We have determined, by immunohistochemical and biochemical techniques, the distribution of an endogenous β-D-galactoside-blnding lectin between the early primitive streak stage and the 5th day of embryonic development of the chick. The lectin, which was purified from the pectoral muscle of 16-day-old chick embryos, migrates on SDS–PAGE as a single polypeptide of relative molecular mass 15 × 103. Antibodies to this pure lectin interact with the 15K (K=103Mr) polypeptide as well as with a 6.5K polypeptide; this second component appears to be antigenically related to the 15K lectin, as antibodies affinity purified on the 15K band recognize both polypeptides.

In early stages of development, lectin immunoreactivity was present in most cells of the epiblast and hypoblast in the region of the primitive streak, while towards the edge of the area pellucida the epiblast was stained less intensely. During gastrulation, strong immunoreactivity was present also in migrating cells and in the mesoblast, while at the margin of the area pellucida the epiblast was negative.

Up to the 10-somite stage, lectin immunoreactivity was present in the somites, neural tube and presumptive cardiac region; the non-neural ectoderm and the extra-cellular matrix were not labeled; the predominant immunoreactive component at this stage of development was the 6.5K polypeptide. Later in development, the lectin immunoreactivity gradually disappeared from the dermamyotome and nervous system to reappear conspicuously as soon as a differentiated myotome could be detected. Immunoreactivity was very high in the myotome, skeletal and cardiac muscles and transient in smooth muscles. The only region of the nervous system that continued to express the lectin throughout development was the trigeminal (semilunar) ganglion; in all other regions of the nervous system, the lectin immuno-reactivity disappeared early in development to be re-expressed only much later. The lining epithelium of the digestive tract and other endodermal derivatives expressed the lectin transiently. In the extraembryonic membranes, immunoreactivity to the lectin was observed in the yolk sac and in both layers of the amnion.

The striking regulation of the expression of this endogenous lectin suggests that its functions are linked to cell proliferation and/or to the selective expression of a developmentally-timed cell phenotype.

The correct morphological development and histological differentiation of an embryo is assured by the coordinated spatiotemporal modulation of the expression of a number of molecular elements that determine the mode of interaction of each cell with its local environment and/or its differentiation into a final phenotype. During the last decade, interest has developed on a class of β-D-galactoside-binding lectins that are expressed in critical moments of vertebrate embryonic development (Teichberg, 1978; Barondes, 1984; Zalik et al. 1987). These proteins are worthy of the attention of developmental biologists as they share some characteristics with a number of morphoregulatory proteins. (1) The peak of their expression is often coincident with critical events in embryogenesis such as gastrulation (Zalik et al. 1987), myoblast fusion (Podleski and Greenberg, 1980), the establishment of neuro-muscular contacts (Teichberg, 1978) and development of sensory neurons (Regan et al. 1986). (2) They are often secreted in the extracellular space (Beyer and Barondes, 1982a; Barondes and Haywood-Reid, 1981) and possess binding sites for carbohydrates present at the cell surface and in the extracellular matrix (Levi and Teichberg, 1981; Beyer and Barondes, 1980). (3) Exogenously-added lectin as well as its competitive inhibitors affect intercellular interactions (Cook et al. 1979; Milos and Zalik, 1981). (4) They are strongly conserved during evolution (Levi and Teichberg, 1982; Paroutaud et al. 1987; Hirabayashi et al. 1987; Hira-bayashi and Kasai, 1988; Raz et al. 1988) and some of them apparently derive from a process of gene duplication that has created several tissue-specific variants (Ohyama and Kasai, 1988). (5) Malignant cell lines in culture overproduce some of these lectins (Gabius, 1987; Raz et al. 1987a,b).

These and other properties have brought about speculations as to the possible involvement of these proteins in a number of processes such as cell adhesion (Harrison and Chesterton, 1980; Beyer and Barondes, 1982a; Meromsky et al. 1986), myoblast fusion (Gartner and Podleski, 1975; MacBride and Przybylski, 1980), synapse formation (Regan et al. 1986), chondrogenesis (Matsutani and Yamagata, 1982), dermal condensation (Kitamura, 1981) and thymocyte maturation (Levi and Teichberg, 1985). However, because of the lack of convincing evidence, the biological functions of animal β-D-galactoside-specific lectins are still largely unknown.

In the chick, at least four different β-D-galactoside-binding lectins have been described. The first to be identified is present at high levels in the muscle of 16-day-old chick embryos (Nowak et al. 1977; Den and Malinzak, 1977). Known as chicken lactose lectin I (CLLI) or 15K β-D-galactoside-specific chicken lectin, it is composed of two identical subunits of relative molecular mass 15X103 assembled in an homodimer of apparent relative molecular mass 30–31×103. The amino acid sequence of the 15K chicken lectin has not yet been established, but it is expected to share some homology with the sequence of electrolectin, a lectin present in the electric organ of the electric eel Electrophorus electricus with which it cross-reacts immunologically (Levi and Teichberg, 1982). A second lectin, initially identified in the adult chicken intestine (Beyer et al. 1980), has a relative molecular mass of 14×103 and is known as chicken lactose lectin II (CLLII) or 14K β-galactoside-specific lectin. CLLI and CLLII are different proteins as they do not cross-react immunologically (Beyer et al. 1980) and have a similar but not identical carbohydrate substrate specificity (Barondes, 1984). During embryonic development, the 14K lectin is found in several tissues but is especially enriched in the kidney and is not expressed by muscle tissues (Beyer and Barondes, 19826). The complete amino acid sequence of the 14K chicken lectin (Hirabayashi et al. 1987) and that of its encoding gene have been established (Ohyama and Kasai, 1988); it contains three internal repeats coded by three different exons; this has led to the suggestion that the 14K lectin-encoding gene has developed by a process of gene replication from an ancestral smaller gene.

In addition to the 14K and the 15K lectins, the existence of two other lectins, appearing at very early stages of embryonic development, has also been reported (Zalik et al. 1987). They have relative molecular masses of 6.5×103 and 70×103 respectively, are secreted by the èxtraembryonic endoderm and have possible membrane binding sites in cells from the area pellucida and extra embryonic endoderm. Their structure and possible relations with the other chicken lectins have not yet been established.

Although the distribution of some of these lectins has been studied at specific stages of development and in specific organs (Beyer and Barondes, 1982a,b; Zalik et al. 1982), no detailed description of the modulation of their expression during critical moments of morphogenesis and organogenesis has yet appeared.

In this paper, we report the complete immunohisto-chemical map of expression of the 15K β-D-galactoside-specific lectin during the first part of chicken embryonic development. Although our antibodies were prepared against the pure 15K molecule, they did also recognise a 6.5K immunoreactive component that was predominant at very early stages of development, and was still expressed, but at lower levels, in more differentiated tissues. The 15K and 6.5K polypeptides appear to be antigenically related as they are both recognized by antibodies affinity purified on the 15K band. Our results shed light on the tissue-specific changes in expression of the 15K β-D-galactoside-specific lectin; particularly striking are the high levels of expression of the lectin in myogenic tissue since their first differentiation. This lectin is not present in tissues where the 14K lectin is prevalent such as the skin, kidney and thymus, suggesting the existence of different regulatory processess of expression of lectins sharing similar carbohydrate specificity.

