣3: 9.0 (29.0%), INT␣6: 29.9 (72.6%), INT␣7: 4.8 (1.7%); INT␣V: 6.0 (4.9%), INT␤2: 302.0 (92.9%), INT␣L: 214.4 (92.7%), INT␣M: 35.7 (34.3%), INT␣X: 5.9 (7.7%), INT␤3: 5.6 (3.4%), INTIIb: 5.3 (1.6%), INT␣V␤3: 6.6 (12.4%), INT␤4: 5.0 (1.1%), and INT␣V␤5: 5.1 (0.5%).

CD8⫹ (T cytotoxic) cells (78.0%) but only by a few CD19⫹ (B) cells (6.3%). INT␤1 was detected in almost all CD4⫹ cells (87.1%) and CD8⫹ cells (95.5%) but only on half of the CD19⫹ cells (49.6%), whereas a low percentage of CD4⫹ cells, CD8⫹ cells, and CD19⫹ cells expressed INT␣3 (16.1%, 15.1%, and 10.8%, respectively; mean value of percentage of positive cells from three donors).

Lm-511 is highly adhesive for blood lymphocytes via ␣6␤1 INT To determine whether Lm-511, a major endothelial BM protein [34], may interact with lymphocytes during extravasation, a cell adhesion assay under static conditions was first performed. Previous studies have demonstrated the lymphocyte adhesive property of Lm-411, another endothelial BM isoform [34, 37]. Here, immobilized rLm-511 and rLm-411 were compared in cell adhesion assays with isolated blood lymphocytes. These

recombinant proteins share Lm␤1 and -␥1 chains but differ in their ␣ chain. Natural mLm-111 (EHS-Lm), human pFN, and HSA were also used. The constitutive cell adhesion to all proteins was minimal, if any (data not shown). However, following stimulation of the cells with phorbol ester for 1 h, the Lm isoforms were more adhesive than HSA, and Lm-511 was more active than Lm-411, followed by Lm-111 (P⬍0.01; Fig. 4A). Correlation of OD values of the adhesion assay with the cell number indicated that nearly 23% of the blood lymphocytes adhered on Lm-511. Under similar experimental conditions, IL-2 (at 1, 5, and 25⫻103 units/ml) and SDF-1␣ [at 0.5, 2.5, and 8 ␮g/ml (1 ␮M)] were unable to induce lymphocyte adhesion (data not shown). Lm-211 and Lm-332 were less adhesive than Lm-111 and hence, poorly active (data not shown). When compared with 1 h, lymphocyte adhesion on Lm-511 slightly increased at 3 h but drastically decreased at 12 h (data not shown). To identify the adhesive receptor, Gorfu et al. Laminin isoforms in lymphocyte biology

5

Lm-511 predominantly promotes blood lymphocyte migration, and migrating lymphocytes use ␣6␤1 INT as a primary Lm receptor

Fig. 4. Adhesion of isolated blood lymphocytes to Lm-511 and to other Lm isoforms and participation of INT. (A) Adhesion of isolated blood lymphocytes to plastic surfaces coated with HSA, Lm-111, Lm-411, rLm-511, and pFN in the presence of phorbol ester (200 nM TPA). Adhesion was measured after incubating the cells (200⫻103 cells/well) in the coated plates for 1 h at 37°C and shown as OD. Plastic surfaces were coated with proteins at 20 ␮g/ml and blocked with 2% PVP. (B) Effect of mAb to INT on lymphocyte adhesion to plastic surfaces coated with HSA, Lm-111, Lm-411, rLm-511, and pFN. mIgG (negative control), mAb IB4 (INT␤2), mAb 6S6 (INT␤1), and mAb GoH3 (INT␣6). Adhesion blocking mAb were used at 20 ␮g/ml. Mean and SD of four to six experiments are shown. P values (*, P⬍0.05; **, P⬍0.01) compare (A) Lm isoforms with HSA and (B) mAb with mIgG on each Lm isoform.

blocking mAb to INT␤2 (IB4), INT␤1 (6S6), and INT␣6 (GoH3) were tested on lymphocyte adhesion to Lm-111, Lm411, Lm-511, and pFN. Statistically significant inhibition (P⬍0.01; Fig. 4B) was obtained with mAb GoH3 and 6S6 on Lm-411 and Lm-511 only, indicating a role of ␣6␤1 INT as a receptor for both vascular Lm isoforms. Although mAb IB4 to INT␤2 was partially inhibitory on all substrata, including HSA, its effect alone did not reach statistical significance. Of INTs analyzed in Figure 3, cell-surface expression of ␣M␤2 was almost doubled by the TPA treatment, whereas expression of all other INTs was slightly reduced or not affected (data not shown). 6

