JHCXXX10.1369/0022155416667928Lectin Binding Giant Danio and Zebrafish heartsManalo et al.

Article Journal of Histochemistry & Cytochemistry 2016, Vol. 64(11) 687­–714 © 2016 The Histochemical Society Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1369/0022155416667928 jhc.sagepub.com

Differential Lectin Binding Patterns Identify Distinct Heart Regions in Giant Danio (Devario aequipinnatus) and Zebrafish (Danio rerio) Hearts Trina Manalo,* Adam May,* Joshua Quinn,* Dominique S. Lafontant,* Olubusola Shifatu, Wei He, Juan M. Gonzalez-Rosa, Geoffrey C. Burns, Caroline E. Burns, Alan R. Burns, and Pascal J. Lafontant

Department of Biology, DePauw University, Greencastle, Indiana (TM, AM, JQ, DSL, OS, WH, PJL); Division of Cardiology, Department of Medicine, Cardiovascular Research Center, Massachusetts General Hospital, Charlestown, Massachusetts (JMG-R, GCB, CEB); and College of Optometry, University of Houston, Houston, Texas (ARB)

Summary Lectins are carbohydrate-binding proteins commonly used as biochemical and histochemical tools to study glycoconjugate (glycoproteins, glycolipids) expression patterns in cells, tissues, including mammalian hearts. However, lectins have received little attention in zebrafish (Danio rerio) and giant danio (Devario aequipinnatus) heart studies. Here, we sought to determine the binding patterns of six commonly used lectins—wheat germ agglutinin (WGA), Ulex europaeus agglutinin, Bandeiraea simplicifolia lectin (BS lectin), concanavalin A (Con A), Ricinus communis agglutinin I (RCA I), and Lycopersicon esculentum agglutinin (tomato lectin)—in these hearts. Con A showed broad staining in the myocardium. WGA stained cardiac myocyte borders, with binding markedly stronger in the compact heart and bulbus. BS lectin, which stained giant danio coronaries, was used to measure vascular reconstruction during regeneration. However, BS lectin reacted poorly in zebrafish. RCA I stained the compact heart of both fish. Tomato lectin stained the giant danio, and while low reactivity was seen in the zebrafish ventricle, staining was observed in their transitional cardiac myocytes. In addition, we observed unique staining patterns in the developing zebrafish heart. Lectins’ ability to reveal differential glycoconjugate expression in giant danio and zebrafish hearts suggests they can serve as simple but important tools in studies of developing, adult, and regenerating fish hearts. (J Histochem Cytochem 64:687–714, 2016) Keywords bulbus, concanavalin A, coronary, development, giant danio, heart, lectin, regeneration, WGA, zebrafish

Introduction Protein and lipid glycoconjugates expressed on the surface of animal cells serve diverse and essential biological functions.1 Their distributions across tissue types and across species are highly heterogeneous. Specific carbohydrate moieties within glycoconjugates are recognized by an important family of proteins: the lectins.2,3 Plant-based lectins in particular have long-established roles as tools for biochemical and histochemical studies in animals,4–6 and lectin

binding patterns reflect differential glycoconjugate expression in a tissue, the evolutionary history of organisms, developmental processes, or injury responses. Received for publication June 30, 2016; accepted August 11, 2016. *These authors contributed equally to this study. Corresponding Author: Pascal J. Lafontant, Department of Biology, DePauw University, 1 E. Hanna St., Olin 258, Greencastle, IN 46135, USA. E-mail: [email protected]

Journal of Histochemistry and Cytochemistry

667928

research-article2016

688 Lectins allow identification of specific cell and tissue types such as endothelial cells and their characterization as a result of differential cell membrane glycoconjugate expression.7,8 Depending on the lectin, binding to extracellular matrix proteins, cells, and tissue types can be species-specific or can be observed across species. As an example, Ulex europaeus agglutinin (UEA) binds selectively to human vascular endothelium but not to the endothelium of several other species tested.7,9,10 On the other hand, Bandeiraea simplicifolia lectin (BS lectin) displays a broader reactivity by binding to various, but not all, vascular beds across several species. Developmental and adult cardiovascular diseases remain a significant and global public health burden.11 Mammalian animal models such as the mouse, as well as non-mammalian models such as the newt and the axolotl, have contributed significantly to our understanding of cardiac development, growth, repair, and regeneration.12–16 In addition, a remarkable body of developmental and physiological research with relevance to mammalian cardiac biology has been performed in a variety of fish species.17–21 Among fish species, the zebrafish serves as a robust model for studies aimed at elucidating the molecular basis of cardiac development22,23 and regeneration.24,25 Heart repair and regeneration have also been studied in fish species such as the giant danio,26 medaka,27 and the goldfish.28 Lectins are routinely used in cardiovascular studies in human and mammalian models of cardiac biology29,30 as exemplified by their utilization in studies of cardiac myocytes, endocardium, and coronary endothelium. Various lectins have been used in the study of brain, kidney, and hematopoietic tissue of zebrafish.31–35 Yet lectins have rarely been applied to cardiovascular studies of non-mammalian models including the zebrafish. In studies where lectin reactivity with fish heart tissue has been tested, little to no binding was found.36 By contrast, we have successfully used wheat germ agglutinin (WGA) lectin to image cardiac myocyte cell borders in the adult zebrafish heart37 and BS lectin in giant danio heart sections to visualize the coronary vessel profiles.26 However, to date, no comprehensive studies of glycoconjugate expression and lectin binding patterns in the cardiovascular system of these fish species have been reported. In this study, we sought to determine the binding pattern of lectins in the hearts of the zebrafish and giant danio, two fish where cardiac regeneration has been documented, to characterize the distribution pattern of glycoconjugates. Specifically, we investigated the binding patterns of WGA, concanavalin A (Con A), BS lectin, Ricinus communis agglutinin I (RCA I), Lycopersicon esculentum agglutinin (tomato lectin), and UEA, which

Manalo et al.  are commonly used lectins in cardiac biology. We demonstrate differential lectin binding in zebrafish and giant danio hearts, suggesting lectins can serve as important tools in the study of fish cardiac biology.

