Adhesion and migration of mouse fetal liver (FL) cells to the thymus were investigated using cells from green fluorescent protein transgenic (GFP+) mice. FL cells from GFP+ embryos at 12 gestational days (E12) of mice were incubated with 2′-deoxyguanosine-treated fetal thymus lobe (from E14) by thymic repopulation (hanging drop) culture methods. GFP+ cells were observed in the thymus lobe at the end of the repopulation culture period. A large part of the infiltrated cells expressed CD44 until day 2 of culture on a permeable membrane, then lost the expression. CD25 expression was observed from day 1 to day 4. Around day 8, GFP+ cells became both CD4+ and CD8+. The results support the early observation of the sequential expression of CD44, CD25, and CD4/8 during the early stages of thymocyte development. When anti-CD44 mAb was added at the beginning of the repopulation culture period, GFP+ FL cells adhered to the surface of the thymus lobe but did not migrate into the thymus. Pretreatment of the thymus with hyaluronidase or hyaluronate produced results similar to the results of anti-CD44 treatment. On the other hand, the addition of anti-integrin α4 mAb inhibited adhesion to the thymus, and almost no GFP+ cells were seen on the surface of the thymus lobe. The data suggest that integrin α4 and CD44 play different roles, i.e., integrin α4 is required for the adhesion of FL cells to the thymus lobe and CD44 is required for the migration of the cells into the thymus.

The interaction of leukocytes with the endothelium has received considerable attention during the last decade (1, 2, 3, 4). Lymphocyte adhesion to the endothelium and migration into inflammatory sites are controlled by several adhesion molecules including selectins, integrins, and CD44, and their respective ligands. Chemokines are also known to be mediators of lymphocyte recruitment and infiltration during inflammation. Chemokines act not only as chemotactic factors for lymphocyte migration, but also as activation molecules for the recruited lymphocytes (5).

Most T cells develop in the thymus. T cell precursors come from the fetal liver (FL)3 during the embryonic stage and from bone marrow in adults. In the fetal thymus, prothymocytes enter the nonvascularized thymic rudiment by the capsule. It is also believed that several adhesion molecules participate in the adhesion and migration to the fetal thymus (6, 7). During T lymphocyte development, various types of cellular interaction between stroma and thymic lymphocytes play crucial roles (8, 9, 10). In order to study the role of cell adhesion molecules, thymic repopulation experiments have been conducted using FL cells and 2′-deoxyguanosine (dGuo)-treated fetal thymus lobe (11). However, because regular thymic repopulation assays require a certain amount of intrathymic cells for flow cytometric study, it is difficult to analyze the earliest stages of thymocyte development, such as the adhesion and migration of FL cells to the thymus lobe.

Ikawa et al. (12) established a line of transgenic mice carrying the green fluorescent protein (GFP) gene of Aequorea victoria. Adult GFP-transgenic (GFP+) mice express GFP in most tissues, and it is thought they can provide a powerful tool for tracing cellular movements. In order to investigate the roles of various adhesion molecules during the early phases of thymocyte development, thymic repopulation assays using FL cells from GFP+ mice added to normal embryonic thymus were performed. Tracing the GFP-tagged cells by confocal laser scanning microscopy enabled us to detect the early adhesion and migration of FL cells to the thymus. The present paper describes integrin α4 as a key molecule in adhesion, and CD44 in migration into the thymus.

Specific pathogen-free C57BL/6 mice aged 6 to 8 wk were purchased from Charles River Japan (Tokyo, Japan). Heterozygous GFP+ mice of C57BL/6 background (12) were maintained in our animal facility by mating with normal C57BL/6 mice. The mice were mated at night, and females were examined the next morning. The day on which a vaginal plug was found was considered as day 0 of embryonic development.

Rat mAbs to mouse CD44 (KM201) (13), c-kit (ACK2) (14), and integrin α4 subunit (TAS-5 and CAS-9) (15) were gifts from Dr. K. Miyake (Saga Medical School, Saga, Japan), Dr. S.-I. Nishikawa (Kyoto University, Kyoto, Japan), and Dr. T. Kina (Kyoto University), respectively. Hamster mAbs to mouse integrin α1 (HMα1), α2 (HMα2), α5 (HMα5-1), and α6 (HMα6) (16, 17) were gifts from Dr. H. Yagita (Juntendo University School of Medicine, Tokyo, Japan). PE-conjugated mAbs, such as anti-CD4 (RM4-5), anti-CD8 (53-6.7), anti-CD25 (PC61), and anti-CD44 (IM7) were purchased from PharMingen (San Diego, CA). Anti-CD24 (J11d) and mouse anti-rat κ-chain (MAR18.5) mAbs (from American Type Culture Collection, Manassas, VA) were purified from hybridoma culture supernatants. Various mAbs were biotinylated using biotinyl N-hydroxysuccinimide ester (E-Y Labs, San Mateo, CA), and MAR18.5 was labeled with FITC in our laboratory. R-PE-Streptavidin, PE-Cy5-Streptavidin, and FITC-conjugated goat anti-hamster IgG were purchased from Molecular Probes (Eugene, OR), Cedarlane (Westbury, NY), and Dako (Tokyo, Japan), respectively.

