DEVELOPMENTAL DYNAMICS 2M437-446 (1996) Tenascin-C Lines the Migratory Pathways of Avian Primordial Germ Cells and Hematopoietic Progenitor Cells KRISTIN K. ANSTROM AND RICHARD P. TUCKER Department of Neurobiology and Anatomy, Bowman Gray School of Medicine, Medical Center Blud, Winston-Salem, North Carolina 27157 (K.K.A.), and Department of Cell Biology and Human Anatomy, University of California, Davis, California 95616 (R.P.T.) ABSTRACT Tenascin-C is a large hexameric extracellular matrix glycoprotein associated with epithelial-mesenchymal interactions, connective tissue development, and the formation of the central nervous system. Tenascin-C also lines the pathways followed by migrating avian neural crest cells, although its role in neural crest morphogenesis remains unclear. In vitro, tenascin-C interferes with cell-fibronectin interactions, and promotes the motility of many cell types including the neural crest. To determine if tenascin-C is a consistent component of matrices through which invasive embryonic cells migrate, we have investigated if tenascin-Cis associated with 2 additional populations of motile, embryonic cells: primordial germ cells and hematopoietic progenitor cells. We have found that HNK-1, a monoclonal antibody used as a marker of neural crest, also stains avian primordial germ cells. Double-label immunohistochemistry reveals that tenascin-C is found in the mesenchyme adjacent to the ventral half of the dorsal aorta where the primordial germ cells penetrate the vessel wall, and both tenascin-C and fibronectin are present in the extracellular matrix through which the primordial germ cells migrate to reach the genital ridges. Unlike fibronectin, which is found throughout the splanchnic mesoderm, tenascin-C is concentrated in the proximal part of the splanchnicregion where the primordial germ cells are concentrated. In embryos where the gonadal anlagen are surgically removed before the primordial germ cells leave the bloodstream, ectopic primordial germ cells were found exclusively in head and trunk mesenchyme containing tenascin-C. Like primordial germ cells, a subset of hematopoietic progenitor cells migrate through the mesenchyme ventral to the dorsal aorta where they form hematopoietic clusters. Others bud directly into the lumen of the aorta. Anti-tenascin-C stains the mesenchyme surrounding the migrating cells as well as the basal surfaces of the cells that appear to be budding into the lumen. In situ hybridization with a tenascin-C-specificcDNA probe shows that the major sources of the tenascin-C mRNA in this region are the hematopoietic progenitor cells themselves as well as the cells in the wall of the ventral aorta. mRNAs encoding 3 major 0 1996 WILEY-LISS,INC. splice variants of tenascin-C were identified by reverse transcriptase polymerase chain reaction (PCR) in the embryonic aorta and adjacent mesenchyme dissected from both the region of primordial germ cell and hematopoietic precursor cell migration. These experiments indicate that tenascin-Cis a component of the migratory environment for many motile cells in the early embryo, where it has the potential to mediate cell-fibronectin interactions. o 1996 Wiley-Lisa, Inc. Key words: Tenascin, Development, Extracellular matrix, Cytotactin, Cell motility, Primordial germ cell, Hematopoietic progenitor cell INTRODUCTION Extracellular matrix (ECM) molecules play an essential role during cell migration (Thiery et al., 1985; Adams and Watt, 1993). These molecules are known to affect cell proliferation and differentiation, as well as promote cell motility and define migratory pathways. For example, immunohistochemistry has revealed that the glycoprotein tenascin-C is found along the migratory pathways of avian neural crest cells, where it is concentrated in the rostra1 half of each somite at a time when the neural crest invades this structure (Tan et al., 1987; Mackie et al., 1988b). In vitro studies have shown that neural crest cells can migrate on substrata coated with tenascin-C, and that tenascin-C added to the culture medium will cause neural crest cells that have attached to adhesive ECM to round up (Halfter et al., 1989). These observations and others (ChiquetEhrismann et al., 1988; Lotz et al., 1989) have led to the notion that tenascin-C promotes motility by helping cells detach from their substratum and move on. Studies on a variety of pathologies involving invasive cell behavior also support this hypothesis. For example, tenascin-C is upregulated at the margins of healing wounds (Mackie et al., 1988a) and is found in the Received November 30, 1995;accepted February 16,1996. Address reprint requests/correspondence to R.P. Tucker, Dept of Cell Biology and Human Anatomy, School of Medicine, University of California, Davis, CA 95616-8643. 