Ulex Europaeus Agglutinin-1 Is a Reliable Taste Bud Marker for In Situ Hybridization Analyses

Abstract

Taste signals are received by taste buds. To better understand the taste reception system, expression patterns of taste-related molecules are determined by in situ hybridization (ISH) analyses at the histological level. Nevertheless, even though ISH is essential for determining mRNA expression, few taste bud markers can be applied together with ISH. Ulex europaeus agglutinin-1 (UEA-1) appears to be a reliable murine taste bud marker based on immunohistochemistry (IHC) analyses. However, there is no evidence as to whether UEA-1 can be used for ISH. Thus, the present study evaluated UEA-1 using various histochemical methods, especially ISH. When lectin staining was performed after ISH procedures, UEA-1 clearly labeled taste cellular membranes and distinctly indicated boundaries between taste buds and the surrounding epithelial cells. Additionally, UEA-1 was determined as a taste bud marker not only when used in single-colored ISH but also when employed with double-labeled ISH or during simultaneous detection using IHC and ISH methods. These results suggest that UEA-1 is a useful marker when conducting analyses based on ISH methods. To clarify UEA-1 staining details, multi-fluorescent IHC (together with UEA-1 staining) was examined, resulting in more than 99% of cells being labeled by UEA-1 and overlapping with KCNQ1-expressing cells.

Introduction

Taste buds are involved in receiving and transmitting taste signals. In mammals, taste buds consist of approximately 50–100 taste cells, mainly located at fungiform, foliate, and circumvallate papillae on the tongue. Additionally, taste buds are also distributed on the soft palate and laryngeal epithelium (Finger 2005; Yarmolinsky et al. 2009; Roper 2013).

Taste cells are classified by their morphological features into four categories. Spindle-shaped cells that are elongated toward the taste pore are called type I, type II, and type III taste cells. Conversely, round-shaped cells existing on the bottom of the taste buds are called type IV taste cells. These type IV cells are thought to be undifferentiated cells. Types I, II, and III cells are thought to express at least one type of taste receptor, and therefore are called the taste receptor cells (TRCs; Lindemann 1996). These TRCs can be classified into at least five functional cell types, and each cell is specialized to receive one of the five basic tastes (i.e., sweet, bitter, sour, salt, and umami; Chandrashekar et al. 2006; Yarmolinsky et al. 2009; Liman et al. 2014).

Some taste-related molecules involved in taste reception and transduction for sweet, umami, and bitter have been recently elucidated, whereas the signaling pathways for sour (Kurokawa et al. 2015) and salt perception have not been fully clarified (Oka et al. 2013). Moreover, the molecular mechanisms involved in the differentiation, maturation, and death of taste cells remain to be determined. Information regarding expression patterns and/or the intracellular localization of certain taste-related molecules would help clarify these mechanisms. In situ hybridization (ISH) and immunohistochemistry (IHC) are generally employed to determine expression patterns at the histological level. To obtain high resolution images in ISH and/or IHC methods, histological sections often require chemical and physical treatments such as aldehyde fixation, enzymatic digestion, acetylation, and heat treatments. Although ISH is an essential method for determining mRNA expression, few taste bud markers can be applied together with this method. Furthermore, it is often difficult to judge whether target molecules are overlapping during ISH, since TRCs are spindle-shaped (Kurokawa et al. 2015). Therefore, taste bud markers would be useful for determining these expression patterns when visualizing TRC membranes during ISH.

