- Email: [email protected]
The tight junction component protein, claudin-4, is expressed by enteric neurons in the rat distal colon
Available online at www.sciencedirect.com
Neuroscience Letters 428 (2007) 88–92
The tight junction component protein, claudin-4, is expressed by enteric neurons in the rat distal colon Shin-ichiro Karaki, Izumi Kaji, Yasuko Otomo, Hideaki Tazoe, Atsukazu Kuwahara ∗ Laboratory of Physiology, Institute for Environmental Sciences, and Graduate School of Nutritional and Environmental Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan Received 25 July 2007; received in revised form 11 September 2007; accepted 16 September 2007
Abstract The expression of a tight junction (TJ) component protein, claudin-4, in the enteric neurons was investigated in the rat distal colon by immunohistochemistry and RT-PCR. Claudin-4 immunoreactivity was detected in almost all neurofilament-positive enteric neurons both of the submucosal and the myenteric plexuses, and both of the cell bodies and the neurofibers. The immunoreactivity of enteric neurons for claudin-4 was divided into two types: strongly and weakly positive neurons. Especially in the myenteric plexus, the stained neurons were classified by Dogiel’s morphological classification of enteric neurons. The strongly stained claudin-4 positive neurons show Dogiel type II morphology, while the weakly stained claudin-4 positive neurons show Dogiel type I morphology. These immunohistochemical data were supported by mRNA expression in the muscle plus submucosa preparation containing the submucosal and myenteric plexuses, as well as mucosa preparation. The physiological function of claudin-4 expressed on enteric neurons is unclear up to now. It is however suggested that claudin-4 expressed on enteric neurons might play roles for the neural activity, for example as insulation between neurofibers. In conclusion, the present study clearly shows that claudin-4 is expressed by enteric neurons. This is the first evidence that the neuron itself expresses the TJ component protein, claudin-4, in the nervous system. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Tight junction; Claudin; Enteric nervous system; Neurofilament; Immunohistochemistry; RT-PCR
Tight junction (TJ) is one of the intercellular junctional structures forming the reticular-paired strands to separate the partition of the body fluid compartments especially in the epithelial tissues (e.g. skin, renal tubules, and gastrointestinal (GI) epithelia) and the endothelial tissues (e.g. blood vessels). The most primary functions of TJs are considered to prevent from passing fluid, ions and some molecules paracellularly as a barrier. However, TJs not only prevent from passing but also TJs expressed in diverse tissues are known to function for selective pores (or channels) to pass diverse ions and molecules, etc. [9] Therefore, the investigation of TJs is currently considered to be important for both basic and clinical aspects [2,10,12], for example, drug delivery system [4,8]. In the nervous system, the TJ-like structures have previously been observed both in central nervous system (CNS) and peripheral nervous system (PNS). In the CNS, TJs of the brain endothelial cells is reported to function as the blood–brain barrier [13,14].
∗
Corresponding author. Tel.: +81 54 264 5707; fax: +81 54 264 5707. E-mail address: [email protected] (A. Kuwahara). URL: http://physiology.u-shizuoka-ken.ac.jp/ (A. Kuwahara).
0304-3940/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2007.09.059
The TJ strands are composed of the assemblies of cell–cell interlinked transmembrane proteins including claudins and occludin, and cytoskeleton-anchoring intracellular proteins including ZO-1–ZO-3, etc. (for review as [3]). Claudins are four-transmembrane ∼23 kDa proteins, and to date, 24 isoforms (claudin-1–claudin-24) are identified [2]. The distinct claudins are expressed by different cells, and are considered to have different properties. For example, the claudin-4 expression is reported to decrease paracellular conductance through a selective decrease in sodium (Na+ ) permeability [11], while claudin-2 expression is reported to increase cation permeability [1]. The claudin-11 and claudin-19 have been reported to be expressed by myelinating oligodendrocytes and Schwann cells in CNS [6] and PNS [5], respectively. However, there is no report for the expression of any claudins in the neuron itself both in CNS and PNS. At first, we have designed to investigate the expression and the distribution of TJ proteins including claudin-1–claudin-4, occludin, and ZO-1 in the epithelia of the rat distal colon during the course of physiological study. At that time, we found that the TJ proteins were expressed by epithelium in specific distribution (data not shown) similar to the previous report [7]. In
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