438 ANSTROM AND TUCKER stroma surrounding many types of tumors (Mackie et al., 1987; Borsi et al., 1992). Here, we have attempted to understand the function of tenascin-C during cell migration by studying 2 populations of motile cells in the embryonic day (E) 3 chick that share many properties with the neural crest: the primordial germ cells (PGCs), and hematopoietic progenitor cells (HPCs). PGCs originate extraembryonically, and were first identified in the germinal crescent of the E l chick where they enter the forming bloodstream (Swift, 1914). They circulate passively until E3 when they adhere to a region of the ventral wall of the dorsal aorta. They eventually pass through the wall of the vessel and migrate through the splanchnic mesoderm en route to the genital ridges. In the E3 chick, the ventrolateral walls of the dorsal aorta are thickened by basophilic cells (DieterlenLihvre and Martin, 1981). As determined by immunohistochemistry, these cells are HPCs (Cormier et al., 1986). In E4 chick embryos, dense paraaortic hematopoietic foci are scattered in the mesenchyme from the mesonephros to the lung rudiment (Dieterlen-Lievre and Martin, 1981). ChicWquail chimeras (Martin et al., 1980) and aortic cultures (Cormier et al., 1986) show that the aortic HPCs are the source of the hematopoietic clusters. The cells from the paraaortic foci subsequently migrate to colonize the thymus of the E6 chick. The thymus emits a chemoattractant to guide the foci cells (Ben Slimane et al., 1983), but the early pathway from the aorta is as yet uncharacterized. Here, we have used immunohistochemistry, in situ hybridization, and reverse-transcriptase polymerase chain reaction (RT-PCR), to determine if tenascin-C is part of the ECM associated with PGCs and HPCs in the chick embryo. We found that tenascin-C is concentrated in the ECM through which these populations of embryonic migratory cells pass, implicating this glycoprotein in a general role associated with promoting cell motility. RESULTS The Antibody HNK-1 is a Marker of Chick PGCs The established marker for avian PGCs is the PAS reaction (Clawson and Domm, 1962; Fujimoto et al., 1976). The PAS reaction, however, does not allow for double-labeling studies with tenascin-C antibodies. We have found that the monoclonal antibody HNK-1 can be used as a marker for chick PGCs. Whole blood smears from stage 13-14 chick embryos contain cells with the typical size and granulated appearance of PGCs that are positive for HNK-1 (Fig. lA,B). Cells that display the “teardrop” shape of a white blood cell are not positive for the HNK-1 antigen (Fig. lC,D). In addition, PGCs were isolated on a Ficoll gradient; half of the isolated cells were stained with PAS and the other half with HNK-1. Large, granular cells positive for PAS (Fig. 1E) are also positive for HNK-1 (Fig. lF,G). The HNK-1 staining is punctate, and appears to be on the cell surface. Chick PGCs are also positive for HNK-1 in situ. When sections containing PGCs were found by PAS staining, adjacent sections were stained with HNK-1. Similarly shaped cells in the same regions are positive for both PAS and HNK-1 (Fig. lH,J). The HNK-1stained cells exhibit the cell-surface, punctate pattern observed in vitro. The staining is not as intense as the staining of nearby neural crest cells, but is consistent and well above background. Thus, HNK-1 can be used to identify not only the neural crest in the avian embryo, but also PGCs. Tenascin Lines the Pathways of Normal and Ectopic Chick PGCs Cross-sections of E3 chicks were stained with PAS to locate PGCs. Adjacent sections were double-labeled with a combinations of either anti-tenascin-C and HNK-1, or tenascin-C and fibronectin antibodies. HNK-Utenascin-C double-labeling shows PGCs in the mesenchyme ventral to the dorsal aorta and in the splanchnic mesoderm (Fig. 2A,D). Tenascin-C is abundant in the ECM around the ventral wall of the aorta and continues ventrally into the splanchnic mesoderm where it is concentrated in or near the endoderm basement membrane (Fig. 2B). Higher magnification demonstrates that tenascin-C is in the splanchnic mesoderm corresponding to areas where PGCs are found (Fig. 2D,E). Anti-fibronectin staining shows that fibronectin is also present, but is not confined to the PGC pathway (Fig. 2C). Controls with secondary antibody alone show no staining (Fig. 2F). Thus, tenascin-C is concentrated in the ECM in regions where PGCs are found, in contrast to the widespread distribution of fibronectin. If the caudal third of stage 12-14 embryos is removed in ovo, and the embryo is allowed to develop for 1 more day, PGCs leave the bloodstream elsewhere (Nakamura et al., 1991). To determine if tenascin-C is an element common to both normal and ectopic PGC pathways, the distribution of tenascin-C around the “lost” cells was examined. Transverse sections through the hemi-embryos stained with PAS show that the PGCs leave the bloodstream and move into the ectopic areas as described by Nakamura et al. (1991). If adjacent sections are stained with tenascin-C antibodies, it is apparent that the PGCs are always located in tenascin-C-rich matrices. For example, PGCs were found in patches of tenascinC-rich matrix in the perioptic mesenchyme (Fig. 3A,B). Near the developing hindbrain, an ectopic PGC