2024 Volume 57 Issue 6 Pages 189-197
Cell-to-cell communications are desirable for efficient functioning in endocrine cells. Gap junctions and paracrine factors are major mechanisms by which neighboring endocrine cells communicate with each other. The current experiment was undertaken to morphologically examine gap junction expression and developmental changes in rat adrenal medullary chromaffin (AMC) cells. The expression of connexin 43 (Cx43) was conspicuous in the rat adrenal cortex, but not detected immunohistochemically in neonatal or adult AMC cells. Consistent with the morphological findings, the phosphorylated and non-phosphorylated forms of Cx43 were predominantly and faintly detected by immunoblotting in the adrenal cortical and medullary homogenates, respectively. In contrast to Cx43, Cx36-like immunoreactive (IR) material was detected in neonatal AMC cells, a fraction of which were in the process of migration to the center of the adrenal gland, but this was not seen in adult AMC cells. The current results raise the possibility that the mechanism for cell-to-cell communication changes in a developmental manner in rat AMC cells.
Endocrine cells secrete hormones into the blood stream in response to humoral and/or neuronal inputs, and a high degree of synchrony between cells is desirable for efficient functioning. Thus, gap junctions generally develop as a mechanism for cell-to-cell communication in endocrine cells [4, 40]. The adrenal gland, one of the endocrine organs, consists of the adrenal cortex, which secretes steroid hormones, and the adrenal medulla, which secretes catecholamines. Adrenal cortical cells have been functionally and morphologically demonstrated to express gap junctions [40], which are assembles of 12 connexins, such as connexin 43 (Cx43) [29]. On the other hand, the expression of gap junctions in adrenal medullary chromaffin (AMC) cells seems to depend upon species, sex, and age [7, 14, 15, 33]. Adult mouse AMC cells were immunohistochemically shown to express Cx36 [10] and Cx43 [39] and found to be electrically or dye coupled between neighboring cells [20]. Furthermore, freeze fracture electron microscopy revealed the expression of gap junctions in mouse AMC cells [14]. On the other hand, the expression of gap junctions in adult rat AMC cells is not straight. Although electrical or dye coupling has been reported to occur between adult rat AMC cells [32], freeze fracture electron microscopy did not reveal the expression of gap junctions in adult rat AMC cells [14]. In addition, immunohistochemistry showed that Cx43 is dominantly expressed in the rat adrenal cortex, but not in the adrenal medulla [38]. These morphological findings suggest that gap junctions are not expressed in adult rat AMC cells. The absence of expression might be understandable given that the synaptic connection with preganglionic fibers is well developed in adult rat AMC cells [47]. The excitation of multicellular AMC cell networks will occur through synaptic input to each of them without the involvement of gap junction-mediated communication [31].
Rat AMC cells before birth secrete catecholamines in response to humoral factors, such as a decrease in blood O2 tension [43]. Innervation by preganglionic sympathetic nerve fibers in rat AMC cells is not completed before the first 7 days of life [38]. Thus, catecholamine secretion in response to hypoxia results from the detection of hypoxia by AMC cells in the perinatal period [12, 43]. The perinatal AMC cells are assumed to function, as adrenal cortical cells do in response to humoral inputs. Thus, gap junctions might be expressed in rat AMC cells before the completion of innervation. The aim of the current experiment is to elucidate whether or not Cx43 and Cx36, typical isoforms of gap junction proteins [41], are expressed in rat AMC cells and gap junction expression changes in a developmental manner. If gap junction expression changes developmentally, we will discuss the physiological implication of the change.
Eleven male Wistar rats weighing 150–300 g and three pregnant Wistar rats were obtained from Kyudo Company (Tosu, Japan). Rats were housed in standard cages with free access to food and water. Animals were kept under a light-dark cycle of 12 hr light and 12 hr darkness. All procedures for the care and treatment of animals were carried out according to the Japanese Act on the Proper Conduct of Animal Experiments issued by the Science Council of Japan. The experimental protocols were approved by the Institutional Animal Care and Use Committee of the University of Occupational and Environmental Health (AE15-014). All efforts were made to minimize the potential for pain, stress, or distress.
