Vertebrate lectins have shown to mediate intracellular protein trafficking, innate immunity, and cell adhesion and communication. The Xenopus laevis oocyte cortical granule lectin (XCGL) is unique in that it plays crucial roles in fertilization and early embryonic development (Wyrick et al. 1974; Roberson and Barondes 1983; Nishihara et al. 1986; Quill and Hedrick 1996). The cDNAs for XCGL and a similar cortical granule lectin in Xenopus tropicalis were isolated and characterized (Lee et al. 1997; GenBank accession nos. X82626, AY079196). In later studies, several other XCGL-like molecules have been cloned in Xenopus (DDBJ accession nos. AB061238, AB061239) and mammals (Komiya et al. 1998; Lee et al. 2001; Tsuji et al. 2001). Despite the structural similarity, little is known about physiological roles of these XCGL-like lectins. We report here a novel molecule in the same family that is expressed exclusively in a subset of epidermal cells in Xenopus embryos.

During screening to identify ligand molecules of a neuronal receptor-type protein tyrosine phosphatase, RPTPβ (Nagata et al. 2001), we accidentally isolated cDNA clones encoding a novel protein, XEEL (see below), from a Xenopus embryo library. The largest cDNA clone is 1,158 bp long and contains an open reading frame potentially encoding an amino acid sequence of 342 residues. BLAST homology search of the DDBJ database revealed several vertebrate lectins containing similar amino acid sequences. Figure 1 shows an alignment of the predicted XEEL translate with those of three other Xenopus lectins, XCGL, the serum lectin (XSL) and the type 2 serum lectin (XSL2). The XEEL translate exhibits 64–70% amino acid identity with these molecules and has a shared fibrinogen-like motif found in the Ficolin/Opsonin p35 family, which probably serves as a carbohydrate-binding domain (Sugimoto et al. 1998). Two N-glycosylation sites are also conserved except for one of the sites in XSL2. The XEEL translate also shows about 62–64% amino acid identity with those of mouse intelectin (Komiya et al. 1998), human HL-1/intelectin and HL-2 (Lee et al. 2001; Tsuji et al. 2001; data not shown).

Fig. 1.
figure 1

Alignment of predicted amino acid sequences of XEEL, XCGL, XSL and XSL2 translates. Dots and bars in the sequences are the same residues and the gaps, respectively. Underlined sequences represent putative signal peptides for secretion and those in a shaded box are fibrinogen-like motifs found in the Ficolin/Opsonin p35 family. Potential N-glycosylation residues are shown in black boxes. The sequences shown in this figure are deposited in the DDBJ/GenBank/EMBL database: XEEL, AB105372; XCGL, X82626; XSL, AB061238; XSL2, AB061239. The database contains an additional sequence for XCGL (U86699; named XL35) that differs in 5 amino acids of 313 residues from that of X82626

Full size image

To examine developmental expression of XEEL, XCGL, XSL and XSL2, total RNA fractions were isolated from various stages of embryos and RT-PCR analysis was performed using primer sets specific to each molecule (Fig. 2). The XEEL and XSL2 transcripts are detectable in early gastrula embryos at stage 10 (Nieuwkoop and Faber 1994) and quickly increase thereafter, whereas the XSL transcript is not detectable before stage 15. In contrast, the XCGL transcript is found in fertilized eggs (stage 1) and the amount decreases in early development to become undetectable after stage 20. It seems therefore that the XCGL mRNA is deposited maternally during oogenesis and that no zygotic expression occurs in embryos. An additional RT-PCR analysis showed no XEEL expression in any adult tissues examined (data not shown).

Fig. 2.
figure 2

RT-PCR analyses of XEEL, XCGL, XSL and XSL2 expression during embryonic development. Total RNA fractions were isolated from embryos at the indicated developmental stages and the fragment of each transcript amplified by RT-PCR using a specific primer set. Ornithine decarboxylase (ODC) is the loading control (–RT reverse transcription control without the reverse transcriptase)

Full size image

To visualize spatial patterns of the XEEL expression, whole-mount in situ hybridization was performed using a digoxigenin (DIG)-labeled RNA probe. The XEEL transcripts are detectable in a subset of epidermal cells of late neurula embryos at stage 16 (Fig. 3a, e). In contrast, cells in the neural crest and the neural plate are XEEL-negative. In early tailbud embryos at stage 23, epidermal expression of XEEL is much stronger, and about the same number of XEEL-positive and XEEL-negative cells distribute throughout the body surface (Fig. 3b). A similar spatial distribution of XEEL transcripts is seen in late tailbud embryos at stage 35/36 (Fig. 3c, f). XEEL-positive cells are also found in the fin epidermis, but the hatching gland (Yoshizaki 1973), the olfactory placodes and the cement gland are devoid of XEEL expression (Fig. 3d). Examination of serial transverse sections of the paraffin-embedded stage 35/36 embryos demonstrated no XEEL expression in internal tissues (Fig. 3g). In addition, the XEEL-negative cells in epidermis are ciliated cells (Fig. 3h). In stage 45 larvae, XEEL expression is still found in limited areas of the belly epidermis (Fig. 3i), but becomes undetectable at stage 47 (data not shown).

Fig. 3a–i.
figure 3

Whole-mount in situ hybridization analyses of XEEL expression. a A late neurula embryo at stage 16, showing XEEL-positive cells at the lateral trunk epidermis. b Early tailbud embryos at stage 23 hybridized with the antisense (top) or the sense (bottom) probe, showing a specificity of hybridization. c A late tailbud embryo at stage 35/36, showing strong epidermal XEEL expression all over the embryo except for the cement gland (cg) and the olfactory placode (op). d Anterior view of the embryo in c, showing the XEEL-negative cement gland, olfactory placode and hatching gland (hg). e Enlargement of the dorso-lateral part of the embryo in a, showing sparsely distributed, strongly-labeled cells and a checkered pattern of weakly labeled and non-labeled cell distribution. f Enlargement of the dorso-lateral part of the embryo in c, showing distribution of strongly labeled cells in the trunk and dorsal fin (df) epidermis. g A representative transverse paraffin section of the trunk region of the hybridized embryo, showing epidermis-specific expression of XEEL. h Enlargement of a part of the section in g, showing XEEL-positive non-ciliated cells and XEEL-negative ciliated cells (arrows). i An anterior part of a stage 45 embryo, showing XEEL expression in the belly epidermis. a–c, e, f, i Anterior to the left and dorsal to the top. Scale bars 200 µm (a–g, i); 50 µm (h)

Full size image

These results suggest that XEEL is a novel member of the XCGL family of lectins secreted from embryonic epidermis. It may have a role in anti-pathogen activity as suggested for human HL-1/intelectin (Tsuji et al. 2001). Thus, the discovery of XEEL gives us new insights into (1) a possible novel function of embryonic epidermis and (2) the molecular evolution of the XCGL-like lectins in vertebrates.