From www.bloodjournal.org by guest on November 20, 2014. For personal use only. 1996 87: 2513-2517 A critical role of VLA-4 in erythropoiesis in vivo K Hamamura, H Matsuda, Y Takeuchi, S Habu, H Yagita and K Okumura Updated information and services can be found at: http://www.bloodjournal.org/content/87/6/2513.full.html Articles on similar topics can be found in the following Blood collections Information about reproducing this article in parts or in its entirety may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests Information about ordering reprints may be found online at: http://www.bloodjournal.org/site/misc/rights.xhtml#reprints Information about subscriptions and ASH membership may be found online at: http://www.bloodjournal.org/site/subscriptions/index.xhtml Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036. Copyright 2011 by The American Society of Hematology; all rights reserved. From www.bloodjournal.org by guest on November 20, 2014. For personal use only. A Critical Role of VLA-4 in Erythropoiesis In Vivo By Keisuke Hamamura, Hironori Matsuda, Yumiko Takeuchi, Sonoko Habu, Hideo Yagita, and KO Okumura Hematopoiesis requires specific interactions with the microenvironments, and VLA-4 has been implicated in these interactions bared on in vitro studies. To study the role of VLA4 in hematopoiesis in vivo, we performed in utero treatment of mice with an anti-VLA-4 monoclonal antibody. Although all hematopoietic cells in fetal liver expressed VLA4, the treatment specifically induced anemia. lt had no effect on the development of nonerythroidlineage cells, ineluding lymphoids and myeloids. In the treated liver almost no erythroblast was detected, whereas the erythroid progenitors, which give riseto erythroid colonies in vitro, were present. These results indicate that VLA-4 playsa criticalrole in erythropoiesis, while it is not critical in lymphopoiesis in vivo. 0 7 9 9 6 by The American Societyof Hematology. D analyzed on FACScan (Becton Dickinson, Mountain View, CA). MoAbs against mouse cell surface markers included ACK-4 (antic-kit): RM4-5 (anti-CM) (Pharmingen, San Diego, CA). 53-6.7 (anti-CD8),I0 RM2-1 (anti-CD2),“ RA3-6B2 (anti-B220),12MlnO (anti-Mac-]),” RB6-8C5 (anti-Gr-1),I4 TER119,’’and PSD.3 (anti-VLA-4).6 The Lin cocktail of MoAbs against lineage markers contained RM4-5, 53-6.7, RM2-1, RA3-6B2, MlnO, RB6-8C5, and TERl19. Conjugation of MoAbs with FITC or biotin was performed by standard methods. Optimal concentrations of labeled antibodies were determined by preliminary experiments. Immunohistochemical analysis. Fetal livers were fixed in periodate-lysine-paraformaldehyde (PLP) containing 4% padormaldehyde at 4°C overnight, embedded in OCT (Miles, Elkhart, IN), and stored at -80°C. Frozen sections were reacted with 5pg/mL primary at 4°C overnight. After rinsing with antibody, PS/2.3 or WK-2:’ PBS, they were incubated with 2.5 &mL biotinylated antirat IgG (Vector, Burlingame, CA) for 60 minutes at room temperature, rinsed, and then reacted with avidin-biotin-peroxidase complex (Vector) for 60 minutes at room temperature. The staining was developed by incubating the specimens in a reaction buffer (0.01% H202, 0.3 mg/mL diaminobenzidine in 50 mmoVL Tris-HCI, pH 7.6) for 5 minutes at room temperature. The specimens were washed and then counterstained in 0.1% Mayer’s hematoxilin. In controls, primary antibodies were substituted with PBS. In utero treatment. Pregnant mice were administered 1 mg/d of PS/2.3 or M/K-2 intraperitoneally from day 7 of gestation, when hematopoiesis begins in the yolk sac of fetuses, until the day of analysis. Trans-placental delivery of PSL2.3 to the fetuses was checked by flow cytometric analysis for the presence of