Materials

Rabbit antibodies to purified chicken skin type I and III collagens were purchased from the Pasteur Institute (Lyon, France). A rabbit antibody against Keyhole Limpet Hemo-cyanin was kindly provided by Dr François Radvanyi (Ecole Normale Superieure, Paris). All other reagents were from Sigma, France.

Embryos

White Leghorn chicken embryos were used throughout the study. Eggs were incubated at 37 °C in a humidified air chamber with periodic rotation. Development of embryos with full number of somites was determined by reference to the staging series of Hamburger and Hamilton (1951). At other stages, ages were estimated by the number of somite pairs.

Lectin purification and antibody preparation

The lectin was purified from extract of pectoral muscles of 16-day-old chick embryos as described (Levi and Teichberg, 1981; Nowak et al. \9T1). Briefly, pectoral muscles of 16-day-old chick embryos were excised and homogenized in buffer A (75 mM-Na2HPO4, 75mM-NaCl, 4 mM-EDTA, 300mM-lactose, 14 mM-2-mercaptoethanol, pH7.4) at 4 °C in a Sorvall omnimixer (100ml of Buffer A per 100g of tissue). The homogenate was centrifuged at 30000g for 30 min, and the supernatant further centrifuged for 1 h at 100 000 g. The clear supernatant was dialyzed five times against 20 volumes of 0.9% NaCl, 14 mM-2-mercaptoethanol in order to eliminate the lactose. During dialysis, some material precipitated and was removed by centrifugation (100 000 g, 60min). The clear supernatant was then applied to a lactosyl-Sepharose 4B affinity column (Levi and Teichberg, 1981). After loading, the column was washed with 0.9% NaCl, 14 mM-2-mercapto-ethanol until no absorbance at 280 nm could be detected in the effluent. The lectin was then eluted from the column upon application of a solution of 0.3 M-lactose, 0.9% NaCl, 14mw-2-mercaptoethanol. The elution of the lectin was followed by measuring the absorbance at 280 nm of the eluted fractions. The fractions corresponding to the protein peak were collected and either kept in the cold or extensively dialyzed against 0.9% NaCl, 14mM-2-mercaptoethanol. The agglutinating activity was measured on trypsinized rabbit erythrocytes as described (Levi and Teichberg, 1981).

Anti-lectin antibodies were prepared by injecting a solution of the purified lectin (100 βg), emulsified in complete Freunds adjuvant, into New Zealand white rabbits. Each rabbit was injected subcutaneously in five sites and intraperitoneally once. A second immunization was administered 2 weeks later using the same procedure. The antigen was then readministered at 1 month intervals emulsified in incomplete Freunds adjuvant. The rabbits were bled 2 weeks after the third immunization. The antisera were kept frozen at −20°C. For immunohistochemistry, a purified IgG fraction prepared from these antisera (Brackenbury et al. 1977) was used.

Affinity-purified anti-lectin antibodies were prepared according to Olmsted (1981) and Zalik et al. (1987). The purified lectin was resolved by SDS–PAGE (Laemmli, 1970) and transferred to nitrocellulose paper (Towbin et al. 1979). A vertical nitrocellulose strip was cut and stained with 0.1% Amido Black to localize the position of the 15K lectin. For affinity purification the area of nitrocellulose containing the 15K lectin was cut out, blocked with 5 % BSA in PBS (PBS–BSA) for 1 h and incubated with the antiserum diluted 1:100 in PBS–BSA at room temperature overnight. After washing once with PBS-BSA, antibodies were eluted from the strip with 200 mM-glycine–HCl, pH2.8 for 5min; the eluted material was immediately brought to pH 7.4 with 1 M-NaOH. The procedure was repeated twice and the eluted material was pooled and dialyzed against PBS.

Immunohistochemistry

All embryos, with the exception of those at the latest stages of development, were embedded in paraffin and prepared for staining as described (Levi et al. 1987). Briefly, whole embryos were frozen in isopentane cooled in liquid nitrogen and immediately immersed in methanol at −70 °C. After 72 h, the embryos were serially transferred to methanol equilibrated at −20°, 4°, and 20°C leaving them to equilibrate for at least 3h at each step. The samples were then immersed twice in xylene for 15 min and transferred to a solution of 50% Paraplast (Monoject Scientific, St. Louis, MO, USA) in xylene at 45°C for 30min, infiltrated twice with Paraplast in a vacuum oven at 45°C for 1 h, and embedded in Paraplast. Sections 10pm thick were cut at room temperature, floated on a 37°C water bath and collected on uncoated glass slides. Older embryos (13 day) were fixed at room temperature with 2.7% formaldehyde in phosphate-buffered saline (PBS; 137mM-NaCl, 3mM-KCl, 8mM-Na2HPO4, 1.5mM-KH2PO4, pH7.4), infiltrated overnight with 18% sucrose in PBS, frozen in Tissue Tek (Lab-Tek Products) with liquid nitrogen and sectioned in a Jung Reichert Frigocut 2800 cryostat. 10μm sections were collected on gelatine-coated slides. For immunofluorescent staining, the deparaffinized sections were incubated sequentially with 5% normal goat serum in PBS, the primary antibody (1:100 dilution of a 5 mg ml−1 IgG solution in PBS, 5 % normal goat serum) and a Rhodamine-conjugated goat anti-rabbit second antibody (1:100 in PBS, 5% normal goat serum). Sections were observed with a Leitz epifluorescence microscope. Controls were performed either substituting the primary antibody with other rabbit sera directed against nonrelevant antigens or with the diluted primary antibody preincubated with the purified lectin (10 βg ml−1 final dilution) for 1 h.

Western blots

To determine the specificity and molecular weights of the proteins recognized by the antibodies, tissues were dissected and directly homogenized in a glass-to-glass Dounce homogenizer with boiling SDS sample buffer (Laemmli, 1970) not containing 2-mercaptoethanol. The homogenate was centrifuged for 3 min on a table 5415C Eppendorf centrifuge at 14 000 rev min−1, and the protein concentration of the clear supernatant determined and normalized to 1 mg ml−1 by addition of further SDS sample buffer. After addition of 2-mercaptoethanol and bromophenol blue, samples were boiled three minutes, resolved by SDS-PAGE (Laemmli, 1970) and transferred to nitrocellulose paper (Towbin et al. 1979). After blocking the nonspecific binding sites with 3% w/w ovalbumin (Sigma) in PBS supplemented with 0.1% Triton X-100 (blocking solution), the presence of immuno-reactive species was revealed by autoradiography after incubating 6h with 50 μl of anti-lectin antiserum in 20 ml of blocking solution, followed by 3h with 2×106 ctsmin−1 l25l-protein A (Amersham) in 20 ml blocking solution.