Bovine hyaluronidase (HAase) (320 U/mg) and human hyaluronate (HA) were purchased from Sigma (St. Louis, MO). dGuo was purchased from Wako Pure Chemicals (Osaka, Japan).

Recolonization of FL cells into dGuo-treated thymus rudiments was performed according to the methods of Kingston et al. (11). RPMI 1640 medium (Flow Laboratories, Irvine, U.K.) containing 10% FCS (Boehringer Mannheim, Australia) with kanamycin (100 mg/L, Meiji Pharmaceutical, Tokyo, Japan) was used as the culture medium. Gestational day 14 (E14) fetal thymi from C57BL/6 mice were cultured with dGuo (final concentration was 1.35 mM) for 5 days at 37°C in 5% CO2 in air. The thymus lobes were transferred to dishes containing medium for 2 h to diffuse the dGuo away from the rudiments.

E12 fetuses were obtained from pregnant GFP+ heterozygous mice time-mated with normal C57BL/6 male mice. GFP+ heterozygous fetuses were identified by brief exposure to UV light of 365 nm as described (12). Single cell suspensions of E12 FL cells from GFP+ embryos were prepared and cocultured with dGuo-treated thymi under hanging drop conditions using Terasaki plates (Sumitomo Bakelite, Tokyo, Japan) for 2 days with or without various mAbs (final concentrations 5–10 μg/ml). Thymus lobes were then transferred to Nuclepore membrane filters (Nuclepore-Corning, Pleasanton, CA), and fetal thymus organ culture was performed for several days at 37°C in 5% CO2 in air. In the present experiments, day 0 indicates the time at which the reconstituted thymus lobes were transferred to the membrane filters.

Cultured thymus lobes were fixed with 4% paraformaldehyde, and cryostat sections were examined for GFP+ cells under a confocal laser scanning microscope (model MRC1024ES, Bio-Rad Laboratories, Hercules, CA). In some experiments, sections were stained with various PE-labeled or biotinylated mAbs (plus R-PE-Streptavidin) and examined under a confocal laser scanning microscope. Excitation/emission wavelengths for GFP and PE were 488 nm/522 nm and 488 nm/585 nm, respectively.

GFP fluorescence was measured at the same excitation/emission wavelength as FITC. Flow cytometric profiles were analyzed with a FACSCalibur analyzer and CellQuest software (Becton Dickinson Immunocytometry Systems, Mountain View, CA).

In order to investigate the role of adhesion molecules in the adhesion and migration of FL cells to the thymus, the expression of integrins and other molecules on normal FL cells was first examined. E12 FL cells from normal C57BL/6 embryos were stained with various mAbs. Among integrin α-chains, α4 was found to be expressed strongly on FL cells (Fig. 1,C), with α5 and α6 showing intermediate levels of expression (Fig. 1, D and E), but no α1 or α2 expression was observed (Fig. 1, A and B). CD44 was also stained strongly on FL cells (Fig. 1,F). c-kit is known to be expressed on hemopoietic cells in FL and adult bone marrow. We then examined the expression of c-kit, CD44, and integrin α4 by three-color flow cytometry. Fluorescence gating was set to include c-kit-positive fractions (Fig. 1,G) and the expressions of CD44 and α4 were examined. As shown in Fig. 1 H, c-kit-positive FL cells express both CD44 and α4.

FIGURE 1.

Expression of various adhesion molecules on FL cells from normal C57BL/6 mice. A–F, E12 FL cells from normal C57BL/6 embryos were stained with mAbs to mouse integrin α1 (HMα1) (A), α2 (HMα2) (B), α4 (TAS-5) (C), α5 (HMα5-1) (D), α6 (HMα6) (E), and CD44 (KM201) (F), together with either FITC-conjugated goat anti-hamster IgG (A, B, D, and E) or FITC-conjugated MAR18.5 (C and F). The viable cell fraction was gated and single-color flow cytometric patterns are shown. G and H, Fluorescence gating was set to include the c-kit-positive (FITC-anti-c-kit) E12 FL cell fraction (G), and three-color flow cytometry was performed by staining with biotynyl anti-α4 (plus PE-Cy5-Streptavidin) and PE-anti-CD44 (IM7) (H). Thin lines indicate background stainings (second Ab alone).