was found in ECM where tenascin-C is concentrated (Fig. 3D,E). Outside the head region, the PGCs were found in the anterior half of the somites (Fig. 3G) where the neural crest are also found (Fig. 31). As previously described (Tan et al., 1987; Mackie et al., 1988b), tenascin-C is restricted to the anterior half of the somites at this time. In all of these regions, fibronectin has a widespread distribution (Fig. 3C,F); PGCs were not found in mesenchyme stained by anti-fibronectin and not anti- TENASCIN AND PRIMORDIAL GERM CELLS 439 Fig. 1. HNK-1 is a marker of chick PGCs. A-D: Whole blood smear from a stage 13 chick embryo. A: A PGC recognized by morphological criteria viewed with phase optics. B: The same cell, but not surrounding cells in the smear, is stained by the monoclonal antibody HNK-1. C: A white blood cell in the same smear. D: This cell is not positive for HNK-1. E-G: PGCs isolated on a Ficoll density gradient are stained by PAS. E: The isolated cells are PAS-positive. F: Another cell from the same frac- tion viewed with phase contrast optics shows the characteristic size and granular appearance of a PGC. G: The cell shown in F is also positive for HNK-1. H: PGCs (arrows) that have just left the bloodstream are darkly stained by PAS in a cross-section through the notochord (n) and paired aorta (a) of an E3 chick embryo. I: In an adjacent section, cells in the same region with the same shape as the PAS-positive cells are stained by HNK-1 (arrows). tenascin-C. Note that the tenascin-C immunoreactivity seen in each of the regions where ectopic PGCs were found was identical to that reported previously (Tan et al., 1987; Mackie et al., 1988b; Tucker, 1991). gion. Consistent with our immunohistochemistry, a hybridization signal was seen in the endothelium around the entire perimeter of the aorta, but was more prevalent in the ventral half (Fig. 4E).In sections containing ventrolateral hematopoietic foci, the hybridization signal was strongest over the foci themselves, indicating that the HPCs or immediately adjacent cells are a major source of tenascin-C (Fig. 4F). Control sections incubated with plasmid DNA show background hybridization levels (Fig. 4G). Tenascin-C is Found Near Migrating HPCs Anti-HNK-1 transiently marks HPCs found in the ventrolateral clusters in the wall of the dorsal aorta (Thiery et al., 1985). Most of the cells in the intraaortic foci bud into the lumen of the aorta upon maturation (Fig. 4A,C), but a subset emigrate from the aorta into the surrounding mesenchyme (Fig. 4C). Immunohistochemistry shows that tenascin-C is concentrated in the mesenchyme through which the HPCs migrate (Fig. 4B,D). Tenascin-C also cradles the HPCs that are detaching from the wall of the aorta (Fig. 4D). In situ hybridization with a tenascin-C-specific cDNA (cTn8) was carried out to determine the cellular origins of tenascin-C in the aortic hematopoietic re- Several Tenascin-C Splice Variants are Present in the ECM of the Two Migratory Pathways Tenascin-C has several splice variants that have different properties in vitro (for reviews see Tucker et al., 1994b; Leprini et al., 1994). To determine which splice variants may be expressed near migrating PGCs and HPCs, RT-PCR was done on poly(A) RNA isolated from the migratory mesenchyme of the PGCs and HPCs. In 440 ANSTROM AND TUCKER Fig. 2. Tenascin-C lines the PGC migratory pathway. A: A low-magnification overview of a frozen cross section through the trunk of an E3 chick embryo stained with HNK-1. HNK-1 stains the neural crest cells on either side of the neural tube (nt) and the notochord (no). PGCs are in the splanchnic mesoderm just ventral to the paired aorta (a). The region in the box is shown at higher magnification in 0. 8: The same section shown in A was also stained with an antiserum to tenascin-C. Tenascin-C is part of the ECM surrounding the neural crest, as well as in the mesoderm surrounding the paired aorta (a). There is no tenascin-C immunoreactivity in the extraembryonic membranes (arrowhead) or in the splanchnic mesoderm ventral to the level occupied by PGCs (arrow). The latter regions are shown at higher magnification in E. C: An adjacent section stained with anti-fibronectin. Fibronectin is widely distributed throughout mesenchymal ECM, including extraembryonic membranes (arrowhead) and the splanchnic mesoderm ventral to the PGCs (arrow). D:The PGC migratory pathway stained with HNK-1. PGCs are present in the splanchnic mesoderm (arrow). E: The same section shown in D stained with anti-tenascin-C. Tenascin-C is present in or near the basement membranes lining the splanchnic mesoderm (arrowheads). The PGC shown in D is near or in contact with a patch of tenascin-C-rich ECM (arrow). F: The same region of an adjacent section stained with secondary antibody alone to control for background fluorescence and nonspecific staining. both regions, PCR products corresponding to 3 splice variants of tenascin-C were found (Fig. 5). cent and Thiery, 1984; Tucker et al., 1984; Rickman et al., 1985; Bronner-Fraser, 