Taste buds are derived from the surrounding local epithelium in mammals (Stone et al. 1995; Stone et al. 2002; Okubo et al. 2009; Kapsimali and Barlow 2013). In fact, taste buds are embedded into epithelial tissues; this is one reason why the distinction between taste buds and surrounding epithelium is difficult under optical microscopic observation. Therefore, any reliable taste bud cell marker should be able to distinguish taste bud cells from the surrounding epithelial cells. Finally, it would be preferable if such markers could be applied together with various histological methods utilizing simple procedures. Lectin is a general term for carbohydrate-binding proteins, and UEA-1, a type of lectin that binds to α-fucose, does label the membrane of rodent taste bud cells (Kano et al. 2001; Wakisaka 2005; Taniguchi et al. 2008). However, the taste cell types that are labeled by UEA-1, and whether it could be employed as a taste bud marker in combination with various histochemical methods (including ISH) has not yet been revealed. Thus, the present study, used lectin histochemistry to examine whether UEA-1 is a suitable taste bud cell marker in ISH and in other histological methods (including double-labeled ISH or in simultaneous detection using both ISH and IHC methods). Additionally, the taste cell type labeled using UEA-1 was examined by combining multi-fluorescence IHC and UEA-1 staining. The goal of the present study was to reveal a method for visualizing taste bud cellular membranes and to distinguish taste bud cells from the surrounding epithelial cells during various histological methods (especially ISH).

Materials & Methods

Animals and Section Preparation

Adult (>8 weeks) C57BL/6J male mice were sacrificed by cervical dislocation to allow dissection of circumvallate papillae (CvP) from the tongue. The tissues containing CvP were embedded in O.C.T. compound (Sakura Finetek; Tokyo, Japan) and immediately frozen with liquid nitrogen. The tissues were sectioned at 7 μm with a cryostat (CryoStar NX70; Thermo Scientific, Waltham, MA). The sections were placed on MAS-coated glass slides (Matsunami Glass; Kishiwada, Japan) and were stored at −80°C until use. The Institutional Animal Care and Use Committee at the University of Tokyo approved the experimental protocols.

Lectins

Biotinylated-lectins were commercially purchased from Vector Laboratories, Inc. (Burlingame, CA) or J-Oil Mills Inc. (Tokyo, Japan). The glycan-binding specificities and concentrations used in our histological staining are shown in Table 1 (modified from Taniguchi et al. 2008).

Table 1.

Glycan-Binding Specificities of Lectins Used in the Present Study.

Abbreviation	Lectin	Binding Specificity	Concentration (μg/ml)*
UEA-1	Ulex europaeus agglutinin-1	α-1, 2 fucose	8
DSL	Datura stramonium lectin	N-acetylglucosamine	2
LEL	Lycopersicon esculentum lectin	N-acetylglucosamine	2
STL	Solanum tuberosum lectin	N-acetylglucosamine	3
WGA	Wheat germ agglutinin	N-acetylglucosamine	3
RCA-I	Ricinus communis agglutinin-I	β-galactoside	3
Jacalin	Jacalin	galactocyl(β-1,3)GalNAc	4
PHA-L	Phytohemagglutinin-L	oligosaccharide (leucoagglutinin)	2
LTL	Lotus tetragolonobus lectin	Fucose(α-1, 2/4 GlcNAc)	5
AAL	Aleuria aurantia Lectin	Fucose(α-1, 2/4 GlcNAc)	5

^*Concentration of each lectin used in our histological study.

Lectin Histochemistry

For lectin histochemistry, fresh-frozen sections were fixed with 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS) for 10 min. Sections were washed with PBS and then incubated with 2.5% normal horse serum (NHS; Vector Laboratories, Inc.) in PBS for 1 hr to block non-specific reactions. The sections were then incubated with biotinylated-lectins buffered with PBS for 1 hr. After washing the sections three times with PBS, the sections were incubated for 30 min with Alexa Fluor 488-conjugated streptavidin (Life Technologies) diluted with PBS (1:200). The sections were then rinsed with PBS and stained with 4’,6-diamidino-2-phenylindole, dihydrochloride (DAPI; Molecular Probes) for 5 min to detect the nuclei. Following rinsing with ultra-pure water, the sections were mounted with Fluoromount (Diagnostic BioSystems; Pleasanton, CA). The negative control was performed using 0.1 M α-fucose incubation, which competes with UEA-1 binding. All staining steps were carried out at room temperature.