Reactivity of the antibodies

We have purified to homogeneity the 15K β-D-galactoside-specific lectin by affinity chromatography from an extract of 16-day chick embryo pectoral muscle. The lactose-eluted fraction contained only one polypeptide component of 15K, as revealed by Coomassie Blue staining (Fig. 1, lane 1). Following dialysis, this fraction displayed most (95%) of the lactose-blockable agglutinating activity applied onto the affinity column. This fraction was used to prepare a polyclonal antibody that was later utilized to study the distribution of immunoreactive material (i.e. of the lectin) during chick embryo development.

Fig. 1.

Lane 1: material retained on lactosyl-Sepharose 4B from an extract of 16-day chick embryonic pectoral muscle, resolved on a 17.5 % acrylamide gel and revealed by Coomassie Blue staining. The lectin appears as a single component of 15K. Lanes 2,3,5: immunoblots with anti-15K IgGs. Purified 15K chick lectin (lane 2); protein extract of 16-day chick embryo pectoral muscle (lane 3); extract of 10-somite chick embryos (lane 5). Lane 4: immunoblot of protein extract of 16-day chick embryo pectoral muscle with affinity-purified anti-15K antibodies. Proteins were separated by SDS–PAGE on a 17.5% polyacrylamide gel, transfered to nitrocellulose paper and revealed with 125I-Protein A and autoradiography. M. markers are in the order: chymotrypsinogen, 25.7×103; lactalbumine, 18.4×103 and lysozyme, 14.5×103.

Fig. 1.

Lane 1: material retained on lactosyl-Sepharose 4B from an extract of 16-day chick embryonic pectoral muscle, resolved on a 17.5 % acrylamide gel and revealed by Coomassie Blue staining. The lectin appears as a single component of 15K. Lanes 2,3,5: immunoblots with anti-15K IgGs. Purified 15K chick lectin (lane 2); protein extract of 16-day chick embryo pectoral muscle (lane 3); extract of 10-somite chick embryos (lane 5). Lane 4: immunoblot of protein extract of 16-day chick embryo pectoral muscle with affinity-purified anti-15K antibodies. Proteins were separated by SDS–PAGE on a 17.5% polyacrylamide gel, transfered to nitrocellulose paper and revealed with 125I-Protein A and autoradiography. M. markers are in the order: chymotrypsinogen, 25.7×103; lactalbumine, 18.4×103 and lysozyme, 14.5×103.

The anti-lectin IgGs were found to recognize a single polypeptide of 15K in Western blots of the purified material against which it had been prepared, further demonstrating the purity of the lectin preparation (Fig. 1 lane 2). However, when this antibody was tested on blots of chick pectoral muscle proteins separated by SDS–PAGE (Fig. 1, lane 3), it recognized the 15K polypeptide as well as a second low relative molecular mass component of 6.5×103. A similar, but much fainter, pattern of immunoreactivity was observed in blots of heart and intestine proteins (not shown) from 13-day-old embryos, while virtually no reactivity was detected in immunoblots of gizzard and brain proteins from 16-day embryos. Interestingly, when extracts of 10-somite chick embryos were tested in Western blots (Fig. 1 lane 5) the predominant immunoreactive material was the 6.5K polypeptide, while the 15K polypeptide was barely visible; a third, but faintly labeled component at 7.5K was present in these extracts.

In order to better understand the relation between the 15K and 6.5K polypeptides, we have resolved the 15K pure protein by SDS–PAGE, transfered the protein on a nitrocellulose sheet and affinity purified some of the antibodies on the isolated band. When tested on blots of chick pectoral muscle proteins these 15K-specific antibodies still recognized both the 15K and 6.5K components, indicating that the two polypeptides are antigenically related (Fig. 1, lane 4).

Early expression of anti-15K lectin immunoreactivity; distribution during gastrulation

We have observed the presence of a strong lectin immunoreactivity at the earliest stages of development investigated. At stage 3 of Hamburger and Hamilton, the lectin was present in cells of the epiblast and in cells that had ingressed from the developing primitive streak (Fig. 2A); the epiblast, however, was positive only around the primitive streak and was negative more laterally. At later stages of gastrulation, lectin immuno-reactivity continued to be present on the epiblast surrounding the primitive streak and on ingressed mesoblastic cells (Fig. 2 C). Laterally to the primitive streak, the mesoblast and the endoblast were positive, while the epiblast was not (Fig. 2 D). Within the limits of resolution permitted by light immunohistofluorescence, the staining appeared to be predominantly intracellular. Most of the immunoreactivity was confined to the area pellucida; only low levels of staining could be detected between yolk granules in the area opaca.

Fig. 2.

Distribution of the 15K lectin immunoreactivity during gastrulation. (A,B), ahort primitive streak (stage 3 of Hamburger and Hamilton); (C,D), regressing primitive streak (stage 5 of Hamburger and Hamilton). (A–C), Areas within the primitive streak; (D), areas lateral to the primitive streak at the border between the area pellucida and the area opaca. Lectin immunoreactivity was present in the epiblast, in ingressed mesoblastic cells and in the endoblast in the region of the primitive streak (A,C); more laterally the mesoblast and the endoblast were still positive while the epiblast was not (D). (A,C,D) Stained with anti-15K lectin antibodies; (B) stained with anti-KLC antibodies, ao, area opaca; en, endoblast; ep, epiblast; m, mesoblast; ps, primitive streak. Bar, 50 μm.

Fig. 2.

Distribution of the 15K lectin immunoreactivity during gastrulation. (A,B), ahort primitive streak (stage 3 of Hamburger and Hamilton); (C,D), regressing primitive streak (stage 5 of Hamburger and Hamilton). (A–C), Areas within the primitive streak; (D), areas lateral to the primitive streak at the border between the area pellucida and the area opaca. Lectin immunoreactivity was present in the epiblast, in ingressed mesoblastic cells and in the endoblast in the region of the primitive streak (A,C); more laterally the mesoblast and the endoblast were still positive while the epiblast was not (D). (A,C,D) Stained with anti-15K lectin antibodies; (B) stained with anti-KLC antibodies, ao, area opaca; en, endoblast; ep, epiblast; m, mesoblast; ps, primitive streak. Bar, 50 μm.

Distribution of anti-15K lectin immunoreactivity up to the 10-somite chick embryo

We have observed a strong and selective immunoreactivity to anti-lectin antibodies in most cells of the somites, neural tube and notochord up to the 10-somite embryo, while the non-neural ectoderm and endoderm were only faintly stained (Fig. 3A,C). The splanchnic mesoderm in the cardiac region and the gut endoderm were brightly stained (Fig. 3E), but the somatic mesoderm and the extracellular matrix were not. As the 6.5K immunoreactive polypeptide is predominant at this stage of development, it is reasonable to assume that most of the staining signal originates from this polypeptide. The staining pattern of the 15K-specific affinity-purified antibodies was, however, virtually identical to that of the total IgGs (Fig. 3A,C).