1986; Loring and Erickson, 1987; Erickson e t al., 1989) and HPCs (Thiery et al., 1985);here we show that avian PGCs are stained with HNK-1, both in vitro and in situ. Double-label immunohistochemistry with HNK-1 and tenascin-C antibod- DISCUSSION The monoclonal antibody HNK-1 is a n established marker of neural crest cells (Vincent et al., 1983; Vin- TENASCIN AND PRIMORDIAL GERM CELLS Fig. 3. Ectopic PGCs are found in tenascin-C-rich ECM. A-C: Sections through the perioptic mesenchyme of an E3 chick embryo following surgical removal of the gonadal anlagen. A: PAS staining was used to identify ectopic PGCs (arrowheads) in the mesenchyme (m) near the pigment epithelium (pe), neural retina (nr) and diencephalon (di). B: An adjacent section stained with an antibody to tenascin-C shows that the ectopic PGCs (approximated by asterisks) are found in tenascin-C-rich ECM. C: The same section shown in B was stained with anti-fibronectin. Fibronectin is also part of the ECM surrounding the ectopic PGCs. D-F: Cross-sections through the developing hindbrain (hb), cardinal vein (c) and nearby notochord (n) of an E3 embryo following removal of the gonadal anlagen. D: PAS reveals an ectopic PGC (arrowhead) in the 44 1 mesenchyme (m) near the cardinal vein. E: An adjacent section (slightly compressed) stained with tenascin-C antibody shows tenascin-C in the ECM surrounding the ectopic PGC (approximated by the asterisk) adjacent to the cardinal vein. F: The same section stained with anti-fibronectin. G-I: Frontal frozen sections through the trunk of an E3 embryo following ablation of the gonadal anlagen. G: PAS staining reveals an ectopic PGC (arrowhead) in the anterior sclerotome (a), but not the posterior sclerotome (p) or dermamyotome (dm) of a somite. H: An adjacent section stained with tenascin-C antibodies shows tenascin-C in the anterior sclerotome. I: The same section stained with HNK-1 shows the neural crest cells in the anterior sclerotome. 442 ANSTROM AND TUCKER Fig. 4. A subset of HPCs make tenascin-C and migrate into tenascinC-rich mesenchyme surrounding the aorta. A-D: Double-label immunohistochemistry with HNK-1 and tenascin-C antiserum on a frozen crosssection through an E3 chick embryo. A: HPCs (arrowhead) lining the ventral wall of the aorta (a) are stained by HNK-1. Neural crest cells on either side of the aorta are also brightly stained, as is the notochord (n). These HPCs are shown at a higher magnification in C. 6: The same section stained with a tenascin-C antibody shows tenascin-C in the ECM surrounding the aorta, especially in the mesenchyme just ventral to the aorta. C: A higher magnification of the HPCs stained by HNK-1. Many of the cells are lining the aorta, where they appear to be budding into the lumen of the vessel (arrowhead). Other HPCs appear to be migrating into the adjacent mesenchyme (arrows). 0: The same region shown in C stained with anti-tenascin-C. Tenascin-C lines the basal surface of the HPCs (arrowhead) and is present in the ECM through which the HPCs migrate. E-G: Darkfield images of nearby sections following in situ hybridization and autoradiography. E: The tenascin-C cDNA probe cTn8 labels the notochord (n) and the endothelium of the aorta (a). F: In a region where 2 ventrolateral clusters of HPCs are found, the tenascin-C cDNA probe cTn8 labels the clusters intensely (arrows). G: A third section incubated with a control DNA (pUC) shows low background levels. ies were done to characterize the ECM through which the PGCs migrate. Both monoclonal and polyclonal tenascin-C antibodies reveal tenascin-C in the ECM surrounding migrating PGCs. This staining is particularly strong around the ventral side of the aorta, but also stretches into the splanchnic mesoderm ventral to the aorta where the PGCs are migrating to the genital ridges. We were unable to use in situ hybridization to determine the cellular origins of this tenascin-C because of unusually high background levels in this part TENASCIN AND PRIMORDIAL GERM CELLS I I I I 3 1060 4 4 540 4 264 I Fig. 5. Transcripts encoding 3 splice variants of tenascin-C were identified in poly(A) RNA isolated from the HPC and PGC migratory pathways by RT-PCR. Primers spanning the tenascin-C variable domain were used to amplify products with the anticipated sizes of 264 b.p., 540 b.p. and 1,060 b.p., correspondingto no variable repeats, one variable repeat, and 3 variable repeats, respectively. RT-PCR products amplified from poly(A) RNA isolated from the HPC migratory pathway (lane 1) and the PGC migratory pathway (lane 2) show that the splice variant profile is similar in both regions. The 1,060 b.p. band is present in lane 2 but appears very faint in this micrograph. Hash marks on the left indicate size standards. From top to bottom: 2,072 b.p., 600 b.p., 300 b.p., 100 b.p. of the embryo (results not shown). However, RT-PCR demonstrates that the mRNAs encoding at least three tenascin-C splice variants are present in the PGCs andlor the tissues lining their migratory pathway. Tenascin-C in the PGC migratory