Synthesis of Labeled cRNA Probes

The synthesis of labeled cRNA probes was performed as described previously (Ishimaru et al. 2005). The cDNA fragments for Plcb2 (NM_177568), Tas1r3 (NM_031872), and Ggust (NM_001081143) were obtained via reverse transcriptase-polymerase chain reaction (RT-PCR) from C57BL/6J CvP total RNA and subcloned into a pBluescript II SK (+) vector (Addgene; Cambridge, MA). Plasmids were then used for synthesis of labeled antisense cRNA probes. For RT-PCR, the primer set 5’-GCGGAATTCCATATCTGAGCCAAGGGG-3’ and 5’-CGCGCGGCCGCCAGGTGTCCCAGAGAAGAGT-3’ were used for Plcb2, 5’-GCGGAATTCGTCTGTCACAGCAATTCAAGG-3’ and 5’-CGCGCGGCCGCGCAGCCAAAGAAGCACATAG-3 for Tas1r3, and 5’-GCGGAATTCCGAGCACCGCTCATATGTC-3’ and 5’- CGCGCGGCCGCTCAGAAGAGCCCACAGTCTTT-3’ for Ggust. The Plcb2 and Ggust probes were labeled with digoxygenin (DIG), and Tas1r3 with fluorescein.

Double Staining of Single-colored ISH and Lectin Histochemistry

The procedures for single-colored ISH were slightly modified from a method described previously (Ishimaru et al. 2012). Fresh-frozen sections were fixed with 4% PFA/PBS for 10 min and treated with 0.1% diethylpyrocarbonate (DEPC) in PBS for 15 min. The sections were treated with 1 μg/ml proteinase K in Tris-buffered saline (TBS; pH 7.4) for 5 min at 30°C. Following an additional treatment with 0.1% DEPC/PBS for 15 min, the sections were acetylated with 0.25% acetic anhydride in 10 mM triethanolamine for 10 min. Pre-hybridization was performed with a mixture of 50% formamide, 5× SSC and 40 μg/ml fish sperm DNA (Roche; Indianapolis, IN) for 2 hr at 58°C. For hybridization, sections were incubated in DIG-labeled antisense Plcb2 cRNA probe prepared in a solution of 50% formamide, 5 × SSC, 5 × Denhardt’s solution, 500 μg/ml fish sperm DNA, 250 μg/ml torula tRNA (Nakalai Tesque; Kyoto, Japan), 1 mM dithiothreitol. Hybridization was performed for 18–42 hr at 58°C. The sections were then washed twice with 0.2× SSC for 30 min at 58°C, followed by incubation in 0.5% blocking reagent (Roche) in PBS for 1 hr to prevent the non-specific reactions. Then, the sections were incubated with Anti-DIG-POD Fab fragments (1:100; Roche) for 1 hr. After washing with PBS, the sections were incubated with Alexa Fluor 555 Tyramide (1:100; Life Technologies). After the hybridization, the sections were incubated in biotinylated-lectins, washed and then incubated in Alexa Fluor 488 conjugated to streptavidin (1:200). Nuclei staining and mounting was performed as described above.

Simultaneous Detection of Double-labeled ISH and UEA-1 Lectin Histochemistry

The procedures for double-labeled ISH were performed similar to that described above until the end of the pre-hybridization step. For hybridization, the sections were incubated in a mixture of DIG-labeled antisense Ggust cRNA probe and fluorescein-labeled antisense Tas1r3 cRNA probe. The sections were then washed and blocked, and then incubated with anti-DIG-POD fab fragments (1:100) and Alexa Fluor 555 Tyramide (1:100). Next, the sections were treated with 3% H₂O₂ in PBS for 30 min for the inactivation of the peroxidase conjugated to the Fab fragments. The sections were then washed with PBS, and incubated in polyclonal rabbit anti-FITC-HRP (1:100; DAKO, Carpinteria, CA) and Alexa Fluor 488 Tyramide (1:100; Life Technologies) for secondary detection. Taste buds were stained with biotinylated-UEA-1 and Alexa Fluor 647-conjugated streptavidin (Life Technologies). Nuclei staining and mounting was performed as described above.