antibody on fetal liver MNC, as detected by FITC-conjugated antirat IgG (Caltag, San Francisco, CA). The delivery of MK-2 through the placenta was expected based on preliminary experiments in which isotype-matched anti-Pgp-l MoAb 1M7.8.1I6 was similarly detectable on fetal thymoctytes after administration to pregnant mice. In both of these experiments, normal rat IgG (Sigma, St Louis, MO) or saline were administered as controls. Histological study. The livers from control or PSR.3-treated neonates were fixedin phosphate-buffered 10% formalin (pH 6.8), EVELOPMENT OF hematopoietic cells depends on the interaction with hematopoietic microenvironments that consist of a variety of components, including stromal cells, cytokines, and extracellular matrix (ECM) protein^."^ Among various ECM, it has been demonstrated that hematopoietic progenitor cells adhered selectively to fibronectin (FN)but not to collagen, laminin, or proteoglycans, indicating a possibility that FN contributes to hematopoiesis! Currently known major FN receptors are VLA-4 and VLA5, which belong to the D l integrin subfamily of adhesion molecules. Integrins not only mediate cellular adhesion, but also transduce signals that regulate cellular response^.^ Recent studies using in vitro hematopoietic culture systems demonstrated the importance of VLA-4 in B and T lymphopoie~is.6.~ In these studies, specific antibodies or peptides interfering with VLA-4-mediated adhesion inhibited lymphoid colony formation in bone marrowculture and stromal cell-dependent thymocyte differentiation. In addition, we recently demonstrated that an anti-VLA-4 monoclonal antibody (MoAb) inhibited stromal cell-dependent erythropoiesis in vitro? These in vitro results raise the question of whether these observations are relevant to hematopoiesis in vivo. To address this issue, we carried out in vivo treatment with a MoAb against mouse VLA-4. A critical contribution of VLA-4 to erythropoiesis in vivo was found. MATERIALS AND METHODS Mice. Timed pregnant C57BU6 mice were purchased from Japan SLC Inc (Shizuoka, Japan). the day of observation of a vaginal plug was designated as day 0 of gestation. Monoclonal antibodies. The hybridoma cells producing MoAb against a4 subunit of VLA-4 (PS12.3) and that against mouse VCAM-I (“2) were kind gifts from Dr Kensuke Miyake (Department of Immunology, Saga Medical School, Japan). The MoAbs were purified from ascites by affinity chromatography on protein GSepharose column (Pharmacia, Uppsala, Sweden). Preparation of cells. Thymocytes were released by pressing thymic lobi between two frosted slide glasses, passed through nylon mesh, and suspended in a-modified Eagles medium (a-MEM) (GIBCO, Gaithersburg, MD). Livers were mashed and suspended in a “ , passed through nylon mesh, and layered on lymphocyteseparating medium (JIMRO, Gumma, Japan) followed by a centrifugation at 1,200g for 10 minutes at room temperature. Mononuclear cells (MNC) at the interface were collected, washed, and resuspended in a “ E M . Flow cytometric analysis. Cells, lo6per sample, were incubated at 4°C for 20 minutes with appropriate dilutions of FJTC-, PE- or biotin-labeled MoAbs and washed twice with phosphate buffered saline (PBS). When the cells were reacted with biotinylated antibody, they were M e r incubated with Red613-streptoavidin at 4°C for 15 minutes, and washed twice with PBS. Immunofluorescence was Blood, Vol 87, No 6 (March 15), 1996: pp 2513-2517 From the SecondDepartment of Intern1 Medicine, Tokyo University; the Department of Immunology. Juntendo University School of Medicine. Tokyo: and the Department of Immunology, Tokai University School of Medicine, Isehara-shi, Japan. Submitted April 10, 1995; accepted October 16, 1995. Address reprint requests to KO Okwnura, MD, Department of Immunology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113, Japan. