The use of lectins as markers for differentiated secretory cells in planarians

Abstract

Freshwater planarians have reemerged as excellent models to investigate mechanisms underlying regeneration. The introduction of molecular tools has facilitated the study of planarians, but cell- and tissue-specific markers are still needed to examine differentiation of most cell types. Here we report the utility of fluorescent lectin-conjugates to label tissues in the planarian Schmidtea mediterranea. We show that 16 lectin-conjugates stain planarian cells or tissues; 13 primarily label the secretory cells, their cytoplasmic projections, and terminal pores. Thus, we examined regeneration of the secretory system using lectin markers and functionally characterized two genes expressed in the secretory cells: marginal adhesive gland-1 (mag-1) and Smed-reticulocalbin1 (Smed-rcn1). RNAi knockdown of these genes caused a dramatic reduction of secretory cell lectin staining, suggesting a role for mag-1 and Smed-rcn1 in secretory cell differentiation. Our results provide new insights into planarian secretory system regeneration and add new markers for labeling several planarian tissues.

INTRODUCTION

The regenerative capacity of freshwater planarians makes them excellent subjects to explore basic mechanisms underlying tissue replacement (Newmark and Sánchez Alvarado, 2002; Reddien and Sánchez Alvarado, 2004). Planarians are endowed with a population of adult pluripotent stem cells that serve to replace cells lost during normal cell turnover or after injury. The molecular tools available to study the planarian Schmidtea mediterranea (Sánchez Alvarado and Newmark, 1999; Sánchez Alvarado et al., 2002; Newmark et al., 2003; Zayas et al., 2005; Robb et al., 2008) are facilitating molecular dissection of diverse problems such as tissue regeneration (Newmark and Sánchez Alvarado, 2002; Reddien and Sánchez Alvarado, 2004; Sánchez Alvarado, 2006), patterning (Forsthoefel and Newmark, 2009; Adell et al., 2010), and homeostasis (Pellettieri and Sánchez Alvarado, 2007); tumorigenesis (Pearson and Sánchez Alvarado, 2008; Oviedo and Beane, 2009); and the mechanisms underlying inductive germ cell specification (Newmark et al., 2008). To address these and other problems in planarians, markers of all the differentiated cell types are needed for analyzing the cellular and molecular events that take place during regeneration.

In this study, we tested plant lectins as candidate molecules to label tissues in the planarian Schmidtea mediterranea. Lectins are sugar-binding proteins that bind with high specificity to carbohydrate moieties found on glycoproteins and function primarily as recognition molecules in diverse biological contexts (Sharon and Lis, 2004; Sharon, 2007). Because of the sugar specificity of lectins and the enormous diversity of glycan conjugates, these molecules have been widely used in clinical and experimental applications such as blood typing, cell identification and separation, and in histochemical and cytochemical studies (Lis and Sharon, 1986).

Here we show that fluorescent lectin-conjugates are useful as markers of several tissue types in S. mediterranea, and predominantly stain the planarian secretory system. Initially described by Iijima (1884; see Pedersen, 1963), gland cells in planarians are insunken epithelial cells. These specialized cells are located in the mesenchyme, and are characterized by elongated cell bodies with long cytoplasmic projections that terminate between or directly penetrate epithelial cells and form secretory pores at the surface (Hyman, 1951; Pedersen, 1963). Although the exact chemical composition of the gland cell secretions has not been fully determined, the mucus-like substances are thought to play an important role in the animal's ability to adhere to substrates, glide during locomotion, defend itself, and capture prey (Hyman, 1951; Pedersen, 1963). Based on histochemical studies, the gland cells have been divided into two groups: acidophilic (eosinophilic) and cyanophilic (basophilic) gland cells (von Graff, 1912-1917; Hyman, 1951; Skaer, 1961; Pedersen, 1963). The acidophilic gland cells can be found anywhere in the body but primarily form clusters of subepidermal marginal adhesive gland cells near the edges of the animal; the pores of these cells form a striking pattern encircling the ventral surface referred to as the marginal adhesive zone. The cyanophilic gland cells are found near the ventral mesenchyme, the pharynx, and in the dorsal region of the head.