FIGURE 1.

Expression of various adhesion molecules on FL cells from normal C57BL/6 mice. A–F, E12 FL cells from normal C57BL/6 embryos were stained with mAbs to mouse integrin α1 (HMα1) (A), α2 (HMα2) (B), α4 (TAS-5) (C), α5 (HMα5-1) (D), α6 (HMα6) (E), and CD44 (KM201) (F), together with either FITC-conjugated goat anti-hamster IgG (A, B, D, and E) or FITC-conjugated MAR18.5 (C and F). The viable cell fraction was gated and single-color flow cytometric patterns are shown. G and H, Fluorescence gating was set to include the c-kit-positive (FITC-anti-c-kit) E12 FL cell fraction (G), and three-color flow cytometry was performed by staining with biotynyl anti-α4 (plus PE-Cy5-Streptavidin) and PE-anti-CD44 (IM7) (H). Thin lines indicate background stainings (second Ab alone).

Close modal

Although the data are not shown, we examined the cellularity and immunological competence of T cells from GFP+ heterozygous mice in preliminary studies. The proliferative responses of T cells from GFP+ mice to either allogeneic stimulator cells or anti-CD3ε stimulation were equivalent to those of T cells from wild-type C57BL/6 mice. The Ab response of GFP+ mice to OVA, and their thymic and splenic lymphocyte numbers also did not differ from wild-type mice. When the thymocyte and splenic T or B lymphocyte fractions were stained with PE-conjugated anti-CD4, anti-CD8, or anti-B220, the results were also found to be equivalent to wild-type mice (our manuscript in preparation).

E12 FL cells from GFP+ mice were stained with PE-anti-CD44, and biotinyl-anti-α4 plus PE-Cy5-Streptavidin, and three-color flow cytometry was performed. As shown in Fig. 2,B, while only 31.9% of FL cells express GFP (Fig. 2,A), GFP+ cells express both α4 and CD44. Similarly, E12 FL cells from GFP+ mice were stained with PE-anti-CD44 together with biotinyl-anti-c-kit (Fig. 2, C and D) or biotinyl-anti-α4 (Fig. 2, E and F) (plus PE-Cy5-Streptavidin), and three-color flow cytometry patterns were examined. As shown in Fig. 2,D, c-kit+/GFPint-low (R2, 12.1%), and c-kit+/GFPhigh (R3, 13.0%) were both stained by anti-CD44. The α4+/GFPhigh (Fig. 2,E, R5, 27.7%) fraction and a large part of the α4+/GFPint-low (Fig. 2,E, R4, 17.6%) fraction were stained by anti-CD44 (Fig. 2 F, thick line). These results suggest that GFP+ FL cells simultaneously express c-kit, CD44, and integrin α4 on their surface.

FIGURE 2.

Expression of c-kit, integrin α4, and CD44 on FL cells from GFP+ mice. A and B, The GFP+ fraction of E12 FL cells (A) (R1, 31.9%) was stained with PE-anti-CD44 (IM7) and biotinyl anti-α4 (TAS-5) (plus PE-Cy5-Streptavidin) (B). C–F, Fluorescence gatings were set to include the c-kit+ (biotinyl anti-c-kit plus PE-Cy5-Streptavidin) and GFPint or high fractions (C and D), or the α4+ (biotinyl anti-α4, TAS-5, plus PE-Cy5-Streptavidin) and GFPint or high fractions (E and F), and their CD44 (PE-anti-CD44) (IM7) expressions were examined: R2 (12.1%) (thin line) and R3 (13.0%) (thick line) (D), or R4 (17.6%) (thin line) and R5 (27.7%) (thick line) (E). Cells from day 1 thymi were collected, and CD44+ cells of GFP+ fraction were analyzed by flow cytometry (G).

FIGURE 2.