Fig. 3.

Distribution of the 15K lectin immunoreactivity in a 10-somite chick embryo. (A,C,E) Strong immunoreactivity is present in the somites, neural tube, notochord and splanchnic mesoderm of the cardiac primordia. The ectoderm is only faintly stained and the endoderm is positive only in the portal region; the extracellular matrix is not stained. A and E are stained with the total anti-15K IgGs while C is stained with affinity-purified anti-15K antibodies. (B) A section adjacent to that of A stained with anti-collagen type 1 antibodies to reveal the presence of the extracellular matrix. (D and F) Sections adjacent to those of C and E stained respectively with anti-KLH antibodies (D), and with anti-!5K lectin antibodies preadsorbed with the purified lectin (F). e, ectoderm; en, endoderm; n, notochord;nt, neural tube;s, somite; sm, splanchnic mesoderm. Bar: 50 μm

Fig. 3.

Distribution of the 15K lectin immunoreactivity in a 10-somite chick embryo. (A,C,E) Strong immunoreactivity is present in the somites, neural tube, notochord and splanchnic mesoderm of the cardiac primordia. The ectoderm is only faintly stained and the endoderm is positive only in the portal region; the extracellular matrix is not stained. A and E are stained with the total anti-15K IgGs while C is stained with affinity-purified anti-15K antibodies. (B) A section adjacent to that of A stained with anti-collagen type 1 antibodies to reveal the presence of the extracellular matrix. (D and F) Sections adjacent to those of C and E stained respectively with anti-KLH antibodies (D), and with anti-!5K lectin antibodies preadsorbed with the purified lectin (F). e, ectoderm; en, endoderm; n, notochord;nt, neural tube;s, somite; sm, splanchnic mesoderm. Bar: 50 μm

In order to test the specificity of the staining pattern, a series of control experiments was performed: antibodies to chick collagen type I used at the same dilution as the anti-lectin antibodies were found to stain the extracellular matrix exclusively and did not stain the structures stained by the anti-lectin antibodies (Fig. 3B) (Duband and Thiery, 1987). In addition, neither anti-Keyhole Limpet Hemocyanin antibodies nor anti-lectin antibodies preabsorbed with purified 15K chicken lectin gave fluorescent signals above background (Fig. 3D,F). It is difficult to decide, at the level of resolution of fluorescent optical microscopy, if the anti-lectin staining was associated with the cell membrane or/and with the cytoplasm.

Development of the myotome

The patterns of lectin immunoreactivity were followed during the various phases of development of the myotome. During somite epithelialization and disaggregation, the staining intensity observed in the earlier stages of development (see Fig. 3) was found to decrease gradually so that only low levels of reactivity to the lectin could be observed in the caudal somites of 18-somite to 45-somite embryos (Figs4B, 5D). More rostrally, a strong immunoreactivity was observed only in somites that had developed a differentiated myotome: in the 24-somite embryo only the first 4-5 somites were positive (Fig. 4A), while in the 35-somite embryo, the first 15 somites were positive (Fig. 4C-E). The somite counting was performed in stained serial para-sagittal sections of the embryos. At the 35-somite stage, the sclerotome and dermatome were negative in the most rostral somites (Fig. 4C) but maintained a low level of immunoreactivity in more caudal somites (Fig. 4 D,E). In the 45-somite embryo, the anti-lectin antibodies labeled exclusively the myotomal portion of the somites (Fig. 5A,B,C) while no reactivity was present in the dermatome, sclerotome, notochord and skin. Identical patterns of reactivity were observed using 15K-specific affinity-purified antibodies or the total anti-lectin IgG fraction. The staining observed in the myotome appears to be predominantly intracellular and associated with developing primary myotubes. However, the exact cellular location of the immuno-reactive material could not be determined with the light microscope.

Fig. 4.

Distribution of the 15K lectin immunoreactivity in the myotome of a 24-somite embryo (A,B) and a 35-somite embryo (C–E). In the 24-somite embryo, the first 4–5 somites have developed a clearly differentiated myotome with a strong staining for the lectin; more caudal somites (B) show only a weak and diffuse pattern of staining; some staining persists in the neural tube and notochord. In the 35-somite embryo, only the somites that have developed a clearly differentiated myotome are strongly stained (C–D); the dermatome and sclerotome of these somites are not stained; the dermamyotome of more caudal somites and the sclerotome display a diffuse faint staining pattern. The neural tube is stained only in its most caudal regions (E), while in more rostral sections some staining persists only ventrally, d, dermatome; dm, dermamyotome; nt, neural tube; my, myothome; sc, sclerotome; arrows, limit of the myotome. Bar: 50 μm.

Fig. 4.

Distribution of the 15K lectin immunoreactivity in the myotome of a 24-somite embryo (A,B) and a 35-somite embryo (C–E). In the 24-somite embryo, the first 4–5 somites have developed a clearly differentiated myotome with a strong staining for the lectin; more caudal somites (B) show only a weak and diffuse pattern of staining; some staining persists in the neural tube and notochord. In the 35-somite embryo, only the somites that have developed a clearly differentiated myotome are strongly stained (C–D); the dermatome and sclerotome of these somites are not stained; the dermamyotome of more caudal somites and the sclerotome display a diffuse faint staining pattern. The neural tube is stained only in its most caudal regions (E), while in more rostral sections some staining persists only ventrally, d, dermatome; dm, dermamyotome; nt, neural tube; my, myothome; sc, sclerotome; arrows, limit of the myotome. Bar: 50 μm.

Fig. 5.

Distribution of the 15K lectin immunoreactivity in somites and nervous system of a 45-somite embryo. At this stage of development, the staining is confined predominantly to the myotome; no staining is observed in the dermatome, sclerotome or dermamyotome of more caudal somites (D). The nervous system is mostly unstained; only few cells are stained in dorsal root ganglia and in the ventral region of the neural tube. (A,C) Transverse section; (B) horizontal section; (A,B,D) stained with total anti-15K IgGs, (C) stained with affinity-purified anti-15K IgGs. d, dermatome; dm, dermamyotome; drg, dorsal root ganglion; my, myotome; nt, neural tube; sc, sclerotome. Bar: 50 μm.

Fig. 5.

Distribution of the 15K lectin immunoreactivity in somites and nervous system of a 45-somite embryo. At this stage of development, the staining is confined predominantly to the myotome; no staining is observed in the dermatome, sclerotome or dermamyotome of more caudal somites (D). The nervous system is mostly unstained; only few cells are stained in dorsal root ganglia and in the ventral region of the neural tube. (A,C) Transverse section; (B) horizontal section; (A,B,D) stained with total anti-15K IgGs, (C) stained with affinity-purified anti-15K IgGs. d, dermatome; dm, dermamyotome; drg, dorsal root ganglion; my, myotome; nt, neural tube; sc, sclerotome. Bar: 50 μm.