Journal of Leukocyte Biology Volume 84, September 2008

To further analyze the interactions between lymphocytes and Lm isoforms, lymphocyte migration on Lm-111, Lm-211, Lm332, Lm-411, and Lm-511 (recombinant and from placenta) was studied in cell transmigration assays, and participation of INT receptors was determined by using function-blocking mAb. Compared with HSA, rLm-411 and rLm-511 significantly promoted lymphocyte migration in the presence of SDF-1␣ as a chemoattractant (Fig. 5A; P⬍0.01 and P⬍0.001, respectively). The level of migration on rLm-511 was by far higher than that on Lm-111, rLm-411, and pFN, and the difference with rLm-411, the other major endothelial Lm isoforms [34], was highly significant (P⬍0.001). In a second group of experiments, placenta Lm-511 was tested and compared with Lm111, rLm-211, Lm-332, and rLm-411. This natural Lm-511 was also found to strongly promote migration of blood lymphocytes (Fig. 5B; P⬍0.001) and to a higher level than the other Lm isoforms. Natural and rLm-511 promoted migration of 80 –100 ⫻ 103 lymphocytes, which correspond to 23–24% of the initial lymphocyte population (i.e., 70% of 500⫻103 PBMCs). Natural Lm-211 (merosin) behaved similarly to rLm-211 (data not shown). In decreasing order, the migration-promoting activity of the various Lm isoforms could be summarized as follows: Lm-511 ⬎ Lm-332 ⬎ Lm-411 ⬎ Lm-111 ⬎ Lm-221, indicating that Lm-511 predominantly promotes migration of blood lymphocytes compared with other Lm isoforms. Interestingly, Lm-511 was also more migration-promoting than pFN. Analysis of lymphocyte populations by immunofluorescence flow cytometry before and after transmigration demonstrated that CD4⫹ (Th) cells and CD8⫹ (cytotoxic T) cells but not CD19⫹ (B) cells readily migrated on Lm-511, as indicated by a shift in the percentage of CD4⫹ cells, CD8⫹ cells, and CD19⫹ cells from 34.6%, 23.9%, and 12.9% to 49.6%, 23.2%, and 1.8%, respectively (mean value of three different donors). Whereas 12.4% of CD4⫹ cells (15,168 out of 122,688) and 8.4% of CD8⫹ cells (7093 out of 84,575) migrated on Lm-511, only 1.3% of CD19⫹ cells (537 out of 45,785) migrated on this Lm isoform. The effect of functionblocking mAb to INTs indicated that migration of blood lymphocytes on Lm-511 was largely mediated by ␣6␤1 INT, as antibodies to INT␣6 and INT␤1 chains exerted a major and highly statistically significant inhibition (Fig. 5C). Antibodies to INT␣3 alone were inactive, but together with INT␣6 antibodies, they exerted 10% more inhibition than INT␣6 antibodies alone, suggesting a small contribution of ␣3␤1 INT (data not shown). On Lm-411, mAb to INT␣6 were also inhibitory. In contrast, mAb to ␤2 INT exerted partial inhibition on both Lm isoforms, but the effect was not statistically significant. On Lm-332, mAb to INT␣3 and INT␤1 were inhibited significantly, as well as mAb to INT␣M and INT␤2, indicating participation of ␣3␤1 and ␣M␤2 INTs in the cell migration (Fig. 5D). On the other hand, antibodies to INT␣6 were modestly inhibitory, and their effect did not reach statistical significance (Fig. 5D). http://www.jleukbio.org

Fig. 5. Lm-511 strongly promotes migration of blood lymphocytes via ␣6␤1 INT and costimulates T cell proliferation. (A–D) Cell migration assay; (E) T cell proliferation assay. Lymphocyte transmigration through 3 ␮m pore filters precoated with proteins was measured after 3 h incubating. (A) HSA, Lm-111, Lm-411, rLm-511, and pFN. (B) HSA, Lm-111, Lm-211, Lm-332, Lm-411, and Lm-511 (placenta). (C) Effect of blocking mAb to INTs on lymphocyte transmigration through filters precoated with Lm-411 and rLm-511. mIgG (negative control), mAb IB4 (INT␤2), mAb 6S6 (INT␤1), and mAb GoH3 (INT␣6). (D) Effect of blocking mAb to INTs on lymphocyte transmigration through filters precoated with Lm-332. mIgG, mAb P1B5 (INT␣3), mAb GoH3 (INT␣6), mAb 13 (INT␤1), mAb IB4 (INT␤2), and mAb 60.1 (INT␣M). (E) Proliferation of isolated CD4 T cells on immobilized mAb CD3 alone, Lm alone, or in combinations: CD3 (mAb UCHT1), HSA, Lm-111, Lm-411, rLm-511, Lm-511 (placenta). Cell proliferation was measured by H3-thymidine incorporation after culturing cells for 4 days. In all figures, mean and SD of at least four experiments are shown. P values (*, P⬍0.05; **, P⬍0.01, ***, P⬍0.001) compare Lms with HSA (A, B, and E) or mAb with mIgG, defined as 100% in all experiments (C and D).

Lm-511 costimulates T cell proliferation Recently, Lm-411 was found to costimulate proliferation of T cells [37]. This Lm isoform is expressed by monocytic cells, lymphoid B cells, and endothelial cells [9, 37, 43]. To investigate whether additional Lm isoforms were able to provide a similar costimulatory signal, Lm-511 and mAb UCHT1 to CD3 were coimmobilized, and isolated CD4⫹ T cells were incubated on the coated wells for 4 days. Under these experimental conditions, extensive cell proliferation was induced by placenta Lm-511 and rLm-511 (Fig. 5E), as measured by 3Hthymidine incorporation. Lm-511 or mAb to CD3 failed to induce any significant cell proliferation on its own (Fig. 5E).

LN cell number in Lm␣4-deficient mice compared with WT do not differ significantly As an initial approach to investigate the role of Lms in lymphocyte biology in vivo, the total cell number of inguinal LNs was compared in Lm␣4-deficient and WT mice. This parameter appears to be influenced by entrance/exit and proliferation/

survival of lymphocytes. Mean value and SD of the total cell number were 405 ⫾115 (⫻103) and 464 ⫾186 (⫻103) for Lm␣4-deficient and WT mice, respectively. This difference was not statistically significant (P⫽0.67). Six animals were tested in each group. Histological examination of Peyer’s patches, a lymphoid tissue that is highly exposed to antigenic stimulation, demonstrated that Lm␣4-deficient mice retained the ability to develop germinal centers (data not shown).

Blood lymphocytes contain and following stimulation, secrete Lm- 511 and Lm-521 To investigate expression of Lm␣5 by blood lymphocytes, the presence of mRNAs encoding for Lm-511 and Lm-521 chains was first studied by RT-PCR, using pairs of primers based on the reported cDNA sequences of hLm␣5, -␤1, -␤2, and -␥1 chains. Amplified products with the expected size were detected with primers for ␣5 (655), ␤2 (632), and ␥1 (687) after 30 cycles of PCR with reverse-transcribed transcripts from blood lymphocytes (Fig. 6A). Results for ␤1 transcripts were inconclusive (data not shown). Gorfu et al. Laminin isoforms in lymphocyte biology