Materials and Methods Animals and Surgical Procedures Giant danio, zebrafish, pearl danio, medaka, Buenos Aires tetra, Blue and Snakeskin gouramis, Chinese algae eater, Clownfish, rubbernose and Trinidad plecos, Red Glass Rosy barb, koi, and goldfish were obtained from thatfishplace.com (Lancaster, PA), and Segrest Farms (Gibsonton, FL). For developmental studies, wild-type TuAb and Tg(cmlc2:nucDsred)38 embryo, larvae, and adults were used. Adult zebrafish, giant danio, and pearl danio were raised in 10- and 20-gallon aquaria at 28C. All other fish were maintained in 10-gallon tanks at 25C. All fish were kept on 14/10 hr day/night light cycles. Experimental procedures were approved by the Committee for the Care and Use of Laboratory Animals at DePauw University. Fish were euthanized in 0.2% MS Tricaine (SigmaAldrich, St. Louis, MO) in deionized water. Once opercular movements ended, the fish were weighed and their length recorded. The hearts were quickly removed by cutting the ventral aorta proximal to the bulbus, lifting the ventricle and severing the atrium. The hearts were briefly washed in PBS to remove the luminal blood. The heart (ventricle and bulbus) was gently drained of excess buffer, weighed, and transferred to ice cold fixative (500 µl of PBS buffer containing 4% paraformaldehyde) for 3 hr or overnight at 4C. Cautery injury to the apical region of the giant danio heart was performed as described previously.26 Briefly, giant danios were removed from tanks and anesthetized with 0.02% MS Tricaine for 1 to 2 min. The fish were placed ventral side up in a slit cut in a sponge block, under a Leica ZOOM 2000 dissecting microscope (Leica Microsystems, Bannockburn, IL). The skin, pectoral muscles, and the pericardium were gently dissected midline to reveal the heart. The tip of the heated wire was gently applied to the apex of the heart. Following apical cauterization, fish were returned to a new freshwater tank. Fish were sacrificed at designated time points using 0.2% MS Tricaine.

Tissue Preparation Fixed hearts were washed in PBS and transferred to PBS containing 30% sucrose for cryoprotection, 3 hr to overnight. Next, the hearts were washed in 1X PBS and embedded in 13-mm-diameter aluminum seal

689

Lectin Binding Giant Danio and Zebrafish Hearts cups (Wheaton, Millville, NJ) filled with freezing medium Tissue-Tek O.C.T. (Ted Pella, Torrance, CA), frozen, and stored at −80C. The hearts were sectioned (10 µm thick) sagitally and mounted on TruBond 360 (Electron Microscopy Sciences, Hatfield, PA) or Superfrost Plus slides (Fisher Scientific, Waltham, MA). For whole-mount studies, the adult hearts and zebrafish embryos were fixed in 4% paraformaldehyde as previously described and permeabilized with 0.5% Triton X-100 in 1X PBS overnight at 4C. The hearts were washed and stained immediately.

Lectin Staining Procedures Fluorescein-conjugated and rhodamine-conjugated BS lectin, Con A, WGA, RCA I, UEA, and tomato lectin (Vector Laboratories, Burlingame, CA), and FITCconjugated BS lectin and Texas Red–conjugated tomato lectin (Sigma-Aldrich, St. Louis, MO) were used. For lectin binding and detection, sections were incubated for 2 hr at room temperature or overnight at 4C in lectin-containing TBS (50-mM Tris, 150-mM NaCl, 1-mM CaCl2, 1-mM MgCl2, pH 7.6) or Hepes buffer (10-mM Hepes, 0.15-M NaCl, 0.1-M CaCl2, pH 7.5) at the following final lectin concentrations: BS lectin (1:400, 5 µg/ml), Con A (1:1000, 2 µg/ml), WGA (1:1000, 2 µg/ml), RCA I (1:500, 4 µg/ml), tomato lectin (1:200, 10 µg/ml), and UEA (1:500, 4 µg/ml). For control, N-acetylglucosamine, N-acetylgalactosamine, and α-methylmannoside (Vector Laboratories) were diluted to 500-mM final concentration in TBS. WGA and tomato lectin were incubated in N-acetylglucosamine, BS lectin with N-acetylgalactosamine, RCA I with d-galactose and N-acetylgalactosamine, and Con A with α-methylmannoside at a final lectin concentration of 2 to 10 µg/ml, for 1 hr at room temperature. Each preabsorbed lectin solution was reacted with giant danio heart sections for 1 to 2 hr at room temperature in parallel with the unabsorbed lectin solution at the same concentration. Preabsorption with the various sugars either abrogated lectin binding, or markedly decreased binding and showed minimal trace binding to the fish myocardium. Following lectin incubation, the slides were washed in PBS, and cell nuclei were stained with Hoechst stain (Invitrogen, Eugene, OR), washed in PBS, and coverslipped using PermaFluor mounting medium (Thermo Fisher Scientific, Fremont, CA). For whole-mount staining of adult fish hearts and developing zebrafish embryos, the heart and embryos were incubated in 400 µl of lectin at the same concentration described above. Adult hearts and embryos were mounted in glass-bottom tissue culture dishes (MatTek, Ashland, MA) in 2% low-melting agarose (Sigma-Aldrich).

Tissue Imaging and Image Analysis Stained tissues were imaged using a Zeiss Axio Imager 2 equipped with an ApoTome.2 for optical sectioning (Zeiss, Göttingen, Germany) or a Nikon A1R confocal laser scanning microscope (Nikon Instruments, Melville, NY). Maximum intensity projections of .czi and .nd2 image stacks were performed using Fiji.39 For quantitation of vascular density in whole-mount control and injured hearts, maximum intensity projections of image stacks were displayed on a Dell 17-inch monitor and overlaid with a point-counting grid with 280 equidistant point intercepts as previously described.26 Two to three fields per heart were evaluated. Vascular density was calculated based on the fraction of point intercepts landing randomly on vessels to the total number of point intercepts covering the projected field.

Statistics Data are expressed as means and standard errors. A one-way ANOVA followed by a Tukey post hoc test was used to assess differences. A p value of ≤0.05 was considered significant.

Results The overall results of lectin staining of zebrafish and giant danio hearts are summarized in Table 1.