addition to the previous findings, we found out the claudin-4immunoreactive neurons in the enteric nervous system (ENS). Furthermore, we have checked the immunoreactivities for the other TJ proteins (claudin-1–claudin-3, occludin and ZO-1), but we did not find the expression of these TJ proteins in the ENS. Therefore, in the present study, we have investigated the expression and the distribution of claudin-4 immunoreactive neurons in the ENS in more detail. In the present study, rat distal colon was taken from male Wister rats (SLC; Hamamatsu, Shizuoka, Japan; or Chars River Laboratories Japan, Yokohama, Japan) after ether anesthetization followed by guillotine decapitation. The handling and the sacrifice of animals were in accordance with the Guidelines for the care and use of laboratory animals of the University of Shizuoka. In immunohistochemistry, the following three kinds of preparations were used: (1) 10-m thin frozen-sections, and whole mount preparations of (2) submucosa containing submucosal plexus (SMP) and (3) myenteric plexus on the longitudinal muscle layer (MP-LM). For preparing the frozen-sections, the flesh segments of distal colon were rapidly frozen with optimal cutting temperature (OCT) compound (TissueTek, Sakura Finetechnical, Tokyo, Japan) in liquid nitrogen, cut into 10-m thin sections in a cryostat. The frozen sections stuck on glass-slides were dried by cool blowing, and fixed in cold methanol (MeOH) at −30 ◦ C for 10 min. While, for preparing whole-mount preparations, the flesh segments were immersed in phosphate buffered saline (PBS) containing 10−6 M nicardipine (PBSN) at 4 ◦ C, cut open into the flat sheet, and pined with mucosa-up on the siliconrubber filled Petri dish in ice-cold PBSN. Mucosa was removed by scraping by using a blade, and the submucosa was exfoliated by using a fine tweezers. Furthermore, circular muscle was removed along the fiber to make the MP-LM preparations. The SMP and MP-LM preparations were pinned on the siliconrubber filled Petri dish, and fixed in MeOH at −30 ◦ C for 30 min. The fixed whole-mount preparations were stored in PBS containing 0.1% sodium azide (NaN3 ) at 4 ◦ C until use. Before incubation with primary antibodies, sections and whole-mount preparations were incubated with 10% normal donkey serum, 1% Triton X-100 and 0.1% NaN3 in PBS at room temperature for 1 h to prevent nonspecific immunoreaction. The preblocked samples were incubated with primary antibodies (Table 1) overnight at 4 ◦ C, and washed in PBS, 3 × 10 min. Then, the samples were incubated with secondary antibodies (Table 1) for 1 h at room temperature, washed in PBS, 3 × 10 min, and mounted on glass-slides in case of whole-mount preparations. Negative control experiment using each secondary antibody have previously been performed, resulting no staining was confirmed (data not shown). The samples were cover-slipped with a mounting medium (DakoCytomation, Glostrup, Denmark), and observed by a fluorescence microscope (IX70, Olympus, Tokyo, Japan). The fluorescence images were captured by a cooled chargecoupled device digital camera system (AxioVision 135; Zeiss, Munich-Hallbergmoos, Germany). For reverse transcriptase/polymerase chain reaction (RTPCR) analysis, the scraped-mucosa from the cut-opened flat sheets of the rat distal colon, and the smooth muscle
89
Table 1 Antibodies Primary antibody
Host
Dilution
Source
Anti-claudin-4 Anti-NFsa Anti-GFAP
Goat Rabbit Mouse
1:500 1:5000 1:2000
Santa Cruz (sc-17664) BIOMOL (NA1297) CHEMICON (MAB360)
Secondary antibody
Label
Dilution
Source
Anti-goat IgG
Alexa Fluor 488
1:200
Anti-rabbit IgG
Alexa Fluor 594
1:200
Anti-mouse IgG
Alexa Fluor 594
1:200
Molecular probes (A21203) Molecular probes (A21207) Molecular probes (A11058)
a This antibody is a rabbit polyclonal antiserum cocktail to three neurofilaments (NFs) including NF-L (62 kDa), NF-M (102 kDa) and NF-H (110 kDa).
with submucosal layer completely without mucosa were immersed immediately in a ribonucleic acid (RNA) stabilization regent (RNAlater; Qiagen K.K., Tokyo, Japan). Tissue samples were freeze-ground by using a grinding mill (SK-100; Tokken, Kashiwa, Chiba, Japan). Total RNAs were isolated by RNeasy Micro Kit (Qiagen K.K.). “Deoxyribonuclease (RT Grade) for Heat Stop” (Nippon Gene Co. Ltd., Tokyo, Japan) was used to remove the genome deoxynucleic acid (DNA), and the complementary DNA (cDNA) was synthesized by using first Strand cDNA Synthesis Kit for RT-PCR (AMV) (Roche Applied Science, Mannheim, Germany). The synthesized cDNAs were stored at −30 ◦ C until use. Primers for RT-PCR of claudin-4 were based on the rat claudin-4 messenger RNA (mRNA) sequence (GenBank: NM 001012022): 5 -CTCTCGCCTCCACGTTACTC-3 (forward) and 5 -AGGGTAGGTGGGTGGGTAAG-3 (reverse)
Fig. 1. Immunoreactivity for claudin-4 in the 10-m thin sections of rat distal colon. Claudin-4 immunoreactivity was observed in surface epithelium, submucosal plexus (arrow) and myenteric plexus (arrowhead). Bar = 100 m. Abbreviations: CM, circular muscle layer; LM, longitudinal muscle layer; MM, muscularis mucosa; muc, mucosa; SM, submucosa. Inset: magnification of the claudin-4 immunoreactive myenteric neurons. Arrow in the inset picture indicates axon-like staining. Bar = 50 m.