Simultaneous Detection of IHC, ISH, and UEA-1 Lectin Histochemistry

Fresh-frozen sections were fixed with 4% PFA/PBS for 10 min. The sections were transferred to 0.1% DEPC/PBS and treated for 15 min. After rinsing with PBS, the sections were blocked with 0.2% bovine serum albumin (Sigma-Aldrich; St. Louis, MO) containing 100 U of recombinant RNase Inhibitor (Takara Bio Inc.; Shiga, Japan) for 1 hr at 4°C. The sections were then treated with a mixture of rabbit polyclonal anti-KCNQ1 antibody (1:5000; Millipore, Billerica, MA) and 100 U of recombinant RNase Inhibitor and incubated overnight at 4°C. The sections were then washed with PBS at 4°C, and incubated with 0.5 μg/ml HRP-conjugated anti-rabbit antibody (Life Technologies) for 2 hr at 4°C. After washing, the sections were stained with Alexa Fluor 488 Tyramide (1:200). Then, the ISH steps were continued beginning from the proteinase K treatment step. After ISH, UEA-1 and nuclei staining were performed as described above.

Simultaneous Detection of Multi-fluorescence IHC and UEA-1 Lectin Histochemistry

Multi-fluorescence IHC was performed together with UEA-1 staining. Fresh-frozen sections were fixed with 4% PFA/PBS for 10 min. After washing with PBS, the sections were incubated with 2.5% NHS for 1 hr to block non-specific reactions. A mixture of primary antibodies and biotinylated-UEA-1 was applied and incubated for 1 hr. Next, the sections were washed three times with PBS, and fluorescent dye-conjugated secondary antibodies and another fluorescent dye-conjugated streptavidin were applied together and incubated for 1 hr. Following washing with PBS, nuclei staining and mounting were performed as described above.

For triple fluorescence staining of KCNQ1, ZO-1, and UEA-1, both rabbit polyclonal anti-KCNQ1 antibody (1:5000; Millipore) and mouse monoclonal anti-ZO1 antibody (1:200; Life Technologies) were used as primary antibodies. The antibodies were applied together with biotinylated-UEA-1. For secondary antibodies, Alexa Fluor 488 donkey anti-rabbit antibody (1:200; Life Technologies) and Alexa Fluor 555 donkey anti-mouse antibody (1:200; Life Technologies) were used together with Alexa Fluor 647 conjugate streptavidin (1:200) for signal detection.

Microscopy and Image Processing

The sections were examined using a confocal laser-scanning microscope (FV10i; Olympus, Tokyo, Japan), and pseudo-colored images were produced by a FV10-ASW viewer (Olympus). Brightness and contrast levels were optimized in Photoshop (Adobe Systems; San Jose, CA).

Calculation of Overlapping Cells Stained with UEA-1 and Immunoreactive for KCNQ1

Using triple staining, merged images of UEA-1, KCNQ1-immunoreactive (IR), and nuclei (the signal of DAPI) were obtained by the confocal laser-scanning microscope and processed with Photoshop. Nuclei, including UEA-1-positive cells, KCNQ1-IR, and double-positive cells of the merged images, were counted. Nuclei numbers were used to determine the percentages of overlapped cells, calculated as the “overlapping ratio”.

Results

Characterization of Lectins Used as a Taste Bud Cell Marker for Single-colored ISH

To evaluate whether UEA-1 could be used as a taste bud marker during ISH, lectin histochemistry was performed separately and together with single-colored ISH. In addition to UEA-1, we selected Datura stramonium lectin (DSL), Lycopersicon esculentum lectin (LEL), Solanum tuberosum lectin (STL), wheat germ agglutinin (WGA), Ricinus communis agglutinin-1 (RCA-1), jacalin, and phytohemagglutinin-L (PHA-L)—all of which have already been reported to bind to the cell membranes of mice taste bud cells (Taniguchi et al. 2008)—for histological staining of murine CvP tissues to compare with UEA-1. The glycan-binding specificities of these lectins are shown in Table 1. When each biotinylated lectin was labeled, and visualized by Alexa-conjugated streptavidin, all lectins stained for the taste bud cell membranes, as reported previously (Fig. 1A–1H). However, the images differed in terms of cell staining specificity. Although UEA-1 clearly showed boundaries between the onion-shaped taste buds and the surrounding epithelial cells (Fig. 1A), the other seven lectins gave indistinct images, as they also labeled part of the surrounding epithelial cells (Fig. 1B–1H).