Con A Staining in Adult Giant Danio and Zebrafish Hearts We studied lectin binding patterns in various anatomic regions, segments, or tissue components of the zebrafish and giant danio hearts, including the bulbus (outflow tract), the compact, the spongy, and the junctional regions of their ventricles, the valves, and coronary vasculature. Con A displayed nearly uniform and strong staining of the bulbus (Fig. 1A), compact heart, and valves of the giant danio (Fig. 1B and C). In the spongy heart, Con A reactivity was higher in the endocardial cells bordering the trabeculae than in the membrane of the cardiac myocytes (Fig. 1D). In the compact hearts of both fish, Con A reactivity was strong in the cardiac myocyte borders, and stronger in the compact heart coronary vessels, the epicardium, and in the junctional region between the compact and spongy hearts (Fig. 1F). Whether the staining was present in the basement membrane, the junctional space between the compact and spongy layers, or both, could not be resolved. A similar staining pattern was also observed in the zebrafish, with clear delineation of cardiac myocyte borders in the spongy heart

690

Manalo et al. 

Table 1.  BS Lectin Binding in Adult Zebrafish and Giant Danio Heart Regions.

Zebrafish   Compact myocardium   Spongy myocardium   Bulbus arteriosus  Valves   Junctional region Giant danio   Compact myocardium   Spongy myocardium   Bulbus arteriosus  Valves   Junctional region

WGA

Con A

BS Lectin

UEA

RCA I

Tomato

++ + ++++ +++ +++

+++ ++ +++ +++ ++++

− − − − −

− − − − −

++ − + ++ +++

− − ++ ++ +

++ + ++++ +++ +++

+++ ++ +++ +++ ++++

+++ (*) + (#) +++ (*) + (#) −

− − − − −

++++ − + ++ ++++

++++ ++ ++ ++++ ++

(*) vascular and bulbus endothelium; (#) endocardium; (+) weak, (++) moderate, (+++) strong, (++++) very strong. Abbreviations: BS, Bandeiraea simplicifolia; WGA, wheat germ agglutinin; Con A, concanavalin A; UEA, Ulex europaeus agglutinin; RCA I, Ricinus communis agglutinin I.

(Fig. 1E) and an overall higher staining level in the compact heart than in the trabeculae (Fig. 1G).

WGA Staining in Adult Giant Danio and Zebrafish Hearts Differential patterns of WGA lectin staining were observed in the giant danio and zebrafish myocardia and their respective outflow tracts. The strongest reactivity to FITC-conjugated WGA lectin was present in the bulbus (Fig. 2A and B) and extended into the bulboventricular (BV) and atrioventricular (AV) valves of the zebrafish (Fig. 2C and D). At higher magnification, differential WGA staining was seen in the zebrafish ventricle. WGA staining was markedly stronger in the compact heart as well as the in the junctional region at the interface of the spongy and compact hearts of the zebrafish (Fig. 2E). WGA staining was also observed but with lesser intensity in the trabeculated myocardium compared with that in the compact myocardium, the valves, and the bulbus. At higher magnification, WGA binding allowed visualization of the borders of compact and trabeculated cardiac myocytes in the ventricle (Fig. 2F) and of cardiac myocytes in the atrium (not shown). A similar staining pattern was also observed in the giant danio with staining strongest in the bulbus and valves (Fig. 2G). In the ventricle, staining of the compact heart was markedly higher than that of the spongy heart (Fig. 2H and I).

BS and UEA Lectin Staining in Adult Giant Danio and Zebrafish Hearts We confirmed previous findings26 that BS lectin strongly stains the coronary vasculature of the adult giant danio compact heart (Fig. 3A). Less intense BS

staining was also observed in the ventricular myocardium. We also found intense BS lectin staining on endothelia of the AV valve (Fig. 3B) as well as the BV valve. The staining in the BV valve was continuous with the endothelium of the bulbus, whereas no reactivity was seen in the bulbus mesenchyme (Fig. 3C and D). In the zebrafish, we were unable to detect BS lectin in the compact myocardium and its vessels (Fig. 3E), nor in the spongy myocardium and endocardium, the bulbus, or the valves (Fig. 3F, Table 1). We further investigated whether the visualization of the coronary vasculature in the giant danio and zebrafish could be performed using UEA, another marker of mammalian endothelium. We found that UEA did not stain the coronary vasculature of either the zebrafish or the giant danio (Table 1). To determine whether the BS lectin staining pattern in coronary vasculature was unique to the giant danio, we studied several additional fish species under a dissecting microscope and in sections double stained with Con A and BS lectin (Table 2). In the majority of these fish species, the coronary vasculature could readily be seen through direct observation of whole-mount tissue. In two of these species, a vasculature could not be observed. In koi, ventricular coronary vessels could be easily visualized in whole-mount (Fig. 4A) and in Con A–stained sections (Fig. 4B). However, BS lectin did not react with the koi coronary vessels (Fig. 4C). Similarly, coronary vessels were readily observed in the Buenos Aires tetra in whole-mount (Fig. 4D) and in 10-µm sections stained with Con A (Fig. 4E). Yet no BS lectin reactivity was observed in these coronary vasculatures (Fig. 4F). In addition, we also found that BS lectin strongly reacted to the heart of two groups of fish: the gouramis and the plecos. BS lectin showed intense staining along the bulbus and valve endothelia of

Lectin Binding Giant Danio and Zebrafish Hearts

691

Figure 1.  Differential Con A lectin binding in the giant danio and zebrafish hearts. Con A-fluorescein lectin binding (green) is observed throughout the giant danio heart with the strongest staining seen on the bulbus (A), the CH (B), and the valves (C; nuclei, blue). At higher magnification, Con A binds to the trabecular bundles in the spongy myocardium and to cardiomyocyte membranes (D). Con A binding is comparatively stronger in the CH compared with that in the SH (E). Within the compact myocardium, Con A binding is particularly strong in the epicardium (arrow) and the coronary vasculature (*). The junctional region (arrowhead) is equally stained, forming a boundary line between the CH and SH, with small regions of discontinuity at regular intervals. A similar pattern is seen in the zebrafish heart with Con A staining cardiomyocyte borders of the SH (F) and CH (G). Abbreviations: Con A, concanavalin A; SH, spongy heart; CH, compact heart. Scale bars: A–G = 10 µm.

692

Manalo et al. 

