Characteristic Localization of Neuronatin in Rat Testis,Hair Follicle,Tongue,and Pancreas

Abstract

Neuronatin (Nnat) is expressed in the pituitary, pancreas, and other tissues; however, the function of NNAT is still unclear. Recent studies have demonstrated that NNAT is localized in the sex-determining region Y-box 2-positive stem/progenitor cells in the developing rat pituitary primordium and is downregulated during differentiation into mature hormone-producing cells. Moreover, NNAT is widely localized in subcellular organelles, excluding the Golgi. Here, we further evaluated NNAT-positive cells and intracellular localization in embryonic and postnatal rat tissues such as the pancreas, tongue, whisker hair follicle, and testis. Immunohistochemistry revealed that NNAT was localized in undifferentiated cells (i.e., epithelial basal cells and basement cells in the papillae of the tongue and round and elongated spermatids of the testis) as well as in differentiated cells (insulin-positive cells and exocrine cells of the pancreas, taste receptor cells of the fungiform papilla, the inner root sheath of whisker hair follicles, and spermatozoa). In addition, NNAT exhibited novel intracellular localization in acrosomes in the spermatozoa. Because the endoplasmic reticulum (ER) is excluded from spermatozoa and sarco/ER Ca²⁺-ATPase isoform 2 (SERCA2) is absent from the inner root sheath, these findings suggested that NNAT localization in the ER and its interaction with SERCA2 are cell- or tissue-specific properties.

Keywords

acrosome differentiation endoplasmic reticulum hair follicle Neuronatin pancreas taste receptor

Introduction

Neuronatin (NNAT) is a small protein consisting of two splice variants encoding 81 (NNATα) and 58 amino acids (NNATβ) and is well conserved evolutionally.^1,2 NNAT has attracted attention from many researchers because of its high expression in the neonatal brain^3–5 and pituitary gland.⁶ Long before, Wijnholds et al. showed a transient expression in the hindbrain development⁵ but the developmental pattern in the embryonic brain is not yet well characterized. The role of NNAT in vivo is also poorly understood. Recently, we found that NNAT first appears in and occupies stem/progenitor cells in the rat embryonic pituitary primordium and is then downregulated during differentiation into mature hormone-producing cells, suggesting that NNAT may be involved in pituitary development.⁷ Moreover, we found that NNAT localizes in various subcellular organelles in addition to the endoplasmic reticulum (ER), as reported previously.⁸ Furthermore, the expression pattern of Nnat during mouse eye development at embryonic and postnatal periods had been observed.⁹

Several important roles of NNAT have been reported in the past decade. For example, Lin et al.¹⁰ demonstrated that NNAT promotes neural differentiation of embryonic stem cell and that NNAT interacts with sarco/ER Ca^2⁺-ATPase isoform 2 (SERCA2). Involvement of NNAT in the differentiation of keratinocytes¹¹ and adipose tissue¹² has also been suggested. Moreover, several studies have demonstrated diverse roles of NNAT, such as insulin secretion,⁸ synaptic plasticity,¹³ calcium-induced cell migration,¹⁴ stress responses,^15,16 and others.^17–19 Thus, NNAT performs diverse functions according to its expression in particular tissues, although the details have not yet been reported. Recently, the involvement of NNAT in disease development and regression, including tissue degeneration and cancer, has been reported,²⁰ particularly in relation to downstream signaling of NNAT, such as via increase of intracellular Ca^2⁺ level through the regulation of Ca^2⁺ATPase in the ER,¹² mediation of the high glucose-induced apoptosis of pancreatic β-cells,⁸ reduction in glucose-mediated insulin secretion,²¹ increase of the inflammatory response in the endothelial cells,²² and decrease of the inflammatory signaling pathway.²³

Accordingly, in this study, we analyzed the localization of NNAT-positive cells using immunohistochemistry in embryonic and postnatal tissues, with a particular focus on the testes, whisker hair follicles, and tongue papillae, which are regulated by Ca^2⁺ signaling, and the pancreases, regulated by glucose signaling. We re-evaluated the intracellular localization of NNAT in specific subcellular organelles other than the ER. Our observations provide valuable insights into the function of NNAT in tissue development.

Materials and Methods

Animals

Wistar rats were housed individually in a temperature-controlled room under a 12-hr light/12-hr dark cycle. Adult rats were matched, and the day on which a vaginal plug was observed was designated as embryonic day 0.5 (E0.5). The present study was approved by the Institutional Animal Care and Use Committee of Meiji University and was performed based on National Institutes of Health (NIH) Guidelines for the Care and Use of Laboratory Animals.