We took advantage of these new gland cell markers to examine regeneration of the planarian secretory system. Labeling animals with fluorescent lectin-conjugates allowed us to visualize the formation of secretory projections and their terminal pores during regeneration. Using these markers, we also investigated the role of two genes, marginal adhesive gland-1 (mag-1) and Smed-reticulocalbin1 (Smed-rcn1), which are specifically expressed in subepidermal gland cells. RNAi knockdown of mag-1 or Smed-rcn1 showed that the function of these genes is required for the normal differentiation of secretory structures during regeneration. Taken together, our results demonstrate the utility of lectins as markers of planarian tissues, particularly the secretory system. These markers enabled us to examine gland cell morphogenesis during regeneration and will be useful to study in greater detail the biology of planarian gland cells.

RESULTS

Labeling of planarian tissues with fluorescent lectin-conjugates

We tested 22 commercially available fluorescent lectin-conjugates (see Experimental Procedures) to examine if these reagents could be used as markers of specific planarian cell types and/or tissues. We found that 16 lectin conjugates stained planarian tissues (Table 1), including the CNS, epidermis, intestine, muscle, and the pharynx. Most notably, 13 lectins strongly stained different secretory cell types (see Table 1). For example, Erythrina cristagalli lectin (ECL) specifically labeled the subepidermal marginal adhesive gland cells, including the processes projecting from the cell bodies in the mesenchyme to the epidermis, and the secretory pores (Figure 1A). By contrast, wheat germ agglutinin (WGA) broadly stained several secretory cell types, including their cytoplasmic projections, and terminal pores (Figures 1B-C). The cytoplasmic projections become branched, giving them a rosette appearance characteristic of the marginal adhesive zone pore clusters. Co-labeling planarians with WGA and anti-phospho-tyrosine revealed the membrane junctions of lectin-positive pores and marked individual pore rosettes (Figure 1C). In addition to staining the marginal adhesive zone, Vicia villosa agglutinin (VVA) stained secretory cells located dorsal and posterior to the photoreceptors, and secretory cells surrounding the pharynx (Figure 1D). The cells behind the photoreceptors (Figure 1D), which also stained with WGA (Figure 2), project to the anterior tip of the head. These cells likely form the adhesive organ that the animals use to capture prey (Hyman, 1951; Pedersen, 1963). Lectins such as Griffonia simplicifolia lectin I (GSL-I), Soybean agglutinin (SBA), and VVA (Figure 1E and Supplementary Figure S1) label the pores of the ventral secretory projections, which release mucus and facilitate cilia-driven locomotion (Hyman, 1951; Pedersen, 1963). Staining with Lens culinaris agglutinin (LCA) was ubiquitous (Figure 1F), but also revealed some interesting features such as the rhabdome (Figure 1G), the site where rhabdomeres projecting from photosensitive cells contact the pigment cup. In addition, LCA stained cells within the planarian intestine (Figure 1H); based on their location and highly vacuolar appearance these are likely goblet cells (Baguñà and Romero, 1981). Concanavalin A (Con A) stained the junctions between epidermal cells (Figure 1I). In addition to the subepidermal gland cells, Ricinus communis agglutinin I (RCA I) stained dorsal secretory pores and ventral secretory cells and their projection to the ventral surface (Figure 1J). Our observations of planarians stained with fluorescent lectin-conjugates show that these sugar-binding proteins are useful markers of differentiated planarian tissues.

Table 1.

Tissues labeled with several fluorescent lectin-conjugates tested in asexual Schmidtea mediterranea.