Later development

In the 5-day chick embryo, the lectin immunoreactivity was predominantly expressed in developing muscles. The myotome still displayed a prominent level of staining (Fig. 6A). The mesenchyme of the developing limbs showed a low level of diffuse immunoreactivity above which groups of cells differentiating into skeletal muscles were brightly fluorescent (Fig. 6B). This immunoreactivity appeared again to be mostly intracellular, as developing myotubes were labeled uniformly independently of the plane of the section. The high level of expression of the lectin in the muscle persisted at least up to day 13 of embryonic development. At this stage, all skeletal muscles displayed a strong immunoreactivity (Fig. 6C,D) localized apparently to the intracellular compartment. Other authors have reported that after birth the lectin is eventually externalized and is found in the extracellular spaces between muscle fibers (Barondes and Haywood-Reid, 1981).

Fig. 6.

Distribution of the 15K lectin immunoreactivity at later stages of muscle development. In the 5 day embryo (A,B), very intense staining is observed in the myotome, while the dermatome and sclerotome are not stained (A). A strong immunoreaction can be observed in condensing muscles in the developing wing (B). In the 13-day embryo, all skeletal muscles are labeled; skin and thymus are negative. (C), neck muscle; (D), wing muscle, d, dermatome; ms, skeletal muscle; my, myotome; sc, sclerotome; t, thymus. Bar=50 μn.

Fig. 6.

Distribution of the 15K lectin immunoreactivity at later stages of muscle development. In the 5 day embryo (A,B), very intense staining is observed in the myotome, while the dermatome and sclerotome are not stained (A). A strong immunoreaction can be observed in condensing muscles in the developing wing (B). In the 13-day embryo, all skeletal muscles are labeled; skin and thymus are negative. (C), neck muscle; (D), wing muscle, d, dermatome; ms, skeletal muscle; my, myotome; sc, sclerotome; t, thymus. Bar=50 μn.

In the 13-day embryo, no staining was observed in the skin, kidney and thymus, organs in which the 14K chicken lectin is expressed at high levels (Beyer and Barondes, 1982b).

Cardiac muscle and smooth muscle

Up to the fifth day of development, the chick embryo heart showed a strong immunoreactivity to the anti-15K lectin antibodies. As already observed in the 10-somite embryo, the heart primordia was stained (Fig. 3E). Later in development, all the cardiac muscle tissues continued to be immunopositive together with the anterior and posterior mesocardium (Fig. 7A,B). Western blot analysis of 13-day chick embryonic heart proteins showed, as observed in skeletal muscle, the presence of a predominant band at 15K and a fainter component at 6.5K (not shown).

Fig. 7.

Distribution of the 15K lectin immunoreactivity in cardiac and smooth muscle. (A,B) 5-day chick embryo. All parts of the cardiac muscle are strongly labeled by anti-15K lectin antibodies; (A), mesocardium; (B), ventricule. The endocardium is not stained. Smooth muscles surrounding the gut, gizzard and esophagous are positive until the 45-somite stage (C, gizzard) then gradually lose the lectin immunoreactivity. Smooth muscles surrounding blood vessels are transiently expressing the lectin immunoreactivity. (D) 13-day chick embryo aortic arch in which concentric layers of smooth muscles are intercalated with connective tissue, ec, endocardium; me, mesocardium; vt, ventricule; sm, smooth muscle. Bar: 50 μm.

Fig. 7.

Distribution of the 15K lectin immunoreactivity in cardiac and smooth muscle. (A,B) 5-day chick embryo. All parts of the cardiac muscle are strongly labeled by anti-15K lectin antibodies; (A), mesocardium; (B), ventricule. The endocardium is not stained. Smooth muscles surrounding the gut, gizzard and esophagous are positive until the 45-somite stage (C, gizzard) then gradually lose the lectin immunoreactivity. Smooth muscles surrounding blood vessels are transiently expressing the lectin immunoreactivity. (D) 13-day chick embryo aortic arch in which concentric layers of smooth muscles are intercalated with connective tissue, ec, endocardium; me, mesocardium; vt, ventricule; sm, smooth muscle. Bar: 50 μm.

Smooth muscles were stained only transiently. The splanchnic mesoderm was immunopositive in early embryos (Fig. 3E) and remained faintly positive in regions surrounding the gut and the esophagus (see Fig. 8B). The gizzard was brightly positive in all stages up to the 45-somite stage (Fig. 7C), but the staining was found to be strongly diminished in the 5-day embryo and was below detection by day 13 of development. Smooth muscles surrounding blood vessels were also immunopositive; Fig. 7D shows the staining of a number of concentric layers of smooth muscle intercalated with connective tissue in the aortic arch of a 13-day embryo.

Fig. 8.

Distribution of the 15K lectin immunoreactivity in endodermal derivatives. (A,B,C) 45-somite chick embryo, sections at different levels of the digestive tract; the epithelium lining the gut and related organs displays clear lectin immunoreactivity, ip, intestinal portal; g, gut; ph, pharynx; sm, smooth muscle; arrows, lung buds. Bar: 50 μm.

Fig. 8.

Distribution of the 15K lectin immunoreactivity in endodermal derivatives. (A,B,C) 45-somite chick embryo, sections at different levels of the digestive tract; the epithelium lining the gut and related organs displays clear lectin immunoreactivity, ip, intestinal portal; g, gut; ph, pharynx; sm, smooth muscle; arrows, lung buds. Bar: 50 μm.

Endodermal derivatives

The epithelium lining the gut and related internal organs displayed a clear anti-lectin immunoreactivity since their first formation (Fig. 3E) and up to at least the 45-somite embryo (Fig. 8A,B,C). The staining appeared to be membrane-associated and displayed a clear polarization, being brighter in the luminal side of the epithelium. The lining of the intestinal portal (Fig. 8A), gut (Fig. 8B), gizzard (Fig. 7C), pharynx (Fig. 8C) and lung buds (Fig. 8B) were clearly positive. Liver and pancreas were only weakly positive.

Nervous system

After a transient immunopositivity at early stages of development (see Figs 2 and 3), the nervous system was generally negative up to the 5th day of embryonic development. Most of the immunoreactivity in the nervous system had disappeared in the 45-somite embryo although some cells in the ventral horn of the spinal cord and in the dorsal root ganglia displayed a low level of immunoreactivity (Fig. 5A). TTie only exception was the trigeminal (semilunar) ganglion that was positive since its first condensation in the 24-somite embryo, and remained positive throughout development (Fig. 9A,B,C). In later stages (13-day embryo), a diffuse immunoreactivity was observed in some regions of the central nervous system. The presence of a β-D-galactoside-specific lectin in the developing chick optic tectum has been previously reported; it is not present until the 7th day of embryonic development, but thereafter gradually increases in neurites and cell bodies (Gremo et al. 1978).

Fig. 9.

Distribution of the 15K lectin immunoreactivity in the semilunar ganglion. Staining of the trigeminal (semilunar) ganglion at different stages of development of the chick embryo. (A) 24-somite; (B) 45-somite; (C) 5 day embryo, tg, trigeminal ganglion. Bar: 50 μm.