Immunohistochemical Analyses

Pregnant rats were euthanized at the appropriate time point by cervical dislocation under anesthesia. Whole embryos at E16.5, E20.5, and E21.5 (n=3 each) were surgically removed. Male rats at postnatal day 30 (P30) and P60 (n=3 each) were administered anesthetic and then subjected to perfusion fixation using 4% paraformaldehyde in 20 mM HEPES (pH 7.5).Then appropriate tissues were dissected. Whole embryos and tissues were fixed in 4% paraformaldehyde in 20 mM HEPES (pH 7.5) overnight at 4C, followed by immersion in 30% trehalose in 20 mM HEPES (pH 7.5) for cryoprotection. Tissues were then embedded in Tissue-Tek OCT compound (Sakura Finetek Japan, Tokyo, Japan) and frozen immediately. Frozen sections (4 or 10 µm thick) on glass slides (Matsunami, Osaka, Japan) were activated as required using Immunosaver (Nisshin EM, Tokyo, Japan) for 60 min at 80C and washed three times in 20 mM HEPES (pH 7.5) containing 100 mM NaCl (HEPES buffer) for 15 min. This was followed by blocking with 0.4% Triton X-100 (Sigma-Aldrich Corp., Tokyo, Japan) and 10% fetal bovine serum in HEPES buffer (pH 7.5) for 60 min at room temperature. Sections were then incubated for 16 hr at 4C with the following primary antibodies: rabbit IgG against mouse NNAT (1:250 dilution; ab27266; Abcam, Cambridge, UK), guinea pig IgG against human insulin (1:100 dilution; ab7842; Abcam), goat IgG against human sex-determining region Y-box 2 (SOX2; 1:500 dilution; GT15098; Neuromics, Edina, MN), mouse IgG against human E-cadherin (1:200 dilution; BD Bioscience, San Jose, CA), and mouse monoclonal antibody against rat protein disulfide isomerase (PDI; 1:1000 dilution; ab2792; Abcam). After being washed with HEPES buffer three times, sections were incubated for 2 hr at room temperature with Cy3, Cy5, or fluorescein isothiocyanate (FITC)–conjugated donkey anti-mouse, anti-rabbit, anti-guinea pig, or anti-goat IgG as secondary antibodies (Jackson ImmunoResearch, West Grove, PA), followed by washing with HEPES buffer three times. This was followed by enclosing in Vectashield mounti-ng medium containing 4′,6-diamidino-2-phenylindole (DAPI; Vector, Burlingame, CA) along with observation using fluorescence microscopy BZ-9000 (Keyence, Osaka, Japan) and fluorescence confocal microscopy FV1000 (Olympus, Tokyo, Japan).

The specificity of anti-mouse NNAT polyclonal antibody used in this study had been confirmed by pre-incubation with NNAT peptide (ab30332, Abcam) using the cephalic part on E16.5 as a positive control, as described previously.⁷

Hematoxylin and Eosin (HE) Staining

Before staining, frozen sections (10 µm thick) on glass slides were washed with water for 15 min, stained for 10 min with hematoxylin (Wako, Osaka, Japan), washed with water for 1 hr, and stained with eosin (diluted 5-fold with 95% ethanol; Muto Pure Chemicals Co., Ltd., Tokyo, Japan). Finally, after dehydration, the sections were sealed for observation.

Cell Count

The proportions of immune-positive cells were recorded for eight areas (each 19–36 cells/area) in the testis and six areas (each 129–495 cells/area) in the pancreas. The data are presented as mean ± SE.

Results

Immunohistochemistry for NNAT in the Testes

The testes are reproductive tissues derived from the mesoderm and are composed of the seminiferous tubules and interstitium. The former is involved in spermatogenesis, in which spermatozoa are produced from spermatogonia, and the latter is involved in androgen production by Leydig cells.

In immature rat testes on P30, no NNAT-positive cells were observed (Fig. 1A1 and A2). On P60, NNAT-positive cells appeared in the interstitium and seminiferous tubules of the testis (Fig. 1C). In the interstitium, granular NNAT signals were observed in Leydig cells, and the proportion of NNAT-positive cells was about 60% (Fig. 1D and E). In the seminiferous tubules, NNAT signals were absent from the spermatogonia, spermatocytes, and Sertoli cells (Fig. 1F1–M, SG, SP, and SC, respectively) during spermatogenesis. NNAT signals were first observed at acrosome vesicles in round spermatids after meiosis (Fig. 1L, RS) and exhibited aggregated localization across the crescent acrosome in elongated spermatids (Fig. 1N, ES). Even after spermatozoa moved to the epididymis, NNAT exhibited similar localization with slight attenuation of signal intensity between the caput epididymis and cauda epididymis (Fig. 1P and Q, respectively).

Figure 1.

Immunohistochemistry for NNAT in immature and mature testes. Immunostaining for NNAT (red, Cy3) was performed for frozen sections prepared from immature (A1 and A2) and mature testes (C and D) on postnatal day 60. Hematoxylin and eosin (HE) staining on P60 is shown in C. Numbers and proportions of NNAT-positive cells in the testicular interstitium were recorded for eight areas (19–36 cells each) and are shown in E. Data are presented as mean ± SE. Developmental stages I to XIV in spermatogenesis (F1–J1) and merged images with nuclear staining with 4′,6-diamidino-2-phenylindole (DAPI, blue, F2–J2) are shown. Enlarged images of NNAT-positive cells during acrosome formation (K–N) and spermatozoa (O, merged image with bright field, NNAT and DAPI) in the epididymal head (P) and cauda epididymis (Q) are shown. Scale bars = 50 µm (B, C), 5 µm (A1, A2, D, E, F1, F2, G1, G2, H1, H2, I1, I2, J1, J2, K–Q). Abbreviations: NNAT, Neuronatin; ES, elongated spermatid; SG, spermatogonia; RS, round spermatid; SP, spermatocyte; SC, Sertoli cell.

Immunohistochemistry for NNAT in Whisker Hair Follicles

The hair follicles and tongues are ectodermal appendages formed by interactions between epithelial cells derived from the surface ectoderm and the underlying mesenchyme. We next analyzed the whisker hair follicles. Hair follicle morphogenesis begins from the formation of the hair placode at the middle embryonic stage and continues until birth. Adult hair follicles are maintained by cycling through three phases of active hair follicle regeneration, that is, the hair growth (anagen), destruction (catagen), and rest (telogen). During full anagen phase, hair growth is active by differentiation of matrix progenitor cells of the hair bulb into the hair shaft, which is composed of three layers (i.e., the medulla, hair cortex, and hair cuticle), and the inner root sheath (IRS), which is composed of three layers (i.e., the IRS cuticle, Huxley’s layer, and Henle’s layer).