ImmunohistochemistryAdult rats were anaesthetized with sodium pentobarbital (50 mg Kg−1 i.p.), with 30 ml of saline perfused through the ascending aorta, followed by 250 ml of 4% paraformaldehyde in phosphate-buffered saline (PBS: pH 7.2), and newborn rats were killed by cervical dislocation. The adrenal glands were removed from the rats and fixed with 4% paraformaldehyde in PBS overnight at 4°C. After fixation and rinsing in PBS, they were dehydrated through a graded ethanol series and embedded in O.C.T. Compound (Tissue-Tek: Sakura Finetek, Tokyo, Japan) at the temperature of liquid nitrogen. Thin sections 5 μm in thickness were obtained with a microtome/cryostat (Tissue-Tek Polar: Sakura Finetek) at −20°C, mounted on a glass slide (MAS-coated Superfrost: Matsunami, Kishiwada, Japan), and dried for 1 hr at room temperature. Next, the sections were rinsed for 10 min in PBS. After treatment with Protein Block Serum-Free (Agilent, Santa Clara, CA, U.S.A.) for 60 min to reduce non-specific binding, the sections were incubated overnight with primary antibodies (Abs), and then secondary Abs conjugated with Alexa Fluor 488 or 546 (Molecular Probes, Eugene, OR, USA). The immunostaining was observed with a confocal laser scanning microscope (LSM5 Pascal: Carl Zeiss, Tokyo, Japan). Excitation wavelength and emission filters were a 488 nm laser and 510–560 nm filter for Alexa Fluor 488 (FITC-like fluorescence) and a 543 nm laser and 560 nm long-pass filter for Alexa Fluor 546 (rhodamine-like fluorescence). To identify Cx43, a rabbit anti-Cx43 Ab at a dilution of 1 : 400 (c6219: Sigma-Aldrich, Tokyo, Japan) was used, whereas for the identification of Cx36 and chromogranin A (CgA) a rabbit anti-Cx36 Ab at a dilution of 1 : 100 (36-4600: Thermo Fisher Scientific, Waltham, MA, USA) and goat anti-CgA Ab at a dilution of 1 : 50 (sc-1488: Santa Cruz Biotechnology, Santa Cruz, USA) [34] were used, respectively. The specificity of the anti-Cx43 and anti-Cx36 Abs for immunoblotting and/or immunohistochemistry has well been documented [26, 30, 44]. Some of the sections were treated with 4',6'-diamidino-2-phenylindole dihydrochloride (DAPI) at a dilution of 1 : 10000 (Thermo Fisher Scientific) to stain the nucleus, and the fluorescence was visualized using a fluorescence microscope equipped with a UV light source and a blue/cyan filter (~460 nm). For the diaminobenzidine (DAB) reaction, endogenous peroxidase activity was inhibited by pre-treatment with 0.1% hydrogen peroxidase in methanol for 20 min. The immunoreaction was examined with the indirect immunoperoxidase method (Histofine Simple Stain Max-PO: Nichirei, Tokyo, Japan). The peroxidase complex was visualized by treatment with a freshly prepared solution of DAB tetrahydrochloride (DAB substrate kit: Nichirei), and some sections were counterstained with hematoxylin. The sections were air-dried, and mounted with a cover glass.