Lectin	CNS	Epidermis	Intestinal1	Parenchyma	Muscle2	Secretory	Pharynx	Photoreceptors3
Con A (n = 6)		+++
DBA (n = 6)						+++a,b,d,e	++
DSL (n = 9)						+++c	++
ECL (n = 16)					+	+++b,d	++
GSL I (n = 11)						+++a,e
GSL II (n = 9)						+++b,d	+
Jacalin (n = 11)						+++d,e	+
LCA (n = 23)	+		++		++			++
LEL (n = 18)		+++
PNA (n = 6)						+++d	++
PSA (n = 14)	+	+++	++	+++	++			++
RCA I (n = 9)						+++a,d,e	++
SBA (n = 9)						+++d,e
sWGA (n = 11)						+++a,b,d	++
VVA (n = 19)						+++b,c,d,e
WGA (n = 15)						+++b,d,e	+

Staining: + weak, ++ intermediate, +++ strong

¹goblet cells;

²staining observed in the extracellular space surrounding muscle;

³staining observed in rhabdome and adjacent to the pigment cups;

^adorsal secretory pores;

^bpre-pharyngeal secretory cells;

^cpharyngeal secretory cells;

^dsubepidermal marginal adhesive glands;

^eventral secretory pores

Figure 1.

Labeling of planarian tissues with fluorescent lectin conjugates. (A) ECL labels the marginal adhesive zone (arrowheads) formed by cytoplasmic projections (arrows) from subepidermal marginal adhesive gland cells. (B) WGA extensively labels the subepidermal secretory projections (arrows) and the adhesive zone (arrowheads). (C) Higher magnification of adhesive zone individual pore rosettes from a planarian stained with WGA (green) and counterstained with anti-phospho-tyrosine (magenta). Cytoplasmic projections from subepidermal gland cells (arrows) terminate in the exterior surface of the epidermis and form pores (arrowheads) demarcated by the anti-phospho-tyrosine staining (n = 6). (D) VVA labels the subepidermal gland cells and dorsal secretory cells (arrow with asterisk) projecting to the anterior tip of the head (arrow). In addition, VVA labels secretory cells surrounding and sending projections to the pharynx (arrowheads). (E) VVA strongly labels the ventral secretory pores. The marginal adhesive zone is also visible, including a concentration of gland cytoplasmic projections (arrow) terminating at the anterior tip of the head. The inset shows a higher magnification image of the margin near the tip of the head. Arrowheads mark the secretory ducts formed by the VVA-positive projections. (F) Montage of single confocal optical sections of a planarian stained with LCA. The lumen of the intestine is readily visible; goblet cells were observed throughout the primary intestinal branches. (G) Higher magnification image of photoreceptors (boxed region near the head in F). Labeling is observed in a punctate pattern near the pigment cups and the rhabdome (arrows). (H) Higher magnification of the goblet cells (boxed region anterior to the pharynx in F; arrowheads) found throughout the primary branches of the intestine. (I) Con A strongly stains the epithelial cell junctions. Inset shows a higher magnification image of the epidermis of a planarian stained with Con A (green) and counterstained with the nuclear dye TOTO-3 (magenta; n = 5). (J) High magnification of a planarian stained with RCA I, which stains subepidermal gland cell projections (white arrowheads) that penetrate the epidermis and terminate externally (yellow arrowheads). Anterior is up (A, B, G) or to the left. Scale bars: in (A-B) 200 μm; (C) 25 μm; (D-E) 200 μm; inset in (E) 50 μm; (F) and (I) 200 μm; inset in (I) 50 μm; (J) 100 μm.

Figure 2.

(A-F) Lectin staining of regenerating planarians. Animals were transected anterior to the pharynx, fixed at specific time intervals post-amputation, and stained with ECL- or WGA-FITC. After 3 days of regeneration, the marginal adhesive zone was not detectable with either ECL (n = 7) or WGA (n = 6). Regeneration of the marginal adhesive zone (arrowheads) in the new head was visualized with either ECL- (n = 7) or WGA-FITC (n = 7) after 5 days of regeneration. By day 7, when regeneration of the head is complete, the stereotypical pattern of the marginal adhesive zone was restored (ECL, n = 6; WGA, n = 5) compared to controls (ECL, n = 7; WGA, n = 6; not shown). In addition, WGA strongly stained the cytoplasmic projections from the cluster of secretory cells located dorsally, behind the photoreceptors (arrow; WGA at 7 days). Animals are ventral with the blastema facing up. Scale bars = 200 μm.