Expression of c-kit, integrin α4, and CD44 on FL cells from GFP+ mice. A and B, The GFP+ fraction of E12 FL cells (A) (R1, 31.9%) was stained with PE-anti-CD44 (IM7) and biotinyl anti-α4 (TAS-5) (plus PE-Cy5-Streptavidin) (B). C–F, Fluorescence gatings were set to include the c-kit+ (biotinyl anti-c-kit plus PE-Cy5-Streptavidin) and GFPint or high fractions (C and D), or the α4+ (biotinyl anti-α4, TAS-5, plus PE-Cy5-Streptavidin) and GFPint or high fractions (E and F), and their CD44 (PE-anti-CD44) (IM7) expressions were examined: R2 (12.1%) (thin line) and R3 (13.0%) (thick line) (D), or R4 (17.6%) (thin line) and R5 (27.7%) (thick line) (E). Cells from day 1 thymi were collected, and CD44+ cells of GFP+ fraction were analyzed by flow cytometry (G).

Close modal

We then performed the thymic repopulation assay using FL cells from GFP+ embryos. From day 0 to 1 of culture on a membrane filter, GFP+ cells could be observed in the thymus lobe (Fig. 3, A and B). When the same sections were stained with anti-CD44, a large proportion of the GFP+ cells were found to be CD44+ (Fig. 3, E and F), but the fluorescence intensities of PE (anti-CD44 mAb) on days 0 and 1 (Fig. 3, E and F) appeared weaker than those of GFP on days 0 and 1 (Fig. 3, A and B). Recall that most of the E12 FL cells from GFP+ mice stain for CD44 by three-color flow cytometry (Fig. 2, C–F). Therefore, the apparent difference between GFP and CD44 expression in the confocal images implies that down-regulation of the CD44 Ag occurred on the FL cells during the hanging drop culture. This hypothesis was confirmed by flow cytometry, which showed that only 50–60% of GFP+ cells extracted from day 1 thymi still expressed detectable CD44 (Fig. 2,G). Furthermore, the confocal images indicate that CD44 expression by GFP+ cells decreased further by day 3 (Fig. 3,G) and finally disappeared by day 6 of culture (Fig. 3,H). Thus the expression of CD44 decreases rapidly in the thymus from the early stages of development. A small number of CD25+ cells were observed on day 0 (Fig. 3,I), and the number increased on days 1 and 2 (Fig. 3, J and K). The expression reached a peak from day 2 to day 4 (Fig. 3, K and L), and then decreased and disappeared by day 6 (data not shown). CD24+ cells could be observed from day 0 to day 1; thereafter the number decreased slightly and then remained unchanged until day 8 (Fig. 3, M–P). Neither CD4 nor CD8 could be detected on day 6 (data not shown), but the infiltrated cells start to express CD4 or CD8 on day 8 (Fig. 3, Q–T). These results indicate that FL cells undergo a sequential change in their surface phenotype during the early stages of intrathymic differentiation. These results support the earlier observation of Godfrey et al. (18). In other words, FL cells from GFP+ mice provide an appropriate model for thymic repopulation experiments.

FIGURE 3.

Expression of various marker molecules on the surface of GFP+ cells in the thymus. Intrathymically populated FL cells from GFP+ mice were examined under confocal laser-scanning microscopy, and GFP (A–D, Q, and S) and PE (E–P, R, and T) fluorescence is shown. A–D, GFP (A, day 0; B, day 1; C, day 4; and D, day 9). E to H, Stained with PE-anti-CD44 (IM7) (E, day 0; F, day 1; G, day 3; H, day 6). I–L, Stained with PE-anti-CD25 (PC61) (I, day 0; J, day 1; K, day 2; L, day 4). M–P, Stained with biotinyl-anti-CD24 (J11d) plus R-PE-Streptavidin (M, day 0; N, day 1; O, day 2; P, day 8). R, Stained with PE-anti-CD8 (53-6.7) (Q and R, day 8). T, Stained with PE-anti-CD4 (RM4-5) (S and T, day 8). A and E, B and F, Q and R, and S and T show the same sections. In each case, one representative result from 10 to 20 examinations is shown.

FIGURE 3.

Expression of various marker molecules on the surface of GFP+ cells in the thymus. Intrathymically populated FL cells from GFP+ mice were examined under confocal laser-scanning microscopy, and GFP (A–D, Q, and S) and PE (E–P, R, and T) fluorescence is shown. A–D, GFP (A, day 0; B, day 1; C, day 4; and D, day 9). E to H, Stained with PE-anti-CD44 (IM7) (E, day 0; F, day 1; G, day 3; H, day 6). I–L, Stained with PE-anti-CD25 (PC61) (I, day 0; J, day 1; K, day 2; L, day 4). M–P, Stained with biotinyl-anti-CD24 (J11d) plus R-PE-Streptavidin (M, day 0; N, day 1; O, day 2; P, day 8). R, Stained with PE-anti-CD8 (53-6.7) (Q and R, day 8). T, Stained with PE-anti-CD4 (RM4-5) (S and T, day 8). A and E, B and F, Q and R, and S and T show the same sections. In each case, one representative result from 10 to 20 examinations is shown.