Figure 2.  Differential wheat germ agglutinin (WGA) lectin binding in the zebrafish and giant danio hearts. In the zebrafish, the strongest WGA lectin binding was observed in the bulbus mesenchyme (A, B) with the ventricle appearing minimally stained by comparison, and in the valves (C, D). At higher magnification, WGA could be seen staining the compact heart, the trabecular bundles of the spongy heart (E), and the cardiomyocyte borders (F). In the giant danio, the staining pattern was similar with the strongest staining level in the bulbus mesenchyme (G) with the ventricle appearing minimally stained. When imaged without the bulbus, WGA staining is seen throughout the ventricle but with higher levels in the compact heart and the valves (H). At higher magnification, WGA was observed to stain cardiac myocyte borders as well as the junctional region (arrowhead) of the giant danio heart (I). Scale bars: A–I = 50 µm.

gouramis (Fig. 4G) as well as a population of bulbus subepicardial cells (Fig. 4H). Strong BS lectin reactivity was observed in the ventricular endocardium (Fig. 4I). Moreover, in the pleco, we found that that BS lectin strongly stained the bulbus endothelium (Fig. 4J) as well as a population of bulbus subepithelial cells (Fig. 4K). Furthermore, BS lectin strongly stained the plecos’ ventricular endocardium as well as the coronary vessel endothelium of the compact heart (Fig. 4L).

Whole-Mount Staining and Imaging of Fish Coronary Vasculature The pattern of BS lectin binding in the giant danio, blue gouramis, and pleco coronary vessels observed in 10-µm sections indicated that these vessels are restricted to the compact hearts of these fish. We performed whole-mount staining to generate three-dimensional images of the entire coronary vasculature. We

693

Lectin Binding Giant Danio and Zebrafish Hearts

Figure 3.  Differential BS lectin binding in the giant danio and zebrafish hearts. BS lectin strongly stained the endothelium of the coronary vasculature of the CH, with minimal binding seen in the ventricular endocardium (A). Strong BS lectin binding was seen in the endothelium of the valves (B). BS lectin strongly stained the bulbus (C). Higher magnification show that the staining was localized in the bulbus endothelium folds (D). No BS lectin binding was observed in the zebrafish ventricle (E), including the valves (F). Abbreviations: BS, Bandeiraea simplicifolia; CH, compact heart. Scale bars: A–F = 50 µm. Table 2.  BS Lectin Staining in Fish Hearts. BS Lectin Reactivity Species

Coronary Vessels Observed

Coronary Endothelium

Bulbus Endothelium

Yes Yes No Yes Yes Yes Yes Yes Yes No Yes No Yes

− + − + + − − − + − + − −

− + − + + − − − − − + − −

Buenos Aires tetra (Hyphessobrycon anisitsi) Blue gourami (Trichopodus trichopterus) Chinese algae eater (Gyrinocheilus aymonieri) Clown pleco (Panaqolus maccus) Giant danio (Devario aequipinnatus) Goldfish (Carassius auratus) Koi (Cyprinus carpio) Pearl danio (Danio albolineatus) Rubbernose pleco (Chaetostoma milesi) Red Glass Rosy barb (Puntius conchonius) Snakeskin gourami (Trichogaster pectoralis) Trinidad pleco (Hypostomus plecostomus) Zebrafish (Danio rerio) (+) present; (−) absent. Abbreviation: BS, Bandeiraea simplicifolia.

found that BS lectin staining could be achieved in the adult whole-mount giant danio heart (Fig. 5A) but not

in the zebrafish heart (Fig. 5B). Similar result was also obtained in the gourami and pleco (Fig. 5C and D).

694

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

695

Editor’s Highlight

(continued)

696

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

697

Editor’s Highlight

(continued)

698

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

699

Editor’s Highlight

(continued)

700

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

701

Editor’s Highlight

(continued)

702

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

703

Editor’s Highlight

(continued)

704

Manalo et al. 

Editor’s Highlight

(continued)

Lectin Binding Giant Danio and Zebrafish Hearts

705

Editor’s Highlight

(continued)

706

Manalo et al. 

Editor’s Highlight Figure 4.  Differential BS lectin binding in various fish species hearts. Coronary vessels observed in the koi ventricle under dissecting microscope (A). Con A staining of 10-µm section of the koi ventricle with coronaries (*) (B). No BS lectin was observed in the vessels nor in the ventricle of koi hearts (C). Coronary vessels observed in the BAT ventricle under dissecting microscope (D). Con A staining of 10-µm section of the BAT ventricle (E; *, vessel). No BS lectin was observed in the vessels nor in the ventricle of BAT hearts using BS lectin-fluorescein, or BS lectin-rhodamine (F). Strong BS lectin binding in the gourami bulbus endothelium folds as well as in the ventricular endocardium (G). BS lectin also strongly binds to cell profiles on the bulbus external layer (H). At higher magnification, BS lectin binding was seen in the endocardium lining trabecular myocardial bundles (I). In the rubbernose pleco, BS lectin strongly stained the bulbus endothelium and the ventricular endocardium (J). At higher magnification, staining of cells in the external layer of the bulbus (K), the endocardium, epicardium, and subepicardium was observed (L). Abbreviations: BS, Bandeiraea simplicifolia; Con A, concanavalin A; BAT, Buenos Aires tetra; SH, spongy heart; CH, compact heart. Scale bars: B, C, E–L = 50 µm; A, D = 1 mm.

Figure 5. Whole-mount staining and imaging of fish coronary vasculature. Widefield imaging of Bandeiraea simplicifolia (BS) lectinfluorescein stained adult giant danio heart where large and small coronary vessels can be visualized (A). Widefield imaging of BS lectinfluorescein stained zebrafish heart showing absence of BS lectin staining (B). Maximum projection of optical sections showing a vascular network on the gourami bulbus (C) and in the pleco ventricle (D). Maximum projection of optical sections showing a vascular network of uninjured giant danio heart (E) and during regeneration at days 3 (F) and 30 (G) following apical ventricular cautery injury. Dashed line highlights the border zone between injured and non-injured areas of the ventricle. Quantitation of coronary vascularization (H) during regeneration. Scale bars: A–G = 500 µm. GD = giant danio.

Using this approach, we performed BS lectin staining to assess the loss of coronary vasculature following injury and to evaluate reconstruction of the coronary vascular network in regenerating giant danio hearts. We performed BS lectin staining in postfixed and permeabilized control and cauterized hearts. The intact pattern of coronary vessels was observed in the control ventricle (Fig. 5E). Following cauterization, the BS lectin–based vascular pattern was absent at the apical region of the ventricle. By contrast, the vascular pattern was maintained in areas remote to the injury, in

the mid-myocardium, resulting in a clearly identifiable border zone (Fig. 5F). The myocardial apical region became progressively vascularized and appeared complete by 30 days postinjury (Fig. 5G and H).