In addition to staining intact planarians, we tested the feasibility of using lectins to label tissues during regeneration. The majority of the lectins we tested strongly labeled the secretory system (Table 1); thus, we used these markers to examine regeneration of the marginal adhesive zone. We amputated planarians anterior to the pharynx and allowed them to regenerate, and then we fixed the animals at different time points and stained them with ECL or WGA. Detectable staining of the regenerating trunk pieces with either ECL or WGA was not observed in the new tissue until day 5, when lectin-positive secretory pores were detected (Figure 2). After 7 days of regeneration, when formation of a new head is largely complete, the pattern of lectin staining resembled that of intact planarians. This staining pattern was also evident in animals stained with WGA, in which the prominent projections from secretory cells posterior to the photoreceptors were detected (Figure 2; arrow at 7 days). Our results show that lectin labeling can be used to visualize the secretory projections and secretory pores of differentiated gland cells in newly regenerated tissues. Little is known about how gland cells are regenerated or maintained in planarians. Thus, we decided to further analyze regeneration of the subepidermal marginal adhesive gland cells using genes that are expressed specifically in these cells.

Characterization of genes expressed in the subepidermal marginal adhesive gland cells

Previous whole-mount in situ hybridization studies in asexual S. mediterranea revealed that the mRNAs corresponding to asexual cDNA clones H.1.3b and H.10.8a are expressed specifically in the subepidermal marginal adhesive gland cells (Sánchez Alvarado et al., 2002) (Figure 3). The partial H.1.3b cDNA shares similarity with SCO-spondin (Blastx; e-value=2×10⁻²⁰). The cDNA is predicted to encode two Thrombospondin type 1 repeat (TSP1) domains, a Trypsin Inhibitor-like cystein-rich (TIL) domain, and numerous low-complexity regions. Analysis of the predicted H.1.3b sequence deduced from the S. mediterranea genome sequence did not reveal any additional functional domains. Due to a lack of clear similarity to other known genes, we have named this gene marginal adhesive gland-1 (mag-1) based on its restricted expression in the marginal adhesive gland cells. The H.10.8a cDNA sequence is homologous to the ER-resident protein Reticulocalbin (Blastx; e-value=3×10⁻⁴⁰, S. mansoni). The predicted protein H.10.8a sequence contains a signal peptide, all six EF-hand calcium-binding motifs, and a C-terminal luminal ER sorting sequence (DDEL). Based on homology, we have named clone H.10.8a Smed-reticulocalbin1 (rcn1).

Figure 3.

Whole-mount in situ hybridization to mag-1 and Smed-rcn1 mRNAs in intact and regenerating planarians. mag-1 (A) and Smed-rcn1 (B) mRNAs are expressed in subepidermal marginal adhesive gland cells located near the body margin (n = 10 per gene). Both mRNAs were also detected in secretory cells located in the dorsal surface, posterior to the photoreceptors (arrows in A and in B). After head amputation, expression of mag-1 and Smed-rcn1 (A and B, 1-10 days after amputation; n = 10 for each time point, respectively) was detected in the blastema, near the body margin, as early as 2 days post-amputation (arrowheads at 2 days). mRNAs for both genes were also detected in the blastema, closer to the wound site (arrows at 3 days). Expression in the blastema, body margin (arrowheads at 5 days) and the area posterior to the photoreceptors (arrows at 5 days) were consistently observed throughout subsequent days of regeneration. The stereotypical pattern observed in uninjured planarians was re-established after 10 days (A and B). Diffuse signal from Smed-rcn1 was also detected in anterior and marginal secretory cell projections (arrow; 10 days in B). Anterior is to the left in A and B. Blastemas are facing up or to the left. Scale bars = 200 μm.