ImmunoblottingAdult rats were killed by cervical dislocation, and the heart and the adrenal glands were excised and immediately put into ice-cold Ca2+-deficient balanced salt solution in which 1.8 mM CaCl2 was simply omitted from standard saline. The standard saline contained 137 mM NaCl, 5.4 mM KCl, 1.8 mM CaCl2, 0.5 mM MgCl2, 0.53 mM NaH2PO4, 5 mM D-glucose, and 5 mM HEPES, and the pH of the solution was adjusted to 7.4 with 4 mM NaOH. The adrenal cortex was removed from the adrenal gland using microscissors and forceps under stereoscopic observations. The preparations were minced and homogenized with a Potter-Elvehjem homogenizer in 10 volumes of a solution containing 10 mM Tris-HCl (pH 7.4), 150 mM NaCl, and a proteinase inhibitor cocktail (set 1: Calbiochem, San Diego, CA, USA). Homogenates were centrifuged at 500 g for 10 min at 4°C to remove the nuclei, then the post-nucleus supernatants were mixed with equal volumes of an SDS buffer containing 25 mM Tris-HCl (pH 6.8), 4% SDS and 20% glycerol. Protein concentrations in samples were determined using a BSA protein assay kit (Pierce, Rockford, IL, USA). After the addition of 2-mercaptoethanol (final concentration, 5% v/v) and bromophenol blue (0.05% v/v) to the sample, proteins were separated by 10% (w/v) SDS-PAGE, and then transferred to a polyvinylidene difluoride (PVDF) membrane. The membrane was blocked with 5% (w/v) fat-free powdered milk dissolved in PBS-T solution, which contained 2 mM NaH2PO4, 145 mM NaCl, and 0.1% Tween 20. The PVDF membrane was incubated with an anti-Cx43 Ab or anti-actin Ab (MAB1501: Sigma-Aldrich). The immunoreaction was detected by incubating the membrane with a secondary Ab linked to horseradish peroxidase (Amersham, Buckinghamshire, UK), and then with ECL-Plus (Amersham). Alkaline phosphatase treatments were performed by adding SDS (final concentration, 0.2% (w/v)) to the post-nuclear homogenate of the rat heart and treating the sample with 100 units ml−1 alkaline phosphatase at 37°C for the indicated durations [44].
Cell cultureMIN6 β cells [37] were cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Gibco: Life Technologies, Tokyo, Japan) supplemented with 10% fetal bovine serum (FBS) (172012: Sigma-Aldrich). The cells were fixed with 4% paraformaldehyde in PBS for 30 min at room temperature. After washing three times with PBS, the cells were incubated in PBS containing 0.1% Triton X-100 for 30 min and then with PBS containing 1% FBS for 1 hr at room temperature. The cells were treated with a rabbit anti-Cx36 Ab and then with anti-rabbit IgG Ab conjugated with Alexa Fluor 543. The coverslips were mounted in 50% glycerol containing 1 mg ml−1 4-diaminobenzene.
Sources of reagentsMIN6 β cell line was provided by Dr Miyazaki (Osaka University, Osaka, Japan); alkaline phosphatase was obtained from Takara Bio Inc. (Kusatsu, Japan).
StatisticsAll statistical analyses were performed with Sigma Plot (v13.0: Systat Software, San Jose, CA, USA). The data are represented in means ± S.E.M. When data showed a normal distribution (Shapiro-Wilk test), the statistical significance of differences was determined by the unpaired Student’s t-test. Otherwise, the Mann-Whitney rank sum test was used.
The fact that synaptogenesis is not functionally completed before the first 7 days of life [43] suggests the possibility that the adrenal medulla does not fully develop at birth. To explore this possibility, AMC cells in the adrenal glands of developing rats were immunohistochemically identified (Fig. 1A). CgA, which is a major protein constituent in chromaffin granules or large-dense core vesicles (LDCVs) [19, 49], is focused as a marker protein for AMC cells. To our surprise, CgA-like immunoreactive (IR) material was located not only in the core of the adrenal gland, but also in the surrounding layer in one-day old rats (Fig. 1A). The amount of CgA-like IR material in the surrounding layer was markedly diminished in five-day old glands, and almost all CgA-like IR material in two-week old adrenal glands was located in the core, as is also the case with adult adrenal glands. In addition, the fraction of AMC cells in the adrenal gland tended to decrease while the size of the adrenal gland markedly increased from the neonatal period to adult life (Fig. 1, B and C). The results indicate that AMC cells in the adrenal gland is migrating to the core during the first week of life.