Close modal

The effect of various mAbs on the adhesion and migration of FL cells to the fetal thymus was examined. Anti-CD44 mAb (KM201) was added throughout the repopulation culture period, after which the fluorescence of the thymus lobes was examined. As shown in Fig. 4,A, GFP+ cells are observed on the surface of the thymus lobe, while few GFP+ cells can be seen inside the thymus. This suggests that the anti-CD44 mAb does not prevent the adhesion of FL cells to the thymus lobe, but prevents their migration into the thymus. The anti-CD44-treated thymi were examined for their expression of CD44 using PE-conjugated anti-CD44 (IM7). The adhering GFP+ cells were all positive to anti-CD44 (Fig. 4,B). On day 2 (Fig. 4,C) and day 5 (Fig. 4,D), GFP+ cells remained on the surface of the thymus lobe. Very few but still a significant number of GFP+ cells were observed inside the thymus on days 2 (Fig. 4,C) and 5 (Fig. 4 D). These cells appeared weakly CD44 positive on day 2 and negative on day 5 (data not shown), suggesting that a small number of cells escape the anti-CD44-mediated migratory inhibition.

FIGURE 4.

Inhibition of adhesion and migration of FL cells to the thymus. The adhesion and migration of FL cells from GFP+ mice to the thymus were examined. A–D, GFP+ FL cells were incubated with dGuo-treated thymus lobes in the presence of anti-CD44 (KM201). A, C, and D, GFP expression, incubated with anti-CD44 (A, day 0; C, day 2; D, day 5). B, CD44 expression, incubated with anti-CD44 (KM201) and stained with PE-anti-CD44 (day 0). A and B show the same section. E–H, fetal thymus was incubated with dGuo and HAase (1500 U/ml) for 5 days and subjected to repopulation assay. Fluorescence was examined on day 0. E, GFP. F, PE-anti-CD44. G, Mixed color of E and F. H, Visible light transmission picture. I–L, dGuo-treated thymus was subjected to repopulation assay in the presence of HA for 2 days. Fluorescence was examined on day 0. I, HA (100 μg/ml). J, HA (500 μg/ml). K and L, HA (1000 μg/ml). M–T, Fetal thymus was incubated with dGuo for 5 days and subjected to repopulation assay (hanging drop culture) for 2 days with CAS-9 (anti-integrin α4) (M and N) or TAS-5 (O and P), or with anti-α5 (HMα5-1) (Q and R), or anti-α6 (HMα6) (S and T). Fluorescence was examined on day 0. K to T, each shows two independent cultures. A typical result from 10 to 20 examination is shown.

FIGURE 4.

Inhibition of adhesion and migration of FL cells to the thymus. The adhesion and migration of FL cells from GFP+ mice to the thymus were examined. A–D, GFP+ FL cells were incubated with dGuo-treated thymus lobes in the presence of anti-CD44 (KM201). A, C, and D, GFP expression, incubated with anti-CD44 (A, day 0; C, day 2; D, day 5). B, CD44 expression, incubated with anti-CD44 (KM201) and stained with PE-anti-CD44 (day 0). A and B show the same section. E–H, fetal thymus was incubated with dGuo and HAase (1500 U/ml) for 5 days and subjected to repopulation assay. Fluorescence was examined on day 0. E, GFP. F, PE-anti-CD44. G, Mixed color of E and F. H, Visible light transmission picture. I–L, dGuo-treated thymus was subjected to repopulation assay in the presence of HA for 2 days. Fluorescence was examined on day 0. I, HA (100 μg/ml). J, HA (500 μg/ml). K and L, HA (1000 μg/ml). M–T, Fetal thymus was incubated with dGuo for 5 days and subjected to repopulation assay (hanging drop culture) for 2 days with CAS-9 (anti-integrin α4) (M and N) or TAS-5 (O and P), or with anti-α5 (HMα5-1) (Q and R), or anti-α6 (HMα6) (S and T). Fluorescence was examined on day 0. K to T, each shows two independent cultures. A typical result from 10 to 20 examination is shown.