pathway could be doing several things, including marking the region of the aorta where the PGCs leave the bloodstream, thus triggering their active phase of migration. Alternatively, tenascin-C could be interacting with other ECM molecules such as fibronectin to facilitate PGC migration. The first role seems unlikely. In vitro studies have suggested that the factor responsible for guiding the PGCs is a chemoattractant emitted by the genital ridges (Kuwana et al., 1986; Godin et al., 1990) and that TGF-P1 can mimic this effect (Godin and Wylie, 1991). Paraaortic tenascin-C immunoreactivity is not specific to the area of PGC attachment and migration, but instead is present around the dorsal aorta at all axial levels examined. Thus, it is unlikely that tenascin-C is the signal for PGCs to leave the bloodstream and begin their active phase of migration. It is possible that specific variants of tenascin-C could guide the PGCs; however, RT-PCR shows that each of the major splice variants is expressed a t axial levels where the PGCs leave the aorta as well as at more rostra1 levels (i.e., where the HPCs are forming). It is possible that tenascin-C contributes to a matrix that supports PGC migration, perhaps by interfering with cell-fibronectin interactions (Chiquet-Ehrismann et al., 1988; Lotz et al., 1989). Fibronectin has been 443 found in the PGC pathway in chick (Urven et al., 1989), Xenopus (Heasman et al., 1981) and mouse (ffrenchConstant et al., 1991). Here, we show that anti-fibronectin staining was not confined to the PGC pathway, but was widespread throughout the chick embryo. In contrast, individual HNK-1-positive PGCs were frequently seen close or in contact with strands of antitenascin-C-stained ECM. Little or no anti-tenascin-C staining was seen ventral to the pathways followed by the PGCs. In addition, ectopic PGCs are encountered exclusively in regions stained by both tenascin-C and fibronectin antibodies. The location of the ectopic cells supports the hypothesis that a combination of tenascin-C and fibronectin forms an ideal substratum for cell migration. However, one must consider that most of the tenascin-C in the head region is probably expressed by the cranial neural crest (Tucker and McKay, 1991). It is possible that the PGCs are responding to the same migratory cues as the neural crest, and that the colocalization of tenascin-C and ectopic PGCs is a coincidence of this comigration. This hypothesis could be tested in a future study by a double ablation of the neural crest and the genital ridges. Finally, it has been shown that premigratory PGCs respond to fibronectin differently than postmigratory PGCs (ffrench-Constant, 1991). It is possible that PGCs express tenascin-C upon activation, thus changing their affinity for fibronectin. The HPCs also migrate from the ventral wall of the aorta into the surrounding mesenchyme. Anti-tenascin-C staining is robust throughout the ventral mesenchyme a t this axial level. We have also shown that like the neural crest (Tucker and McKay, 1991), HPCs are potentially the source of much of the surrounding tenascin-C, thus making it unlikely that tenascin-C actually directs the pathways followed by the HPCs. Most HPCs bud into the lumen of the aorta; one role oftenascin-C made by HPCs may be to help them detach from the fibronectin-rich substratum and/or each other as they mature, thus regulating the release of cells into the bloodstream. Previous studies have shown that tenascin-C surrounds migrating neural crest cells (Tan et al., 1987; Mackie et al., 1988b; Epperlein et al., 1988; BronnerFraser, 1988). This tenascin-C is probably made by the neural crest cells themselves, as tenascin-C can be detected in culture medium conditioned by the neural crest, and in situ hybridization shows that migrating neural crest cells are positive for tenascin-C transcripts (Tucker and McKay, 1991). Here, we show tenascin-C in the ECM surrounding normal and ectopic PGCs, as well as migrating HPCs. In the case of the PGCs, tenascin-C is present in a subset of the ECM stained with antibodies to fibronectin. These observations demonstrate that tenascin-C is a component common to the ECM surrounding diverse populations of migratory embryonic cells, thus supporting a role for tenascin-C in promoting cell motility in a general way, perhaps by interfering with cell-fibronectin interactions. 