Developmental change in immunostaining for chromogranin A in rat adrenal glands. (A) Immunohistochemical staining for chromogranin A in adrenal gland sections. Right, enlargement of the square shown in left. Thin slices of adrenal glands of indicated ages (D: day; W: week) were subjected to immunohistochemistry for chromogranin A. Immunoreaction was visualized as DAB reaction. The sections were counterstained with hematoxylin. (B and C) Summaries of fractions (%) of Adrenal medullary chromaffin (AMC) cells in the adrenal gland and areas of the adrenal gland, respectively. AMC cells were immunohistochemically identified as chromogranin A-positive cells. Data represent means ± SEM of 4 slices for each newborn and adult rat. Newborn rats include 1- and 5-day old rats; adult rats represent 8 to 12-week-old rats.
The functional analysis of cell-to-cell communication in the rat adrenal medulla suggested that the number of gap junctions decreases during development [33], and the expression of Cx43 and Cx36 at the mRNA level was reported in 20–25% of adult AMC cells [33]. Thus, expression at the protein level was first explored in newborn and adult rat adrenal glands. Immunohistochemistry clearly showed that Cx43-like IR material was absent in the core of one- and seven-day-old adrenal gland slices (n = 5 and n = 5, respectively) and conspicuous in the surrounding layer (Fig. 2A). In addition, the immunostaining of adult adrenal glands showed that immunoreactivity for Cx43 was absent in the medulla and conspicuous in the cortex in adrenal slices (n = 5) (Fig. 2B). The detailed inspection of stain images revealed that immunostaining decreased in intensity toward the putative zona glomerulosa [9, 40], which apparently lacked immunoreactivity.
Immunostaining and immunoblotting for connexin 43. (A) Immunostaining for connexin 43 (Cx43) of adrenal slices of 1- and 7-day old rats. Image in the 2nd row represents the enlargement of a square in the 1st row. (B) Immunostaining for Cx43 of an adrenal slice of the adult rat. The immunoreaction was visualized as DAB reaction. (C) Immunoblots of rat adult heart (H), adrenal cortex (AC) and adrenal medulla (AM) homogenates for Cx43 and actin. The same amount of proteins (5 μg) was loaded in each lane. (D) Immunoblot for Cx43 of heart homogenates treated (AP(+)) or not treated (AP(−)) with alkaline phosphatase. AP(+) denotes the preparations treated with alkaline phosphatase for the indicated periods (min) at 36°C, whereas AP(−) denotes the preparation treated for 50 min at 36°C without alkaline phosphatase. Arrows indicate three forms of Cx43.
The expression of Cx43 was further examined with immunoblotting. Cx43 is known to be phosphorylated at multiple sites [27–29]. In fact, three bands with apparent molecular weights of 39, 43, and 46 kDa were detected in rat heart homogenates by immunoblotting for Cx43 (Fig. 2C, D). The treatment of the homogenates with alkaline phosphatase led to successive decreases in the 43 and 46 kDa bands and commemorate increases in the 39 kDa band, and only the 39 kDa band was detected in the homogenates treated for 50 min (Fig. 2D). The 39 kDa band and bands with the higher apparent molecular weights constituted 24.4 ± 2.2% (n = 6) and 75.6 ± 2.2% (n = 6) of the total amount of the recognized bands. In contrast to the heart homogenates, the 39 and 43 kDa bands were recognized in homogenates of the adrenal cortex with the 39 kDa band comprising 68.8 ± 2.7% (n = 6) of the total amount. These two bands were also faintly recognized in homogenates of the adrenal medulla. The total amount of the two bands in the adrenal medulla was significantly smaller (P < 0.05) than that in the adrenal cortex, being just 35.0 ± 5.0% (n = 6) of the latter. The 39 kDa band represent 61.8 ± 1.0% (n = 6) of the total amount, a value which did not differ from that of the adrenal cortex (P = 0.065).