7

Fig. 6. Blood lymphocytes contain Lm-511 and Lm-521 and secrete Lms␣4 and -␣5. (A) RT-PCR of Lm-521 components in isolated lymphocytes. Amplification of Lm␣5, -␤2, and -␥1 cDNA fragments by RT-PCR of total RNA. GAPDH and buffer were used as positive and negative controls, respectively. (B) Detection of Lm␣5, -␤1, -␤2, and -␥1 chains in blood lymphocytes by immunofluorescence flow cytometry. Reactivity of mAb with nonpermeabilized and permeabilized cells is shown. mIgG (shaded peak) and mAb (open peak): 9.6 (CD2), 15H5 (Lm␣5), LN-26 (Lm␤1), C4 (Lm␤2), and LN-82 (Lm␥1). CD2 was used as a lymphocyte marker. (C) Western blot analysis of Lm␣5 mAb (4C7) immunoaffinity-purified protein from isolated blood lymphocyte lysate (upper panel) and of total secreted material from isolated blood lymphocytes stimulated with 200 nM TPA for 20 min (lower panel). Name of the antibodies and their specificity (in parentheses) are shown. Arrows indicate bands of 350, 230, 220, and 190 kDa, corresponding to Lm␣5, -␤1, -␥1, and -␤2, respectively (upper panel), and bands of 300/280, 230, 220, 190, and 180, corresponding to Lm␣5, -␤1, -␥1, -␤2, and -␣4 chains, respectively (lower panel). *, A band of approximately 200 kDa reactive with mAb 15H5 (Lm␣5), which was observed occasionally. Proteins were separated by SDS gel electrophoresis under reducing conditions (5% acrylamide).

To further address the presence of Lms␣5 in blood lymphocytes, immunofluorescence flow cytometry was performed. Lymphocytes, gated by forward- and side-scatter from PBMCs, displayed reactivity with mAb to CD2, a marker of most lymphocytes. mAb 15H5 (Lm␣5), LN-26 (Lm␤1), C4 (Lm␤2), and LN-82 (Lm␥1) did not react, or only minimally, with intact lymphocytes, whereas practically all cells were stained with the antibodies following cell permeabilization (Fig. 6B). To further investigate expression of Lm␣5, the cell lysate of isolated blood lymphocytes was immunopurified on a Lm␣5 (4C7) antibody column, and the purified material was then analyzed by Western blotting. This technique allows identification of Lm chains and demonstrates their physical association. Under reducing conditions, mAb DG10 (Lm␤1), C4 (Lm␤2), 22 (Lm␥1), and 15H5 (Lm␣5) recognized polypeptides of 230, 190, 220, and 350 8

Journal of Leukocyte Biology Volume 84, September 2008

kDa, respectively (Fig. 6C, upper panel). The intensity of Lm␤1 and -␤2 bands suggested that Lm-521 was more abundant than Lm-511. In the total supernatant of blood lymphocytes stimulated with phorbol ester for 20 min, Lm␤1, -␤2, and -␥1 were also detected, in addition to the Lm␣4 chain (Fig. 6C, lower panel), and Lm␣5 appeared as smaller 300/280 kDa bands (Fig. 6C, lower panel). Occasionally, an additional band (*) of approximately 200 kDa, reactive with mAb 15H5 (Lm␣5), was observed in the secreted material, suggesting further proteolytic cleavage.

DISCUSSION In the present study, Lms of HEVs and RFs of lymphoid tissue and their INT receptors in blood lymphocytes are identified, http://www.jleukbio.org

and the effect of Lm isoforms covering all ␣ chains on the lymphocytes is described. Moreover, a strong and predominant, migration-promoting effect of Lm␣5 on blood lymphocytes, particularly T cells, is reported for the first time, as well as synthesis and secretion of Lm␣5 by lymphocytes. In spite of its apparent relevance to lymphocyte recirculation, the role and composition of HEV BMs are poorly understood. Early morphological studies reported a thick, dense, and strongly negatively charged BM, distinct from that of flat endothelium [44 – 46]. Our confocal microscopy studies suggested the expression of Lm-311/321, Lm-411/421, and Lm511/521 by the HEV BM. Neither Lm-332 nor Lm␣1 or -␣2 was detected in these vessels. As all HEVs expressed the same Lm chains, each HEV BM should contain simultaneously at least three different Lm isoforms. The HEV BMs may provide an adhesive and migration-promoting substrate for the lymphocytes, as demonstrated for Lms in the present study. Moreover, the Lms may also be able to bind and present chemokines and other cytokines, as reported for other BM proteins [5]. In contrast to Lm␣4 and -␣5, Lm␣3 is rarely found in blood vessels, as only vasculatures of lymphoid and a few other tissues express this Lm chain [28, 34] (to be published). Among the lymphoid vessels, selective staining of the capillaries in the lymphoid follicles by antibodies to ␤3 and ␥2 Lm chains indicated the presence of Lm-332 in this vasculature but not in HEVs. Others [35] have reported similar results. Expression of Lm␣5 by LN blood vessels, including HEV, has been described in mice [47], and broad staining of human LN vasculature, without specifying HEVs, with antibodies to Lm␣3, -␣4, and -␣5 chains has been reported recently [12]. As major components of the reticular network, RFs connects the subcapsular sinus to HEVs in lymphoid tissues, allowing the transport of small molecules from the afferent lymph in a conduit system [13, 14, 36]. With a core of collagen types I and III and surrounded by BM components, these fibers are largely, but not fully, ensheathed by fibroblastic reticular cells [11, 13, 36]. Nearly 10% of the surface of these interfollicular fibers appears to be accessible to migratory cells, allowing physical interaction with lymphocytes, dendritic cells, and macrophages [7]. In the present study, the confocal microscopy suggested expression of Lm-211/221, Lm-311/321, Lm-411/421, and Lm-511/521, but not Lm␣1, by the RFs. Lack of reactivity with mAb to ␤3 and ␥2 Lm chains indicated that RFs in LNs, in contrast to thymus [48], do not express Lm-332. Although Lms appear to be more accessible to lymphocytes in HEV BMs than in RFs, Lms and other outer components of the RFs may still function, not only as cell-adhesive substrates but also as migration-promoting proteins for lymphocytes and other migrating immune cells. Others [12, 13, 36] have also reported expression of Lm␣2, -␣3, -␣4, and -␣5 but not -␣1 chains by LN RFs. Among Lm-binding INTs expressed by blood lymphocytes, ␣6␤1 was the most abundant, followed by ␣1␤1, ␣3␤1, and ␣V␤3 at much lower levels. Analysis of lymphocyte populations indicated that ␣6␤1 INT was expressed by most T cells, including CD4⫹ cells and CD8⫹ cells, and that most B cells lacked this INT, and INT ␣3␤1 was expressed by a minority of cells in each lymphocyte population. INT ␣M␤2, which has been proposed to be a Lm receptor [24], was detected in half of