Figure 1.

Lectin histochemistry in murine CvP. (A–J) Transverse sections of murine CvP were stained for biotinylated lectins and visualized using Alexa Fluor 488-conjugated streptavidin. (K-T) Lectin staining was performed after single-colored ISH procedures using a Plcb2 cRNA probe. Biotinylated lectins were also used. (A and K) UEA-1, (B and L) DSL, (C and M) LEL, (D and N) STL, (E and O) WGA, (F and P) RCA-1, (G and Q) Jacalin, (H and R) PHA-L, (I and S) LTL, and (J and T) AAL. Staining signals for lectins are shown in green and those for Plcb2 mRNA in magenta. Scale, 20 μm.

Lectin histochemistry was applied together with single-colored ISH analysis to confirm whether chemical and physical treatments during ISH procedures affected lectin staining specificities. When biotinylated lectins were applied after ISH for the detection of Plcb2 (the taste-related molecule mRNA), these lectins also detected part of the cell membrane in this region (Fig. 1K–1R). It should be noted that UEA-1 retained the staining properties and specificities, which distinctly indicated taste bud cell positioning; the signal intensity of UEA-1 staining did not remarkably change (Fig. 1K). In contrast, the other seven lectins also detected the surrounding epithelial cells as well as taste bud cells (Fig. 1L–1R), indicating that they were not reliable taste bud markers during the ISH analysis.

Additionally, two lectins—Lotus tetragolonobus lectin (LTL) and Aleuria aurantia lectin (AAL)—that show α-fucose-binding specificities similar to UEA-1 (Yariv et al. 1967; Kochibe and Furukawa 1980; Clark and Mao 2012), were also examined in our experimental conditions. Although they could indicate taste bud cell membranes when applied solely (Fig. 1I, 1J), the staining specificity dramatically changed when used after ISH. LTL staining appeared nonspecifically in the nuclei of several cells (Fig. 1S), and although the staining specificity of AAL to taste bud cells remained almost the same, the signal appeared to relate to the surrounding epithelial cells (Fig. 1T). The negative control staining for UEA-1 using adsorbed biotinylated-UEA-1 (0.1 M α-fucose) resulted in the disappearance of UEA-1 signals (Fig. 2).

Figure 2.

Incubations using UEA-1 adsorbed by 0.1 M α-fucose. (A) Transverse sections of murine CvP were stained using UEA-1 histochemistry or (B) UEA-1 adsorbed by 0.1 M α-fucose as a control. (C) UEA-1 staining was performed after single-colored ISH using a Plcb2 cRNA probe. (D) Staining using UEA-1 adsorbed by 0.1 M α-fucose was performed after single-colored ISH as in (C). Staining signals for UEA-1 are shown in green and those for Plcb2 mRNA in magenta. All panels show merged images of fluorescent and phase-contrast views. Scale, 20 μm.

Application of UEA-1 Staining for Various Methods Based on ISH

Next, UEA-1 staining was examined together with various histochemical methods based on ISH analyses. We performed UEA-1 staining in combination with double-fluorescence labeled ISH (Fig. 3A–3C) or with both ISH and IHC methods (Fig. 3D–3F). Tas1r3 (Zhao et al. 2003; Baldwin et al. 2014) and Ggust (Kusakabe et al. 2000; Ishimaru et al. 2012) are the taste-related molecules participating in G protein-coupled receptor (GPCR) signaling pathways. We show that these were expressed in a subset of taste bud cells in mice CvP, with signals detected in the cytosol of a portion of the taste bud cells via double-labeled ISH (Fig. 3A). In this case, UEA-1 staining after double-labeled ISH procedures clearly detected the cellular membranes (Fig. 3B, 3C), and the signals distinctly indicated boundaries between taste bud cells and the surrounding epithelial cells.