RCA I Lectin Staining in Adult Giant Danio and Zebrafish Hearts In the zebrafish, RCA I lectin weakly stained the bulbus endothelium and mesenchyme (Fig. 6A). Little reactivity was observed in the spongy myocardium. By contrast,

707

Lectin Binding Giant Danio and Zebrafish Hearts

Figure 6.  Differential RCA I lectin binding in giant danio and zebrafish hearts. Low level of RCA I lectin staining was seen in the zebrafish bulbus mesenchyme (A). Strong RCA I lectin binding is seen in the zebrafish CH, and particularly in the junctional region (B). At higher magnification, RCA I staining can be observed in the epicardium, cardiomyocyte borders, and in coronary vessel profiles, albeit at a lower level than the junctional region (C; *, vessel). A similar binding pattern was seen in the giant danio with strong binding in the CH (D) and little to no reactivity in the spongy myocardium. At higher magnification, RCA I is observed to bind strongly in the epicardium, the junctional region, cardiomyocyte borders, and coronary vessel profiles (E; *, vessel). Abbreviations: RCA I, Ricinus communis agglutinin I; SH, spongy heart; CH, compact heart. Scale bars: A–E = 20 µm.

strong reactivity was seen in the compact heart, in particular at the junctional region between the compact and spongy ventricles (Fig. 6B). Close inspection confirmed that the strongest reactivity appeared in the junctional region, the epicardium, and the capillary endothelium of the coronary vessels, with weaker staining on cardiac myocyte borders (Fig. 6C). In the giant danio, RCA I staining pattern was similar to that of the zebrafish and was particularly evident in the compact myocardium with little to no reactivity in the spongy heart (Fig. 6D). In addition, high RCA I reactivity was seen in the basal aspect of the transitional cardiac myocytes of the junctional region, and the epicardium. RCA I binding was also strong in the endothelium of coronary vessels and compact cardiac myocyte borders (Fig. 6E). RCA I lectin staining was also strong in the AV and BV valves of both fish (not shown).

Tomato Lectin Staining in Adult Giant Danio and Zebrafish Hearts In the giant danio, tomato lectin reactivity was observed in the bulbus endothelium and the adjacent mesenchyme, as well as the AV and BV valve mesenchymes

(Fig. 7A and B). Tomato lectin reactivity was intense in the compact heart (Fig. 7C), outlining compact cardiac myocyte borders (Fig. 7D) similar to that observed with WGA staining. Tomato lectin binding could also be observed in the spongy myocyte borders (Fig. 7E) but to a much lesser extent than in the compact heart myocytes’ borders. Tomato lectin displayed differential but relatively weak binding in sections of the zebrafish heart, with the exception of the bulbus endothelium and portion of the mesenchyme (Fig. 7F), and the AV and BV valve regions (Fig. 7G). Reactivity was also observed in the membrane of the transitional cardiac myocytes (also called primordial) of the zebrafish junctional region (Fig. 7H and I).

Lectins Staining in the Developing Zebrafish Heart (WGA, Con A, and RCA I) Given the potential of using lectins to study cardiac development, we performed whole-mount staining in zebrafish embryos. In the absence of trabeculae at this early stage, WGA relatively weakly stained cardiac

708

Manalo et al. 

Figure 7.  Differential tomato lectin binding in giant danio and zebrafish hearts. Tomato lectin staining was present in the bulbus endothelium and mesenchyme, and strong in the valves (A, B; higher magnification of bulbus) of the giant danio. Binding was also strong in the CH, staining the cardiomyocyte borders (C, D; higher magnification). A similar binding pattern was seen in the SH but weaker (E). In the zebrafish, tomato lectin staining was present in the bulbus (F) and the valve (G). Little to no reactivity was seen in the ventricle except for low but discernable binding in the junctional region (H) and to the borders of the transitional cardiac myocytes (I) bridging the compact and spongy myocardia. Abbreviations: SH, spongy heart; CH, compact heart. Scale bars: A–I = 20 µm.

myocyte borders and the endocardium of the singlecell-layered compact heart. Intense WGA staining was observed in the developing bulbus by 72 hr postfertilization (Fig. 8A and B), but the valve could not be readily visualized. Staining with Con A showed strong binding in the epicardium, the endocardium, and cardiac myocyte borders (Fig. 8C and D). RCA I staining was strongly positive in the developing bulbus, the endocardium, and the epicardium (Fig. 8E and F). In contrast to WGA, Con A, and RCA I, we could not detect BS lectin, tomato lectin, or UEA staining in the developing zebrafish heart.

Discussion This is the first comprehensive investigation of the lectin binding patterns in the giant danio and zebrafish hearts. It reveals the differential expression, level, and distribution of glycoconjugates in the hearts of these two closely related fish species used in cardiac regeneration studies.

We selected a panel of plant-based lectins most commonly used in functional, histological, and biochemical studies of mammalian29,30,40,41 and non-mammalian36,42,43 hearts. As lectins recognize specific carbohydrate moieties on membrane, cytosolic, and extracellular glycoconjugates, and different lectins can bind the same carbohydrate moieties, they can be categorized into specificity groups.44,45 For these studies, we tested the binding pattern of six lectins belonging to five specificity classes that include mannose/glucose-binding lectin (Con A), N-acetyl-glucosamine (and N-acetylneuraminic acid)-binding lectin (WGA, tomato lectin), galactose/ N-acetylgalacto-samine-binding lectin (BS lectin), d-galactose-binding lectin (RCA I), and fucose-binding lectin (UEA). Observation of Con A staining indicates that mannose residues are ubiquitously expressed in the ventricle, endocardium, myocytes, the valves, and bulbus mesenchymes, as well as in the endothelium of coronary

Lectin Binding Giant Danio and Zebrafish Hearts

709

Figure 8.  Differential lectin binding in developing zebrafish heart. WGA lectin staining of zebrafish heart 72 hr postfertilization shows strong staining of the bulbus mesenchyme (A, B), and little to no staining of the ventricle. Con A stains the endocardium and epicardium of the developing heart (C) as well as cardiac myocyte borders (D). RCA I stains strongly the epicardium and myocardium (E) as well as the developing bulbus (F). Abbreviations: WGA, wheat germ agglutinin; Con A, concanavalin A; RCA I, Ricinus communis agglutinin I. Scale bars: A–F = 10 µm.