the lymphocyte population. NK cells, CD8⫹ T cells, and B cells are known to express this INT [49]. In functional assays of the present study, Lms were adhesive for blood lymphocytes after stimulation of the cells with TPA. Lm␣5 was the most active isoform, and vascular Lm␣4 and -␣5 used the ␣6␤1 INT as a primary receptor with some contribution of (␣M)␤2 INT. Through protein kinase C activation and protein phosphorylation, TPA may induce INT activation by cytoskeleton-mediated clustering of cell membrane INTs and/or conformational changes of these adhesive receptors. ␣M␤2 INT may be responsible for the relatively high, “nonspecific” lymphocyte adhesion, as reported for neutrophils [24]. The extent of adhesion of stimulated lymphocytes to HSA, heat-denatured HSA, and OVA was similar (data not shown). When compared with our early study that focused on the interaction of lymphocytes with Lm-411 [37], the present adhesion results with stimulated cells were largely similar, but the adhesion to Lm-411, although higher than on HSA and lower than on Lm-511, did not reach statistical significance. Moreover, specific adhesion of the cells to Lms was undetectable in the absence of stimulus. These differences may be explained by the use of isolated blood CD4 T cells in the early study, in contrast to total blood lymphocytes (Percoll-isolated) in the present study. The later preparation contains, in addition to CD4⫹ T cells, NK cells, CD8⫹ T cells, and B cells, which in contrast to CD4⫹ T cells, express considerable amounts of ␣M␤2 INT [49]. Participation of this INT apparently decreased the adhesive ratio between Lms and HSA. A predominant, cell-adhesive property of Lm␣5, when compared with other isoforms, has also been reported for hematopoietic cell lines, encephalitogenic T cell lines, malignant lymphoid cells, a subpopulation of thymocytes, and bone marrow progenitor cells and was similarly mediated via ␣6␤1 INT [27, 33, 50 –52]. Correspondingly, migration of bone marrow progenitor cells and malignant lymphoid cells was higher on Lm␣5 than on other Lm isoforms via the same INT receptor [32, 50]. In our present study, the migration-promoting activity of Lm␣5 on blood lymphocytes was superior to that of Lm␣1, -␣2, -␣3, and -␣4, as well as to that of FN, probably the most extensively used ECM protein in lymphocyte studies. Interestingly, analysis of lymphocyte populations demonstrated that the CD4⫹ cells and the CD8⫹ cells, but not the B cells, readily migrated on Lm-511, suggesting a migration-promoting activity of this Lm isoform preferentially on T cells. Studies in progress investigate the effect of various Lm isoforms in several lymphocyte populations and the contribution of INTs. In our early Lm-411/lymphocyte study [37], a commercial placenta Lm preparation, referred to as LN-10/11 (Lm-511/521), was used for comparison and found not to promote lymphocyte migration. We did not molecularly characterize the batches of this preparation. Recent studies from our group have demonstrated major molecular and functional heterogeneity among different batches of the same commercial Lm product [16]. In the present study, only those batches reactive with a mAb 4C7 to Lm␣5 globular domain [16], which is recognized by ␣3␤1 and ␣6␤1 INTs [53], were used as Lm-511. These molecularly characterized preparations behaved similarly to rLm-511 in the cell migration assays. Gorfu et al. Laminin isoforms in lymphocyte biology

9

The statistically significant, inhibitory effect of mAb to INT␣6 on the adhesion and migration of the lymphocytes on Lm-411 and Lm-511 and a similar effect of mAb to INT␤1 identified ␣6␤1 INT as a primary receptor for these vascular Lms. Interestingly, expression of this INT on blood CD4⫹ (Th) cells, CD8⫹ (cytotoxic T) cells, and CD19⫹ (B) cells roughly correlated with the ability of these lymphocyte populations to migrate on Lm-511. Moreover, the additive effect of mAb to INT␤2 suggested an additional contribution of the leukocyte INTs. Treatment of rats with rabbit antibodies to a rat Lm has been found to decrease the accumulation of adoptively transferred lymphocytes in peripheral LNs [54]. The role of Lms in lymphocyte migration in vivo is similarly supported by studies with mice lacking ␣6A INTs, the Lm-binding INT splice variant expressed by the immune cells [55]. In these INT␣6A KO mice, lymphocytes displayed decreased migration on a Lm␣1 substrate and a reduction in the number of T cells isolated from peripheral LNs. However, lymphocyte homing to the LNs, which involves several receptor-ligand interactions, was not affected in these animals. As a classical Lm receptor, ␣6␤1 INT binds Lms covering all ␣ chains but with some preference for Lm␣5 [53]. Antigen-specific activation of T cells is known to require not only CD3/TCR engagement by antigen/MHC but also additional costimuli. In a previous study, we showed a synergistic effect of Lm-411 and a placenta Lm-511 preparation together with mAb to CD3 to drive CD4⫹ T cells to proliferate, in line with previous reports [1, 37]. In the present study, we reproduced these findings and observed, in addition, a stronger, synergistic effect of Lm-511 when compared with Lm-411. The higher costimulatory effect of placenta Lm-511, when compared with rLm-511, may be a result of in vivo processing of the natural form. Inasmuch as T cells normally recognize antigen present on cells (with MHC molecules) rather than on ECMs, it is unlikely that Lms may exert such an effect as ECM components. Instead, Lms may be costimulatory to T cells when exposed on the surface of monocyte/macrophages and other APCs. In accordance, we recently found the presence of Lm␣4 and -␣5 in monocytes, and some of these moieties were localized on the cell surface [43] (to be published). In contrast to other Lms, Lm-332, which is expressed by the thymus, has been described not to be comitogenic to thymocytes [31]. On the other hand, this isoform was reported to promote thymocyte migration via ␣3␤1 INT and to be poorly adhesive for blood lymphocytes, which expressed low amounts of the latter INT [28, 30]. In the present study, Lm-332 was similarly poorly adhesive for blood lymphocytes but significantly promoted migration of these cells via ␣3␤1 and ␣M␤2 INTs. Recently, we reported strong migration-promoting activity of Lm-332 and Lm-511 in glioma cells via ␣3␤1 INT [56]. It should be noted that we use Lm-332 as a source of Lm␣3, as Lm-311/321 are not available. INTs ␣3␤1, ␣6␤1, and ␣6␤4 are known to bind the Lm globular tandem of the Lm␣3 chain [9], which may also be important for the interaction of newly extravasated lymphocytes with epithelial BMs [28]. When compared for adhesion, migration, and costimulation, rLm-511 was more active than rLm-411. As both Lm isoforms share ␤1 and ␥1 chains, their different biological effects should be ascribed to their distinct ␣ chain. Lm ␣5, the last ␣ 10