444 ANSTROM AND TUCKER EXPERIMENTAL PROCEDURES Animals, Sectioning and Immunostaining Fertile chicken eggs (Hubbard Farms, Statesville, NC) were kept in a humid incubator at 37°C for 55-72 hours. For immunohistochemistry, in situ hybridization and PAS staining whole embryos were fixed in ice-cold 4% paraformaldehydetphosphate-bufferedsaline (PBS) solution for 5 hours and then cryoprotected in 20% sucrose in PBS overnight. After embedding in Tissue Tek (Miles, Elkhart, IN) using a slurry of dry ice and 2-methylbutane, 10-15 pm sections were cut on a Bright cryostat and put on poly-L-lysine (Sigma, St. Louis, Mobcoated slides. The sections were air-dried for 20 minutes and then stored a t -70°C until needed. All antibodies used in this study have been previously characterized. The monoclonal antibody M1 recognizes a constant region of tenascin-C (Chiquet and Fambrough, 1984; Hybridoma Bank, Rockville, MD). Undiluted hybridoma supernatants were used. The rabbit anti-chicken tenascin-C polyclonal antiserum (Chiquet-Ehrismann et al., 1986) used in double-labeling studies was a generous gift from Dr. Eleanor J. Mackie (RVC, London). The mouse monoclonal antibody HNK-1 recognizes a carbohydrate epitope and can be used to identify neural crest cells (e.g., Vincent and Thiery, 1984). The hybridoma was acquired from the American Type Culture Collection (Rockville, MD). Undiluted hybridoma supernatants were used. The anti-human plasma fibronectin polyclonal antibody was purchased from Accurate Chemical and Scientific (Westbury, NY) and diluted 1500, while TRITC and FITC-conjugated secondary antibodies were obtained from Hyclone Laboratories. Sections to be stained were rehydrated in PBS for 5 minutes and then blocked in 0.5% bovine serum albumin (BSA) in PBS for 15 minutes before secondary antibody, diluted 1 5 0 in PBS, was added. After 5 hours the slides were rinsed for 15 minutes in PBS and mounted in 50% glycerol in PBS containing 1 pg ml-' Hoechst nuclear dye (Boehringer Mannheim, Idianapolis, IN). If double-labeling studies were done, appropriate quantities of both primary and then secondary antibodies were applied t o the sections a t the same time. PAS staining was performed as described by Mowry (1963). Immunostaining of isolated PGCs was done in a manner similar to the immunohistochemistry described above, except the cells were stained in a microcentrifuge tube. HNK-1 was added to the tube for 1 hour; secondary antibodies were allowed to incubate for 20 minutes. The cells were rinsed in PBS by gentle centrifugation. pellet was then added to 200 pl of a 6.3% FicolUmedium 199 solution and layered on 900 pl of 16% Ficoll/medium 199 in a glass culture tube. To isolate the PGCs, the gradient was centrifuged at 800 x g for 30 minutes in a swinging bucket rotor. The PGCs collected from the interface were diluted 1.5-fold with medium 199 and centrifuged for 3 minutes at 400 x g. The PGC pellet was rinsed 3 times with medium 199 by centrifugation. In Situ Hybridization The chick tenascin-C-specific cDNA probe cTn8 (a generous gift, from Dr. Ruth Chiquet-Ehrismann, Basel) was previously described by Pearson et al. (1988). In situ hybridization was carried out as described by Tucker et al. (1994a). In brief, sections were cut as described above and prehybridized for 1 hour at room temperature in prehybridization buffer (Sigma) diluted 1:1with autoclaved deionized water and 20 mM 2-mercaptoethanol (Sigma). Sections were then dehydrated by dipping into 100% ethanol and air-dried. cTn8 was labeled with [35SldCTP(Amersham, Arlington Heights, IL) using a Prime-a-Gene kit (Promega, Madison, WI). Unincorporated nucleotides were removed using a G-50 spin column (Worthington Biochemical, Freehold, NJ) 5 x lo6 c.p.m. of boiled probe was suspended in 20 pl hybridization buffer (Sigma) containing 50%deionized formamide (Fisher Scientific, Pittsburgh, PA), 20 mM Tris pH 8.0, 0.1% sarkosyl (Fisher), and 1%dextran sulfate (Sigma), and applied to the section. The sections were allowed to incubate at 42°C for 18 hours. After removing the coverslips, the sections were washed with 1 x SSC and 20 mM 2-mercaptoethanol for 1 hour a t room temperature and 1 hour a t 42°C. The slides were allowed t o dry and then were processed by dipping in NTB2 autoradiography emulsion (Kodak, Rochester, NY). After developing, the slides were mounted in 50% glycerol in PBS and 1 pg ml-' Hoechst nuclear dye. Surgeries To dissect the HPC and PGC migratory mesenchyme, E3 chicks were taken from the egg, rinsed in PBS and placed in an agar-coated 10-cm petri dish. The aorta can be easily followed throughout the embryo a t this stage. To isolate the HPC migratory region, the mesenchyme around the aorta rostra1 to the vitelline vein was dissected free using tungsten needles. The mesenchyme through which the PGCs move was removed in a similar manner, but only from axial levels caudal t o the vitelline veins. The dissected tissue was transferred immediately to an autoclaved microcentriPGC Isolation fuge tube and placed in liquid nitrogen. Poly(A) RNA PGCs were isolated from stage 13-14 (Hamburger was isolated from the dissected tissue using a Micro and Hamilton, 1951) chick blood using methods modi- Fast Trak kit (Invitrogen, San Diego, CA). To induce ectopic PGC migration, the gonadal anlafied from Chang et al. (1992). Blood from the vitelline veins was collected using an ultrafine glass pipette and gen were removed prior to PGC colonization. Eggs were suspended in 1ml of medium 199 (Sigma). The suspen- set to obtain stage 12-13 embryos, and candled to desion was centrifuged at 200 x g for 3 minutes. The termine the position of the blastodisc within the egg. TENASCIN AND PRIMORDIAL GERM CELLS 445 ral crest cell migration in avian embryos using monoclonal antiWindows were made over this site using an electric body HNK-1. Dev. Biol. 115:44-53. sander, keeping the shell membrane intact. After M. (1988) Distribution and function of tenascin durswabbing the perimeter of the window and the mem- Bronner-Fraser, ing cranial neural crest development in the chick. J . Neurosci. h s . brane with 70% ethanol, the shell membrane was re21:135-147. moved. To improve the contrast between the embryo Chang, I.