Whole-mount in situ hybridization to mag-1 and Smed-rcn1 confirmed that these genes are expressed in the subepidermal marginal adhesive gland cells, along the body margin posterior to the photoreceptors, and in groups of dorsal cells located behind the photoreceptors (Sánchez Alvarado et al., 2002) (Figures 3A-B). Smed-rcn1 mRNA was also weakly detected in cells surrounding the pharynx (Figure 3B). We next investigated the expression patterns of these genes in regenerating trunks after the head was amputated pre-pharyngeally. During anterior regeneration, mag-1 expression was not observed in 1-day blastemas (Figure 3A). At day 2, mag-1-expressing cells appeared in the blastema underneath the wound epithelium, and after 3 days, mag-1-expressing cells appeared along the margin of the anterior tip of the head and in posterior medial regions of the body. As regeneration proceeds and the head grows, mag-1-positive cells were primarily restricted to the body margin (behind the photoreceptors) and there was a marked increase in cells expressing mag-1 in the dorsal region posterior to the photoreceptors, corresponding to the dorsal cluster of secretory cells (see Figure 1D). Finally, by day 10 of regeneration the newly differentiated head shows a pattern similar to that found in uninjured planarians. No or few mag-1-positive cells were seen in the head margin anterior to the photoreceptors. Analyses with Smed-rcn1 showed similar expression pattern changes during anterior regeneration (Figure 3B).

To investigate whether the expression of mag-1 and Smed-rcn1 correspond to the same population of cells, we performed double fluorescent in situ hybridization (FISH) experiments. As we expected, FISH to mag-1 and Smed-rcn1 showed these mRNAs were expressed in subepidermal marginal adhesive gland cells (Figure 4). Interestingly, fluorescent detection revealed Smed-rcn1 mRNAs were also present in cells adjacent to the photoreceptors (Figure 4; Smed-rcn1). Superposed images of mag-1 and Smed-rcn1 showed that the corresponding mRNAs were largely co-localized to the same population of gland cells. We conclude that mag-1 and Smed-rcn1 mRNAs are expressed in differentiated marginal adhesive gland cells.

Figure 4.

mag-1 and Smed-rcn1 mRNA expression is co-localized in dorsal and marginal adhesive gland cells. We performed double whole-mount fluorescent in situ hybridization to mag-1 (A, B) and Smed-rcn1 (C, D) (n = 5). The merged images (E, F) showed that expression of these mRNAs overlaps in the cells posterior to the photoreceptors (head in A, C, and E) and in the marginal adhesive gland cells (tail in B, D, F). Expression of Smed-rcn1 was also observed in cells in close proximity to the photoreceptors (C). Scale bars = 200 μm.

RNAi knockdown of mag-1 or Smed-rcn1 eliminates lectin staining of the secretory pores

To test whether mag-1 and Smed-rcn1 play a role in the regeneration of the secretory system, we treated animals with dsRNA corresponding to each gene. As a first step, we generated non-overlapping dsRNA and riboprobes for either mag-1 or Smed-rcn1 (see Experimental Procedures) and tested the specificity of the RNAi knockdowns. After three consecutive days of injecting dsRNA into planarians, animals were amputated at the pre-pharyngeal level and allowed to regenerate for seven days. The expression of both of these genes was not affected by control injections (Figures 5A and D). By contrast, mag-1(RNAi) abolished expression of mag-1 mRNA (Figure 5B). The expression of Smed-rcn1, however, was still detected in mag-1(RNAi) planarians (Figure 5E), suggesting the marginal adhesive gland cells are replaced during regeneration and remain present in the uninjured tissues. Similarly, in Smed-rcn1(RNAi) planarians, the mag-1 transcript was readily detected (Figure 5C), but detectable expression of the Smed-rcn1 message was eliminated (Figure 5F). Thus, RNAi knockdown of either mag-1 or Smed-rcn1 was specific and had no discernable effect on the expression of the other gene. We then tested if mag-1(RNAi) or Smed-rcn1(RNAi) had any effect on the regeneration of the gland cells' cytoplasmic projections or secretory pores using lectin markers (Figure 6). Animals were treated with dsRNA as described above, allowed to regenerate for 7 days, and then fixed and stained with ECL or WGA. As we initially observed, labelling with either of these lectins in control animals injected with water showed the marginal adhesive zone was fully re-established after 7 days of regeneration (Figures 6A and D). Strikingly, we found that ECL staining of the secretory pores was completely eliminated after mag-1(RNAi) or Smed-rcn1(RNAi) (Figures 6B-C). In contrast to ECL, WGA was detected in the gland cells and cytoplasmic projections, but staining of the secretory pores was dramatically reduced after either gene was knocked down (Figures 6E-F). We conclude that mag-1 and Smed-rcn1 do not appear to be required for gland cell fate determination. These genes, however, may play a role in the maturation of the gland cell pores and regeneration of the marginal adhesive zone.