Journal of Leukocyte Biology Volume 84, September 2008

chain to be identified, is the most widely distributed ␣ chain and is highly expressed by epithelia and endothelia [57]. It has an apparent molecular weight of 350 kDa, in contrast with the shorter 200 kDa Lm␣4 chain, and it is recognized by ␣3␤1, ␣6␤1, and ␣6␤4 INTs [53]. Interestingly, its vascular expression appears to be developmentally regulated, as it is widely expressed in blood vessels in adult life [34, 47]. Lm␣5 genetic deletion leads to death late in embryogenesis, probably as a result of defects in the placental vasculature and other developmental abnormalities [58]. In contrast, Lm␣4 KO mice survive, in spite of transient hemorraghes at birth, which reflect impaired microvessel maturation [23]. Although we have previously reported defective recruitment of neutrophils in a peritonitis model in Lm␣4 KO mice [24], we were unable to detect significant alterations in the LN cell number of these animals in the present study. Considering that lymphocyte trafficking (homing and egress), proliferation, and survival in the lymphoid tissue may all affect the LN cell number, this result does not support a pivotal role for Lm␣4 in these lymphocyte activities. It may, instead, suggest a major role of Lm␣5. A recent study showing colocalization of extravasating mononuclear leukocytes with Lm␣4- but not Lm␣5-expressing brain blood vessels in experimental autoimmune encephalitis in mice has led to the assumption that the Lm␣4 promotes cell extravasation, whereas Lm␣5 is nonpermissive and even inhibitory [52]. A more recent study of normal and pathological (multiple sclerosis) human brain has not confirmed this exclusive expression of Lm␣4 or -␣5 by brain vessels but demonstrated instead simultaneous expression of the two isoforms by the vasculature and accumulation of leukocytes around vessels expressing Lm␣5 [59]. Whereas the role of vascular Lm isoforms may differ in inflammatory and homeostatic lymphocyte extravasation, our results do not support the concept of Lm␣5 as a nonpermissive or inhibitory substrate for blood lymphocyte migration. On the contrary, the functional studies, the Lm isoform recognition by ␣6␤1 INT, and the Lm isoform expression by HEVs strongly suggest a predominant role of Lm␣5 in lymphocyte extravasation. We have described previously synthesis and secretion of Lm␣4, as Lm-411, by lymphoid cells [37]. Now we report, for the first time in leukocytes the presence of Lm␣5, mainly as Lm-521 with some Lm-511, in blood lymphocytes and secretion of Lm␣5 following stimulation of the cells. As recently reported by us for platelets [40], the Lm␣5 chain associates more to the Lm␤2 chain than to the Lm␤1 chain. The weak Lm bands obtained from the blood lymphocytes by RT-PCR, which may reflect a low amount of transcripts present in these resting cells, suggest synthesis of Lm␣5. Immunofluorescence flow cytometry indicated that practically all blood lymphocytes expressed Lm␣5 but at relatively low amounts and mostly intracellularly. Lm␣4 and/or -␣5 may correspond to the reactivity of polyclonal anti-Lm-111 antibodies with mouse, rat, and human NK cells demonstrated in early studies [37]. The function of the endogenous Lms of lymphocytes is presently unknown. Interestingly, Lm␣4 and -␣5 were secreted following stimulation of the cells, and the larger Lm␣5 chain appeared to suffer proteolytic cleavage. In the absence of BM and RF Lms, Lm secreted and deposited by activated lymphocytes may be used for cell migration. On the other hand, Lms expressed on http://www.jleukbio.org

the lymphocyte cell surface may allow interaction of activated lymphocytes with other cells. Studies in progress address these issues. Altogether, the present results suggest that during lymphocyte extravasation into secondary lymphoid tissues, Lm␣5 and to a lower extent, Lm␣3 and -␣4 found in HEV BM contribute to lymphocyte migration through the vessel wall. In the RFs, these Lms may constitute pathways for the migration of lymphocytes and other cell types within the lymphoid tissue. This principle may also apply to HEV-like vessels found in chronically inflamed, nonlymphoid tissues of patients with rheumatoid arthritis and other inflammatory diseases. On the other hand, HEV and RF Lms may also contribute to the extravasation of cancer cells and the migration of these cells throughout the lymphoid tissue, respectively, promoting hematogenous and lymphatic dissemination of cancer. In the absence of exogenous Lms, stimulated lymphocytes may secrete, use, and provide Lm␣5. When compared with other Lm isoforms and ECM proteins, Lm␣5 appears to exert the strongest migrationpromoting activity for blood lymphocytes, particularly T cells.

ACKNOWLEDGMENTS This research work was supported by the Swedish Cancer Society and Karolinska Institutet (M. P.) and Engineering Virtual Organization research grants from the Helsinki University Hospital (I. V.). The authors thank Drs. Eva Engvall, Kiyotoshi Sekiguchi, Samuel Wright, Patrick Beatty, Sergei Smirnov, and Peter Yurchenco for providing mAb and recombinant proteins, Dr. Bjo¨rn Rozell for histological examination of mouse lymphoid tissues, and Ms. Ingegerd Andure´n and Mr. Herna´n Concha for excellent technical assistance.