Close modal

The major ligand of CD44 is known to be a HA. To test the role of CD44 in adhesion, thymus lobes were treated with HAase (1500 U/ml) together with dGuo, then subjected to repopulation assay. Similar to the result of anti-CD44 treatment (Fig. 4, A–D), FL cells from GFP+ mice adhered to the thymus lobe, although a small number of CD44+ cells were found in the HAase-treated thymus (Fig. 4, E–G). When the thymus lobes were treated at a lower enzyme concentration, similar adhesion was observed, but significant numbers of GFP+ cells were also seen in the thymus (data not shown). To confirm the interaction between CD44 and HA, various concentrations of HA (100–1000 μg/ml) were added to the repopulation assay mixtures. As shown in Fig. 4, I–L, cell migration was prevented at higher HA concentrations, and most of the GFP+ cells remained on the surface of the thymus. This supports the notion that CD44 plays a key role in the efficient migration of FL cells into the thymus, but that the CD44-HA interaction does not affect the adhesion of cells to the thymus.

The effects of anti-integrin mAbs were then examined. As shown in Fig. 4, M–P, the addition of an anti-integrin α4 mAb, either CAS-9 (Fig. 4, M and N) or TAS-5 (Fig. 4, O and P), to the culture prevented the appearance of GFP+ cells on both the surface and interior of the thymus. Integrin α5 is intermediately expressed on FL cells (Fig. 1,D). mAb to the integrin α5 subunit (HMα5-1) caused a partial inhibition of adhesion, but data varied among experiments (two typical results are shown in Fig. 4, Q and R). Anti-integrin β1 mAb (HMβ1-1) (17) (a gift from Dr. H. Yagita) produced results similar to those of anti-α5 (data not shown). Integrin α6 was also found to be expressed intermediately on the surface of FL cells (Fig. 1,E), but had no effect on the adhesion or migration of FL cells to the thymus (Fig. 4, S and T). Other anti-integrin mAbs, such as anti-α1 (HMα1) and anti-α2 (HMα2), whose target integrins are not expressed on the surface of FL cells (Fig. 1, A and B), had no effect (data not shown).

In order to exclude the possibility that the anti-integrin α4 mAbs (CAS-9 or TAS-5) have cytotoxic effects on FL cells, FL cells were incubated in the presence of anti-integrin α4 mAbs or other mAbs for 48 h in vitro. When direct cytotoxicity was examined by staining the cells with 7-amino actinomycin D (7-AAD) (Sigma) after 48 h in culture, the percentage of 7-AAD-stained cells range from 4.8 to 5.5% in either the absence or presence of control IgG, anti-integrin α4 mAbs (CAS-9 or TAS-5), or anti-CD44 (KM201). Viable cell recoveries in the presence of these mAbs or anti-c-kit (ACK2) after 48 h in culture were also equivalent to recoveries from cultures containing control IgG or medium alone. We conclude that the mAbs used in these experiments have no direct cytotoxicity on FL cells in vitro.

Taken together, our results show integrin α4 to be a key molecule in the adhesion of FL cells to the fetal thymus, and CD44 to be necessary for the migration of adhered cells into the thymus lobe.

Developmental pathway of intrathymic T cells has been investigated extensively during the last decade (9, 10). Thymic repopulation assays, using dGuo-treated fetal thymus lobes and the most immature CD3CD4CD8 triple-negative thymocytes from adult mice have revealed the maturational sequence to be: CD44+CD25→CD44+CD25+→CD44CD25+→CD44CD25 (18). Because regular repopulation assays require certain numbers of intrathymic cells for flow cytometry, it is very difficult to analyze the earliest stages of thymocyte development, such as the adhesion of FL cells to the thymus lobe. In the present study, in order to analyze the earliest stages of thymocyte development from FL cells in the thymus, thymic repopulation assays using FL cells from mice to normal embryonic thymus were performed. GFP+ cells were observed in thymus lobes cocultured with FL cells for 2 days by the hanging drop method on day 0. We then determined the sequential expression of CD44 and CD25 in GFP+ cells. Sequential phenotypic changes observed during early thymocyte development from FL cells were in agreement with earlier observations made using adult immature thymocytes (18).

Godfrey et al. also used a fetal thymus repopulation assay to show that intrathymic CD4+CD8+ cells arise from adult CD44CD25 thymocytes (18). This result suggests that CD44 are not necessary for the migration of adult thymocytes into the thymus. Wu et al. also reported that HA is not a ligand for CD44 in adult thymocytes (19). In the present study, treatment with anti-CD44 or HAase inhibited the migration of FL cells into the fetal thymus. The discrepancy in the role of the CD44-HA interaction in migration to the thymus is not yet resolved, but it is possible to speculate that immature thymocytes from adult mice use different adhesion molecules than FL cells. Besides the discrepancy, repopulation assays using adult thymocytes to fetal thymus are not physiological, and FL cells might be a better choice for this type of assay.