710 vessels. Mannose is essential to N-linked glycosylation of proteins.46 This could explain the broad, strong, and ubiquitous staining of the heart by Con A. However, Con A binding sites were markedly enriched in the compact myocardium in both the giant danio and zebrafish. The differential level of staining between the spongy and compact hearts supports the notion of phenotypic heterogeneity within the zebrafish heart. The compact heart in the zebrafish arises not during development but during the juvenile stage. Indeed, the compact heart emerges from a number of cardiac myocyte clones that originate from the base of the ventricle,47 and its accretion is progressive into adulthood. The different level of glycosylation using mannose may be another difference underlying or reflecting the heterogeneity in the zebrafish and giant danio ventricles. Enhanced Con A staining was also observed in the junctional region and indicated that mannose was enriched in the abluminal aspect of the transitional (primordial) cardiac myocytes, including the basement membrane. We have previously shown that the compact and spongy hearts are separated by a junctional space containing fibroblasts forming a network framed by a set of flat elongated transitional cardiac myocytes that subtend the spongy trabeculae and anchor the compact myocardium to the spongy myocardium.37 The transitional cardiac myocyte abluminal surfaces for the most part are not directly apposed to the compact heart cardiac myocytes; instead, they make contacts with adjacent cardiac myocytes at regular intervals through mostly narrow contact bridges, a pattern also observed in other fish species.48 We note that in the junctional region, Con A staining was absent at regular intervals. These gaps in the Con A staining may represent the region of contact between transitional (primordial) myocytes and the adjacent cardiac myocyte in the compact (cortical) myocardium, where desmosomes and fascia adherentes are abundant and where basement membrane may be absent. Observation of WGA and tomato lectin staining indicated that N-acetylglucosamine is expressed throughout the myocardium and bulbus of the giant danio and zebrafish hearts. However, although the binding was similar in some regions of the heart, the staining pattern and intensity diverged in other areas, and differentially across the two species. The highest WGA binding was observed in the bulbus mesenchyme, valves, and compact hearts of the two species. The WGA binding pattern overlapped with that seen with Con A staining, suggesting a higher level of expression of various glycoconjugates in the compact heart as compared with that in the spongy heart, further supporting the notion of phenotypic differences between compact and

Manalo et al.  spongy heart myocardia. However, by contrast to Con A, WGA did not appear to react with coronary vessels or the bulbus endothelium of the adult giant danio and zebrafish hearts. In contrast to WGA, tomato lectin patterns diverged in the giant danio and zebrafish hearts. On the one hand, the binding pattern of tomato lectin was similar to that observed with WGA in the giant danio, and this pattern was consistent with their shared glycoconjugate binding specificities. However, tomato lectin reactivity was generally lower than that of WGA. This may reflect differences in the affinity of the two lectins for the glycoconjugates. On the other hand, in the zebrafish heart, tomato lectin displayed a different pattern than that observed in the giant danio, as well as WGA-stained hearts. Indeed, the binding of tomato lectin was discernable but weak in the ventricle and was mostly restricted to the transitional cardiac myocyte borders in the junctional region. Whether these differences could be explained by the affinity of these lectins for these residues within each species, or by differential patterns of expression on the glycoconjugates, or by both, is not clear. Nevertheless, our findings suggest glycoconjugate expression is heterogeneous across species. The pattern of BS lectin staining in the giant danio heart indicates that N-acetylgalactosamine/galactose conjugates are highly expressed in the bulbus and valve endothelia, consistent with that reported in mammalian coronary vessels.49 Staining of the endocardium in the giant danio was much lower by comparison. Surprisingly, BS lectin did not stain the coronary endothelium, the endocardium, the myocardium, the interstitium, nor the bulbus of the zebrafish heart under the tested conditions. This finding suggests an absence or inaccessibility of N-acetylgalactosamine moieties, despite the close phylogenic relationship to the giant danio. BS lectin reactivity was also not seen in koi, another closely related cyprinid, and was consistent with our observation in the goldfish heart28 where little to no staining was seen on the uninjured heart. By contrast, BS lectin reactivity was detected in the coronary endothelia of phylogenetically distant fish such as plecos and gouramis. However, unlike the giant danio, BS lectin staining was present at high level in both the endocardium and coronary endothelium, underlying the complexities of glycoconjugate patterns of expression. In these fish, the relatively superficial localization of the coronary vessels enabled us to develop a novel whole-mount staining protocol to visualize the three-dimensional architecture of the entire adult coronary vascular network that obviates the need for microtome sectioning. Using this approach,

711

Lectin Binding Giant Danio and Zebrafish Hearts we were able to quantify coronary vessels in the giant danio during repair and regeneration. Similar to BS lectin, RCA I has been used to stain and image mammalian vascular beds, particularly the brain.50,51 RCA I stained the coronary vessels in both the giant danio and the zebrafish. However, intense staining was also seen along the cardiac myocyte borders, the epicardium, and the junctional region, making it less suitable for specifically visualizing coronary vessels. Finally, our study provides preliminary insights into the expression of glycoconjugates during heart development in zebrafish. Changes in lectin binding patterns during development have been reported in chick52,53 and mouse hearts.54,55 Our results suggest mannose (Con A), N-acetylglucosamine (WGA), and d-galactose (RCA I) are expressed in the developing myocardium, whereas N-acetylgalac-tosamine (BS lectin) and fucose (UEA) are not. The lack of BS lectin staining is consistent with our observation in the adult zebrafish heart. N-acetylglucosamine (WGA) and d-galactose (RCA I) are also highly expressed in the zebrafish bulbus, which is analogous to the pattern observed in the adult fish. This suggests that the pattern of expression of glycoconjugates is developmentally regulated and maintained in adulthood. However, although wholemount imaging allowed us to visualize the regional patterns of glycoconjugate expression in the developing heart, we were limited in our ability to precisely locate and discriminate between binding to the endocardium and binding to cardiac myocytes, or to the epicardium and cardiac myocytes, or to basement membranes. This level of localization requires the use of gold-labeled lectins and electron microscopy. Nevertheless, our studies in the zebrafish heart demonstrate concordance in the binding pattern of key lectins in both the developing and adult zebrafish hearts. In conclusion, this is the first comprehensive study of lectin binding patterns in fish hearts, in which we demonstrate that the expression of glycoconjugates varies in different heart regions of adult zebrafish and giant danio. This study also shows that the spatial distribution of glycoconjugates can be analogous and consistent, or heterogeneous and divergent within and across species. Importantly, we document the early expression of glycoconjugates in the developing zebrafish heart. The functional significance of the temporal and spatial expression patterns of glycoconjugates and their potential role in development and regeneration are yet to be explored. Nevertheless, these studies establish lectin histochemistry as an invaluable imaging tool for investigating the development, growth, and regeneration in the hearts of giant danio and zebrafish.