Figure 3.

Effects of combined histochemical procedures on UEA-1 staining. (A–C) Transverse sections of murine CvP were histochemically stained for UEA-1 and subjected to double-labeled ISH using probes for Ggust and Tas1r3; or (D–F) with combined IHC staining, using anti-KCNQ1, and ISH, using a probe for Plcb2. All panels show merged images of fluorescent and phase-contrast views. Left and right panels show merged images of multi-colored fluorescent views of the combination indicated in the panels. Scale, 20 μm.

KCNQ1 is a member of the voltage-gated potassium channel, which is expressed on the cellular membrane of nearly all taste bud cells with the exception of a few Type IV basal cells in mice CvP tissues (Ohmoto et al. 2006; Wang et al. 2009). In the simultaneous detection of IHC (anti-KCNQ1) and ISH (Plcb2), together with UEA-1 staining, UEA-1 signals almost completely overlapped with KCNQ1-IR cells, and the expression of Plcb2 mRNA was also clearly detected in a subset of taste bud cells (Fig. 3D–3F). Thus, UEA-1 staining could be applicable in these complex histological studies based on ISH analyses.

Characterizing UEA-1-positive Taste Cell Types in Mice CvP and Subcellular Localization of UEA-1 Signals

To identify the cell types detected by UEA-1 staining in murine CvP, multi-fluorescence IHC using an anti-KCNQ1 antibody was performed together with UEA-1 staining (Fig. 4). When transverse and sagittal sections were observed at low magnification, UEA-1 positively stained taste bud cells, cornified epithelium, and Ebner’s glands (Fig. 4A, 4D). In the horizontal sections, UEA-1 specifically stained taste bud cells; cornified epithelium and Ebner’s glands were not included in these sections (Fig. 4G). As for the region around the taste bud cells, UEA-1-positive cells almost completely overlapped with those expressing KCNQ1 (Fig. 4C, 4F, 4I). From the high-magnification images of transverse sections (Fig. 4J–4L), we calculated the overlapping ratios of UEA-1-positive cells with KCNQ1 immunoreactivity, and found that more than 99% of cells overlapped (Table 2).

Figure 4.

Taste cell type specificity for UEA-1 histochemistry. (A–C) Transverse, (D–F) sagittal, and (G–I) horizontal sections of murine CvP stained using UEA-1 histochemistry and anti-KCNQ1 IHC. (J–L) High-magnification images of the squared areas in (A–C). Nuclei were counterstained with DAPI. Staining signals for UEA-1 are shown in green, those for KCNQ1-IR in red, and those for DAPI in blue. Left and center panels show merged images of fluorescent and phase-contrast views. Right panels show merged images of multi-colored fluorescent views. Abbreviations: TBs, taste buds; CE, cornified epithelium; EGs, Ebner’s glands. Scale (A–I) 100 μm; (J–L) 20 μm.

Table 2.

Overlapping Ratio of the Cells Stained by UEA-1 and Cells with KCNQ1 Immunoreactivity.

	UEA1+ and KCNQ1+/KCNQ1+	UEA1+ and KCNQ1+/UEA1+
Mouse 1	652 / 655	652 / 654
Mouse 2	940 / 949	940 / 947
Mouse 3	732 / 744	732 / 734
Mouse 4	625 / 630	625 / 628
Total	2949 / 2978	2949 / 2967
Overlapping ratio ± SE	99.0 ± 0.2%	99.4 ± 0.2%

Conversely, to verify the subcellular localization of UEA-1 signals, we performed multiple-fluorescence staining using UEA-1, anti-KCNQ1, anti-ZO-1, and DAPI (Fig. 5). ZO-1 is a tight-junction protein that separates the apical and basolateral plasma membranes in taste bud cells (Schneeberger and Lynch 2004). Judging from high-magnified images, we noted that the subcellular localization of UEA-1 signals was only slightly different from that of KCNQ1-IR. UEA-1 signals could be observed in both the basolateral and apical membranes of taste bud cells (Fig. 5A, 5D), whereas KCNQ1-IR seemed to be strictly localized in the basolateral membrane (Fig. 5B, 5E).