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact. 0 19% by The American Society of Hematology. 0006-4971A6/8706-0025$3.at/o 2513 From www.bloodjournal.org by guest on November 20, 2014. For personal use only. 2514 HAMAMURA ET AL Fig 1. Expmssion of VIA4 and VCAM-1 in fetal (A) Exp d o n of VIA4 on fetal liver MNC. Embryonic day 15 fetuses lii. wafedhectedfromtimedpregnant mice, and mononuclear calk were isdated. They were stained by FlTccOnjugated PS/ 2.3, PEconjugated ACK45, and the cocktail of biotinylatedMoAlw against lineage markers (Lin). folowed by streptoavidinRedS13. and analyzed on FACSmn. , autofluomseeme; *--, PS123 (anti-VIA41 staining of total liver MNC; , PS12.3 staining of the elecbonially gated c-kit+/Lin-cells. Ver- ----- ticalscalerepresentsrdativodl number and horizontalscale represents log Ruores#~lce. (B and C) I m m u n o h i iical staining of embryonic day 15 fetal liver with PS123 oc WK-2. E m bryonic day 15 fetal livers were ~ e in d perkhe-lysinsparaformaldehyde (PLP) containing 4% paraformaldehyde. Frozen sectiomweremactedwithaprimary antibady, PS12.3 or M/K-2, followed by biotinylated antirat I g G and avidin-biotin-peroxidase complex. The staining was & velOpea by diaminobenzidine, and the n was done with Mayer's hematoxilin. Results the stainingwith pS12.3 (B) and WK-2 (C). - Fig 3. Histological study of newborn liver treated in utero with PS/2.3. The livers from control (A) or PS/Z.3-treated (B)neonates were f w d in ph-hate buffered 10% formaline (pH6.8).embeddedin paraffin, and the t i w e ~e&iom were stainedwith 0.1% Mayer's hamoxilin and 0.025% eosin. Arrows indicateerythroblasts, and arrow heads indicate myeloid cells. From www.bloodjournal.org by guest on November 20, 2014. For personal use only. 2515 ROLE OF VIA-4 IN ERYTHROPOIESIS embedded in paraffin, and the tissue sections were stained with 0.1% Mayer's hematoxilin and 0.025% eosin. Colony u s s q . Semi-solid methyl cellulose culture of hematopoietic cells was performed accordingto the previously described technique" with some modifications. 'Ihe livers were isolatedfrom control or PSf2.3-treated neonates and were mashed by pressing them betweenfrostedslideglasses. By this manipulationhepatocytes coaggregated, and dissociated hematopoietic cells became a single cell suspension. These cells (2 X 104 to I X 10s) were cultured in 1 mL of Iscove's modified Dulbecco's medium (IMDM, Kyokuto, Tokyo, Japan) containing 1.0% methyl cellulose (Dow Chemical, Midland, MI), 30% fetal calfserum (JRHBiosciences, Lenexa, KS), 1% deionized fractionV bovine serum albumin (BSA; Sigma), and 0.1 mmoVL 2-mercaptoethanol in the presence of 6 U/mL human recombinanterythropoietin(Chugai, Tokyo, Japan),100 U/mL mouse recombinant IL-3, and50 ng/mL human-recombinant granulocyte colony-stimulating factor (G-CSF; Chugai). The numbers of colony forming unit-erythroid (CFU-E) were scored as cell aggreLog fluorescence gates of more than eight cells at day 3 of culture. Burst forming fig 2. Tmnsphmtal delivory of PS12.3 to Tho pmsenca uniterythroid (BFU-E), colony-forming units granulocyte, macroof PSR.3 on f.M livu MNC wum analyzad by staining with FITG phage, and granulocyte macrophage (CFU-G, CFU-M, CFU-GM), conjug.hd.ntk.t100antibody.""