Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Thyrotropin (thyroid-stimulating hormone [TSH]) is the primary factor that regulates thyroid follicular cell (thyrocyte) function and, ultimately, thyroid hormone secretion. In addition to the secretagogue action of thyroid hormones, TSH is known to increase the binding of growth factors such as epidermal growth factor (EGF) (1) and transforming growth factor (TGF)- (2) to their respective receptors to promote thyrocyte growth. Moreover, TSH appears to have targets other than thyrocytes, including human testicular cells, adrenal glands, adipocytes (3), lymphocytes (4, 5), rat fat cells (6), and intraepithelial T lymphocytes in mouse intestinal mucosa (7). Also, the source of TSH is not limited to the anterior pituitary gland; TSH can even be released from lymphocytes (5, 7) and the mucous epithelium (7) as well. However, the extrathyroidal roles of TSH are largely unknown, although its potential activities in regulating immune responses have been suggested (5, 7).

The airway submucosal gland secretes mucins, various enzyme proteins, and electrolytes (therefore, water), which serve as a nonspecific airway defense mechanism (10). Moreover, the submucosal gland secretes immunoglobulins (Igs) to neutralize and/or eliminate microbes or foreign substances. Plasma cells and other cells containing Igs in the human respiratory tract were found preferentially adjacent to the submucosal gland (11). The Igs released from plasma cells are, in turn, incorporated into and secreted through the glandular cells via secretory component (SC), a polymeric Ig (pIg)-binding segment of the pIg receptor (12, 13). However, the pathways regulating the airway humoral immunity are only partly understood (14). Recently, interferon (IFN)- was reported to upregulate a vectorial transcytosis of SC and/or dimeric IgA across a monolayer of Calu-3 cells (12, 15), a human lung- derived adenocarcinoma cell line (16). The Calu-3 cells seem to highly resemble those of tracheobronchial gland in the expression of a number of RNA transcripts (16). Interestingly, the plasma level of IFN- has been reported to be increased by exogenous TSH in human in vivo (9) and TSH-enhanced Ig production as well, which was intermediated by T lymphocytes (8). Therefore, it is of interest to examine whether TSH affects the secretory activity of airway submucosal gland, one of the candidate major contributors of airway humoral immunity.

Because human airway epithelium is likely to be primarily absorptive (17), a major fraction of the airway fluid seems to be derived from the submucosal gland (20) and follows an active Cl secretion from the glandular acini (21, 22). We reported previously that human and feline tracheal gland acinar cells generated ionic currents in response to cholinergic and -adrenergic stimuli, and that the currents were activated by the cellular Ca²⁺ concentration raised by the neurotransmitters (23, 24). In the present study we used the ionic current as a marker of the acinar cell activation, and examined the effects of TSH (and related agents) on the neurotransmitter-induced responses in freshly isolated and enzymatically dissociated acinar cells in tracheal gland using a patch-clamp technique. Here we report that the responses to the neurotransmitters in tracheal gland acinar cells were strikingly potentiated by TSH in a dose-dependent fashion. This action of TSH seemed not to be mediated by the thyroid-type TSH receptor (TSH-R) nor by the signaling cascade of adenosine 3',5'-cyclic monophosphate (cAMP)-protein kinase (PK) A, but by a protein-tyrosine kinase pathway. This is the first report that describes a regulation of exocrine gland secretion by pituitary hormones, which may have potential importance in airway mucosal immunity.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Preparation

Submucosal glands were isolated either from the tracheae of human surgical specimens from patients with laryngeal cancer or those of cats (2 to 5 kg body wt) anesthetized with intramuscular ketamine hydrochloride (30 mg/kg) and intravenous thiopental sodium (30 mg/kg). The cat trachea was put into an extracellular solution (see ELECTRICAL RECORDINGS) immediately after removal. The external surface of the trachea was cleaned of fat and connective tissues, cut into rings 3 to 4 cm long, and fixed by pins in the extracellular solution with the posterior (membranous) wall side up. Light through a flexible fiber bronchoscope (FBS-1; Machida, Tokyo, Japan) placed inside the tracheal ring was used to transilluminate the membranous portion. The outermost layer and thick smooth-muscle layer were carefully removed. The submucosal gland could then be easily distinguished from the surrounding connective tissue under a stereoscopic microscope (×60 to ×80 magnification). Fresh, unstained submucosal glands were isolated using two pairs of tweezers and microscissors (25). The isolated glands were further dispersed enzymatically into single or clustered acinar cells by incubating them with enzyme solution containing collagenase (200 U/ml), DL-dithiothreitol (0.31 mg/ ml), and trypsin inhibitor (1 mg/ml) for 30 min at 37°C. After dispersion and a wash with centrifugation at 180 × g the cells were resuspended in a standard extracellular solution until use. In a series of experiments investigating the effect of herbimycin A, a protein-tyrosine kinase inhibitor, the cells were incubated in extracellular solution containing herbimycin A (0.5 µg/ml) for 5 h before use.