Fig. 9.

Distribution of the 15K lectin immunoreactivity in the semilunar ganglion. Staining of the trigeminal (semilunar) ganglion at different stages of development of the chick embryo. (A) 24-somite; (B) 45-somite; (C) 5 day embryo, tg, trigeminal ganglion. Bar: 50 μm.

Extraembryonic membranes

The developing chick embryo is surrounded by a system of extraembryonic membranes that ensures a proper compartmentalization in the egg. Each of these membranes differentiates into specific tissue types; for example, the external layer of the amnion becomes a specialized type of smooth muscle (Romanoff, 1960). The amnion and the yolk sack were brightly stained by anti-lectin antibodies at all stages of development (Fig. 10A,B), while the chorion and the allantois remained negative. Both the external layer of the amnion, which eventually becomes smooth muscle, and the internal one, which remains a limiting epithelium, were labeled with anti-lectin antibodies.

Fig. 10.

Distribution of the 15K lectin immunoreactivity in the extraembryonic membranes. 5-day chick embryo. (A) amnion; (B) yolk sack. Both the internal and the external layers of the amnion are labeled with anti-lectin antibodies; at this stage, the external layer (filled arrows), of mesodermal origin, begins to differentiate in smooth muscle, while the internal one (open arrows), of ectodermal origin, becomes a limiting epithelia. The yolk sack is labeled at all levels, a, amnion; sk, skin; ys, yolk sack. Bar: 50 μm

Fig. 10.

Distribution of the 15K lectin immunoreactivity in the extraembryonic membranes. 5-day chick embryo. (A) amnion; (B) yolk sack. Both the internal and the external layers of the amnion are labeled with anti-lectin antibodies; at this stage, the external layer (filled arrows), of mesodermal origin, begins to differentiate in smooth muscle, while the internal one (open arrows), of ectodermal origin, becomes a limiting epithelia. The yolk sack is labeled at all levels, a, amnion; sk, skin; ys, yolk sack. Bar: 50 μm

In this paper, we have described the tissue distribution of the 15K β-D-galactoside-specific lectin during the early development of the chicken embryo. Our results are as follows. (1) Antibodies prepared against the pure 15K. protein recognise the latter protein together with a second component of 6.5K; this low molecular weight component is prevalent at the early stages of development (up to 10 somites), while the 15K prevails in more differentiated tissues. Antibodies affinity purified on the 15K band still recognize both components, indicating that these share common epitopes and are antigenically related. The anti-15K lectin antibodies did not crossreact with the 14K chicken lectin that is strongly expressed in the embryonic kidney and skin (Beyer et al. 1980) or with the 70K lectin detected in early chicken gastrulas (Zalik et al. 1987). The absence of immunological cross-reactivity between the 14K and the 15K chicken lectins has been already reported by other authors (Beyer and Barondes, 19826). (2) Lectin immunoreactivity is present from stage 3 of Hamburger and Hamilton (1951) in the epiblast surrounding the primitive streak and in ingressing mesoblastic cells; more lateral epiblast is negative. (3) Up to the 10-somite stage, all cells of the somites, neural tube, notochord, splanchnic mesoderm and gut endoderm are immunostained. The ectoderm, somatic mesoderm and endoderm are not. This pattern of staining is probably ascribable to the 6.5K immunoreactive protein that is prevalent at this stage of development. (4) The most striking changes in lectin expression occur during somite differentiation; while the sclerotome and dermatome rapidly lose their immunoreactivity, a very bright staining is observed as soon as a differentiated myotome is detected. Thus, in the 24-somite embryo, the first 4–5 somites are strongly labeled and in the 35-somite embryo, the first 15. Skeletal, cardiac and smooth muscles are all strongly lectin-immunopositive. However, the staining in smooth muscles is transient; for example, the gizzard is no longer labeled at the 5th day of embryonic development. At the stages that were studied, the staining in the muscle appears to be predominantly associated with the cytoplasm. It is relevant to mention that the lectin distribution in myogenic tissues is almost exactly coincident with that of the epitopes of 13F4, a monoclonal antibody that recognises a specific marker of early myogenic cells and differenting muscles from the avian embryo (Rong et al. 1987). Both anti-lectin antibodies and 13F4 stain early myotomes, skeletal muscles, cardiac muscle and smooth muscle; furthermore, the number of somites stained at different stages of development is virtually identical. (5) The lectin is expressed in the endoderm-derived lining epithelium of the digestive tract and related organs (e.g. lung buds). The gut endoderm is positive from its first invagination in the anterior portal. The staining is apparently associated with the cell membrane and appears to be polarized, since the luminal aspect of the epithelium is the most intensely labeled region. The liver and the kidney show only very low levels of staining. (6) In the nervous system, the staining of the early neural tube is only transient. However, cells in the ventral region of the spinal cord, in the dorsal root ganglia and most cells and fibers of the trigeminal ganglion continue to express the lectin. Later in development (13-day embryo) the lectin is re-expressed in most of the nervous system but in a diffuse pattern. (7) Both layers of the allantois and the yolk sack are strongly labeled by the anti-lectin antibody, while no staining is observed in the chorion and amnion.

As mentioned above, the patterns of staining can be accounted for by the presence of either the 15K component, the 6.5K component, or both. One may suggest that the 6.5K immunoreactive component recognized by the anti-15K antibody is identical to the 6.5K β-D-galactoside-specific lectin present in the gastrulating chick blastoderm (Zalik et al. 1987). However, one ought to account for the fact that the 6.5K component did not copurify with the 15K lectin. One possibility is that methodological differences in the purification procedures (e.g. prolonged dialysis) prevented the copurification of the two proteins. Furthermore, the pattern of staining that we detect at comparable stages of development is different from that reported for the 6.5K lectin (Zalik et al. 1987) suggesting that the cross-reactive component detected by our antibodies might be different from the described lectin. The existence of a 6.5K protein, immunoreactive with anti-15K antibodies, is of particular interest in view of the recently established structure of the gene encoding the 14K β-galactoside-specific lectin (Ohyama and Kasai, 1988). This gene is composed of a small exon followed by three major exons that share a high degree of homology and are probably derived from a process of gene duplication from a small ancestral gene that may have coded for a 4 to 8K protein. Although the structure of the 15K chicken lectin is not yet known, published data indicates that the 15K and the 14K-β-galactoside-specific chicken lectins must be related: the 15K lectin has a strong immunological cross-reactivity with electrolectin (Levi and Teichberg, 1982), a similar lectin present in the electric organ of the electric eel, which, in its turn, displays more than 39% amino acid sequence homology with the 14K chicken lectin (Paroutaud et al. 1987). It is possible, therefore, that both chick lectins have derived from a process of duplication of the same ancestral gene and that the 6.5K component may result from a process of alternative splicing of these genes. The possibility still exists that the 6.5K component derives from a proteolytic degradation of the 15K polypeptide; this is, however, unlikely in view of the fact that the tissues analyzed in Western blots were homogenized directly in a boiling SDS-containing buffer immediately after dissection.