-, ~ ~ : " " - , ~ n t k . t and CFU-Mix were differentially counted at7 to days 8 as aggregates antiof more than40cells (200 cellsfor BFU-E) under microscopy. Some IgG staining of total liin MNCfrom contrd individuab; -, rat IgG staining of total liin MNC from the PSI2.3-treated brdividucolonies were lifted from the cultures using micropipet, cytocentriab. Vwticd #d. d.lrtivcall . number and horizontaltub fuged, and identified by Diff-Quick (Midori-Juji, Osaka, Japan) or mprasont!s log. R benzidine staining. fetuses. RESULTS AND DISCUSSION We first evaluated the VLA-4 expression in fetal liver, which is the main site of early hematopoiesis. Three-color flow cytometric analysis of fetal liver MNC showed that almost all cells strongly expressed VLA-4 (Fig lA), which implies that VLA-4 can participate in the development of all lineages. The c-kit+/Lin- cells, which are enriched for stem cells: expressed VLA-4 most strongly, suggesting the importance of VLA-4 for these cells. Immunohistochemically, VLA-4 resided on small round cells with a large nucleudcytoplasm ratio and dense chromatin; these are morphological characteristics of hematopoietic cells (Fig IB). By contrast, VCA"1, VLA-4 ligand other than FN, resided on the cells with irregular shape-extending cytoplasmic projections to surrounding hematopoietic cells, which are morphological characteristics of stromal cells (Fig 1C). This implied the possibility that not only FN but also VCAM-l could act as a VLA-4 ligand in the hematopoietic microenvironment, as suggested by the inhibitory effect of an antiVCA"1 MoAb on stromal cell-dependent lymphohematopoiesis in vitro!.'* In vivo treatment was then performed with an a n t i - U 4 MoAb, PSL2.3, which has been demonstrated to inhibit VLA-4-mediated cell binding to both FN andVCAM-1 and also toefficiently inhibit lymphohematopoiesisin vitro? Fetuses were treated with P92.3 from gestational day 7 to birth by administering the antibody into maternal peritonea. The injected antibody was efficiently transferred to the fetuses as estimated by its deposition on fetal liver MNC. Figure 2 shows that rat IgG was present on MNCs from the PSL2.3-treated fetal livers, but it was absent in controls. Because most liver MNCs from the PS/2.3-treated individuals were positive for rat IgG, we could conclude that the amount of transplacentally delivered PSl2.3 was enough to block VLA-4 molecules on fetal liver MNC. The neonates that had beentreated in utero with PSL2.3 looked pale, though they did not show any other morphological abnormality (not shown). In these neonates, a marked anemia was defined by red blood cell counts that were approximately one-fifth of the controls (Table 1). Since the liver is the major site of hematopoiesis in fetuses and neonates,I9phenotypical analysis was carried out with liver-derived MNC. The number of TER119+ cells representing erythroid ~rogenitors'~ was strikingly decreased. In contrast, the number of white cells in the blood and those of Gr-l+ or Mac-l+ myeloid cells in the liver were not significantly decreased.CD4'8', CD4+8-, or CD4-8+ cells in the thymus and B220+ cells in the liver were also not significantly affected. These results indicate that the fetal anti-VLA-4 treatment interfered onlywith erythropoiesis; it did not affect the development of other lineages including granuloid, monocytoid, and lymphoid cells. h addition to the erythroid cells, a significant decrease was also noted for c-kit'&% cells in the liver and CM-8cells in the thymus. Since these cells are the most immature cells within each organ, VLA-4 may also contribute to stem cell homing from the yolk sac to the liver and homing of T progenitor cells to the thymus. We then histologically examined the liver, thymus, and spleen of the PSL2.3-treated neonates as well as major nonhematopoietic organs including the heart,lung, kidney, and intestine. In the control liver (Fig 3A), manyclusters of small round cells with condensed nucleus and scant cytoplasm, characteristic of erythroblasts (arrow), were seen.In contrast, these