90
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
(estimated amplicon size: 190 bp). PCR of -actin was performed to provide a control using the following primers: 5 -GACTACCTCATGAAGATCCT-3 (forward) and 5 (reverse) (estimated CCACATCTGCTGGAAGGTGG-3 amplicon size: 512 bp). PCR was performed by using a RT-PCR kit (Ready-To-Go RT-PCR Beads; Amersham plc, Buckinghamshire, UK), and Takara PCR Thermal Cycler MP (TP3000, TaKaRa Bio Inc., Otsu, Shiga, Japan). Before PCR, reaction mixtures were incubated at 94 ◦ C for 5 min to completely denature the template. PCR cycles consisted of denaturing at 94 ◦ C for 1 min, annealing at 60 ◦ C for 1 min, and extension at 72 ◦ C for 1 min. The reactions were repeated for 35 cycles, followed by extension at 72 ◦ C for 7 min. Finally, amplification products were stored at 4 ◦ C. The amplification products and a DNA size marker (GenRuler 100 bp DNA Ladder Plus; Fermentas, Burlington, Ontario, Canada) were separated by electrophoresis on 1.5% agarose gel in 0.5× TRIS–borate–EDTA buffer containing ethidium bromide. Bands of the amplification products were viewed by ultraviolet light, and the images were taken by GelDoc2000 (Bio-Rad Laboratories, Hercules, CA, USA). Fig. 1 shows the immunoreactivity for claudin-4 in the rat distal colon. Claudin-4-immunoreactivity was observed in the surface epithelium of mucosa as well as the previous reports [7]. Furthermore, we have observed additional claudin-4 immunoreactivity in the ENS of distal colon; the myenteric and submucosal plexuses showed stronger claudin-4-immunoreactivity than that of epithelium (Fig. 1). In higher magnification, claudin-4immunoreactivity was observed as fiber-like morphologies in
both nerve cell soma and axons (Fig. 1, inset). Thus, we performed double staining for claudin-4 and neurofilaments (NFs) or glial filament acidic protein (GFAP). The results showed that the claudin-4 was completely colocalized with NFs (Fig. 2), whereas there was no colocalization between claudin4 and GFAP (Fig. 3). To confirm the results, we tested the following combination of primary and secondary antibodies: rabbit anti-NF + anti-goat IgG Alexa Fluor488, and goat anticaludin-4 + anti-rabbit IgG Alexa Fluor594. The results indicate both negative (data not shown). These indicate that claudin4 expressed by enteric neurons probably connects NFs, but not by enteric glia. For the morphology of enteric neurons, claudin-4-immunoreactive neurons were roughly divided into two groups; one was strongly stained and another was weakly stained (Fig. 2). The strongly stained claudin-4 positive neurons showed the Dogiel type II morphology, that is, smooth cell soma and multiple axonal processes. On the other hand, all Dogiel type I neurons showing short and lamellar dendrites, and single axon were weakly stained for claudin-4. The strongly stained claudin-4 positive neurons were also strongly stained by NFs, and the weakly stained claudin-4 positive neurons were also weakly stained by NFs. This result suggests that the expression of claudin-4 is connected with the expression of NFs. To support the immunohistochemical data of claudin-4 expression in enteric neurons, the expression of claudin-4 mRNA was investigated by RT-PCR analysis. The amplicon of detected bands (near 200 bp for claudin-4 and near 500 bp for -actin) were consistent with each estimated amplicon sizes (190 and 512 bp, respectively). The results showed that claudin-
Fig. 2. Immunoreactivity for claudin-4 and NFs in the submucosal and myenteric plexuses of rat distal colon. Claudin-4- and NF-immunoreactivities were observed in the whole-mount preparations of submucosal (A–C) and myenteric (D–F) plexuses. Claudin-4 immunoreactivity was visualized by the Alexa488-conjugated secondary antibody (A and D, green), and NF-immunoreactivity was by the Alexa594 (B and E, red). Merged images were also shown in D and F, and the yellow indicates colocalization. Numbers in panel F respectively indicates the following characteristics: (1 and 2) representative Dogiel type II neurons, i.e., smooth cell bodies and multiple axons; (3) smooth cell body like Dogiel type II, but only single axon; (4) Dogiel type II morphology, but lower positive in both claudin-4 and NFs than 1–3 neurons; (5–7) Dogiel type I-like neurons, i.e., short lamellar dendrites and single axon; (8) looking like Dogiel type I, but difficult to identify definitely; (9) very lower positive or negative neurons for both claudin-4 and NFs. All bars indicate 50 m.
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
91
Fig. 3. Immunoreactivity for claudin-4 and GFAP of the submucosal and myenteric plexuses. Claudin-4- and GFAP-immunoreactivities were observed in the whole-mount preparations of submucosal (A–C) and myenteric (D–F) plexuses. Claudin-4-immunoreactivity was visualized by the Alexa488-conjugated secondary antibody (A and D, green), and GFAP-immunoreactivity was by the Alexa594 (B and E, red). Merged images were also shown in D and F, but there were no the yellow as colocalization. All bars indicate 50 m.