-K., Tagima, A,, Yasuda, Y., Chikamune, T., and Ohno, T. (1992) Simple method for isolation of primordial germ cells from and the surrounding membranes, the embryo was chick embryos. Cell Biol. Intl. Rep. 16:853-857. stained with neutral red. The vitelline membranes Chiquet, M., and Fambrough, D.M. (1984) Chick myotendinous antiover the operation site were opened. The gonadal angen I. A monoclonal antibody as a marker for tendon and muscle lagen were removed under a stereomicroscope by runmorphogenesis. J. Cell Biol. 98:1926-1936. ning sharp tungsten needles around the perimeter of Chiquet-Ehrismann, R., Mackie, E.J., Pearson, C.A., and Sakakura, T. (1986) Tenascin: An extracellular matrix protein involved in tisthe caudal third of the embryo and making a single sue interactions during fetal development and oncogenesis. Cell cross cut. The excised piece was lifted out using watch47:131-139. makers’ forceps. After moistening the embryo with a Chiquet-Ehrismann, R., Kalla, P., Pearson, C.A., Beck, K., and Chidrop of sterile saline, the egg was sealed with cellotape quet, M. (1988) Tenascin interferes with fibronectin action. Cell 53~383-390. and returned to the incubator. The embryo was incubated for one day, untuned. Control embryos under- Clawson, R.C., and Domm, L.V. (1962) Developmental changes in glycogen content of primordial germ cells in the chick embryo. Proc. went all procedures excluding the actual removal of the SOC.Exp. Biol. Med. 112:533-537. gonadal anlagen. Embryos were processed for section- Cormier, F., De Paz, P., and Dieterlen-Lievre, F. (1986) In uitro detection of cells with monocytic potentiality in the wall of the chick ing and staining as described above. In all, 8 embryos embryo aorta. Dev. Biol. 118:167-175. survived the surgery to be analyzed by PAS and imDieterlen-Lievre, F., and Martin, C. (1981) Diffuse intraembryonic munohistochemical staining. hemapoiesis in normal and chimeric avian development. Dev. Biol. 88:180-191. Reverse Transcriptase PCR Epperlein, H.H., Halfter, W., and Tucker, R.P. (1988) The distribution of fibronectin and tenascin along migratory pathways of the neural To determine which tenascin-C splice variants are crest in the trunk of amphibian embryos. Development 103:743present in the PGC and HPC migratory pathway, RT756. PCR from the isolated poly(A) RNA from the dissected Erickson, C.A., Loring, J.F., and Lester, S.M. (1989) Migratory pathways of HNK-1-immunoreactive neural crest cells in the rat emmesenchyme was performed with primers described in bryo. Dev. Biol. 134:112-118. Tucker et al. (1994b). These primers, TNSVA and ffrench-Constant, C., Hollingsworth, A., Heasman, J., and Wylie, C. TNSVB, correspond to the beginning of the fifth and (1991) Response to fibronectin of mouse primordial germ cells besixth fibronectin-type I11 repeats of chick tenascin-C; fore, during and after migration. Development 113:1365-1373. i.e., they span the variable domain (Tucker et al., Fujimoto, T., Ukeshima, A., and Kiyofuji, R. (1976) The origin, migration and morphology of the primordial germ cells in the chick 1994a). A 264 b.p. product corresponds to a tenascin-C embryo. Anat. Rec. 185:139-154. transcript without the variable domain. Products of Godin, I., and Wylie, C.C. (1991) TGF-PI inhibits proliferation and 540 b.p. and 1,060 b.p. correspond to transcripts encodhas chemotropic effect on mouse primordial germ cells in culture. Development 113:1451-1457. ing tenascin-C with 1 and 3 variable repeats, respectively. All nucleotides, enzymes and buffers were sup- Godin, I., Wylie, C., and Heasman, J. (1990) Genital ridges exert long-range effects on mouse primordial germ cell numbers and diplied in a reverse transcriptase PCR kit (Perkin Elmer, rection of migration in culture. Development 108:357-363. Nonvalk, CT). The reactions were carried out as rec- Halfter, W., Chiquet-Ehrismann, R., and Tucker, R.P. (1989) The efommended by the manufacturer. fect of tenascin and embryonic basal lamina on the behavior and ACKNOWLEDGMENTS This work was done in partial fulfillment of a doctoral degree a t the Department of Neurobiology and Anatomy at the Bowman Gray School of Medicine, with funds provided by the National Science Foundation (BNS-9021124) and the UCD School of Medicine. REFERENCES Adams, J.C., and Watt, F.M. (1993) Regulation of development and differentiation by the extracellular matrix. Development 117: 1183-1198. Ben Slimane, S., Houllier, F., Tucker, G., and Thiery, J.P. (1983) In vitro migration of avian hemopoietic cells to the thymus: Preliminary characterization of a chemotactic mechanism. Cell Diff. 13:l24. Borsi, L., Carnemolla, B., Nicolo, G., Spina, B., Tanara, G., and Zardi, L. (1992) Expression of different tenascin isoforms in normal, hyperplastic and neoplastic human breast tissues. Int. J. Cancer 52: 688-692. Bronner-Fraser, M. (1986) Analysis of the early stages of trunk neu- morphology of neural crest cells in vitro. Dev. Biol. 132:14-25. Hamburger, V., and Hamilton, H.L. (1951) A series of normal stages in the development of chick embryo. J. Morph. 