Looking back to the available data on the β-D-galactoside-specific lectins, it appears that during evolution they have been very strongly conserved (Paroutaud et al. 1987), but that a gradual process of diversification has led to different variants of the same protein expressed in different tissues (Gitt and Barondes, 1986). A similar situation has been recently described for the cadherins, a family of Ca2+-dependent cell adhesion molecules, which during evolution have produced at least three highly homologous variants expressed in different tissues with an exquisitely tissue-specific cell adhesion function (Shirayoshi et al. 1986; Hatta and Takeichi, 1986).

The crucial question, which remains to date unresolved, concerns the relationship between the tissue and cell localization of the β-D-galactoside-specific lectins and their functions.

As mentioned in the Introduction, possible roles of β-D-galactoside-specific lectins in cell adhesion, cell fusion and cell-to-matrix attachment have been investigated without conclusive evidence being gathered. Considering the fact that the lectin is localized intracellularly during most of the embryonic development and is only transiently expressed in developing tissues, one may reiterate the early suggestion (Teichberg et al. 1975) that the lectin functions are related to early intracellular events linked to cell replication and/or to the selective expression of a developmentally timed phenotype. The conspicuous presence of lectins in malignant (Teichberg et al. 1975; Gabius, 1987) and in transformed cells (Raz et al. 1987a,b) lends credence to this suggestion.

The developmental changes affecting the 15K lectin fall in fine with the above concept. The lectin appears transiently in tissues in which cell replication and/or phases of differentiation take place, and declines during maturation.

The most stiking changes in the levels of the 15K lectin during development are observed in cells that differentiate into muscles and which exibit a very strong immunoreactivity. Although not studied here, later embryonic stages are characterized by a marked decline in lectin levels (Nowak et al. 1976).

In skeletal muscles, the expression of the 15K β-D-galactoside-specific lectin appears to parallel that of proteins such as the acetylcholine receptor, which are down-regulated by the muscle activity and its concomitant maturation. Indeed, besides being abundant in embryonic muscle and virtually disappearing in the adult (Kobiler and Barondes, 1977), the 15K lectin is reexpressed in denervated adult muscle (Teichberg, 1978), i.e. under conditions leading to the repetition of early events in development. The precise nature of the cellular events leading to the local regulation of the expression of the lectin remains to be elucidated.

We thank Dr Jean Paul Thiery and Jean-Loup Duband for interesting discussion and criticism. This work was supported by the Centre National de la Recherche Scientifique, the Institute National de la Santé et de la Recherche Médicale, the Fondation pour la Recherche sur la Myopathie, the Gesellschaft fur Biotechnologische Forshung, mbH, FRG and Minerva. G.L. was a recipient of a short-term fellowship from the European Molecular Biology Organization. V.I.T. holds the Louis and Florence Katz-Cohen Professorial Chair in Neuropharmacology.