Immunostaining for NNAT was performed on E16.5 in rats, when the whisker hair follicles are still underdeveloped. All cells in the primordium were negative for NNAT (Fig. 2A, arrow). Longitudinal sections on E20.5 revealed that positive signals for NNAT were present in the domain from the region close to the dermal papilla to the middle portion of the hair follicle (Fig. 2B). Cross-sections revealed NNAT-positive concentric rings of IRS, composed of three layers (the IRS cuticle close to the hair shaft, the Huxley’s layer as middle layer, and the outermost Henle’s layer) essential for the shaping and guidance of the hair to the skin surface by keratinization (Fig. 2C2, Ci, Hu, and He, respectively). The intensity of NNAT signals became stronger on P60 (Fig. 2D2 and E2). Notably, the Henle’s layer in the outermost region and the cuticle layer in the innermost region of the IRS exhibited stronger NNAT signals than those in the middle of Huxley’s layer.

Figure 2.

Immunohistochemistry for NNAT in embryonic and adult whisker hair follicles. Immunostaining for NNAT (red, Cy3) was performed using frozen sections prepared from embryos on embryonic day 16.5 (E16.5, A) and E20.5 (B2 and C2), and from adult tissues on P60 (D2 and E2). HE staining is shown in B1 to E1. Longitudinal sections (A, B, D) and cross-sections (C and E) are shown. Nuclei are stained with DAPI (blue). Immunostaining for NNAT (red, Cy3) and SOX2 (white, Cy5) was performed for frozen sections prepared from tissues on E20.5 (F) and P60 (I). Boxed areas in F and I are enlarged in G1 to G4, H1 to H4, J1 to J4, and K1 to K4. Arrow indicates immature whisker hair follicle. Scale bars = 50 µm (A–F, I), 10 µm (G1–G4, H1–H4, J1–J4, K1–K4). Abbreviations: NNAT, Neuronatin; HE, Hematoxylin and Eosin; He, Henle’s layer; Hu, Huxley’s layer; Ci, cuticle of inner root sheath; ORS, outer root sheath; IRS, inner root sheath.

Double immunostaining for NNAT and SOX2 on E20.5 and P60 revealed that SOX2-positive cells were localized in the dermal papilla (Fig. 2F and I). All NNAT-positive cells in the IRS and NNAT-negative cells in the outer root sheath (ORS) were negative for SOX2 (Fig. 2G1–G4, H1–H4, J1–J4, K1–K4).

In a previous study of the pituitary gland, we observed, for the first time, that NNAT was widely localized in subcellular organelles in addition to the ER.⁷ Consequently, we have re-examined subcellular localization of NNAT in the adult tissues using an antibody against protein disulfide isomerase (PDI), a marker of the ER. Double immunostaining for NNAT and PDI revealed that PDI signals were widely observed in the cytoplasm of cells in the IRS (Fig. 3A). In addition to many signals overlapping with NNAT and PDI (Fig. 3B1–B4, C1–C4, D1–D4, and E1–E4, open arrowheads), NNAT signals negative for PDI and PDI signals negative for NNAT were also observed (Fig. 3B1–B4, C1–C4, D1–D4, and E1–E4, white arrowheads and arrows, respectively).

Figure 3.

Intracellular localization of NNAT and protein disulfide isomerase (PDI) in whisker hair follicles. Double immunostaining for NNAT and PDI, an endoplasmic reticulum (ER) marker, in rat whisker hair follicles on P60 was performed. Boxed areas in A are enlarged in B1 to B4, C1 to C4, D1 to D4, and E1 to E4. NNAT and PDI were visualized with Cy3 (red) and Cy5 (green), respectively, and merged images with DAPI (blue) are shown. White arrowhead: NNAT-single positive signal, open arrowhead: NNAT/PDI-double positive signal, arrow: PDI-single positive signal. Scale bars = 50 µm (A), 5 µm (B–E). Abbreviations: NNAT, Neuronatin; DAPI, 4′,6-diamidino-2-phenylindole.

Immunohistochemistry for NNAT in the Tongues

We next analyzed the anterior part of the tongue, which comprises two papillae, that is, fungiform taste papilla housing taste buds and nontaste filiform papilla (Fig. 4C4 and F4, FFP and Flp, respectively). Immunohistochemistry for SOX2 in the tongue on E21.5 revealed that the basal epithelial cells of both the filiform and fungiform papillae and the taste bud cells in the fungiform papillae were positive for SOX2 (Fig. 4C1), showing weak and strong intensities of SOX2 signals, respectively (Fig. 4C1 and D1, double and single open arrowheads and white arrow, respectively), as previously reported.^24–26 On P60, a number of SOX2-positive cells with elongated nuclei, characteristic of mature taste cells, were detected in the taste buds (Fig. 4F1 and G1, yellow arrow). In addition, SOX2-positive progenitor cells with round nuclei were found in the basal compartment of the taste bud (Fig. 4G1, white arrow).

Figure 4.