4 mRNA was detected in both the scraped-mucosa and the muscle plus submucosa samples (Fig. 4). The detection of claudin-4 mRNA in the muscle plus submucosa samples including myenteric and submucosal neurons supports the present immunohistochemical data. What is the physiological function of claudin-4 expressed by enteric neurons? The claudin-4 expression is reported to decrease paracellular conductance through a selective decrease in sodium (Na+ ) permeability, already mentioned above [11]. Thus, one possibility is considered that they work as insulation between each nerve fiber to prevent from signal crossing. However, claudin-4 was expressed also in the cytoplasm of enteric neurons. It is unclear that the claudin-4 expressed in cytoplasm has some functions there, or only stands by at cell bodies for being transported to axons or dendrites. To date, there is no
evidence about the function of claudin-4 expressed in enteric neurons. Therefore, further study needs to perform in order to clarify the function of claudin-4. In conclusion, the present study indicates that claudin-4 is expressed by enteric neurons, but the function of claudin-4 expressed by enteric neurons is unclear yet. However, this is the first evidence that the neuron itself expresses the TJ component protein, claudin-4 in the nervous system. Acknowledgements This work was partly supported by a Japan Society for the Promotion of Science (no. 18590207) and Smoking Research Foundation to A. Kuwahara; and Promotion of Health and Nutrition from the Danone Institute to S. Karaki. References
Fig. 4. Analysis of expression of claudin-4 mRNA in the rat distal colon by RTPCR. The amplification products of RT-PCR for claudin-4 (lanes 3 and 5) and actin (lanes 2 and 4), and DNA maker (lane 1) were separated by electrophoresis containing ethidium bromide, and viewed by UV light. Bands for Claudin-4 and -actin were detected in both mucosa and muscle plus submucosal samples at near 200 and 500 bp, respectively. RT-PCR for -actin was used as a control.
[1] S. Amasheh, N. Meiri, A.H. Gitter, T. Schoneberg, J. Mankertz, J.D. Schulzke, M. Fromm, Claudin-2 expression induces cation-selective channels in tight junctions of epithelial cells, J. Cell. Sci. 115 (2002) 4969– 4976. [2] M. Furuse, S. Tsukita, Claudins in occluding junctions of humans and flies, Trends Cell. Biol. 16 (2006) 181–188. [3] L. Gonzalez Mariscal, A. Betanzos, P. Nava, B.E. Jaramillo, Tight junction proteins, Prog. Biophys. Mol. Biol. 81 (2003) 1–44. [4] M. Kondoh, A. Takahashi, M. Fujii, K. Yagi, Y. Watanabe, A novel strategy for a drug delivery system using a claudin modulator, Biol. Pharm. Bull. 29 (2006) 1783–1789. [5] T. Miyamoto, K. Morita, D. Takemoto, K. Takeuchi, Y. Kitano, T. Miyakawa, K. Nakayama, Y. Okamura, H. Sasaki, Y. Miyachi, M. Furuse, S. Tsukita, Tight junctions in Schwann cells of peripheral myelinated axons: a lesson from claudin-19-deficient mice, J. Cell. Biol. 169 (2005) 527– 538.
92
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
[6] K. Morita, H. Sasaki, K. Fujimoto, M. Furuse, S. Tsukita, Claudin-11/OSPbased tight junctions of myelin sheaths in brain and Sertoli cells in testis, J. Cell. Biol. 145 (1999) 579–588. [7] C. Rahner, L.L. Mitic, J.M. Anderson, Heterogeneity in expression and subcellular localization of claudins 2, 3, 4, and 5 in the rat liver, pancreas, and gut, Gastroenterology 120 (2001) 411–422. [8] N.N. Salama, N.D. Eddington, A. Fasano, Tight junction modulation and its relationship to drug delivery, Adv. Drug Deliv. Rev. 58 (2006) 15–28. [9] E.E. Schneeberger, R.D. Lynch, Structure, function, and regulation of cellular tight junctions, Am. J. Physiol. 262 (1992) L647–L661. [10] E.E. Schneeberger, R.D. Lynch, The tight junction: a multifunctional complex, Am. J. Physiol. Cell. Physiol. 286 (2004) C1213–C1228.
[11] C. Van Itallie, C. Rahner, J.M. Anderson, Regulated expression of claudin-4 decreases paracellular conductance through a selective decrease in sodium permeability, J. Clin. Invest. 107 (2001) 1319–1327. [12] C.M. Van Itallie, J.M. Anderson, The molecular physiology of tight junction pores, Physiology (Bethesda) 19 (2004) 331–338. [13] D. Virgintino, M. Errede, D. Robertson, C. Capobianco, F. Girolamo, A. Vimercati, M. Bertossi, L. Roncali, Immunolocalization of tight junction proteins in the adult and developing human brain, Histochem. Cell. Biol. 122 (2004) 51–59. [14] H. Wolburg, A. Lippoldt, Tight junctions of the blood–brain barrier: development, composition and regulation, Vascul. Pharmacol. 38 (2002) 323–337.