Figure 5.

Subcellular localization of UEA-1 staining signals. A sagittal section of murine CvP was stained via UEA-1 histochemistry along with double-labeled IHC for anti-KCNQ1 and anti-ZO-1 antibodies. Nuclei were counterstained by DAPI. Staining signals for UEA-1 are visualized as green, those for KCNQ1-IR as red, those for ZO-1-IR as cyan, and those for DAPI as blue. (A–C) Single color images of three-colored fluorescent views. (D) Merged image of (A) and (C). (E) Merged image of (B) and (C). (F) Merged image of (A), (B), and (C) with DAPI staining. Abbreviations: TP, taste pore; TJ, tight junction. Scale, 20 μm.

Additionally, fluorescent dye-conjugated UEA-1 (Alexa 488-UEA-1) was produced from unconjugated-UEA-1 by a commercially available Alexa-conjugation kit, and it was used for staining murine CvP histological sections (Supplemental Fig. S1). The staining properties of biotinylated-UEA-1 and Alexa 488-UEA-1 were the same.

Discussion

The main purpose of the present study was to identify a reliable marker that distinctly defines taste bud cells via histological analysis (based on ISH) utilizing simple procedures. A reliable taste bud marker would be advantageous for molecular analyses using various complex histological methods, such as ISH and double-labeled ISH, as well as for the simultaneous detection of markers using both ISH and IHC methods. The current study focused on UEA-1 lectin histochemistry, and we show that this lectin could be used as a reliable mouse taste bud cell marker in ISH. It has been reported that UEA-1 is one of the α-fucose-binding lectins (Pereira et al. 1978; Loris et al. 1998; Clark and Mao 2012), and is used as a marker for other cell types such as M cells (Jang et al. 2004; Nakato et al. 2009), vascular tumors (Yonezawa et al. 1987), and collecting duct carcinomas in the kidneys (Amin et al. 1997). Although some reports indicate that murine taste cells can also be detected by UEA-1 (Taniguchi et al. 2008; Wakisaka 2005), detailed information regarding whether UEA-1 can be used with ISH (and the taste cell type stained by UEA-1) was not provided. Here, the present study confirms that UEA-1 staining can be applied as a taste bud marker in conjunction with the use of complex histological components based on ISH of murine CvP sections. As described in the Materials & Methods section, more complex procedures were used for double-labeled ISH and the simultaneous detection of markers using ISH and IHC as compared with single-colored ISH or IHC analyses. Furthermore, additional chemical and physical treatments are often required for these sections, which could lead to denaturation of several cellular components, including proteins and sugars. However, UEA-1 clearly stained taste bud cells using these methods, suggesting that the UEA-1 binding target seems not to have changed during our experimental conditions. Moreover, UEA-1 staining could advantageously determine the target molecule expression pattern during ISH, as UEA-1 stained the taste bud cell membranes along the boundary line. Additionally, UEA-1 staining was also revealed at the taste bud cells’ apical membrane region. It has been reported that KCNQ1 (Ohmoto et al. 2006; Wang et al. 2009) and CK19 (Wong et al. 1994; Knapp et al. 1995) are taste bud cell markers in mice CvP sections. However, these proteins are not localized to the taste pore region, indicating that UEA-1 staining is useful for examining the apical localization of taste-related molecules expressed in taste bud cells. KCNQ1 is expressed at the cellular membrane of nearly all taste bud cells, with the exception of a few type IV basal cells in mice CvP tissues (Ohmoto et al. 2006; Wang et al. 2009). Thus, our finding that more than 99% of cells labeled by UEA-1 overlapped with KCNQ1-expressing taste bud cells indicates that UEA-1 is a taste bud marker that labels nearly all taste bud cells. This property can be advantageous for counting the number of taste cells and/or judging the borders between taste buds and the surrounding epithelial cells.