Figure 5.

RNAi knockdown of mag-1 and Smed-rcn1. Animals injected with water (controls) or dsRNA for either mag-1 or Smed-rcn1 were amputated anterior to the pharynx and allowed to regenerate for 7 days (n = 5 animals per group). Trunk fragments regenerating a new head were fixed and processed by in situ hybridization using riboprobes with no overlap to the dsRNAs (see Experimental Procedures). After mag-1(RNAi), expression of this gene was not detected in the newly regenerated heads (B). By contrast, mRNA expression of mag-1 was not affected in animals injected with water (A) or with Smed-rcn1 dsRNA (C). Similarly, the expression of Smed-rcn1 was not detected after Smed-rcn1(RNAi) (F) but was unaffected in animals injected with water (D) or mag-1 dsRNA (E). Anterior is up to the left in all panels. Scale bar = 200 μm.

Figure 6.

RNAi knockdown of mag-1 and Smed-rcn1 dramatically reduces lectin staining of the marginal adhesive zone. Animals injected with water or dsRNA for mag-1 or Smed-rcn1 were amputated at the pre-pharyngeal level and allowed to regenerate for 7 days. Regenerating fragments were fixed at specific time points and stained with either ECL- (A-C) or WGA-FITC (D-F). Control animals regenerate the marginal adhesive zone by 7 days (arrowheads in A and D; n = 29 and 30, respectively). mag-1(RNAi) (n = 33) or Smed-rcn1(RNAi) (n = 28) completely abolished staining with ECL. By contrast, WGA staining was dramatically reduced (mag-1(RNAi), n = 29; Smed-rcn1(RNAi), n = 28) in the marginal adhesive zone but was weakly detected in the neck projections of the subepidermal adhesive gland cells (arrowheads in E). Scale bar = 200 μm.

DISCUSSION

We tested a subset of commercially available fluorescent lectin-conjugates and found that the majority of these molecules stain planarian tissues (Table 1). Apart from the ability to label specific cell types by whole-mount in situ hybridization (Umesono et al., 1997; Sánchez Alvarado et al., 2002; Zayas et al., 2005), there are limited reagents available to label differentiated cell types in planarians (Robb and Sánchez Alvarado, 2002). The markers described here add additional tools to examine regeneration and cell turnover in planarians. Most notably, our results identified 13 lectins that strongly label the secretory system, including both acidophilic and cyanophilic gland cells, and their cytoplasmic projections and terminal pores. Lectins such as ECL, Jacalin, Pisum sativum agglutinin (PSA), and SBA marked the acidophilic subepidermal gland cells and prominently labeled the marginal adhesive zone. Cyanophilic gland cell bodies or processes were primarily visible when animals were stained with lectins such as GSL I, RCA I, and WGA. However, VVA stained both acidophilic and cyanophilic gland cells. Thus, in contrast to the gland cell classification using classic histochemical methods (Pedersen, 1963), secretory cell staining with lectins is most suited for labeling gland cell structures and can be combined with other specific markers.