Neuroscience Letters 428 (2007) 88–92
The tight junction component protein, claudin-4, is expressed by enteric neurons in the rat distal colon Shin-ichiro Karaki, Izumi Kaji, Yasuko Otomo, Hideaki Tazoe, Atsukazu Kuwahara ∗ Laboratory of Physiology, Institute for Environmental Sciences, and Graduate School of Nutritional and Environmental Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan Received 25 July 2007; received in revised form 11 September 2007; accepted 16 September 2007
Abstract The expression of a tight junction (TJ) component protein, claudin-4, in the enteric neurons was investigated in the rat distal colon by immunohistochemistry and RT-PCR. Claudin-4 immunoreactivity was detected in almost all neurofilament-positive enteric neurons both of the submucosal and the myenteric plexuses, and both of the cell bodies and the neurofibers. The immunoreactivity of enteric neurons for claudin-4 was divided into two types: strongly and weakly positive neurons. Especially in the myenteric plexus, the stained neurons were classified by Dogiel’s morphological classification of enteric neurons. The strongly stained claudin-4 positive neurons show Dogiel type II morphology, while the weakly stained claudin-4 positive neurons show Dogiel type I morphology. These immunohistochemical data were supported by mRNA expression in the muscle plus submucosa preparation containing the submucosal and myenteric plexuses, as well as mucosa preparation. The physiological function of claudin-4 expressed on enteric neurons is unclear up to now. It is however suggested that claudin-4 expressed on enteric neurons might play roles for the neural activity, for example as insulation between neurofibers. In conclusion, the present study clearly shows that claudin-4 is expressed by enteric neurons. This is the first evidence that the neuron itself expresses the TJ component protein, claudin-4, in the nervous system. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Tight junction; Claudin; Enteric nervous system; Neurofilament; Immunohistochemistry; RT-PCR
Tight junction (TJ) is one of the intercellular junctional structures forming the reticular-paired strands to separate the partition of the body fluid compartments especially in the epithelial tissues (e.g. skin, renal tubules, and gastrointestinal (GI) epithelia) and the endothelial tissues (e.g. blood vessels). The most primary functions of TJs are considered to prevent from passing fluid, ions and some molecules paracellularly as a barrier. However, TJs not only prevent from passing but also TJs expressed in diverse tissues are known to function for selective pores (or channels) to pass diverse ions and molecules, etc. [9] Therefore, the investigation of TJs is currently considered to be important for both basic and clinical aspects [2,10,12], for example, drug delivery system [4,8]. In the nervous system, the TJ-like structures have previously been observed both in central nervous system (CNS) and peripheral nervous system (PNS). In the CNS, TJs of the brain endothelial cells is reported to function as the blood–brain barrier [13,14].
∗
Corresponding author. Tel.: +81 54 264 5707; fax: +81 54 264 5707. E-mail address: [email protected] (A. Kuwahara). URL: http://physiology.u-shizuoka-ken.ac.jp/ (A. Kuwahara).
0304-3940/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.neulet.2007.09.059
The TJ strands are composed of the assemblies of cell–cell interlinked transmembrane proteins including claudins and occludin, and cytoskeleton-anchoring intracellular proteins including ZO-1–ZO-3, etc. (for review as [3]). Claudins are four-transmembrane ∼23 kDa proteins, and to date, 24 isoforms (claudin-1–claudin-24) are identified [2]. The distinct claudins are expressed by different cells, and are considered to have different properties. For example, the claudin-4 expression is reported to decrease paracellular conductance through a selective decrease in sodium (Na+ ) permeability [11], while claudin-2 expression is reported to increase cation permeability [1]. The claudin-11 and claudin-19 have been reported to be expressed by myelinating oligodendrocytes and Schwann cells in CNS [6] and PNS [5], respectively. However, there is no report for the expression of any claudins in the neuron itself both in CNS and PNS. At first, we have designed to investigate the expression and the distribution of TJ proteins including claudin-1–claudin-4, occludin, and ZO-1 in the epithelia of the rat distal colon during the course of physiological study. At that time, we found that the TJ proteins were expressed by epithelium in specific distribution (data not shown) similar to the previous report [7]. In
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