Electrical Recordings

Ionic currents were measured with a patch-clamp amplifier (EPC9; HEKA Electronic, Lambrecht/Pfalz, Germany), low-pass filtered at 2.9 kHz, and monitored on both a built-in software oscilloscope and a thermal pen recorder (RECTI-HORIZ-8K; Nippondenki San-ei, Tokyo, Japan). Patch pipettes were made of glass capillary with an outer diameter of 1.5 mm using a vertical puller (PP-83; Narishige Scientific Instruments, Tokyo, Japan), and had a tip resistance of 2 to 6 M. The junction potential between the patch-pipette and bath solution was nulled by the amplifier circuitry. After establishing a high-resistance (> 1 G), tight seal, the whole cell configuration was obtained by rupturing the patch membrane with negative pressure applied to the pipette tip. Membrane currents were monitored at two different holding potentials (HPs), i.e., 0 and 80 mV, which roughly corresponded to the Cl- and K⁺-equilibrium potential, respectively, under the present electrolyte conditions. This was accomplished by applying 200-ms voltage pulses of 80 mV at a frequency of 2 Hz to the pipette HP of 0 mV (23, 24, 26). The upward or downward deflection of the current tracing represents outward (I_o) or inward current (I_i), respectively. The solutions employed were of the following compositions (in mM): extracellular (bath) solution, 120 NaCl, 4.7 KCl, 1.13 MgCl₂, 1.2 CaCl₂, 10 glucose, and 10 N-2-hydroxyethylpiperazine-N'-2-ethane sulfonic acid (Hepes); and intracellular (pipette) solution, 120 KCl, 1.13 MgCl₂, 0.5 ethylene glycol-bis(-aminoethyl ether)-N,N,N',N'-tetraacetic acid, 1 Na₂ adenosine triphosphate, 10 glucose, and 10 Hepes. The fluids were superfused over the cell(s) by hydrostatic pressure-driven application (20 to 30 cm H₂O) through polyethylene tubes. All solutions were at pH 7.2, and all experiments were carried out at room temperature (22 to 25°C).

Quantification Procedure

The net electric charge movement across the cell membrane was quantified with a pen recorder chart by measuring the area circumscribed with the current trace and baseline using a digital planimeter (PLACOM KP-92N; Koizumi, Tokyo, Japan), and was converted to picoCoulombs per s (pQ/s). The effects of TSH or other agents on the agonist-stimulated responses were estimated by comparing the net electric charge movements of 20-s duration just before and after treatment with TSH (or other agents) and are expressed as proportions of the pretreatment control values.

Immunohistochemical Study

Surgical specimens of human trachea or thyroid gland were fixed in periodate-lysine-paraformaldehyde at 4°C for 24 h. After washing in sucrose (10%, 15%, 20%)/phosphate-buffered saline for 2 h each, they were embedded in OCT compound (Miles Laboratories, Naperville, IL) in liquid nitrogen and stored at 80°C until use. The staining was performed with the alkaline phosphatase-antialkaline phosphatase method (17, 27). Cryostat sections (6 µm) were stained for human TSH-R using monoclonal mouse antihuman TSH-R IgG2a antibody specific to 211-414 amino acid residue (YLEM, Rome, Italy). The slides were incubated with the primary antibody diluted 25-fold in Tris-buffered saline (0.05 M Tris and 0.15 M NaCl, pH 7.6). After overnight incubation at 4°C, the slides were incubated with antimouse Ig rabbit Igs (Dako Corp., Carpinteria, CA) for 30 min at room temperature followed by an incubation with soluble complexes of alkaline phosphatase and mouse monoclonal antialkaline phosphatase (Dako Corp.) for 30 min at room temperature. Slides were then developed by exposure to the substrate for 12 min with Fast Red substrate system (Dako Corp.) according to the manufacturer's protocol, and counterstained with hematoxylin. As negative controls, slides were evaluated with an irrelevant primary mouse monoclonal antibody (mAb) (anti-Aspergillus niger glucose oxidase; Dako Corp.).