88:49-92. Heasman, J., Hynes, R.O., Swan, A.P., Thomas, V., and Wylie, C.C. (1981) Primordial germ cells of Xenopus embryos: The role of fibronectin in their adhesion during migration. Cell 27:437-447. Kuwana, T., Maeda-Suga, H., and Fujimoto, T. (1986) Attraction of chick primordial germ cells by gonadal anlage in vitro. Anat. Rec. 215:403-406. Leprini, A., Querze, G., and Zardi, L. (1994) Tenascin isoforms: Possible targets for diagnosis and therapy of cancer and mechanisms regulating their expression. Persp. Dev Neurobiol. 2:117-123. Loring, J.F., and Erickson, C.A. (1987) Neural crest cell migratory pathways in the trunk of the chick embryo. Dev. Biol. 121:220-236. Lotz, M.M., Burdsal, C.A., Erickson, H.P., and McClay, D.R. (1989) Cell adhesion to fibronectin and tenascin: Quantitative measurements of initial binding and subsequent strengthening response. J. Cell Biol. 109:1795-1805. Mackie, E.J., Chiquet-Ehrismann, R., Pearson, C.A., Inaguma, Y., Taya, K., Kawarada, Y., and Sakakura, T. (1987) Tenascin is a stromal marker for epithelial malignancy in the mammary gland. Proc. Natl Acad. Sci. 844621-4625. Mackie, E.J., Halfter, W., and Liverani, D. (1988a) Induction oftenascin in wound healing. J . Cell Biol. 107:2757-2767. 446 ANSTROM AND TUCKER Mackie, E.J., Tucker, R.P., Halfter, W., Chiquet-Ehrismann, R., and Epperlein, H.H. (1988b) The distribution of tenascin coincides with the pathways of neural crest cell migration. Development 102:237250. Martin, C., Beaupain, D., and Dieterlen-LiBvre, F. (1980) A study of the development of the hematopoietic system using quail-chick chimeras obtained by blastoderm recombination. Dev. Biol. 75:303314. Mowry, R.W. (1963) The special value of methods that color both acidic and vicinal hydroxyl groups in the histochemical study of mucins with revised directions for the colloidal iron stain, the use of alcian blue 8GX and their combinations with the periodic acidSchiff reaction. Ann. N. Y. Acad. Sci. 106:402-423. Nakamura, M., Kuwana, T., Miyayama, Y., Yoshinaga, K., and Fujimoto, T. (1991) Ectopic colonization of primordial germ cells in chick embryos lacking the gonads. Anat. Rec. 229:109-115. Pearson, C.A., Pearson, D., Shibahara, S., Hofsteenge, J., and Chiquet-Ehrismann, R. (1988)Tenascin: cDNA cloning and induction by TGF-p. EMBO J. 7:2677-2981. Rickmann, M., Fawcett, J.W., and Keynes, R.J. (1985) The migration of neural crest cells and the growth of motor axons through the rostra1 half of the chick somite. J . Embryo]. Exp. Morph. 90:437455. Swift, C.H. (1914) Origin and history of the primordial germ cells in the chick. Am. J. Anat. 15483-516. Tan, S.-S., Krossin, K.L., Hoffman, S., and Edelman, G.M. (1987) Asymmetric expression in somites of cytotactin and its proteoglycan ligand is correlated with neural crest cell distribution. Proc. Natl Acad. Sci. U.S.A. 84:7977-9781. Thiery, J.P., Duband J.L., Tucker, G.C. (1985) Cell migration in the vertebrate embryo: Role of cell adhesion and tissue environment in pattern formation. Annu. Rev. Cell Biol. 1:91-113. Tucker, G.C., Aoyama, H., Lipinski, M., Tursz, T., and Thiery, J.P. (1984) Identical reactivity of monoclonal antibodies HNK-1 and NC-1: Conservation in vertebrates on cells derived from the neural primordium and some leukocytes. Cell Diff. 14:223-230. Tucker, R.P. (1991) The distribution of Jl/tenascin and its transcript during the development of the avian cornea. Differentiation 485966. Tucker, R.P., and McKay, S.E. (1991) The expression of tenascin by neural crest cells and glia. Development 112:1031-1039. Tucker, R.P., Spring, J., Baumgartner, S., Martin, D., Hagios, C., Poss, P.M., and Chiquet-Ehrismann, R. (1994a) Novel tenascin variants with a distinctive pattern of expression in the avian embryo. Development 120:637-647. Tucker, R.P., Brunso-Bechtold,J.K., Jenrath, D.A., Khan, N.A., Poss, P.M., Sweatt, A.J., and Xu, Y. (199413) Cellular origins of tenascin in the developing nervous system. Persp. Dev. Neurobiol. 2:89-99. Urven, L.E., Abbot, U.K., and Erickson, C.A. (1989) Distribution of extracellular matrix in the migratory pathway of avian primordial germ cells. Anat. Rec. 224:14-21. Vincent, M., and Thiery, J.P. (1984) A cell surface marker for neural crest and placodal cells: Further evolution in peripheral and central nervous system. Dev. Biol. 103:468-481. Vincent, M., Duband, J.L., and Thiery, J.P. (1983) A cell surface determinant expressed early on migrating avian neural crest cells. Dev. Brain Res. 9:235-238.