addition to the previous findings, we found out the claudin-4immunoreactive neurons in the enteric nervous system (ENS). Furthermore, we have checked the immunoreactivities for the other TJ proteins (claudin-1–claudin-3, occludin and ZO-1), but we did not find the expression of these TJ proteins in the ENS. Therefore, in the present study, we have investigated the expression and the distribution of claudin-4 immunoreactive neurons in the ENS in more detail. In the present study, rat distal colon was taken from male Wister rats (SLC; Hamamatsu, Shizuoka, Japan; or Chars River Laboratories Japan, Yokohama, Japan) after ether anesthetization followed by guillotine decapitation. The handling and the sacrifice of animals were in accordance with the Guidelines for the care and use of laboratory animals of the University of Shizuoka. In immunohistochemistry, the following three kinds of preparations were used: (1) 10-m thin frozen-sections, and whole mount preparations of (2) submucosa containing submucosal plexus (SMP) and (3) myenteric plexus on the longitudinal muscle layer (MP-LM). For preparing the frozen-sections, the flesh segments of distal colon were rapidly frozen with optimal cutting temperature (OCT) compound (TissueTek, Sakura Finetechnical, Tokyo, Japan) in liquid nitrogen, cut into 10-m thin sections in a cryostat. The frozen sections stuck on glass-slides were dried by cool blowing, and fixed in cold methanol (MeOH) at −30 ◦ C for 10 min. While, for preparing whole-mount preparations, the flesh segments were immersed in phosphate buffered saline (PBS) containing 10−6 M nicardipine (PBSN) at 4 ◦ C, cut open into the flat sheet, and pined with mucosa-up on the siliconrubber filled Petri dish in ice-cold PBSN. Mucosa was removed by scraping by using a blade, and the submucosa was exfoliated by using a fine tweezers. Furthermore, circular muscle was removed along the fiber to make the MP-LM preparations. The SMP and MP-LM preparations were pinned on the siliconrubber filled Petri dish, and fixed in MeOH at −30 ◦ C for 30 min. The fixed whole-mount preparations were stored in PBS containing 0.1% sodium azide (NaN3 ) at 4 ◦ C until use. Before incubation with primary antibodies, sections and whole-mount preparations were incubated with 10% normal donkey serum, 1% Triton X-100 and 0.1% NaN3 in PBS at room temperature for 1 h to prevent nonspecific immunoreaction. The preblocked samples were incubated with primary antibodies (Table 1) overnight at 4 ◦ C, and washed in PBS, 3 × 10 min. Then, the samples were incubated with secondary antibodies (Table 1) for 1 h at room temperature, washed in PBS, 3 × 10 min, and mounted on glass-slides in case of whole-mount preparations. Negative control experiment using each secondary antibody have previously been performed, resulting no staining was confirmed (data not shown). The samples were cover-slipped with a mounting medium (DakoCytomation, Glostrup, Denmark), and observed by a fluorescence microscope (IX70, Olympus, Tokyo, Japan). The fluorescence images were captured by a cooled chargecoupled device digital camera system (AxioVision 135; Zeiss, Munich-Hallbergmoos, Germany). For reverse transcriptase/polymerase chain reaction (RTPCR) analysis, the scraped-mucosa from the cut-opened flat sheets of the rat distal colon, and the smooth muscle
89
Table 1 Antibodies Primary antibody
Host
Dilution
Source
Anti-claudin-4 Anti-NFsa Anti-GFAP
Goat Rabbit Mouse
1:500 1:5000 1:2000
Santa Cruz (sc-17664) BIOMOL (NA1297) CHEMICON (MAB360)
Secondary antibody
Label
Dilution
Source
Anti-goat IgG
Alexa Fluor 488
1:200
Anti-rabbit IgG
Alexa Fluor 594
1:200
Anti-mouse IgG
Alexa Fluor 594
1:200
Molecular probes (A21203) Molecular probes (A21207) Molecular probes (A11058)
a This antibody is a rabbit polyclonal antiserum cocktail to three neurofilaments (NFs) including NF-L (62 kDa), NF-M (102 kDa) and NF-H (110 kDa).
with submucosal layer completely without mucosa were immersed immediately in a ribonucleic acid (RNA) stabilization regent (RNAlater; Qiagen K.K., Tokyo, Japan). Tissue samples were freeze-ground by using a grinding mill (SK-100; Tokken, Kashiwa, Chiba, Japan). Total RNAs were isolated by RNeasy Micro Kit (Qiagen K.K.). “Deoxyribonuclease (RT Grade) for Heat Stop” (Nippon Gene Co. Ltd., Tokyo, Japan) was used to remove the genome deoxynucleic acid (DNA), and the complementary DNA (cDNA) was synthesized by using first Strand cDNA Synthesis Kit for RT-PCR (AMV) (Roche Applied Science, Mannheim, Germany). The synthesized cDNAs were stored at −30 ◦ C until use. Primers for RT-PCR of claudin-4 were based on the rat claudin-4 messenger RNA (mRNA) sequence (GenBank: NM 001012022): 5 -CTCTCGCCTCCACGTTACTC-3 (forward) and 5 -AGGGTAGGTGGGTGGGTAAG-3 (reverse)
Fig. 1. Immunoreactivity for claudin-4 in the 10-m thin sections of rat distal colon. Claudin-4 immunoreactivity was observed in surface epithelium, submucosal plexus (arrow) and myenteric plexus (arrowhead). Bar = 100 m. Abbreviations: CM, circular muscle layer; LM, longitudinal muscle layer; MM, muscularis mucosa; muc, mucosa; SM, submucosa. Inset: magnification of the claudin-4 immunoreactive myenteric neurons. Arrow in the inset picture indicates axon-like staining. Bar = 50 m.