Immunohistochemistry for Cx36As Cx36 was detected in rat and mouse AMC cells at the mRNA [32] and protein levels [10], respectively, its expression was immunohistochemically explored in newborn and adult adrenal medullae. First, the immunoreactivity of a rabbit anti-Cx36 Ab was examined in MIN6 cells, which have been known to express Cx36 at the plasma membrane [6]. As shown in Fig. 3A, Cx36-like IR material was mainly located at the cell periphery, suggesting that the Ab recognizes Cx36 in immunostaining. Thus, the indirect immunoperoxidase method with the Ab was used to examine the expression of Cx36 in the newborn adrenal gland. Compared with a control slice for staining (Fig. 3Ba), the slice treated with the Ab exhibited DAB reaction product at the periphery of the adrenal gland, near the periphery (arrow in Fig. 3Bb), and in the core (arrowhead in Fig. 3Bb). This notion was confirmed by the line scan analysis of DAB reaction intensity (Fig. 3C). The detailed inspection of the periphery (Fig. 3Bc) revealed that the DAB reaction product in the area near the periphery was distributed in a reticular manner, whereas that at the uttermost layer was homogenous.
Immunostaining for Cx36 of the rat adrenal gland. (A) Immunostaining of MIN6 β cells for Cx36. Left, confocal image of Cx36-like immunoreactive (IR) material; right, merge of confocal and differential interference contrast (DIC) images. Immunoreaction was visible as rhodamine-like fluorescence. (B) Immunostaining for Cx36 of an adrenal gland slice of 1-day old rat with the indirect immunoperoxidase method. a, control slice, where the treatment with an anti-Cx36 Ab was omitted; b, immunostaining with the anti-Cx36 Ab; c, enlargement of a square in b. The arrow and the arrowhead indicate the area inside the uttermost layer and the core with significant DAB reaction, respectively. (C) Line scan of DAB reaction intensity along the straight line in Bb. peri and core represent the area inside the uttermost layer (arrow in Bb) and that in the center of the adrenal gland (arrowhead). Intensity is expressed in arbitrary unit (au). (D) Double immunostaining of an adrenal gland slice of a 1-day-old rat for Cx36 and chromogranin A. a, confocal image of merge of Cx36- and chromogranin A-like fluorescence; b, enlargement of a square in a; c and d, confocal Cx36- and chromogranin A-like fluorescence images, respectively. Cx36- and chromogranin A-like IR material were visible as FITC- and rhodamine-like fluorescence, respectively. (E) Immunostaining for Cx36 of an adult rat adrenal gland slice with the indirect immunoperoxidase method. The slice was counterstained with hematoxylin. a, control slice, where the treatment with the anti-Cx36 Ab was omitted; b, immunostaining with the anti-Cx36 Ab.
The presence of Cx36-like IR material near the periphery raises the possible presence of Cx36 in the AMC cells migrating to the core. To explore this possibility, the newborn adrenal gland slice was subjected to double staining for Cx36 and CgA. As is evident in Fig. 3D, Cx36-like IR material was present only in the cells exhibiting immunoreactivity for CgA (5 slices). In contrast, adrenal cortical or AMC cells in adult adrenal slices (n = 5) exhibited no DAB reaction not only in the slices not treated with the anti-Cx36 Ab, but also in those treated with it (Fig. 3E). These immunohistochemical studies suggest the presence of gap junctions made up of Cx36 in newborn, but not adult, rat AMC cells.