Barondes
,
S. H.
(
1984
).
Soluble lectins: A new class of soluble proteins
.
Science
223
,
1259
1264
.
Barondes
,
S. H.
and
Haywood-Reid
,
P. L.
(
1981
).
Externalization of an endogenous muscle lectin with in vivo development
.
J. Cell Biol
.
91
,
568
572
.
Beyer
,
E. C.
and
Barondes
,
S. H.
(
1980
).
Chicken tissue binding sites for a purified chicken lectin
.
J. supratnolec. Struct
.
13
,
219
227
.
Beyer
,
E. C.
and
Barondes
,
S. H.
(
1982a
).
Secretion of endogenous lectin by chicken intestinal goblet cells
.
J. Cell Biol
.
92
,
28
33
.
Beyer
,
E. C.
and
Barondes
,
S. H.
(
1982b
).
Quantitation of two endogenous lactose-inhibitable lectins in embryonic and adult chicken tissues
.
J. Cell Biol
92
,
23
27
.
Beyer
,
E. C.
,
Zweig
,
S. E.
and
Barondes
,
S. H.
(
1980
).
Two lactose binding lectins from chicken tissues: purified lectin from intestine is different from those of liver and muscle
.
J. biol. Chem
.
255
,
4236
4239
.
Brackenbury
,
R.
,
Thiery
,
J. P.
,
Rutishauser
,
U.
and
Edelman
,
G. M.
(
1977
).
Adhesion among neural cells of the chick embryo. I. An immunological assay for molecules involved in cell-cell binding
.
J. biol. Chem
.
252
,
6835
6840
.
Cook
,
G. M. W.
,
Zalik
,
S. E.
,
Milos
,
N.
and
Scott
,
V.
(
1979
).
A lectin which binds specifically to β-D-galactoside groups is present at the earliest stages of chick embryo development
.
J. Cell Sci
.
38
,
293
304
.
Den
,
H.
and
Maunzak
,
D. A.
(
1977
).
Isolation and properties of β-D-galactoside-specific lectin from chick embryo thigh muscle
.
J. biol. Chem
.
252
,
5444
5448
.
Duband
,
J.-L.
and
Thiery
,
J. P.
(
1987
).
Distribution of laminin and collagens during avian neural crest development
.
Development
101
.
461
478
.
Gabius
,
H.-J.
(
1987
).
Endogenous lectins in tumors and the immune system
.
Cancer Investig
.
5
,
39
46
.
Gartner
,
T. K.
and
Podleski
,
T. R.
(
1975
).
Evidence that a membrane bound lectin mediates fusion of L6 myoblasts
.
Biochem. biophys. Res. Commun
.
67
,
972
978
.
Gitt
,
M. A.
and
Barondes
,
S. H.
(
1986
).
Evidence that a human soluble β-galactoside-binding lectin is encoded by a family of genes
.
Proc. natn. Acad. Sci. U.S.A
.
83
.
7603
7607
.
Gremo
,
F.
,
Kobiler
,
D.
and
Barondes
,
S. H.
(
1978
).
Distribution of an endogenous lectin in the developing chick optic tectum
.
J. Cell Biol
.
79
,
491
499
.
Hamburger
,
V.
and
Hamilton
,
H.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morph
.
88
,
49
92
.
Harrison
,
F. L.
and
Chesterton
,
C. J.
(
1980
).
Factors mediating cell-cell recognition and adhesion. Galaptins, a recently discovered class of bridging molecules
.
FEBS Lett
.
122
,
157
165
.
Hatta
,
K.
and
Takeichi
,
M.
(
1986
).
Expression of N-cadherin adhesion molecule associates with early morphogenetic events in chicken development
.
Nature, Lond
.
320
,
447
449
.
Hirabayashi
,
J.
and
Kasai
,
K.
(
1988
).
Complete amino acid sequence of a β-galactoside-binding lectin from human placenta
.
J. Biochem
.
104
,
1
4
.
Hirabayashi
,
J.
,
Kawasaki
,
H.
,
Suzuki
,
K.
and
Kasai
,
K.
(
1987
).
Complete amino acid sequence of a 14 kDa β-galactoside-binding lectin of chick embryo
.
J. Biochem
.
101
,
775
787
.
Kitamura
,
K.
(
1981
).
Distribution of endogenous β-galactoside-specific lectin, fibronectin and type I and III collagens during dermal condensation in chick embryos
.
J. Embryol. exp. Morph
.
65
,
41
56
.
Kobiler
,
D.
and
Barondes
,
S. H.
(
1977
).
Lectin activity from embryonic chick brain, heart and liver changes with development
.
Devl Biol
.
60
,
326
330
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature, Lond
.
227
,
680
685
.
Levi
,
G.
,
Crossin
,
K. L.
and
Edelman
,
G. M.
(
1987
).
Expression sequences and distribution of two primary cell adhesion molecules during embryonic development of Xenopus laevis
.
J. Cell Biol
.
105
,
2359
2372
.
Levi
,
G.
and
Teichberg
,
V. I.
(
1981
).
Isolation and physicochemical characterization of electrolectin, a β-D-galactoside binding lectin from the electric organ of Electrophorus electricus
.
J. biol. Chem
.
256
,
5735
5740
.
Levi
,
G.
and
Teichberg
,
V. I.
(
1982
).
Antigenic cross-reactivity between electrolectins
.
FEBS Lett
.
148
,
145
148
.
Levi
,
G.
and
Teichberg
,
V. I.
(
1985
).
Isolation and characterization of chicken thymic electrolectin
.
Biochem. J
.
226
,
379
384
.
Macbride
,
R. G.
and
Przybylski
,
R. J.
(
1980
).
Purified lectin from skeletal muscle inhibits myotube formation in vitro
.
J. Cell Biol
.
85
,
617
625
.
Matsutani
,
E.
and
Yamagata
,
T.
(
1982
).
Chick endogenous lectin enhances chondrogenesis of cultured chick limb bud cells
.
Devl Biol
.
92
,
544
548
.
Meromsky
,
L.
,
Lotan
,
R.
and
Raz
,
A.
(
1986
).
Implication of endogenous tumor cell surface lectins as mediators of cellular interactions and lung colonization
.
Cancer Res
.
46
,
5270
5275
.
Milos
,
N.
and
Zalik
,
S. E.
(
1981
).
Mechanism of adhesion among cells of the early chick blastoderm. Role of β-D-galactoside-binding lectin on the adhesion of extraembryonic endoderm cells
.
Differentiation
21
,
175
182
.
Nowak
,
T. P.
,
Haywood
,
P. L.
and
Barondes
,
S. H.
(
1976
).
Developmentally regulated lectin in embryonic chick muscle and a myogenic cell line
.
Biochem. biophys. Res. Commun
.
68
,
650
657
.
Nowak
,
T. P.
,
Kobiler
,
D.
,
Roel
,
L. E.
and
Barondes
,
S.
(
1977
).
Developmentally regulated lectin from embryonic chick pectoral muscle
.
J. biol. Chem
.
252
,
6026
6030
.
Ohyama
,
Y.
and
Kasai
,
K.
(
1988
).
Isolation and characterization of the chick 14K β-galactoside-binding lectin gene
.
J. Biochem
.
104
,
173
177
.
Olmsed
,
J. B.
(
1981
).
Affinity purification of antibodies from diazotized paper blots of heterogeneous protein samples
.
J. biol. Chem
.
256
,
11955
11957
.
Paroutaud
,
P.
,
Levi
,
G.
,
Teichberg
,
V. I.
and
Strosberg
,
A. D.
(
1987
).
Extensive amino acid sequence homologies between animal lectins
.
Proc. natn. Acad. Sci. U.S.A
.
84
,
6345
6348
.
Podleski
,
T. R.
and
Greenberg
,
I.
(
1980
).
Distribution and activity of endogenous lectin during myogenesis as measured with antilectin antibody
.
Proc. natn. Acad. Sci. U.S.A
.
77
,
1054
1058
.
Raz
,
A.
,
Avtvi
,
A.
,
Pazerini
,
G.
and
Carmi
,
P.
(
1987a
).
Cloning and expression of cDNA for two endogenous UV-2237 fibrosarcoma lectin genes
.
Expl Cell Res
.
173
,
109
116
.
Raz
,
A.
,
Carmi
,
P.
and
Pazerini
,
G.
(
1988
).
Expression of two different endogenous galactoside-binding lectins sharing sequence homology
.
Cancer Res
.
48
,
645
649
.
Raz
,
A.
,
Meromski
,
L.
,
Zvibel
,
J.
and
Lotan
,
R.
(
1987b
).
Transformation related changes in the expression of endogenous cell lectins
.
Int. J. Cancer
39
,
353
360
.
Regan
,
L. J.
,
Dodd
,
J.
,
Barondes
,
S. H.
and
Jessell
,
T. M.
(
1986
).
Selective expression of endogenous lactose-binding lectins and lactoseries glycoconjugates in subsets of rat sensory neurons
.
Proc. natn. Acad. Sci. U.S.A
.
83
,
2248
2252
.
Romanoff
,
A. L.
(
1960
).
The Avian Embryo. Structural and Functional Development
, pp.
1039
1149
.
The Macmillan Company
,
New York
.
Rong
,
P.-M.
,
Ziller
,
C.
,
Pena-Melian
,
A.
and
Le Douarin
,
N. M.
(
1987
).
A monoclonal antibody specific for avian early myogenic cells and differentiated muscle
.
Devl Biol
.
122
,
338
353
.
Shirayoshi
,
Y.
,
Hatta
,
K.
,
Hosoda
,
M.
,
Tsunasawa
,
S.
,
Saktyama
,
F.
and
Takeichi
,
M.
(
1986
).
Cadherins, cell adhesion molecules with distinct binding specificity share a common structure
.
EMBO J
.
5
,
2485
2488
.
Teichberg
,
V. I.
(
1978
).
Electrolectins: β-D-galactoside-binding proteins
.
Meth. Enzym
.
50
,
291
302
.
Teichberg
,
V. L
,
Silman
,
L
,
Beitsch
,
D. D.
and
Resheff
,
G.
(
1975
).
A β-D-galactoside binding protein from electric organ tissue of Electrophorus electricus
.
Proc. natn. Acad. Sci. U.S.A
.
72
,
1383
1387
.
Towbin
,
H.
,
Staehelin
,
T.
and
Gordon
,
J.
(
1979
).
Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications
.
Proc, natn. Acad. Sci. USA
.
76
,
4350
4354
.
Zalik
,
S. E.
,
Milos
,
N.
and
Ledsham
,
I.
(
1982
).
Distribution of a galactose specific lectin in endoderm cells from early chick embryos
.
Cell Tiss. Res
.
225
,
223
228
.
Zalik
,
S. E.
,
Thompson
,
L. W.
and
Ledsham
,
I. M.
(
1987
).
Expression of an endogenous galactose-binding lectin in the early chick embryo
.
J. Cell Sci
.
88
,
483
493
.