Reverse Transcription/Polymerase Chain Reaction Analysis of TSH-R Messenger RNA

Human submucosal glands isolated from tracheae of surgical specimens were used. A human thyroid gland obtained from the macroscopically normal part of a surgical specimen of thyroid cancer was used as a positive control. Total RNA was extracted from each sample within 1 h of isolation in Isogen (Nippon Gene Co., Ltd., Tokyo, Japan), according to the manufacturer's protocol. The amount of 1 µg total RNA extracted from cells was converted to first-strand complementary DNA (cDNA) with oligo (dT)_12-18 primer and Moloney murine leukemia virus reverse transcriptase (GIBCO BRL, Life Technologies, Inc., Rockville, MD) following the supplier's instructions. Oligonucleotide primers used in polymerase chain reaction (PCR) for TSH-R were 5'-CTGGAGCACCTGAAGGAACTGATAG-3' and 5'-GGTTGCACATGAGAAAGCGGGGGA-3' and amplified 631 base pairs (bp) of the TSH-R cDNA segment. One-hundredth of the cDNA synthesis reaction volume was combined at a final volume of 100 µl for PCR amplification by using each primer and 2.5 U of Taq DNA polymerase (Takara, Tokyo, Japan) for TSH-R cDNA segment amplification. The temperature and cycle for amplification were one cycle of denaturation at 94°C for 3 min, annealing at 52°C for 15 s, and extension at 72°C for 1 min, followed by 44 cycles of denaturation at 94°C for 1 min, annealing at 50°C for 20 s, and extension at 72°C for 1 min. Each 6 µl of PCR product was electrophoresed in a 1.5% agarose gel containing ethidium bromide.

Animal Care and Human Surgical Specimens

This study was approved by the Animal Care and Use Committee and the Ethic Committee on Human Investigations of the Tohoku University School of Medicine. The care and handling of the animals were performed in accordance with National Institutes of Health guidelines (28).

Reagents

Hepes was purchased from Dojin Co. Ltd., Kumamoto, Japan. Collagenase was from Wako Pure Chemicals (Osaka, Japan) and OCT compound from Miles Laboratories. Mouse monoclonal antihuman TSH-R IgG specific to COOH terminus was from YLEM. Antimouse Ig rabbit Igs, mouse monoclonal antialkaline phosphatase, mouse monoclonal anti-A. niger glucose oxidase, and Fast Red substrate systems were from Dako Corp. All other chemicals used were purchased from Sigma (St. Louis, MO).

Statistics

The data are expressed as means ± standard error; n is the number of experiments on different cells. Data were analyzed by Wilcoxon's signed rank test, and significance was accepted at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of TSH on Neurotransmitter-Induced Electric Responses in Tracheal Gland Acinar Cells

The freshly isolated human and feline acinar cells in tracheal submucosal glands generated ionic currents in a qualitatively identical manner in response to various agents including acetylcholine (ACh) and norepinephrine (NE) (23, 24). We previously reported that the neurotransmitter-induced I_o and I_i were carried mainly by K⁺ and Cl, respectively, which were dependent on the cellular Ca²⁺ concentration ([Ca²⁺]_i) (23, 24). The Cl I_i in exocrine acinar cells corresponded to the Cl secretion from the acini, which have been shown to parallel the water secretion physiologically (29).

In the present study, we used low doses of neurotransmitters (ACh 10⁸ M; NE 10⁷ M) to stimulate the electric responses in feline tracheal glands. In the presence of such low concentrations of agonists, some cells generated an oscillatory Cl current (I_i; 33 out of 92 cells tested, 35.9%), some failed to display an apparent response (21 cells, 22.8%), and the rest of the cells showed an oscillatory current imposed on sustained current (38 cells, 41.3%).