Immunohistochemistry for NNAT and SOX2 in embryonic and adult tongue tissues. HE staining of embryonic and adult tongue is shown in A and B. Immunostaining for NNAT (red, Cy3) and SOX2 (white, Cy5) was performed for frozen sections from embryos on E21.5 (C1–C 4) and P60 (F1–F4). Boxed areas in C4 and F4 are enlarged in D1 to D4, E1 to E4, G1 to G4, and H1 to H4. Nuclei are stained with DAPI (blue). White arrowhead: NNAT-single positive cell, open arrowhead: NNAT/SOX2-double positive cell, white arrow: SOX2-single positive cell, yellow arrowhead: NNAT-single positive elongated cell, yellow arrow: mature taste cell, double open arrowhead: basal epithelial cell. Scale bars = 50 µm (A1–A4, C1–C4, F1–F4), 10 µm (D1–D4, E1–E4, G1–G4, H1–H4). Abbreviations: NNAT, Neuronatin; HE, Hematoxylin and Eosin; DAPI, 4′,6-diamidino-2-phenylindole; FFP, fungiform papilla; Flp, filiform papilla.

In the embryonic papillae on E21.5, double immunohistochemistry for NNAT and SOX2 revealed that some NNAT/SOX2-double positive cells intermingled with SOX2-single positive cells in the developing taste buds (Fig. 4D1–D4, open arrowheads and white arrows, respectively). All SOX2-positive basal epithelial cells in the filiform papillae were positive for NNAT (Fig. 4E1–E4, double open arrowheads), as was also observed in the fungiform papillae (Fig. 4D1–D4, double open arrowheads).

In the adult fungiform papillae, various types of cell populations were present, that is, NNAT-single-, NNAT/SOX2-double-, and SOX2-single-positive cells (Fig. 4G1–G4, white arrowheads, open arrowheads, and white arrows, respectively). Cells exhibiting intense NNAT signals extended a characteristic elongated edge toward the taste pore (Fig. 4G1–G4, yellow arrowheads). In contrast, in the adult filiform papillae, although NNAT/SOX2-double positive cells were still observed in the basal epithelium at the bottom (Fig. 4H1–H4, open arrowheads), the upper portion of the basal epithelial layer was negative for SOX2 but positive for NNAT (Fig. 4H1–H4, white arrowheads). Notably, NNAT signals were present in the keratohyalin granules of cells existing in the interpapillary region of the filiform papillae (Fig. 4H1–H4, yellow arrows).

Double immunostaining for NNAT and PDI was then performed to study the relationship between NNAT and the ER (Fig. 5A). Notably, cells in NNAT-positive domains in the filiform and fungiform papillae exhibited immature ER with weak PDI signals (Fig. 5B1–B4 and D1–D4, open arrowheads), whereas intense PDI signals were observed in NNAT-negative keratinocytes in the filiform papilla (Fig. 5B1–B4, arrows) and the fungiform papilla (Fig. 5D1–D4, arrows). NNAT-single positive anucleate cells were present in the corneocytes in the filiform papilla (Fig. 5C1–C4, white arrowheads).

Figure 5.

Intracellular localization of NNAT and PDI in filiform and fungiform papillae. Double immunostaining for NNAT and PDI in the rat filiform and fungiform papillae on P60 was performed. Boxed areas in A are enlarged in B1 to B4, C1 to C4, and D1 to D4. NNAT and PDI were visualized with Cy3 (red) and Cy5 (green), respectively, and merged images with DAPI (blue) are shown. White arrowhead: NNAT-single positive signal, open arrowhead: NNAT/PDI-double positive signal, arrow: PDI-single positive signal. Scale bars = 50 µm (A), 5 µm (B4, C4, D4). Abbreviations: NNAT, Neuronatin; PDI, protein disulfide isomerase; DAPI, 4′,6-diamidino-2-phenylindole.

Immunohistochemistry for NNAT in the Pancreases

Previous studies have reported the expression of Nnat in the pancreas, an endodermal appendage.^8,21 HE staining of embryonic and adult pancreatic tissues was performed to visualize developing pancreatic islets and exocrine cells in addition to the pancreatic ducts and blood vessels (Fig. 6A–C). Immunohistochemistry for NNAT revealed a small number of immunoreactive signals in the region of exocrine cells on E16.5 (Fig. 6A, D, and E, white arrowhead), and granular signals were observed in pancreatic islets (Fig. 6B: PI; Fig. 6F and G: white arrowhead) and exocrine cells on E21.5 (Fig. 6B: EC; Fig. 6F and H: white arrowhead). The signal intensity was increased on P60 (Fig. 6C, I–K, white arrowheads). Strong tube-shaped signals were observed on P60 (Fig. 6I, yellow arrow). Double immunostaining for NNAT and E-cadherin, which is known to be expressed in the pancreatic duct, revealed that cells with tube-shaped NNAT signals were negative for E-cadherin and E-cadherin-positive pancreatic ducts were negative for NNAT (Fig. 6L, M1–M3, yellow arrows and yellow arrowheads, respectively). Because NNAT has been reported to be expressed in the endothelial cells of the pancreatic aorta,²² these tube-shaped signals may represent pancreatic endothelial cells of the artery.

Figure 6.