Immunohistochemistry with anti-adrenaline (AD) and anti-noradrenaline (NA) Abs revealed that there are three phases for histogenesis of the rat adrenal medulla on the cellular level [48]: the first is synthesis and storage of almost exclusively NA; the second, synthesis and storage of NA and AD in a single cell; the third, synthesis and storage of NA and AD in separate cell types. The last phase occurs on and after the second/third postnatal day. The present immunostaining for CgA revealed that rat AMC cells are migrating to the center of the adrenal gland during the first week of life. To our knowledge, the present experiment is the first to demonstrate that a fraction of AMC cells are still in the process of migration to the core of the adrenal gland just after birth [25]. This fact is consistent with no neuronal transmission from the sympathetic preganglionic nerve fiber to the newborn rat AMC cell [43]. Because the synapse has been formed on newborn rat AMC cells [47], no synaptic transmission is thought to be due to the functional immaturity of the synapse. The present immunostaining for CgA suggests that immaturity of AMC cells is also responsible for the absence in synaptic transmission.
It would be worth noting that the fraction of AMC cells in the adrenal gland section decreased from 12% to 8% in growth from the newborn to adulthood, consistent with the previous results [8], whereas the size of the adrenal gland section increased by 10-fold. Thus, a proliferation rate of the adrenal cortical cells [40] is larger than that of the AMC cells after birth [46], and fibroblast growth factor-mediated signaling might be involved in such an increase in size of the adrenal cortex [13, 24, 35].
Cx43 expressionThe expression of Cx43 in adult rat AMC cells has been reported at the mRNA level [32], and a Cx43 band was detected in the homogenate of adult rat adrenal medullae [7]. In contrast to these findings, Cx43-like IR material was not observed in the core of the newborn and adult rat adrenal glands in the present immunohistochemical studies. Thus, it is apparently difficult that the present findings are reconciled with those in the reference. However, the expression at the mRNA level does not mean that at the protein level, as has been demonstrated for several channel proteins [3, 21]. In addition, the homogenate of rat adrenal medullae used in the reference could have been contaminated by the adrenal cortex. In the present homogenate of the adult adrenal cortex, 39 and 43 kDa bands were conspicuously detected with an anti-Cx43 Ab, whereas the both bands were faint in that of the adrenal medulla and the fraction of the 39 kDa band in the total amount of bands in the adrenal medulla was similar to that in the adrenal cortex. In contrast to the adrenal gland, three bands of 39, 43, and 46 kDa were detected in rat heart homogenates by immunoblotting. The treatment of the heart preparation with alkaline phosphatase resulted in the disappearance of the 43 and 46 kDa bands and the commemorate increase in the 39 kDa band. Based on these results, the 39, 43, and 46 kDa bands may correspond to the non-phosphorylated form P0 and the phosphorylated forms P1 and P2 of Cx43 identified in normal rat kidney (NRK) cells [44] and other cells [51]. Thus, the current experiments clearly show that the extent of phosphorylation in Cx43 differs between cardiac myocytes and adrenal cortical cells and the fraction of the non-phosphorylated form in the total amount of the recognized phosphorylated and non-phosphorylated Cx43 differs between the heart and the adrenal cortex. These results suggest that differences in humoral and/or neuronal inputs, which are involved in the phosphorylation of Cx43 proteins, are responsible for those in the phosphorylation event of Cx43. NA and adrenocorticotropin (ACTH) are the major factors for phosphorylation of Cx43 in the heart [42] and the adrenal cortex [28, 39, 42, 45, 51], respectively. Both β1 adrenoceptor and ACTH receptor are coupled with adenylate cyclase through a stimulating G protein (Gs), thus inducing phosphorylation of Cx43 through a cAMP-dependent pathway. On the other hand, if Cx43 was expressed and phosphorylated in rat AMC cells, ACh would be a candidate for mediating phosphorylation [16, 23]. ACh is expected to exert its effect through nicotinic and/or muscarinic receptors. However, none of the muscarinic receptor subtypes (M1 to M5) are coupled with adenylate cyclase through a Gs [11, 17]. Thus, it is likely that the presence of faint bands of 39 and 43 kDa in the adrenal medulla was due to contamination by the adrenal cortex. This notion is supported by the finding that the fraction of the non-phosphorylated form in the total amount of the Cx43 bands in the adrenal medulla was the same as that in the adrenal cortex and differed from that in the heart.