REFERENCES 1. Shimizu, Y., Shaw, S. (1991) Lymphocyte interactions with extracellular matrix. FASEB J. 5, 2292–2299. 2. van den Berg, T. K., van der Ende, M., Dopp, E. A., Kraal, G., Dijkstra, C. D. (1993) Localization of ␤ 1 integrins and their extracellular ligands in human lymphoid tissues. Am. J. Pathol. 143, 1098 –1110. 3. Pribila, J. T., Quale, A. C., Mueller, K. L., Shimizu, Y. (2004) Integrins and T cell-mediated immunity. Annu. Rev. Immunol. 22, 157–180. 4. Bajenoff, M., Egen, J. G., Koo, L. Y., Laugier, J. P., Brau, F., Glaichenhaus, N., Germain, R. N. (2006) Stromal cell networks regulate lymphocyte entry, migration, and territoriality in lymph nodes. Immunity 25, 989 –1001. 5. Miyasaka, M., Tanaka, T. (2004) Lymphocyte trafficking across high endothelial venules: dogmas and enigmas. Nat. Rev. Immunol. 4, 360 – 370. 6. von Andrian, U. H., Mempel, T. R. (2003) Homing and cellular traffic in lymph nodes. Nat. Rev. Immunol. 3, 867– 878. 7. Hayakawa, M., Kobayashi, M., Hoshino, T. (1988) Direct contact between reticular fibers and migratory cells in the paracortex of mouse lymph nodes: a morphological and quantitative study. Arch. Histol. Cytol. 51, 233–240. 8. Miner, J. H., Yurchenco, P. D. (2004) Laminin functions in tissue morphogenesis. Annu. Rev. Cell Dev. Biol. 20, 255–284. 9. Patarroyo, M., Tryggvason, K., Virtanen, I. (2002) Laminin isoforms in tumor invasion, angiogenesis and metastasis. Semin. Cancer Biol. 12, 197–207. 10. Aumailley, M., Bruckner-Tuderman, L., Carter, W. G., Deutzmann, R., Edgar, D., Ekblom, P., Engel, J., Engvall, E., Hohenester, E., Jones, J. C., et al. (2005) A simplified laminin nomenclature. Matrix Biol. 24, 326 – 332.

11. Kramer, R. H., Rosen, S. D., McDonald, K. A. (1988) Basement-membrane components associated with the extracellular matrix of the lymph node. Cell Tissue Res. 252, 367–375. 12. Maatta, M., Liakka, A., Salo, S., Tasanen, K., Bruckner-Tuderman, L., Autio-Harmainen, H. (2004) Differential expression of basement membrane components in lymphatic tissues. J. Histochem. Cytochem. 52, 1073–1081. 13. Sixt, M., Kanazawa, N., Selg, M., Samson, T., Roos, G., Reinhardt, D. P., Pabst, R., Lutz, M. B., Sorokin, L. (2005) The conduit system transports soluble antigens from the afferent lymph to resident dendritic cells in the T cell area of the lymph node. Immunity 22, 19 –29. 14. Gretz, J. E., Norbury, C. C., Anderson, A. O., Proudfoot, A. E., Shaw, S. (2000) Lymph-borne chemokines and other low molecular weight molecules reach high endothelial venules via specialized conduits while a functional barrier limits access to the lymphocyte microenvironments in lymph node cortex. J. Exp. Med. 192, 1425–1440. 15. Geberhiwot, T., Wondimu, Z., Salo, S., Pikkarainen, T., Kortesmaa, J., Tryggvason, K., Virtanen, I., Patarroyo, M. (2000) Chain specificity assignment of monoclonal antibodies to human laminins by using recombinant laminin ␤1 and ␥1 chains. Matrix Biol. 19, 163–167. 16. Wondimu, Z., Gorfu, G., Kawataki, T., Smirnov, S., Yurchenco, P., Tryggvason, K., Patarroyo, M. (2006) Characterization of commercial laminin preparations from human placenta in comparison to recombinant laminins 2 (␣2␤1␥1), 8 (␣4␤1␥1), 10 (␣5␤1␥1). Matrix Biol. 25, 89 –93. 17. Doi, M., Thyboll, J., Kortesmaa, J., Jansson, K., Iivanainen, A., Parvardeh, M., Timpl, R., Hedin, U., Swedenborg, J., Tryggvason, K. (2002) Recombinant human laminin-10 (␣5␤1␥1). Production, purification, and migration-promoting activity on vascular endothelial cells. J. Biol. Chem. 277, 12741–12748. 18. Kortesmaa, J., Doi, M., Patarroyo, M., Tryggvason, K. (2002) Chondroitin sulphate modification in the ␣4 chain of human recombinant laminin-8 (␣4␤1␥1). Matrix Biol. 21, 483– 486. 19. Kortesmaa, J., Yurchenco, P., Tryggvason, K. (2000) Recombinant laminin-8 (␣(4)␤(1)␥(1)). Production, purification, and interactions with integrins. J. Biol. Chem. 275, 14853–14859. 20. Magner, W. J., Chang, A. C., Owens, J., Hong, M. J., Brooks, A., Coligan, J. E. (2000) Aberrant development of thymocytes in mice lacking laminin-2. Dev. Immunol. 7, 179 –193. 21. Rousselle, P., Aumailley, M. (1994) Kalinin is more efficient than laminin in promoting adhesion of primary keratinocytes and some other epithelial cells and has a different requirement for integrin receptors. J. Cell Biol. 125, 205–214. 22. Smirnov, S. P., McDearmon, E. L., Li, S., Ervasti, J. M., Tryggvason, K., Yurchenco, P. D. (2002) Contributions of the LG modules and furin processing to laminin-2 functions. J. Biol. Chem. 277, 18928 –18937. 23. Thyboll, J., Kortesmaa, J., Cao, R., Soininen, R., Wang, L., Iivanainen, A., Sorokin, L., Risling, M., Cao, Y., Tryggvason, K. (2002) Deletion of the laminin ␣4 chain leads to impaired microvessel maturation. Mol. Cell. Biol. 22, 1194 –1202. 24. Wondimu, Z., Geberhiwot, T., Ingerpuu, S., Juronen, E., Xie, X., Lindbom, L., Doi, M., Kortesmaa, J., Thyboll, J., Tryggvason, K., Fadeel, B., Patarroyo, M. (2004) An endothelial laminin isoform, laminin 8 (␣4␤1␥1), is secreted by blood neutrophils, promotes neutrophil migration and extravasation, and protects neutrophils from apoptosis. Blood 104, 1859 –1866. 25. Iwao, M., Fukada, S., Harada, T., Tsujikawa, K., Yagita, H., Hiramine, C., Miyagoe, Y., Takeda, S., Yamamoto, H. (2000) Interaction of merosin (laminin 2) with very late activation antigen-6 is necessary for the survival of CD4⫹ CD8⫹ immature thymocytes. Immunology 99, 481– 488. 26. Kim, M. G., Lee, G., Lee, S. K., Lolkema, M., Yim, J., Hong, S. H., Schwartz, R. H. (2000) Epithelial cell-specific laminin 5 is required for survival of early thymocytes. J. Immunol. 165, 192–201. 27. Kutlesa, S., Siler, U., Speiser, A., Wessels, J. T., Virtanen, I., Rousselle, P., Sorokin, L. M., Muller, C. A., Klein, G. (2002) Developmentally regulated interactions of human thymocytes with different laminin isoforms. Immunology 105, 407– 418. 28. Wayner, E. A., Gil, S. G., Murphy, G. F., Wilke, M. S., Carter, W. G. (1993) Epiligrin, a component of epithelial basement membranes, is an adhesive ligand for ␣ 3 ␤ 1 positive T lymphocytes. J. Cell Biol. 121, 1141–1152. 29. Virtanen, I., Lohi, J., Tani, T., Sariola, H., Burgeson, R. E., Lehto, V. P. (1996) Laminin chains in the basement membranes of human thymus. Histochem. J. 28, 643– 650. 30. Vivinus-Nebot, M., Rousselle, P., Breittmayer, J. P., Cenciarini, C., Berrih-Aknin, S., Spong, S., Nokelainen, P., Cottrez, F., Marinkovich, M. P., Bernard, A. (2004) Mature human thymocytes migrate on laminin-5 with activation of metalloproteinase-14 and cleavage of CD44. J. Immunol. 172, 1397–1406.