Localization of NNAT in embryonic and adult pancreatic tissues. HE staining (A–C) and immunostaining for NNAT (D–K, red, Cy3) was performed using serial frozen sections prepared from embryonic tissues on E16.5 (A and D) and E21.5 (B and F), and from adult tissues on P60 (C and I). Boxes in A to C indicate areas adjacent to boxes in D, F, and I, respectively. Nuclei are stained with DAPI (blue). Double immunostaining for NNAT (red, Cy3) and E-cadherin (green, FITC) was performed on P60 (L, M1–M3). Boxed areas in D, F, I, and L are enlarged in E, G, H, J, K, and M1–M3. White arrowhead: NNAT-positive cell, white arrow: NNAT-negative cell, yellow arrow: blood vessel, yellow arrowhead: pancreatic duct. Scale bars = 50 µm (A–C, D, F, I, L), 10 µm (E, G, H, J, K, M1–M3). Abbreviations: NNAT, Neuronatin; HE, Hematoxylin and Eosin; DAPI, 4′,6-diamidino-2-phenylindole; FITC, fluorescein isothiocyanate.

Next, we performed double immunostaining for NNAT and insulin because some NNAT-negative cells were detected in the pancreatic islets on E21.5 (Fig. 6G, white arrow). The results showed that insulin-positive cells had already emerged in the embryonic pancreas on E16.5 (Fig. 7A) but were negative for NNAT (Fig. 7B1–B4, arrows). In the pancreatic islets on E21.5 (Fig. 7C: PI), cells double positive for NNAT and insulin were observed, as were NNAT-single positive and insulin-single positive cells (Fig. 7D1–D4, open arrowheads, white arrowheads, and arrows, respectively). NNAT-positive cells were also detected in exocrine cells (Fig. 7C: EC, Fig. 7 D1–D4: white arrowheads on the lower side). In adult islets, cells positive for NNAT or insulin alone were still detected (Fig. 7E, F1–F4, white arrowheads and arrows, respectively), whereas most insulin-positive cells were positive for NNAT (Fig. 7E, F1–F4, open arrowheads). NNAT-positive cells were also observed in exocrine cells. Insulin-positive cells accounted for about 59% of pancreatic islet cells (Fig. 7G) and were mostly positive for NNAT on P60. Insulin/NNAT-double positive cells accounted for approximately 96% of insulin-positive cells and 97% of NNAT-positive cells (Fig. 7G), exhibiting a high proportion of colocalization.

Figure 7.

Immunohistochemistry for NNAT and insulin in embryonic and adult pancreatic tissues. Immunostaining for NNAT (red, Cy3) and insulin (green, FITC) was performed for frozen sections prepared from tissues on E16.5 (A), E21.5 (C), and P60 (E). Boxed areas in A, C, and E are enlarged in B1 to B4, D1 to D4, and F1 to F4. The proportions of insulin-positive cells/DAPI and insulin/NNAT-double positive cells in insulin-positive and NNAT-positive cell populations were recorded for six areas (129–495 cells each) and are listed in G. Data are presented as mean ± SE. White arrowhead: NNAT-single positive cell, open arrowhead: NNAT/insulin-double positive cell, arrow: insulin-single positive cell. Scale bars = 50 µm (A, C, E), 10 µm (B1–B4, D1–D4, F1–F4). Abbreviations: NNAT, Neuronatin; FITC, fluorescein isothiocyanate; DAPI, 4′,6-diamidino-2-phenylindole; PI, pancreatic islet; EC, exocrine cell.

Next, we analyzed the localization of NNAT in subcellular organelle other than the ER by double staining for NNAT and PDI (Fig. 8A). The results showed that, whereas most NNAT-positive cells in pancreatic islets were positive for PDI, some signals with positivity for NNAT only were also observed (Fig. 8B1–B5, open arrowheads and white arrowheads, respectively). In addition, exocrine cells outside the pancreatic islets exhibited stronger PDI signals, but overlapping of NNAT and PDI signals was absent (Fig. 8C1–C5). Notably, endothelial cells with strong NNAT signals were negative for PDI (Fig. 8A, yellow arrow).

Figure 8.

Intracellular localization of NNAT and PDI in the pancreas. Double immunostaining for NNAT (Cy3, red) and PDI (Cy5, green) in the rat pancreas on P60 was performed and merged images with DAPI (blue) are shown. Boxed areas in A are enlarged in B1 to B4 and C1 to C4. Orthogonal projections by Z-stack imaging with 0.124-µm-thick slices are shown (B5 and C5). White arrowhead: NNAT-single positive signal, open arrowhead: NNAT/PDI-double positive signal, yellow arrow: blood vessel. Scale bars = 50 µm (A), 5 µm (B, C). Abbreviations: NNAT, Neuronatin; PDI, protein disulfide isomerase; DAPI, 4′,6-diamidino-2-phenylindole.

Discussion

NNAT is known to be involved in diverse intracellular roles in various tissues and/or cell types through Ca^2⁺ signaling, glucose signaling, and inflammatory signaling.²⁰ In contrast to its initial identification as an ER membrane protein,⁸ NNAT localization has also been observed in several subcellular organelles of the pituitary stem/progenitor cells⁷ and in the outer segment of rod photoreceptors, which do not contain an ER.¹⁶ The present immunohistochemical study demonstrated that NNAT was localized in developing tissues and in differentiating cells in the pancreas, tongue, whisker hair follicles, and testis and suggested that NNAT was involved in cell differentiation and in the function of terminally differentiated cells beyond the germ layer. In addition, our findings suggested that NNAT is not a directly ER-associated protein.

In this study, we demonstrated, for the first time, that NNAT emerged in round spermatids, accompanied by acrosome formation, from the early stage in round spermatids to the maturation stage in spermatozoa. The ER is absent from spermatozoa, and the acrosome is derived from the Golgi apparatus.²⁷ Moreover, in our previous study, we found that NNAT is localized in progenitor cells with rudimentary ER in the pituitary gland⁷ and the disk membrane of the outer segment of rod photoreceptors;¹⁶ thus, NNAT may be located in the membrane of appropriate subcellular organelles, not limited to the ER.