90
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
(estimated amplicon size: 190 bp). PCR of -actin was performed to provide a control using the following primers: 5 -GACTACCTCATGAAGATCCT-3 (forward) and 5 (reverse) (estimated CCACATCTGCTGGAAGGTGG-3 amplicon size: 512 bp). PCR was performed by using a RT-PCR kit (Ready-To-Go RT-PCR Beads; Amersham plc, Buckinghamshire, UK), and Takara PCR Thermal Cycler MP (TP3000, TaKaRa Bio Inc., Otsu, Shiga, Japan). Before PCR, reaction mixtures were incubated at 94 ◦ C for 5 min to completely denature the template. PCR cycles consisted of denaturing at 94 ◦ C for 1 min, annealing at 60 ◦ C for 1 min, and extension at 72 ◦ C for 1 min. The reactions were repeated for 35 cycles, followed by extension at 72 ◦ C for 7 min. Finally, amplification products were stored at 4 ◦ C. The amplification products and a DNA size marker (GenRuler 100 bp DNA Ladder Plus; Fermentas, Burlington, Ontario, Canada) were separated by electrophoresis on 1.5% agarose gel in 0.5× TRIS–borate–EDTA buffer containing ethidium bromide. Bands of the amplification products were viewed by ultraviolet light, and the images were taken by GelDoc2000 (Bio-Rad Laboratories, Hercules, CA, USA). Fig. 1 shows the immunoreactivity for claudin-4 in the rat distal colon. Claudin-4-immunoreactivity was observed in the surface epithelium of mucosa as well as the previous reports [7]. Furthermore, we have observed additional claudin-4 immunoreactivity in the ENS of distal colon; the myenteric and submucosal plexuses showed stronger claudin-4-immunoreactivity than that of epithelium (Fig. 1). In higher magnification, claudin-4immunoreactivity was observed as fiber-like morphologies in
both nerve cell soma and axons (Fig. 1, inset). Thus, we performed double staining for claudin-4 and neurofilaments (NFs) or glial filament acidic protein (GFAP). The results showed that the claudin-4 was completely colocalized with NFs (Fig. 2), whereas there was no colocalization between claudin4 and GFAP (Fig. 3). To confirm the results, we tested the following combination of primary and secondary antibodies: rabbit anti-NF + anti-goat IgG Alexa Fluor488, and goat anticaludin-4 + anti-rabbit IgG Alexa Fluor594. The results indicate both negative (data not shown). These indicate that claudin4 expressed by enteric neurons probably connects NFs, but not by enteric glia. For the morphology of enteric neurons, claudin-4-immunoreactive neurons were roughly divided into two groups; one was strongly stained and another was weakly stained (Fig. 2). The strongly stained claudin-4 positive neurons showed the Dogiel type II morphology, that is, smooth cell soma and multiple axonal processes. On the other hand, all Dogiel type I neurons showing short and lamellar dendrites, and single axon were weakly stained for claudin-4. The strongly stained claudin-4 positive neurons were also strongly stained by NFs, and the weakly stained claudin-4 positive neurons were also weakly stained by NFs. This result suggests that the expression of claudin-4 is connected with the expression of NFs. To support the immunohistochemical data of claudin-4 expression in enteric neurons, the expression of claudin-4 mRNA was investigated by RT-PCR analysis. The amplicon of detected bands (near 200 bp for claudin-4 and near 500 bp for -actin) were consistent with each estimated amplicon sizes (190 and 512 bp, respectively). The results showed that claudin-
Fig. 2. Immunoreactivity for claudin-4 and NFs in the submucosal and myenteric plexuses of rat distal colon. Claudin-4- and NF-immunoreactivities were observed in the whole-mount preparations of submucosal (A–C) and myenteric (D–F) plexuses. Claudin-4 immunoreactivity was visualized by the Alexa488-conjugated secondary antibody (A and D, green), and NF-immunoreactivity was by the Alexa594 (B and E, red). Merged images were also shown in D and F, and the yellow indicates colocalization. Numbers in panel F respectively indicates the following characteristics: (1 and 2) representative Dogiel type II neurons, i.e., smooth cell bodies and multiple axons; (3) smooth cell body like Dogiel type II, but only single axon; (4) Dogiel type II morphology, but lower positive in both claudin-4 and NFs than 1–3 neurons; (5–7) Dogiel type I-like neurons, i.e., short lamellar dendrites and single axon; (8) looking like Dogiel type I, but difficult to identify definitely; (9) very lower positive or negative neurons for both claudin-4 and NFs. All bars indicate 50 m.
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
91
Fig. 3. Immunoreactivity for claudin-4 and GFAP of the submucosal and myenteric plexuses. Claudin-4- and GFAP-immunoreactivities were observed in the whole-mount preparations of submucosal (A–C) and myenteric (D–F) plexuses. Claudin-4-immunoreactivity was visualized by the Alexa488-conjugated secondary antibody (A and D, green), and GFAP-immunoreactivity was by the Alexa594 (B and E, red). Merged images were also shown in D and F, but there were no the yellow as colocalization. All bars indicate 50 m.