Acknowledgments We wish to thank Wendy Tomamichel, Lauren Chen, Sarah Glasshagel, and Thomas Shelton for technical support, and Clay Higginbotham for help with animal care.

Competing Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Author Contributions TM, OS, WH performed wheat germ agglutinin, Bandeiraea simplicifolia (BS) lectin, and concanavalin A staining. DSL performed BS, Ricinus communis agglutinin, and tomato lectin staining and imaging. JQ, AM, and PJL performed surgery and lectin staining. JMGR, GCB, CEB, ARB designed experiments and revised manuscript. PJL initiated the study, designed and interpreted experiments, and wrote the manuscript.

Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: P.J.L. is supported by NIH 1 R15 HD084262-01, Donald E. Town Faculty Fellowship, and Faculty Development Funds at DePauw University. A.R.B. is supported by grant NEI P30EY007551. J.M.G.-R. was supported by EMBO Long-term Fellowship ALTF 253-2014 and a Fund for Medical Discovery Fellowship awarded by the Executive Committee on Research at Massachusetts General Hospital. G.C.B. and C.E.B. were supported by National Institutes of Health/National Heart, Lung, and Blood Institute (NIH/NLBLI) grant 5R01HL127067.

Literature Cited 1. Brandley BK, Schnaar RL. Cell-surface carbohydrates in cell recognition and response. J Leukoc Biol. 1986;40:97–111. 2. Debray H, Decout D, Strecker G, Spik G, Montreuil J. Specificity of twelve lectins towards oligosaccharides and glycopeptides related to N-glycosylproteins. Eur J Biochem. 1981;117:41–55. 3. Wu AM, Sugii SJ, Herp A. A guide for carbohydrate specificities of lectins. Adv Exp Med Biol. 1988;228:819–47. 4. Danguy A, Decaestecker C, Genten F, Salmon I, Kiss R. Applications of lectins and neoglycoconjugates in histology and pathology. Acta Anat (Basel). 1998;161:206–18. 5. Roth J. Lectins for histochemical demonstration of glycans. Histochem Cell Biol. 2011;136:117–30. 6. Akimoto Y, Kawakami H. Histochemical staining using lectin probes. Methods Mol Biol. 2014;1200:153–63. 7. Holthofer H, Virtanen I, Kariniemi AL, Hormia M, Linder E, Miettinen A. Ulex europaeus I lectin as a marker for vascular endothelium in human tissues. Lab Invest. 1982;47:60–66.

712 8. Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins. Identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2745–56. 9. Roussel F. [Labelling endothelial cells with the lectins from Cytisus sessilifolius and Ulex europaeus; comparison between human and animal cells]. C R Seances Soc Biol Fil. 1985;179:790–4. 10. Alroy J, Goyal V, Skutelsky E. Lectin histochemistry of mammalian endothelium. Histochemistry. 1987;86: 603–7. 11. Moran AE, Forouzanfar MH, Roth GA, Mensah GA, Ezzati M, Flaxman A, Murray CJ, Naghavi M.The global burden of ischemic heart disease in 1990 and 2010: the Global Burden of Disease 2010 study. Circulation. 2014;129:1493–501. 12. Oberpriller JO, Oberpriller JC. Response of the adult newt ventricle to injury. J Exp Zool. 1974;187:249–53. 13. Bader D, Oberpriller JO. Repair and reorganization of minced cardiac muscle in the adult newt (Notophthalmus viridescens). J Morphol. 1978;155:349–57. 14. Flink IL. Cell cycle reentry of ventricular and atrial cardiomyocytes and cells within the epicardium following amputation of the ventricular apex in the axolotl, Amblystoma mexicanum: confocal microscopic immunofluorescent image analysis of bromodeoxyuridinelabeled nuclei. Anat Embryol (Berl). 2002;205:235–44. 15. Laube F, Heister M, Scholz C, Borchardt T, Braun T. Re-programming of newt cardiomyocytes is induced by tissue regeneration. J Cell Sci. 2006;119:4719–29. 16. Piatkowski T, Muhlfeld C, Borchardt T, Braun T. Reconstitution of the myocardium in regenerating newt hearts is preceded by transient deposition of extracellular matrix components. Stem Cells Dev. 2013;22:1921–31. 17. Hu N, Sedmera D, Yost HJ, Clark EB. Structure and function of the developing zebrafish heart. Anat Rec. 2000;260:148–57. 18. Hu N, Yost HJ, Clark EB. Cardiac morphology and blood pressure in the adult zebrafish. Anat Rec. 2001;264:1–12. 19. Santer RM. Morphology and innervation of the fish heart. Adv Anat Embryol Cell Biol. 1985;89:1–102. 20. Agnisola C, Tota B. Structure and function of the fish cardiac ventricle: flexibility and limitations. Cardioscience. 1994;5:145–53. 21. De Andres AV, Munoz-Chapuli R, Sans-Coma V. Development of the coronary arteries and cardiac veins in the dogfish (Scyliorhinus canicula). Anat Rec. 1993;235:436–42. 22. Stainier DY, Fishman MC. The zebrafish as a model system to study cardiovascular development. Trends Cardiovasc Med. 1994;4:207–12. 23. Stainier DY, Lee RK, Fishman MC. Cardiovascular development in the zebrafish. I. Myocardial fate map and heart tube formation. Development. 1993;119:31–40. 24. Poss KD, Wilson LG, Keating MT. Heart regeneration in zebrafish. Science. 2002;298:2188–90. 25. Kikuchi K, Poss KD. Cardiac regenerative capacity and mechanisms. Annu Rev Cell Dev Biol. 2012;28:719–41.