In the whisker hair follicles, specific localization of NNAT in the IRS was demonstrated. The IRS is known to abundantly produce trichohyalin,²⁸ which mediates keratin filamentous assembly, although trichohyalin is also produced in the medulla of the hair shaft.²⁹ In addition to trichohyalin in the IRS, several molecules involved in keratinization, such as peptidyl arginine deiminase,³⁰ involucrin,³¹ and transglutaminases,³² are produced. However, the localization of these molecules differs from that of NNAT, exhibiting expression in other parts of the hair follicle in addition to the IRS. Notably, keratin K71, which has been reported to exhibit limited localization in the three layers of the IRS,³³ exhibits localization similar to that of NNAT. Deficiency of mK6irs1 protein, the mouse ortholog of keratin K71, causes alopecia.³⁴ Thus, elucidation of the coexistence of NNAT and keratin K71 is essential to clarify the mechanisms mediating the development and keratinization of the IRS.

Sheridan et al.³⁵ further described that the IRS is consistently negative for SERCA2, indicating that the Ca^2⁺ concentration in the ER of the IRS is modulated by a mechanism that does not involve SERCA2. Accordingly, the interaction between NNAT and SERCA2 is cell- or tissue-specific but not universal.

In the tongue, NNAT exhibited distinct localizations in undifferentiated and differentiated cells present in two types of papillae, that is, filiform and fungiform papillae. In the filiform papillae, both SOX2-positive basal epithelial cells in the bottom region and SOX2-negative cells in the upper region were positive for NNAT, whereas filiform keratinocytes were negative for NNAT. These results indicated that NNAT-positive cells in the filiform papillae were stem/progenitor cells and committed cells to differentiate into keratinocytes. Moreover, NNAT was also localized in corneocytes, terminally differentiated keratinocytes. In contrast, in taste bud cells in the fungiform papillae, NNAT was localized in SOX2-positive basal epithelial cells and taste bud cells in addition to SOX2-negative mature taste cells. NNAT-positive taste cells negative for SOX2 revealed extended NNAT signals toward the front edge of the elongated cells. The taste buds contain immature basal cells (type IV) and elongated mature taste cells with three morphologically distinguished types (types I, II, and III), which detect at least five types of tastes (salt, sweet, umami, bitter, and sour).³⁶ Because type IV and type I cells express Sox2 at high levels, whereas type II and type III cells rarely express Sox2,²⁶ NNAT-positive taste receptor cells could be type II and/or III.

The molecular mechanisms of NNAT function are unclear. In type II cells in the taste buds, three tastes (sweet, umami, and bitter) are received through taste transduction mediated by taste receptors, TAS1Rs and TAS2Rs, which belong in the G protein-coupled receptor family. Notably, these receptors are frequently observed outside the gustatory system, such as in the sperm,³⁷ and some are localized in acrosomes,³⁸ in keratinocytes in the skin,³⁹ airway and gut,⁴⁰ and in the placenta.⁴¹

In addition, one of the downstream signaling molecules of taste receptors, TRPM5, a Ca^2⁺-activated cation channel, has been reported to localize in pancreatic β-cells.^42,43 Therefore, to elucidate the role of NNAT, it may be necessary to clarify the signaling pathways involved in keratinization in the filiform papillae of the tongue and the IRS in the hair follicle, in acrosome formation in the spermatozoa, in photoreception in the rod photoreceptors, and in characteristic signals in NNAT-positive stem/progenitor cells.

In the pancreas, we showed that NNAT immunoreactive signals emerged in a few exocrine cells but not in insulin-positive cells in the middle stage of embryonic development, followed by signals in β-cells in the pancreatic islets in the late stage of embryonic development. In the postnatal pancreas, NNAT signals were localized in a few exocrine cells and the majority of β-cells. Millership et al.⁴⁴ recently reported that the C-terminal hydrophilic domain of NNAT, located on the cytosolic side of the ER membrane, interacts with components of the signal peptidase complex to induce translocation of preproinsulin into the ER and to modulate insulin secretion in β-cells. Moreover, NNAT was shown not to be involved in Ca^2⁺-mediated signaling using Nnat-null primary β-cells, suggesting that NNAT modulates insulin secretion at the stage of translocation of preproinsulin into the ER. However, the involvement of NNAT in secretion control may not be common to other endocrine tissues. In the pituitary gland, we observed that NNAT was present in stem/progenitor cells in the developing and developed pituitary but not in terminally differentiated hormone-producing cells.⁷ NNAT may have diverse roles in stem/progenitor cells and terminally differentiated cells, even in such a relevant endocrine tissue. In addition, the roles of NNAT in insulin-negative cells of islet and exocrine cells and subcellular organelles outside of the ER remain unclear.

In summary, our current findings showed that NNAT exhibited various localizations, suggesting diverse cell- or tissue-specific roles. NNAT functions beyond the germ layer and in the differentiating (tongue and testis) and terminally differentiated (pancreas, tongue, whisker hair follicle, and spermatozoa) cells. The observed subcellular localization of NNAT in the acrosome and IRS provides essential insights into the role of NNAT. Future studies are needed to identify molecules that interact with NNAT and explore the related molecular mechanisms.