cells were almost completely absent in the PSD.3treated liver, while other cells of myeloid morphology (arrow head) were maintained (Fig 3B). The PSL2.3-treated thymus From www.bloodjournal.org by guest on November 20, 2014. For personal use only. HAMAMURA ET AL 2516 Table 1. Effect of In Utero Treatment W* Hematopoiesis PW2.3 on Cell Numbers Source of Cells ~ Blood' (x105cells/pL) Liver MNC (xlOScells/organ) Thymocyte (x105cells/organ) Control PS/Z.B-treated ____~ Red cell Whitecell 55.9 -t 7.2 11.0 0.125 t 0.0980.119 TER119' 14.3 Mac-l+ Gr-l+ B220+ 17.6 c-kit'llin- t 2.04 0.79 7.96 t 1.09 8.23 6.98 t 7.92 0.80 t 12.5 0.07 0.378 1.14 t 0.08 CD4-8CD4'8' CD4'8CD4-8' 15.8 25.5 t 11.1 85.9 t 22.3 2.00 t 0.72 10.2 ? 5.9 IT 4.1 2 0.036 t 0.23 t 1.17 5 1.06 ? 1.44 2 0.087 5 1.3 91.0 5 21.0 1.73 2 0.34 9.41 2 5.51 Neonates that hadbeen treated in utero withPS/2.3 were analyzed on the day of birth. Two microliters of the blood was aspirated from the right atrium, diluted appropriately with PBS or TWk's solution, and the numbers of red and white blood cells were counted using Neubauer's hemocytometer. Thymocytes and splenocytes were released from each organ by pressing between two frosted slide glasses; they were suspended in a-MEM. Thymocytes, splenocytes, and liver MNC were stained with labeled antibodies against the indicated cell surface markers and analyzed on FACScan. Data indicate means c SD of threeor four (*) different individuals. and spleen did not show any histological abnormality nor any other nonhematopoietic organs (not shown). Next, we evaluated the colony forming capacity of the PW2.3-treated liver to determine the stage at which the erythropoiesis was affected by the anti-VLA-4 treatment. Table 2 represents the mean colony numbers per control or treated neonatal liver. Erythroid colonies (BFU-E and CFU-E) from the PSR.3-treated liver were reduced to 40% of the normal liver, but this reduction was much less prominent than the almost complete absence of erythroblasts in the treated liver (Fig 3). Although the reduction indicates that erythropoiesis was arrested at the erythroid precursor stage by the antiVLA-4 treatment, erythroid progenitor cells were still present in the treated liver. A similar extent of reduction in the colony forming capacity was also observed with other lineages, including CFU-Mix, which represents multipotential stem cells. This suggests that VLA-4 is also involved in the migration of stem and progenitor cells of all lineages. This notion is consistent with the recent observations by other investigators that VLA-4 appeared to be involved in the lodging of CFU in murine spleens2' and that peripheralization of multiple hematopoietic progenitors was induced by anti-VLA-4 antibody administration in primates?' However, the normalized development of myeloid and lymphoid cells after the anti-VLA-4 treatment (Table 1) suggests that some other VLA-4-independent pathways can compensate for the lymphomyelopoiesis but not the erythropoiesis. In order to determine whether the inhibitory effect of antiVLA-4 on erythropoiesis was exerted by its interference withVLA-4/FN or VLA-4ffCAM-l interaction, we per- formed the in utero treatment with an anti-VCAM- I MoAb (hUK-2), which has been demonstrated to inhibit in vitro lymphopoiesis as efficiently as PS/2.3,6 according tothe same protocol as PW2.3. The W - 2 treatment did not cause anemia (red blood cells, 5.31 ? 