The adhesion and migration of T cells to the endothelium have been extensively investigated (reviewed in Refs. 2–4). CD44-HA interaction-dependent weak adhesion of a human T cell clone to tonsilar stroma has been reported (20). Recently, DeGrendele et al. (21) showed that the migration and extravasation of activated T cells from blood to the peritoneal cavity depend on the CD44-HA interaction. Intraperitoneal injection of staphylococcal enterotoxin B followed by the i.v. injection of anti-CD44 mAb prevented the entrance of Vβ8+ T cells into the peritoneum (22). These reports suggest that the interaction involving CD44-HA occurs at a rolling (weak adhesion) step of lymphocyte adhesion to endothelial cells, and differs from a strong adhesion step. Although it remains unclear whether integrins and CD44 play different roles in the adhesion process in adult endothelium, studies of the molecular mechanisms of the CD44-HA interaction at the migration step will be interesting.

The expression of integrin α4 increases on activated (memory) lymphocytes whose adherence to endothelial cells occurs via VCAM-1 or mucosal addressin cell adhesion molecules (3, 4). Weak adhesion (tethering or rolling) is mediated by members of the selectin family, and strong adhesion is mediated mainly by LFA-1/ICAM-1 or very late antigen-4 (VLA-4)/VCAM-1. Integrin-mediated strong adhesion is followed by the migration of activated T cells through the endothelium into the surrounding tissues. In agreement with our observation, Wada et al. used a unique high oxygen submersion culture technique to show that mAbs to CD44 and integrin α4 inhibit early thymocyte development from FL cells (15). The inhibition of adhesion and migration by these mAbs will result in the inhibition of thymocyte development. It is known that TAS-5 (anti-integrin α4) blocks the binding of VLA-4 with both fibronectin and VCAM-1, whereas CAS-9 (anti-integrin α4) blocks binding with fibronectin (15). These results suggest that the anti-integrin α4-mediated inhibition of binding to the fetal thymus might be mediated by fibronectin. Other mAbs to integrins, except anti-β1 and anti-α5, had no effect (see text). Because an integrin α4 null mutant chimera gave rise to normal thymocyte development in early life (23), α5 or other unidentified adhesion molecules may fulfill the function of α4. Oppenheimer-Marks et al. reported that the adhesion and migration of activated human T cells to umbilical vein epithelial cells are significantly but modestly inhibited by anti-LFA-1 or anti-CD44 mAb (24). The adhesion and migration of an LFA-1-deficient T cell clone to endothelial cells are partly mediated by the VLA-4/VCAM-1 interaction (25). Thus integrins and CD44 might be essential molecules for cell trafficking to various organs.

Using a novel GFP+ mouse system, we have found that CD44 and integrin α4 play different roles in the adhesion and migration of FL cells to the fetal thymus. This system will also enable us to investigate the roles of adhesion molecules and extracellular matrix proteins on cellular movement within the thymus.

We thank Drs. Kensuke Miyake, Tatsuo Kina, Hideo Yagita, and Shin-ichi Nishikawa for providing various mAbs and Dr. Margaret Dooley Ohto for reading this manuscript.

1

This work was supported by grants-in-aid from the Ministry of Education, Science, Sports, and Culture, and the Ministry of Health and Welfare of Japan.

3

Abbreviations used in this paper: FL, fetal liver; dGuo, 2′-deoxyguanosine; GFP, green fluorescent protein; GFP+, GFP-gene transgenic; HA, hyaluronate; HAase, hyaluronidase; VLA-4, very late Ag-4; 7-AAD, 7-amino actinomycin D.