Distinguishing gland cell types will be facilitated by in situ hybridization to detect genes expressed in different secretory cells. For instance, mag-1 and Smed-rcn1 are specifically expressed in the subepidermal adhesive gland cells and the secretory cells posterior to the photoreceptors (Sánchez Alvarado et al., 2002; Figure 3). Here we show that mag-1 and Smed-rcn1 are useful markers to track gland cell differentiation and migration after amputation (Figure 3). In situ hybridization to detect either of these transcripts in planarian regenerates stained cells initially located in the blastema near the epidermis (Figure 3, Day 2). During subsequent days of regeneration, mag-1- or Smed-rcn1-positive cells were observed near the epidermis, but also in cells possibly migrating deeper into the tissue (Figure 3, Days 3-5). These cells appeared to move posterior to the photoreceptors where the gland cells are ultimately located in the fully regenerated worm. Additionally, we examined the functional role of these genes using RNAi. We found that mag-1(RNAi) or Smed-rcn1(RNAi) animals were capable of regenerating gland cells (Figure 5), but proper differentiation of the marginal adhesive zone was disrupted, as indicated by a lack of lectin staining (Figure 6). One possible explanation for the observed reduction in lectin staining is that the carbohydrate modifications found on these gland projections are not present until the gland cells have reached their final position in the animal and become fully differentiated. However, the specific role of these genes in planarian secretory cells is not presently clear; further experimentation will be necessary to elucidate the mechanism underlying the decreased lectin-staining phenotype.

In the future, it will be interesting to also characterize the role of genes with expression patterns in pharyngeal or ventral secretory cells (Sánchez Alvarado et al., 2002). These molecular markers can be combined with fluorescent lectin-conjugates to reveal the morphology of distinct gland cell types (unpublished results). Furthermore, lectins can be combined with bromodeoxyuridine (BrdU) staining of the planarian stem cells (Newmark and Sánchez Alvarado, 2000) to examine the rate of gland cell turnover or renewal. Initial pulse/chase studies with BrdU followed by ECL staining suggest that subepidermal gland cells in intact planarians turn over slowly (Guo, 2007). This co-labeling approach will serve to resolve the extent to which gland cells are replaced during periods of feeding and starvation or after injury, and how far they migrate after differentiating, all of which remain open questions for the majority of cell types in planarians. Lectin staining, in combination with cell-specific gene markers and RNAi, will allow us to determine the contribution of different classes of gland cells to the animals' ability to glide, adhere to substrates, or catch prey. In conclusion, our results show that lectins provide inexpensive and simple-to-use reagents for staining planarians and will help to advance the study of these organisms.

EXPERIMENTAL PROCEDURES

Planarians

Asexual Schmidtea mediterranea (clonal strain CIW4) (Sánchez Alvarado et al., 2002) were reared and maintained as previously described (Cebrià and Newmark, 2005). Planarians 4-6 mm in length and starved for at least one week were used for all experiments.

Fluorescent lectin-conjugate staining

Planarians were treated with 2% HCl for 5 minutes on ice, and then fixed with Carnoy's fixative for 2 hours at 4°C (Umesono et al., 1997). After 1 hour wash in methanol at 4°C, planarians were bleached overnight in 6% H₂O₂ in methanol at room temperature. After bleaching, animals were re-hydrated in 75%, 50%, 25% methanol/PBTX (PBS with 0.3% Triton-X 100) and twice with PBTX for 5 min. Samples were then incubated in Blocking Buffer [0.6% IgG free BSA (Jackson Laboratories); 0.45% fish gelatin (Sigma, St. Louis, MO)] in PBTX for 2-4 hrs while shaking. After blocking, samples were incubated with lectins conjugated to either FITC (Con A, DBA, PNA, RCA I, SBA, UEA I, WGA, DSL, ECL, GSL II, Jacalin, LEL, STL, VVL, SNA) or Rhodamine (GSL I, LCA, PHA-E, PHA-L, PSA, Succinylated WGA, SJA) (Vector Labs) in Blocking Buffer for 4 hrs at room temperature or overnight at 4°C. Fluorescent lectin-conjugates were diluted in Blocking Buffer to concentrations ranging from 0.5-5 μg/ml. Typically, we incubated planarians with 1 μg/ml except for the following lectins that were used at 5 μg/ml: DBA, PNA, LCA, RCA I. After incubation with fluorescent lectins, animals were washed with PBTX six times for 1 hr at room temperature. DAPI (0.2 μg/ml) or TOTO-3 (0.2 μg/ml) was included in PBTX during the last wash or incubated overnight at 4°C. For double immunofluorescence and lectin staining, planarians were incubated overnight with an anti-phospho-tyrosine P-Tyr-100 antibody (Cell Signaling Technology) diluted 1:500 as previously described (Cebrià and Newmark, 2005). After washes, animals were incubated with goat anti-mouse Alexa Fluor 568 IgG secondary antibodies (Invitrogen) and WGA lectin. All samples were mounted in Vectashield (Vector Laboratories) and imaged with a Zeiss SteREO Lumar v.12 stereomicroscope, and either a CARV (BD Biosciences) or a Zeiss LSM510 confocal microscope.