Footnotes

Acknowledgements

We thank Dr. Hideji Yako and Dr. Masashi Higuchi for technical help and continuous support.

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

All authors have contributed to this article as follows: TK and YK developed the study design, coordination, and manuscript drafting; NK and SY performed immunohistochemical and quantitative analyses; TK and YK provided study supervision and critical manuscript revision; and all authors have read and approved the manuscript as submitted.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported by JSPS KAKENHI grants (nos. 21380184 to Y.K. and 24580435 to T.K.), by the MEXT-Supported Program for the Strategic Research Foundation at Private Universities, 2014–2018, and by a research grant (A) to Y.K. from the Institute of Science and Technology, Meiji University. This study was supported by Meiji University International Institute for BioResource Research (MUIIR).

References

Joseph

Dou

Tsang

Neuronatin mRNA: alternatively spliced forms of a novel brain-specific mammalian developmental gene. Brain Res. 1995;690:92–8.

Aikawa

Kato

Elsaeesser

Kato

Molecular cloning of porcine neuronatin and analysis of its expression during pituitary ontogeny. Exp Clin Endocrinol Diabetes. 2003;111:475–9.

Joseph

Dou

Tsang

Molecular cloning of a novel mRNA (neuronatin) that is highly expressed in neonatal mammalian brain. Biochem Biophys Res Commun. 1994;201:1227–34.

Usui

Ichikawa

Miyazaki

Nagai

Kumanishi

Isolation of cDNA clones of the rat mRNAs expressed preferentially in the prenatal stages of brain development. Brain Res Dev Brain Res. 1996;97:185–93.

Wijnholds

Chowdhury

Wehr

Gruss

Segment-specific expression of the neuronatin gene during early hindbrain development. Dev Biol. 1995;171:73–84.

Nishida

Yoshioka

St-Amand

The top 10 most abundant transcripts are sufficient to characterize the organs functional specificity: evidences from the cortex, hypothalamus and pituitary gland. Gene. 2005;344:133–41.

Kanno

Higuchi

Yoshida

Yako

Chen

Ueharu

Nishimura

Mitsuishi

Kato

Expression studies of Neuronatin in the prenatal and postnatal rat pituitary. Cell Tissue Res. 2016;364:273–88.

Joe

Lee

Suh

Han

Lim

Song

Seong

Jung

MH.

Crucial roles of neuronatin in insulin secretion and high glucose-induced apoptosis in pancreatic beta-cells. Cell Signal. 2008;20:907–15.

Sel

Patzel

Poggi

Kaiser

Kalinski

Schicht

Paulsen

Nass

Temporal and spatial expression pattern of Nnat during mouse eye development. Gene Expr Patterns. 2017;23–4:7–12.

10.

Lin

Bell

Uwanogho

Perfect

Noristani

Bates

Snetkov

Price

Sun

YM.

Neuronatin promotes neural lineage in ESCs via Ca(2⁺) signaling. Stem Cells. 2010;28:1950–60.

11.

Dugu

Nakahara

Uchi

Liu

Hirano

Yokomizo

Furue

Neuronatin is related to keratinocyte differentiation by up-regulating involucrin. J Dermatol Sci. 2014;73:225–31.

12.

Suh

Kim

Moon

Hong

Eun

Lim

Choi

Song

Jung

MH.

Ectopic expression of Neuronatin potentiates adipogenesis through enhanced phosphorylation of cAMP-response element-binding protein in 3T3-L1 cells. Biochem Biophys Res Commun. 2005;337:481–9.

13.

Oyang

Davidson

Lee

Poon

MM.

Functional characterization of the dendritically localized mRNA neuronatin in hippocampal neurons. PLoS One. 2011;6:e24879.

14.

Ryu

McDonnell

Choi

Gao

Hahn

Joshi

Park

Catena

Brazin

Vahdat

Silver

Mittal

Suppression of miRNA-708 by polycomb group promotes metastases by calcium-induced cell migration. Cancer Cell. 2013;23:63–76.

15.

Yang

Wei

Wang

Guo

Lee

Metformin induces ER stress-dependent apoptosis through miR-708-5p/NNAT pathway in prostate cancer. Oncogenesis. 2015;4:e158.

16.

Shinde

Pitale

Howse

Gorbatyuk

Neuronatin is a stress-responsive protein of rod photoreceptors. Neuroscience. 2016;328:1–8.

17.

Gburcik

Cleasby

Timmons

JA.

Loss of neuronatin promotes “browning” of primary mouse adipocytes while reducing Glut1-mediated glucose disposal. Am J Physiol Endocrinol Metab. 2013;304:E885–94.

18.

Sharma

Rao

Shankar

Satishchandra

Jana

NR.

Lafora disease ubiquitin ligase malin promotes proteasomal degradation of neuronatin and regulates glycogen synthesis. Neurobiol Dis. 2011;44:133–41.

19.

Sharma

Mukherjee

Rao

Iyengar

Shankar

Satishchandra

Jana

NR.

Neuronatin-mediated aberrant calcium signaling and endoplasmic reticulum stress underlie neuropathology in Lafora disease. J Biol Chem. 2013;288:9482–90.

20.

Pitale

Howse

Gorbatyuk

Neuronatin protein in health and disease. J Cell Physiol. 2017;232:477–81.

21.

Chu

Tsai

M-J.

Neuronatin, a downstream target of BETA2/NeuroD1 in the pancreas, is involved in glucose-mediated insulin secretion. Diabetes. 2005;54:1064–73.

22.