Manalo et al.  26. Lafontant PJ, Burns AR, Grivas JA, Lesch MA, Lala TD, Reuter SP, Field LJ, Frounfelter TD. The giant danio (D. aequipinnatus) as a model of cardiac remodeling and regeneration. Anat Rec (Hoboken). 2012;295: 234–48. 27. Ito K, Morioka M, Kimura S, Tasaki M, Inohaya K, Kudo A. Differential reparative phenotypes between zebrafish and medaka after cardiac injury. Dev Dyn. 2014;243:1106-15. 28. Grivas J, Haag M, Johnson A, Manalo T, Roell J, Das TL, Brown E, Burns AR, Lafontant PJ. Cardiac repair and regenerative potential in the goldfish (Carassius auratus) heart. Comp Biochem Physiol C Toxicol Pharmacol. 2014;163:14–23. 29. Charuk JH, Howlett S, Michalak M. Subfractionation of cardiac sarcolemma with wheat-germ agglutinin. Biochem J. 1989;264:885–92. 30. Emde B, Heinen A, Godecke A, Bottermann K. Wheat germ agglutinin staining as a suitable method for detection and quantification of fibrosis in cardiac tissue after myocardial infarction. Eur J Histochem. 2014;58:2448. 31. Van Asselt E, De Graaf F, Smit-Onel MJ, Van Raamsdonk W. Spinal neurons in the zebrafish labeled with fluoro-gold and wheat-germ agglutinin. Neuroscience. 1991;43:611–22. 32. Matsui H, Namikawa K, Koster RW. Identification of the zebrafish red nucleus using Wheat Germ Agglutinin transneuronal tracing. Commun Integr Biol. 2014;7:e994383. 33. Lin LY, Horng JL, Kunkel JG, Hwang PP. Proton pumprich cell secretes acid in skin of zebrafish larvae. Am J Physiol Cell Physiol. 2006;290:C371–8. 34. Bonsignorio D, Perego L, Del Giacco L, Cotelli F. Structure and macromolecular composition of the zebrafish egg chorion. Zygote. 1996;4:101–8. 35. Biehlmaier O, Neuhauss SC, Kohler K. Double cone dystrophy and RPE degeneration in the retina of the zebrafish gnn mutant. Invest Ophthalmol Vis Sci. 2003;44:1287–98. 36. Jung KS, Ahn MJ, Lee YD, Go GM, Shin TK. Histochemistry of six lectins in the tissues of the flat fish Paralichthys olivaceus. J Vet Sci. 2002;3:293–301. 37. Lafontant PJ, Behzad AR, Brown E, Landry P, Hu N, Burns AR. Cardiac myocyte diversity and a fibroblast network in the junctional region of the zebrafish heart revealed by transmission and serial block-face scanning electron microscopy. PLoS ONE. 2013;8:e72388. 38. Mably JD, Mohideen MA, Burns CG, Chen JN, Fishman MC. Heart of glass regulates the concentric growth of the heart in zebrafish. Curr Biol. 2003;13:2138–47. 39. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. Fiji: an open-source platform for biological-image analysis. Nat Methods. 2012;9:676–82. 40. Chajek T, Stein O, Stein Y. Interaction of con canavalin A with membrane-bound and solubilized

Lectin Binding Giant Danio and Zebrafish Hearts lipoprotein lipase of rat heart. Biochim Biophys Acta. 1976;431:507–18. 41. Stein O, Chajek T, Stein Y. Ultrastructural localization of concanavalin A in the perfused rat heart. Lab Invest. 1976;35:103–10. 42. Del Giacco L, Vanoni C, Bonsignorio D, Duga S, Mosconi G, Santucci A, Cotelli F. Identification and spatial distribution of the mRNA encoding the gp49 component of the gilthead sea bream, Sparus aurata, egg envelope. Mol Reprod Dev. 1998;49:58–69. 43. Myrdal SE, Dehaan RL. Concanavalin A increases spontaneous beat rate of embryonic chick heart cell aggregates. J Cell Physiol. 1983;117:319–25. 44. Cummings RD. Use of lectins in analysis of glycoconjugates. Methods Enzymol. 1994;230:66–86. 45. Lis H, Sharon N. Lectins: carbohydrate-specific pro teins that mediate cellular recognition. Chem Rev. 1998;98:637–74. 46. Alton G, Hasilik M, Niehues R, Panneerselvam K, Etchison JR, Fana F, Freeze HH. Direct utilization of mannose for mammalian glycoprotein biosynthesis. Glycobiology. 1998;8:285–95. 47. Gupta V, Poss KD. Clonally dominant cardiomyo cytes direct heart morphogenesis. Nature. 2012;484: 479–84. 48. Midttun B. Ultrastructure of the junctional region of the fish heart ventricle. Comp Biochem Physiol Part A Physiol. 1983;76:471–4.

713 49. Ren G, Michael LH, Entman ML, Frangogiannis NG. Morphological characteristics of the microvasculature in healing myocardial infarcts. J Histochem Cytochem. 2002;50:71–79. 50. Mojsilovic-Petrovic J, Nesic M, Pen A, Zhang W, Stanimirovic D. Development of rapid staining protocols for laser-capture microdissection of brain vessels from human and rat coupled to gene expression analyses. J Neurosci Methods. 2004;133:39–48. 51. Hunter F, Xie J, Trimble C, Bur M, Li KC. RhodamineRCA in vivo labeling guided laser capture microdissection of cancer functional angiogenic vessels in a murine squamous cell carcinoma mouse model. Mol Cancer. 2006;5:5. 52. Griffith CM, Sanders EJ. Changes in glycoconjugate expression during early chick embryo development: a lectin-binding study. Anat Rec. 1991;231:238–50. 53. Griffith CM, Wiley MJ. N-CAM, polysialic acid and chick tail bud development. Anat Embryol (Berl). 1991;183:205–12. 54. Gros D, Bruce B, Kieda C, Delmotte F, Monsigny M, Schrevel J. Evolution of the surface of mouse myocardial cells during ontogenesis. II. Changes of fluorescent lectin-binding sites. J Mol Cell Cardiol. 1983;15:93–104. 55. Gros D, Bruce B, Challice CE, Schrevel J. Ultrastructural localization of concanavalin A and wheat germ agglutinin binding sites in adult and embryonic mouse myocardium. J Histochem Cytochem. 1982;30:193–200.