Within histological experiments, certain chemical and physical treatments should be conducted to enhance the signal intensities of the probes. For example, proteinase K treatment can be used to improve the permeability of cRNA probes during an ISH procedure. Since various proteins in these sections could be digested by proteinase K, it is difficult to detect any proteins after treatment. Therefore, we stained KCNQ1 protein before the detection of Plcb2 mRNA during the simultaneous IHC, ISH, and UEA-1 lectin histochemistry experiment (see the Materials & Methods section). However, UEA-1 clearly labeled taste bud cells using simpler procedures, especially in the cellular membranes of taste bud cells, without a significant loss in specificity or signal intensity after proteinase K treatment. Additionally, the UEA-1 signals disappeared when UEA-1 was adsorbed by 0.1 M α-fucose. Therefore, strong evidence suggests that UEA-1 bound to α-fucose residues conjugated with molecules other than proteins on taste bud cells. Since the UEA-1 signal was mainly observed on the cellular membrane, we hypothesized that glycolipids are candidates for UEA-1 detection. It is well known that blood-group-ABH antigens localize to glycoproteins and glycolipids in erythrocytes (Mehta 1980; Schenkel-Brunner 1980), and H antigens are stained by UEA-1 due to the presence of α-fucose residues (Audette et al. 2000). H antigens are also observed in rat taste buds (Smith et al. 1994; Pumplin et al. 1999; Wakisaka 2005; Feng et al. 2014). The present results suggest that one UEA-1 candidate binding target is an H antigen localized to the glycolipids of taste bud cells. In fact, UEA-1 signals dramatically decreased when detergent treatment, such as Triton X-100, were carried out for a prolonged period (such as overnight) during our histological procedures (data not shown). This is probably because the detergent treatment damaged membrane lipid structures. Although glycolipids are candidates for UEA-1 detection in mouse CvP taste buds, the mucins also could be the binding targets, as these proteins have several sugar chains (including fucose) and are strongly resistant to proteinase K treatment (Toribara et al. 1997). Mucins are the viscous components of saliva, and it is possible that some kinds of mucins are expressed in taste buds. CAR4 is a type of GPI-anchored glycoprotein and is expressed in type III taste cells as a carbonic acid receptor in mice (Chandrashekar et al. 2009). Thus, some GPI-anchored glycoproteins that have fucose residues are possibly expressed in taste buds. Therefore, in addition to glycolipids, mucins and GPI-anchored glycoproteins are also candidate binding targets.

Two detection systems were employed to visualize UEA-1 signals. One was biotinylated-UEA-1 labeling together with visualization using Alexa-conjugated streptavidin. The other was direct labeling with a fluorescent-dye conjugated UEA-1. These two systems provided nearly identical images in terms of label specificity and signal intensity; however, the staining procedures using a fluorescent-dye conjugated UEA-1 were very simple. Generally, biotin-streptavidin bindings or biotin-avidin complex systems are used for conventional histological methods to ensure binding specificities and/or to enhance staining intensities. Therefore, a fluorescent-dye conjugated UEA-1 would be very advantageous for histological analyses since this marker system could be conducted without biotin-streptavidin binding.

The results from the present study strongly indicate that UEA-1 is a reliable taste bud cell marker in various histological methods (especially ISH). Using UEA-1 staining would be advantageous for analyzing taste-related molecule expression during ISH, IHC, and/or more complex histological methods (e.g., double-labeled ISH and the simultaneous detection using ISH and IHC). Because UEA-1 is a marker for nearly all taste bud cells, exploring the markers of all five functional taste cells is desirable for further research. If such markers are found, and can be used during ISH analyses, the expression of taste-related molecules could be more easily analyzed.