Sequence analysis

cDNA clones H.1.3b (AY067799; mag-1) and H.10.8a (AY068175; Smed-rcn1) (Sánchez Alvarado et al., 2002) were fully sequenced using T3 and T7 primers, or gene specific primers. The complete clone sequences were examined by BLASTX against the non-redundant protein database at NCBI. Protein domains were predicted with the web-based tool SMART (Schultz et al., 1998). The complete cDNA sequences and predicted proteins are available from NCBI: mag-1, HM803280; Smed-rcn1, HM803281.

Whole-mount in situ hybridization

Planarians were fixed and bleached as described above. After bleaching, planarians were loaded into incubation columns in an Insitu Pro automated in situ hybridization instrument (Intavis) and processed as described previously (Sánchez Alvarado et al., 2002). Riboprobes were prepared by using the MAXIscript In Vitro Transcription Kit (Ambion) including digoxigenin (DIG)-UTP (Roche). Samples were visualized with a Leica MZ125 stereomicroscope and captured with a MicroFire digital camera (Optronics). DIC images were captured on a Nikon Eclipse TE200. The images were modified by adjusting the levels in Adobe Photoshop 8.0.

For double fluorescence in situ hybridization experiments, riboprobes were generated as described above or generated including fluorescein (FITC)-UTP (Roche) instead of DIG-UTP. After incubation with 1:100 anti-DIG-POD (Roche) overnight at 4°C, samples were washed 6 × 30 min with MABT (100 mM maleic acid/150 mM NaCl/0.1% Tween 20, pH 7.5) followed by two 10 min washes with PBST (0.01% Tween 20). Samples were pre-incubated with 1:1000 Cy3-tyramide or FITC-tyramide in PBST (0.01% Tween 20) for 5 min then developed by adding 0.001% H₂O₂ for 10 min. After developing, samples were washed with PBST 3 × 20 min. Peroxidase activity was quenched by incubating in 2% H₂O₂ /PBST for 1 hr then washed with MABT 3 × 20 min. After blocking, samples were incubated with 1:100 anti-FITC-POD (Roche) overnight at 4°C. Detection of the second riboprobe was carried out as described above. Samples were washed for 1-2 days with PBST then mounted with Vectashield and imaged as described for the fluorescent lectin staining above. A detailed protocol for Cy3- and FITC-tyramide synthesis is available from http://xenbase.org/methods.

RNA interference

Double-stranded RNA of clones H.1.3b (Smed-rcn1) and H.10.8a (mag-1) were prepared using the MEGAscript High Yield Transcription Kit (Ambion) and delivered by injection as described previously (Sánchez Alvarado and Newmark, 1999). After three consecutive days of injecting dsRNA into the animals, they were amputated at the pre-pharyngeal level and allowed to regenerate for seven days. To test the efficiency and specificity of RNAi, non-overlapping probes were synthesized for RNAi and in situ experiments. Clone H.1.3b (2,898 bp) was digested with XhoI and ClaI; the resulting fragments of 800 and 1,800 bp were subcloned into pBluescript II SK (+) cloning vector and used for synthesizing dsRNA and riboprobe for in situ hybridizations, respectively. Clone H.10.8a (1,121 bp) was digested either with XhoI and ClaI or EcoRI and ClaI; the respective resulting fragments of 311 and 800 bp were subcloned into pBluescript II SK (+) and then used to generate dsRNA and riboprobes.