Mzhavia

Ikeda

Chu

Goldberg

Dansky

HM.

Neuronatin: a new inflammation gene expressed on the aortic endothelium of diabetic mice. Diabetes. 2008;57:2774–83.

23.

Han

Jeong

Lee

Yong

Boldbaatar

Joo

Soh

Park

Lim

Lee

Yang

Neuronatin is associated with an anti-inflammatory role in the white adipose tissue. J Microbiol Biotechnol. 2017;27:1180–8.

24.

Okubo

Clark

Hogan

BL.

Cell lineage mapping of taste bud cells and keratinocytes in the mouse tongue and soft palate. Stem Cells. 2009;27:442–50.

25.

Okubo

Pevny

Hogan

BL.

Sox2 is required for development of taste bud sensory cells. Genes Dev. 2006;20:2654–9.

26.

Suzuki

Expression of Sox2 in mouse taste buds and its relation to innervation. Cell Tissue Res. 2008;332:393–401.

27.

Nakamura

Moriya

Baba

Michikawa

Yamanobe

Arai

Okinaga

Kobayashi

An endoplasmic reticulum protein, calreticulin, is transported into the acrosome of rat sperm. Exp Cell Res. 1993;205:101–10.

28.

Birbeck

Mercer

. The electron microscopy of the human hair follicle. III. The inner root sheath and trichohyaline. J Biophys Biochem Cytol. 1957;3:223–30.

29.

Hamilton

Payne

Jr O’Keefe

EJ.

Trichohyalin: presence in the granular layer and stratum corneum of normal human epidermis. J Invest Dermatol. 1991;96:666–72.

30.

Rogers

Winter

McLaughlan

Powell

Nesci

Peptidylarginine deiminase of the hair follicle: characterization, localization, and function in keratinizing tissues. J Invest Dermatol. 1997;108:700–7.

31.

Thibaut

Candi

Pietroni

Melino

Schmidt

Bernard

BA.

Transglutaminase 5 expression in human hair follicle. J Invest Dermatol. 2005;125:581–5.

32.

Alibardi

Ultrastructural immunolocalization of involucrin in the medulla and inner root sheath of the human hair. Ann Anat. 2012;194:345–50.

33.

Langbein

Yoshida

Praetzel-Wunder

Parry

Schweizer

The keratins of the human beard hair medulla: the riddle in the middle. J Invest Dermatol. 2010;130:55–73.

34.

Peters

Sedlmeier

Bussow

Runkel

Luers

Korthaus

Fuchs

Hrabe

Angelis

Stumm

Russ

Porter

Augustin

Franz

Alopecia in a novel mouse model RCO3 is caused by mK6irs1 deficiency. J Invest Dermatol. 2003;121:674–80.

35.

Sheridan

Hollowood

Sakuntabhai

Dean

Hovnanian

Burge

Expression of sarco/endo-plasmic reticulum Ca2⁺-ATPase type 2 isoforms (SERCA2) in normal human skin and mucosa, and Darier’s disease skin. Br J Dermatol. 2002;147:670–4.

36.

Sullivan

Borecki

Oleskevich

Stem and progenitor cell compartments within adult mouse taste buds. Eur J Neurosci. 2010;31:1549–60.

37.

Taste perception: from the tongue to the testis. Mol Hum Reprod. 2013;19:349–60.

38.

Meyer

Voigt

Widmayer

Borth

Huebner

Breit

Marschall

de Angelis

Boehm

Meyerhof

Gudermann

Boekhoff

Expression of Tas1 taste receptors in mammalian spermatozoa: functional role of Tas1r1 in regulating basal Ca(2)(⁺) and cAMP concentrations in spermatozoa. PLoS One. 2012;7:e32354.

39.

Wolfle

Elsholz

Kersten

Haarhaus

Muller

Schempp

CM.

Expression and functional activity of the bitter taste receptors TAS2R1 and TAS2R38 in human keratinocytes. Skin Pharmacol Physiol. 2015;28:137–46.

40.

Finger

Kinnamon

SC.

Taste isn’t just for taste buds anymore. F1000 Biol Rep. 2011;3:20.

41.

Wolfle

Elsholz

Kersten

Haarhaus

Schumacher

Schempp

CM.

Expression and functional activity of the human bitter taste receptor TAS2R38 in human placental tissues and JEG-3 cells. Molecules. 2016;21:306.

42.

Prawitt

Monteilh-Zoller

Brixel

Spangenberg

Zabel

Fleig

Penner

TRPM5 is a transient Ca2⁺-activated cation channel responding to rapid changes in [Ca2⁺]i. Proc Natl Acad Sci U S A. 2003;100:15166–71.

43.

Philippaert

Pironet

Mesuere

Sones

Vermeiren

Kerselaers

Pinto

Segal

Antoine

Gysemans

Laureys

Lemaire

Gilon

Cuypers

Tytgat

Mathieu

Schuit

Rorsman

Talavera

Voets

Vennekens

Steviol glycosides enhance pancreatic beta-cell function and taste sensation by potentiation of TRPM5 channel activity. Nat Commun. 2017;8:14733.

44.

Millership

da Silva Xavier

Choudhury

Bertazzo

Chabosseau

Pedroni

Irvine

Montoya

Faull

Taylor

Kerr-Conte

Pattou

Ferrer

Christian

John

Latreille

Liu

Rutter

Scott

Withers

DJ.

Neuronatin regulates pancreatic beta cell insulin content and secretion. J Clin Invest. 2018;128:3369–81.