4 mRNA was detected in both the scraped-mucosa and the muscle plus submucosa samples (Fig. 4). The detection of claudin-4 mRNA in the muscle plus submucosa samples including myenteric and submucosal neurons supports the present immunohistochemical data. What is the physiological function of claudin-4 expressed by enteric neurons? The claudin-4 expression is reported to decrease paracellular conductance through a selective decrease in sodium (Na+ ) permeability, already mentioned above [11]. Thus, one possibility is considered that they work as insulation between each nerve fiber to prevent from signal crossing. However, claudin-4 was expressed also in the cytoplasm of enteric neurons. It is unclear that the claudin-4 expressed in cytoplasm has some functions there, or only stands by at cell bodies for being transported to axons or dendrites. To date, there is no
evidence about the function of claudin-4 expressed in enteric neurons. Therefore, further study needs to perform in order to clarify the function of claudin-4. In conclusion, the present study indicates that claudin-4 is expressed by enteric neurons, but the function of claudin-4 expressed by enteric neurons is unclear yet. However, this is the first evidence that the neuron itself expresses the TJ component protein, claudin-4 in the nervous system. Acknowledgements This work was partly supported by a Japan Society for the Promotion of Science (no. 18590207) and Smoking Research Foundation to A. Kuwahara; and Promotion of Health and Nutrition from the Danone Institute to S. Karaki. References
Fig. 4. Analysis of expression of claudin-4 mRNA in the rat distal colon by RTPCR. The amplification products of RT-PCR for claudin-4 (lanes 3 and 5) and actin (lanes 2 and 4), and DNA maker (lane 1) were separated by electrophoresis containing ethidium bromide, and viewed by UV light. Bands for Claudin-4 and -actin were detected in both mucosa and muscle plus submucosal samples at near 200 and 500 bp, respectively. RT-PCR for -actin was used as a control.
[1] S. Amasheh, N. Meiri, A.H. Gitter, T. Schoneberg, J. Mankertz, J.D. Schulzke, M. Fromm, Claudin-2 expression induces cation-selective channels in tight junctions of epithelial cells, J. Cell. Sci. 115 (2002) 4969– 4976. [2] M. Furuse, S. Tsukita, Claudins in occluding junctions of humans and flies, Trends Cell. Biol. 16 (2006) 181–188. [3] L. Gonzalez Mariscal, A. Betanzos, P. Nava, B.E. Jaramillo, Tight junction proteins, Prog. Biophys. Mol. Biol. 81 (2003) 1–44. [4] M. Kondoh, A. Takahashi, M. Fujii, K. Yagi, Y. Watanabe, A novel strategy for a drug delivery system using a claudin modulator, Biol. Pharm. Bull. 29 (2006) 1783–1789. [5] T. Miyamoto, K. Morita, D. Takemoto, K. Takeuchi, Y. Kitano, T. Miyakawa, K. Nakayama, Y. Okamura, H. Sasaki, Y. Miyachi, M. Furuse, S. Tsukita, Tight junctions in Schwann cells of peripheral myelinated axons: a lesson from claudin-19-deficient mice, J. Cell. Biol. 169 (2005) 527– 538.
92
S.-i. Karaki et al. / Neuroscience Letters 428 (2007) 88–92
[6] K. Morita, H. Sasaki, K. Fujimoto, M. Furuse, S. Tsukita, Claudin-11/OSPbased tight junctions of myelin sheaths in brain and Sertoli cells in testis, J. Cell. Biol. 145 (1999) 579–588. [7] C. Rahner, L.L. Mitic, J.M. Anderson, Heterogeneity in expression and subcellular localization of claudins 2, 3, 4, and 5 in the rat liver, pancreas, and gut, Gastroenterology 120 (2001) 411–422. [8] N.N. Salama, N.D. Eddington, A. Fasano, Tight junction modulation and its relationship to drug delivery, Adv. Drug Deliv. Rev. 58 (2006) 15–28. [9] E.E. Schneeberger, R.D. Lynch, Structure, function, and regulation of cellular tight junctions, Am. J. Physiol. 262 (1992) L647–L661. [10] E.E. Schneeberger, R.D. Lynch, The tight junction: a multifunctional complex, Am. J. Physiol. Cell. Physiol. 286 (2004) C1213–C1228.
[11] C. Van Itallie, C. Rahner, J.M. Anderson, Regulated expression of claudin-4 decreases paracellular conductance through a selective decrease in sodium permeability, J. Clin. Invest. 107 (2001) 1319–1327. [12] C.M. Van Itallie, J.M. Anderson, The molecular physiology of tight junction pores, Physiology (Bethesda) 19 (2004) 331–338. [13] D. Virgintino, M. Errede, D. Robertson, C. Capobianco, F. Girolamo, A. Vimercati, M. Bertossi, L. Roncali, Immunolocalization of tight junction proteins in the adult and developing human brain, Histochem. Cell. Biol. 122 (2004) 51–59. [14] H. Wolburg, A. Lippoldt, Tight junctions of the blood–brain barrier: development, composition and regulation, Vascul. Pharmacol. 38 (2002) 323–337.