Gorfu et al. Laminin isoforms in lymphocyte biology

11

31. Vivinus-Nebot, M., Ticchioni, M., Mary, F., Hofman, P., Quaranta, V., Rousselle, P., Bernard, A. (1999) Laminin 5 in the human thymus: control of T cell proliferation via ␣6␤4 integrins. J. Cell Biol. 144, 563–574. 32. Spessotto, P., Gronkowska, A., Deutzmann, R., Perris, R., Colombatti, A. (2003) Preferential locomotion of leukemic cells towards laminin isoforms 8 and 10. Matrix Biol. 22, 351–361. 33. Spessotto, P., Yin, Z., Magro, G., Deutzmann, R., Chiu, A., Colombatti, A., Perris, R. (2001) Laminin isoforms 8 and 10 are primary components of the subendothelial basement membrane promoting interaction with neoplastic lymphocytes. Cancer Res. 61, 339 –347. 34. Hallmann, R., Horn, N., Selg, M., Wendler, O., Pausch, F., Sorokin, L. M. (2005) Expression and function of laminins in the embryonic and mature vasculature. Physiol. Rev. 85, 979 –1000. 35. Jaspars, L. H., De Melker, A. A., Bonnet, P., Sonnenberg, A., Meijer, C. J. (1996) Distribution of laminin variants and their integrin receptors in human secondary lymphoid tissue. Colocalization suggests that the ␣ 6 ␤ 4-integrin is a receptor for laminin-5 in lymphoid follicles. Cell Adhes. Commun. 4, 269 –279. 36. Kaldjian, E. P., Gretz, J. E., Anderson, A. O., Shi, Y., Shaw, S. (2001) Spatial and molecular organization of lymph node T cell cortex: a labyrinthine cavity bounded by an epithelium-like monolayer of fibroblastic reticular cells anchored to basement membrane-like extracellular matrix. Int. Immunol. 13, 1243–1253. 37. Geberhiwot, T., Assefa, D., Kortesmaa, J., Ingerpuu, S., Pedraza, C., Wondimu, Z., Charo, J., Kiessling, R., Virtanen, I., Tryggvason, K., Patarroyo, M. (2001) Laminin-8 (␣4␤1␥1) is synthesized by lymphoid cells, promotes lymphocyte migration and costimulates T cell proliferation. J. Cell Sci. 114, 423– 433. 38. Kouwenhoven, M., Ozenci, V., Gomes, A., Yarilin, D., Giedraitis, V., Press, R., Link, H. (2001) Multiple sclerosis: elevated expression of matrix metalloproteinases in blood monocytes. J. Autoimmun. 16, 463– 470. 39. Vainionpaa, N., Butzow, R., Hukkanen, M., Jackson, D. G., Pihlajaniemi, T., Sakai, L. Y., Virtanen, I. (2007) Basement membrane protein distribution in LYVE-1-immunoreactive lymphatic vessels of normal tissues and ovarian carcinomas. Cell Tissue Res. 328, 317–328. 40. Nigatu, A., Sime, W., Gorfu, G., Geberhiwot, T., Anduren, I., Ingerpuu, S., Doi, M., Tryggvason, K., Hjemdahl, P., Patarroyo, M. (2006) Megakaryocytic cells synthesize and platelets secrete ␣5-laminins, and the endothelial laminin isoform laminin 10 (␣5␤1␥1) strongly promotes adhesion but not activation of platelets. Thromb. Haemost. 95, 85–93. 41. Vizirianakis, I. S., Yao, C. C., Chen, Y., Ziober, B. L., Tsiftsoglou, A. S., Kramer, R. H. (2001) Transfection of MCF-7 carcinoma cells with human integrin ␣7 cDNA promotes adhesion to laminin. Arch. Biochem. Biophys. 385, 108 –116. 42. Kikkawa, Y., Sanzen, N., Sekiguchi, K. (1998) Isolation and characterization of laminin-10/11 secreted by human lung carcinoma cells. laminin10/11 mediates cell adhesion through integrin ␣3 ␤1. J. Biol. Chem. 273, 15854 –15859. 43. Pedraza, C., Geberhiwot, T., Ingerpuu, S., Assefa, D., Wondimu, Z., Kortesmaa, J., Tryggvason, K., Virtanen, I., Patarroyo, M. (2000) Monocytic cells synthesize, adhere to, and migrate on laminin-8 (␣ 4 ␤ 1 ␥ 1). J. Immunol. 165, 5831–5838. 44. Freemont, A. J., Jones, C. J. (1983) Light microscopic, histochemical and ultrastructural studies of human lymph node paracortical venules. J. Anat. 136, 349 –362. 45. Freemont, A. J., Stoddart, R. W., Steven, F., Jones, C. J., Matthews, S. (1986) The structure of the basement membrane of human lymph node high endothelial venules: an ultrastructural, histochemical and immunocytochemical study. Histochem. J. 18, 421– 428.