1
Zimmerman, G. A., S. M. Prescott, T. M. McIntyre.
1992
. Endothelial cell interactions with granulocytes: tethering and signaling molecules.
Immunol. Today
13
:
93
2
Shimizu, Y., W. Newman, Y. Tanaka, S. Shaw.
1992
. Lymphocyte interactions with endothelial cells.
Immunol. Today
13
:
106
3
Springer, T. A..
1994
. Traffic signals for lymphocyte recirculation and leukocyte emigration: the multistep paradigm.
Cell
76
:
301
4
Imhof, B., D. Dunon.
1995
. Leukocyte migration and adhesion.
Adv. Immunol.
58
:
345
5
Nelson, P. J., A. M. Krensky.
1998
. Chemokines, lymphocytes and viruses: what goes around, comes around.
Curr. Opin. Immunol.
10
:
265
6
Dunon, D., B. A. Imhof.
1993
. Mechanisms of thymus homing.
Blood
81
:
1
7
Savino, W., D-M. S. Villa-Verde, J. Lannes-Vieira.
1993
. Extracellular matrix proteins in intrathymic T-cell migration and differentiation.
Immunol. Today
14
:
158
8
van Ewijk, W..
1991
. T-cell differentiation is influenced by thymic microenvironments.
Annu. Rev. Immunol.
9
:
591
9
Shortman, K., L. Wu.
1996
. Early T lymphocyte progenitors.
Annu. Rev. Immunol.
14
:
29
10
Anderson, G., N. C. Moore, J. J. T. Owen, E. J. Jenkinson.
1996
. Cellular interactions in thymocyte development.
Annu. Rev. Immunol.
14
:
73
11
Kingston, R., E. J. Jenkinson, J. J. T. Owen.
1985
. A single stem cell can recolonize an embryonic thymus, producing phenotypically distinct T-cell populations.
Nature
317
:
811
12
Ikawa, M., K. Kominami, Y. Yoshimura, K. Tanaka, Y. Nishimune, M. Okabe.
1995
. Green fluorescent protein as a marker in transgenic mice.
Dev. Growth. Differ.
37
:
455
13
Miyake, K., K. L. Medina, S.-I. Hayashi, S. Ono, T. Hamaoka, P. W. Kincade.
1990
. Monoclonal antibodies to Pgp-1/CD44 block lympho-hemopoiesis in long-term bone marrow cultures.
J. Exp. Med.
171
:
477
14
Nishikawa, S., M. Kusakabe, K. Yoshinaga, M. Ogawa, S. Hayashi, T. Kunisada, T. Era, T. Sakakura, S.-I. Nishikawa.
1991
. In utero manipulation of coat color formation by a monoclonal anti-c-kit antibody: two distinct waves of c-kit-dependency during melanocyte development.
EMBO J.
10
:
2111
15
Wada, K., T. Kina, H. Kawamoto, M. Kondo, Y. Katsura.
1996
. Requirement of cell interactions through adhesion molecules in the early phase of T cell development.
Cell. Immunol.
170
:
11
16
Miyake, S., T. Sakurai, K. Okumura, H. Yagita.
1994
. Identification of collagen and laminin receptor integrins on murine T lymphocytes.
Eur. J. Immunol.
24
:
2000
17
Noto, K., K. Kato, K. Okumura, H. Yagita.
1995
. Identification and functional characterization of mouse CD29 with a mAb.
Int. Immunol.
7
:
835
18
Godfrey, D. I., J. Kennedy, T. Suda, A. Zlotnik.
1993
. A developmental pathway involving four phenotypically and functionally distinct subsets of CD3CD4CD8 triple-negative adult mouse thymocytes defined by CD44 and CD25 expression.
J. Immunol.
150
:
4244
19
Wu, L., P. W. Kincade, K. Shortman.
1993
. The CD44 expressed on the earliest intrathymic precursor population functions as a thymus homing molecule but does not bind to hyaluronate.
Immunol. Lett.
38
:
69
20
Clark, R. A., R. Alon, T. A. Springer.
1996
. CD44 and hyaluronan-dependent rolling interactions of lymphocytes on tonsillar stroma.
J. Cell Biol.
134
:
1075
21
DeGrendele, H. C., P. Estess, M. H. Siegelman.
1997
. Requirement for CD44 in activated T cell extravasation into an inflammatory site.
Science
278
:
672
22
Tchilian, E. Z., J. J. T. Owen, E. J. Jenkinson.
1997
. Anti-α4 integrin antibody induces apoptosis in murine thymocytes and staphylococcal enterotoxin B-activated lymph node T cells.
Immunology
92
:
321
23
Arroyo, A. G., J. T. Yang, H. Rayburn, R. O. Hynes.
1996
. Differential requirements for α4 integrins during fetal and adult hematopoiesis.
Cell
85
:
997
24
Oppenheimer-Marks, N., L. S. Davis, P. E. Lipsky.
1990
. Human T lymphocyte adhesion to endothelial cells and transendothelial migration: alteration of receptor use relates to the activation status of both the T cell and endothelial cells.
J. Immunol.
145
:
140
25
Kavanaugh, A. F., E. Lightfoot, P. E. Lipsky, N. Oppenheimer-Marks.
1991
. Role of CD11/CD18 in adhesion and transmigration of T cells: analysis utilizing CD18-deficient T cell clone.
J. Immunol.
146
:
4149