0.62 and 5.59 -t 0.72 X 10' celVpL,mean -t SD, in the treated or control neonates, respectively), but leukocytosis in the blood was notable (white blood cells, 2.39 2 I .31 and 1.25 2 0.99 X lo4 cells/pL, means +- SD, in the treated or control neonates, respectively). This effect was quite different from that of PS/2.3 (Table l), which suggests that VCAM-I does not play a critical role as the VLA-4 ligand for supporting erythropoiesis in vivo, although it is abundantly expressed in the hematopoietic microenvironment (Fig 1C). VCAM- I may, rather, contribute to the lodging as indicated by leukocytosis after the anti-VCAM-1 treatment. Although the presence of other VLA-4 ligands has not been excluded: FN would act as the critical VLA-4 ligand for supporting erythropoiesis in vivo, as previously suggested by the FN requirement for in vitro differentiation of murine erythroleukemia cells into reticulocytes, which was associated with erythrocyte-specific protein induction.** It has been demonstrated that VLA-4-mediated signaling led to cytokine gene expression and autocrine growth in mature T lymphocytes Interestand ingly, VLA-4 can transmit a unique signal distinct from that of other members of the I integrin s ~ b f a r n i l y It. ~ ~ ~ ~ ~ remains to be determinedhow VLA-4-mediated adhesion and signaling participate in the erythropoiesis in collaboration with c-kitlstem cell factor and erythropoietin receptoderythropoietinsystems, which arealso required for erythropoiesis. Table 2. Colony Forming Capacity of Neonatal Liver Treated In Utero W& PW2.3 Colony Number Colony BFU-E CFU-E CFU-GM CFU-G 33.1 CFU-M CFU-Mix 3.54 (XlO' per organ1 Control PSi2.3-Treated 9.69 t 3.94 276 t 38 6.59 -t 4.73 6.66 -C 4.32 29.0? 7.17 0.990 t 1.41 5 1.48 3.88 (40.0) 113 (40.9) t 33 t 2.05 3.68 (55.8) t 1.60 4.19 (62.9) t 4.0 t(28.0) 0.018 (YoControl) (87.6) Semisolid methyl cellulose cultures, of the hematopoietic cells derived from new born livers from control or in utero PM.3-treated individuals were obtained. Livers were mashed by pressing between frosted slide glasses, and dissociated hematopoietic cells (2 x 10' to 1 x 10s) were cultured in 1 mL of IMDM containing 1.0% methyl cellulose, 30% fetal calf serum, 1% deionized fraction V BSA. and 0.1 mmolR 2-mercaptoethanol in the presence of 6 U/mL human recombinant erythropoietin, 100 U/mL mouse recombinant 11-3. and 50 ng/ mL human-recombinant G-CSF. The numbers of CFU-E were scored as cell aggregates of more than eightcells at day 3 of culture. BFUE, CFU-G, CFU-M, CFU-GM, and CFU-Mix were differentially counted at days 7 to 8 as aggregates of more than40 cells (200 cells for BFUE) under microscopy. Data indicate means t SD of three different individuals. From www.bloodjournal.org by guest on November 20, 2014. For personal use only. ROLE OF VIA-4 IN ERYTHROPOIESIS 2517 Hoffman PM, HartleyJ W , Morse HC: Analysisof neoplasms identified by Cas-Br-M MuLV tumor extracts. J Immunol 137:679, 1986 15. &uta K,Kina T, MacNeil I, Uchida N. Peault B, Chien YH, Weissman IL: A developmental switch in thymic lymphocyte maturation potential occursat the level of hematopoietic stem cells. Cell 62:863,1990 16. Pearse M, Wu L, Egerton M, Wilson A, ShortmanK,Scollay R: A murine early thymocyte development sequence is marked by REFERENCES transient expression of the interleukin 2 receptor. Proc Natl Acad 1. Tavassoli Minguell JJ: Homingof hemopoietic progenitor Sci USA 861614, 1989 cells to the marrow. Proc Soc Exp Biol Med 1%367, 1991 17. Shimamura K, Kaminsky SG, Nakamura I: Acute rejection 2. Zuckermann KS, Wicha MS: Extracellular matrix production of allogeneic hemopoietic progenitorsby genetically resistant mice. by the adherent cells of long-term murine bone marrow cultures. Eur J Immunol 19:1165, 1989 Blood 61540, 1983 18. Miyake K, Medina K, Ishihara K, Kimoto M, Aurerbach R, 3. Campbell AD, Wicha MS: Extracellular matrix promotes the Kincade P W A VCAM-likeadhesionmolecule on murinebone growth and differentiationof murine hematopoietic cells