12

Journal of Leukocyte Biology Volume 84, September 2008

46. Sunami-Kataoka, Y., Akagi, H., Nishizaki, K., Taguchi, T., Murakami, T., Ohtsuka, A. (2001) Chondroitin sulfate proteoglycan at the basal lamina beneath high endothelial cells in human palatine tonsils: a light and electron microscopic study using the cationic colloidal iron method. Arch. Histol. Cytol. 64, 535–543. 47. Sorokin, L. M., Pausch, F., Frieser, M., Kroger, S., Ohage, E., Deutzmann, R. (1997) Developmental regulation of the laminin ␣5 chain suggests a role in epithelial and endothelial cell maturation. Dev. Biol. 189, 285– 300. 48. Drumea-Mirancea, M., Wessels, J. T., Muller, C. A., Essl, M., Eble, J. A., Tolosa, E., Koch, M., Reinhardt, D. P., Sixt, M., Sorokin, L., Stierhof, Y. D., Schwarz, H., Klein, G. (2006) Characterization of a conduit system containing laminin-5 in the human thymus: a potential transport system for small molecules. J. Cell Sci. 119, 1396 –1405. 49. Muto, S., Vetvicka, V., Ross, G. D. (1993) CR3 (CD11b/CD18) expressed by cytotoxic T cells and natural killer cells is upregulated in a manner similar to neutrophil CR3 following stimulation with various activating agents. J. Clin. Immunol. 13, 175–184. 50. Gu, Y. C., Kortesmaa, J., Tryggvason, K., Persson, J., Ekblom, P., Jacobsen, S. E., Ekblom, M. (2003) Laminin isoform-specific promotion of adhesion and migration of human bone marrow progenitor cells. Blood 101, 877– 885. 51. Siler, U., Seiffert, M., Puch, S., Richards, A., Torok-Storb, B., Muller, C. A., Sorokin, L., Klein, G. (2000) Characterization and functional analysis of laminin isoforms in human bone marrow. Blood 96, 4194 – 4203. 52. Sixt, M., Engelhardt, B., Pausch, F., Hallmann, R., Wendler, O., Sorokin, L. M. (2001) Endothelial cell laminin isoforms, laminins 8 and 10, play decisive roles in T cell recruitment across the blood-brain barrier in experimental autoimmune encephalomyelitis. J. Cell Biol. 153, 933–946. 53. Nishiuchi, R., Takagi, J., Hayashi, M., Ido, H., Yagi, Y., Sanzen, N., Tsuji, T., Yamada, M., Sekiguchi, K. (2006) Ligand-binding specificities of laminin-binding integrins: a comprehensive survey of laminin-integrin interactions using recombinant ␣3␤1, ␣6␤1, ␣7␤1 and ␣6␤4 integrins. Matrix Biol. 25, 189 –197. 54. Kupiec-Weglinski, J. W., De Sousa, M. (1991) Lymphocyte traffic is modified in vivo by anti-laminin antibody. Immunology 72, 312–313. 55. Gimond, C., Baudoin, C., van der Neut, R., Kramer, D., Calafat, J., Sonnenberg, A. (1998) Cre-loxP-mediated inactivation of the ␣6A integrin splice variant in vivo: evidence for a specific functional role of ␣6A in lymphocyte migration but not in heart development. J. Cell Biol. 143, 253–266. 56. Kawataki, T., Yamane, T., Naganuma, H., Rousselle, P., Anduren, I., Tryggvason, K., Patarroyo, M. (2007) Laminin isoforms and their integrin receptors in glioma cell migration and invasiveness: evidence for a role of ␣5-laminin(s) and ␣3␤1 integrin. Exp. Cell Res. 313, 3819 –3831. 57. Miner, J. H., Patton, B. L., Lentz, S. I., Gilbert, D. J., Snider, W. D., Jenkins, N. A., Copeland, N. G., Sanes, J. R. (1997) The laminin ␣ chains: expression, developmental transitions, and chromosomal locations of ␣1-5, identification of heterotrimeric laminins 8-11, and cloning of a novel ␣3 isoform. J. Cell Biol. 137, 685–701. 58. Miner, J. H., Cunningham, J., Sanes, J. R. (1998) Roles for laminin in embryogenesis: exencephaly, syndactyly, and placentopathy in mice lacking the laminin ␣5 chain. J. Cell Biol. 143, 1713–1723. 59. van Horssen, J., Bo, L., Vos, C. M., Virtanen, I., de Vries, H. E. (2005) Basement membrane proteins in multiple sclerosis-associated inflammatory cuffs: potential role in influx and transport of leukocytes. J. Neuropathol. Exp. Neurol. 64, 722–729.

http://www.jleukbio.org