Cx36 expressionThe DAB reaction product was detected in the uttermost layer corresponding to the capsule, near the periphery, and at the core of the adrenal gland treated with an anti-Cx36 Ab. This DAB reaction minimally occurred in the adrenal slice not treated with the primary Ab, indicating that the DAB reaction product was due to the binding by the Ab. The detailed inspection of the staining clearly revealed that the DAB reaction product was distributed homogenously and in a reticular manner at the uttermost layer and the area inside the layer, respectively. As Cx36 is expected to be present at the interface between neighboring cells, the homogenous presence of the DAB reaction product in the uttermost layer of the adrenal gland suggests a nonspecific reaction. On the other hand, the reticular pattern of distribution of the DAB reaction product suggests its presence at the interface between neighboring cells. The line scan analysis of immunoreaction showed that DAB reaction occurred at the area near the periphery and in the core, suggesting the presence of DAB reaction product in AMC cells. This notion was confirmed by double immunostaining for Cx36 and CgA. Thus, the DAB reaction product in the area inside the uttermost layer may be located in AMC cells in the process of migration to the core. The treatment with the anti-Cx36 Ab did not produce any DAB reaction product either in the adrenal cortex or the adrenal medulla of adult rats. The results indicate that Cx36 is expressed in newborn rat AMC cells, but not in adult cells.
Interestingly, the expression of Cx36 in the rat spinal cord has been known to decrease rapidly following birth [5, 36]. This rapid diminution of Cx36 expression is ascribed to the suppression of gene transcription, which is due to synaptic transmission and the subsequent increase in the Ca2+-cyclic AMP response element binding protein activity [2]. Thus, a similar mechanism might be involved in the abolition of Cx36 expression in AMC cells after birth. In contrast to rat AMC cells, LacZ reporter gene analysis and immunohistochemistry revealed Cx36 expression in adult mouse AMC cells [10], which is compatible with the high incidence of synchronization of Ca2+ signal in neighboring mouse AMC cells [50]. Thus, Cx36 expression in AMC cells differs between species. However, what factor is responsible for the species difference in Cx36 expression remains an open question.
Functional implicationFinally, the functional significance of the maturation of AMC cells in the rat adrenal gland is worth considering. The findings that Cx36 expression occurred in neonatal, but not adult, rat AMC cells and that AMC cells continued to accumulate in the core of the adrenal gland during the first week of life raise the possibility that adult AMC cells communicate with each other via a mechanism other than gap junctions. One such mechanism would be cell-to-cell communication by paracrine factors. This group [18, 22, 34] and others [1] have reported that GABA is a putative paracrine factor in adult AMC cells. Recently, the expression of glutamic acid dehydrogenase (GAD), an enzyme involved in GABA synthesis, was proven to increase in mouse AMC cells after birth, whereas GABAA receptors were already expressed at birth [18]. These results suggest that the gap junction mechanism, which is mainly involved in cell-to-cell communication in the neonatal period, may be replaced with a paracrine mechanism involving GABA in adult AMC cells, which may function as a facilitatory factor by itself and as an inhibitory factor for neuronally evoked secretion [22, 34].
The authors have no conflict of interests to declare.
We appreciate Dr. J. Miyazaki (Osaka University, Osaka, Japan) for supplying the MIN6 β cell line. This study was supported in part by grants of JSPS KAKENHI (17K08555 to M. I. and 18K06865 to H. M.). T. N. and M. I. designed the study, performed the experiment, and analyzed the data; K.-Y. W., H. M., K. O., Y. E., and K. H. performed the experiment; M. I. wrote the manuscript; all the authors approved the final manuscript.