in vitro. J marrow stromal cells mediates bindingof lymphocyte precursors in Clin Invest 75:2085, 1985 culture. J Cell Biol 114557, 1991 4. GiancottiFG,Comoglio PM,TaroneG:Fibronectin-plasma 19. Metcalf D, Moor MAS: Embryonic aspects of hemopoiesis, membrane interaction in the adhesion of hemopoietic cells. J Cell in Frontiers of Biology, ch 4. North-Holland Publishing, 1971 Biol103:429,1986 20. Williams DA, Rios M, StephensC , Patel VP: Fibronectin and 5. Hynes R 0 Integrins: Versatility, modulation, and signaling in VLA4 in hematopoietic stem cell-microenvironment interactions. cell adhesion. Cell 69:l I, 1992 Nature 352:438, 1991 6. Miyake K, Weissman IL, Greenberger JS, Kincade P W Evi21. Papayannopoulou T, Nakamoto B: Peripheralizationof hemodence for a role of the integrin VLA-4 in lympho-hemopoiesis. J poietic progenitors in primates treated with anti-VLA-4 integrin. Exp Med 173599, 1991 Proc Natl Acad Sci W9374, 1993 7. Utsumi K,SawadaM, Nmmiya S , NagamineJ,Sakata T, 22.PatelVP,Lodish HF: Afibronectinmatrix is required for Iwata S , Kita Y, TeraokaH,HiranoH,Ogata M,HamaokaT, differentiation of murine erythroleukemia cells into reticulocytes. J Fujiwara H: Adhesion of immature thymocytes to thymic stromal CellBiol105:3105,1987 cells through fibronectin molecules and its significance for the induc- 23. Nojima Y, Humphries W ,Mould A P , Komoriya A, Yamada tion of thymocyte differentiation.Proc Natl Acad Sci USA 885685, KM, Schlossman SF, Morimoto C: VLA-4 mediates CD3-dependent 1991 CD4+ T cell activation via theCS 1 alternatively spliced domainof 8. Yanai N, Sekine C, Yagita H, Obinata M: Roles for integrin fibronectin. J Exp Med 172:1185, 1990 VLA-4 in stroma-dependent erythropoiesis. Blood 83:2844, 1994 24.DamleNK, Aruffo A:Vascularcelladhesionmolecule 1 9. Okada S , Nakauchi H,Nagayoshi K,Nishikawa S , Nishikawa induces T-cell antigen receptordependentactivationof CD4' T lymS , MiuraY,Suda T: Enrichmentandcharacterization ofmurine phocytes. Proc Natl Acad Sci USA 88:6403, 1991 hematopoietic cells that express c-kit molecule. Blood 78:1706,1991 25. Anwar A R F , Moqbel R, Walsh GM, Kay AB, Wardlaw AJ: 10. Ledbetter JA, Herzenberg LA: Xenogenic monoclonal antiAdhesion to fibronectin prolongs eosinophil survival. J Exp Med bodies to mouse lymphoid differentiation antigens. Immunol Rev 177:839,1993 47:63,1979 26. Chan BM, Kassner PD. Schiro JA, Byers HR, Kupper TS, 11. Yagita H,Nakamura T, Karasuyama H, Okumura K MonoHemler M E Distinct cellular functions mediatedby different VLA clonal antibodies specific for murine CD2 reveal its presence on B integrin alpha subunit cytoplasmic domains. Cell 68:1051, 1992 as well as T cells. Proc Natl Acad Sci USA 86645, 1989 27. Nojima Y, Rothstein DM, Sugita K, Schlossman SF, Mori12. Coffman RL, Weissman IL: B220: A B cell-specific member mot0 C: Ligationof VLA-4 on T cells stimulates tyrosine phosphoryof the T200 glycoprotein family. Nature 289681. 1981 lation of a 105-kD protein. J Exp Med 175:1045, 1992 13. Springer T, Galfre G , Secher DS, Milstein C:Mac-l: A mac28.FreedmanAS,RhynhartK,Nojima Y, Svahn J, Elise0 L, rophage differentiation antigen identified by monoclonal antibody. Benjamin CD, Morimoto C, Vivier E Stimulation of protein tyrosine Eur J Immunol 9:301, 1979 phosphorylation in human Bcells after ligationof the beta 1 integrin 14.Holmes KL, LangdonWY,Fredrickson TN, Coffman RL, VLA-4. J Immunol 1501645, 1993 ACKNOWLEDGMENT We thank Y. Kojima and Y. Murata for technical assistance on immunohistochemistry,Drs K. Shimamura and T. Suda for advising techniques on colonyassay, Dr K. Miyake for generous gifts of the PSR and M/K-2 hybridoma cells, and Dr Y. Uchida for generous support of the work. MT.
© Copyright 2018