As shown in Figures 1A and 1B, TSH in combination with ACh potentiated the agonist-evoked electric responses in a dose-dependent manner. When TSH close to the serum level (2 ng/ml, 14 µIU/ml) was introduced on top of ACh, both I_o and I_i were increased to 4-fold (10.0 ± 3.0 pQ/s for ACh and 37.3 ± 9.0 for ACh/TSH, P < 0.05; n = 7) and to 3-fold (10.4 ± 5.8 versus 30.2 ± 9.7 pQ/s, P < 0.05; n = 7) the pre-TSH control values, respectively (Figure 2A). A high concentration of TSH (20 ng/ml) further augmented I_o and I_i to 4-fold (34.8 ± 9.8 versus 133.5 ± 23.7 pQ/s, P < 0.01; n = 28) and to 6-fold (17.9 ± 4.1 versus 99.6 ± 25.0 pQ/s, P < 0.01; n = 28) the control values, respectively. TSH also potentiated the NE-induced response (Figures 1C, 1D, and 2B). In the presence of 2 ng/ ml TSH, I_o and I_i were increased to 5-fold (from 19.7 ± 10.1 to 102.5 ± 44.5 pQ/s, P < 0.05; n = 8) and 3-fold (15.6 ± 7.8 versus 42.3 ± 12.6 pQ/s, P < 0.05; n = 11), respectively, and in the presence of 20 ng/ml TSH, to 23-fold (6.0 ± 4.0 versus 135.8 ± 50.5 pQ/s, P < 0.01; n = 5) and to 23-fold (5.0 ± 3.9 versus 117.0 ± 55.8 pQ/s, P < 0.01; n = 5) the pre-TSH control responses, respectively (Figure 2B).

View larger version (56K): [in this window] [in a new window]   Figure 1.   Representative original recordings showing the effect of TSH on the neurotransmitter-induced ionic currents in tracheal submucosal gland acinar cells. A conventional whole-cell patch-clamp technique was used. Membrane currents were monitored at two different HPs, i.e., 0 and 80 mV, which was accomplished by applying 200-ms voltage pulses of 80 mV at a frequency of 2 Hz to the pipette HP of 0 mV. The periodic spikes imposed on the current traces (clearly seen on C and D) are parts of the capacitance currents of the plasma membrane. These artificial currents arise at the moment just after the switching of the holding potential from 0 mV to 80 mV (or the opposite). The upward or downward deflection of the current tracing represents Io or Ii, respectively, as indicated in B. We previously identified Io and Ii to be carried mainly by K+ and Cl, respectively. Agonist concentrations used were 108 M for ACh and 107 M for NE throughout the experiments. TSH potentiated the neurotransmitter-induced ionic currents significantly in a dose-dependent fashion. TSH alone was without effect on baseline currents. Note that the cells that exhibited minimal responses to the agonists acquired a distinctive responsiveness by TSH (D).

View larger version (34K): [in this window] [in a new window]   Figure 2.   Data summarizing the effects of TSH on ACh- (A), and NE- (B) induced ionic currents in tracheal submucosal gland acinar cells. Data obtained from the experiments with FSH and LH are also included in A. Electric charge movements of 20 s duration just before and after introducing TSH were compared, estimating the mean values of pre-TSH responses as 1. *P < 0.05, **P < 0.01. See MATERIALS AND METHODS for the numbers of each experiment.

Interestingly, even the cells that apparently showed no responses to agonist produced large ionic currents when TSH was introduced (e.g., see Figure 1D). TSH alone was without effect on the baseline currents up to a concentration of 20 ng/ml (n = 14). We usually added TSH at a time point of about 30 s after the application of the either neurotransmitter. Although we did not examine the influence of the time course in detail, no qualitative difference was found within the present variance of the timing of the TSH application. Also, a change in the order of drug administration showed apparently no effect on the potentiation effects.

Analyses Investigating the Involvement of TSH-R

To investigate the physiologic relevance of the TSH-mediated upregulation of the electrolyte secretion, we tested the effects of other pituitary glycoprotein hormones, follicular stimulating hormone (FSH) or luteinizing hormone (LH), on tracheal glands. Like TSH, both of the gonadotropins per se induced no apparent response. As exemplified in Figure 3A, however, both FSH and LH did affect the ACh-evoked currents at their maximal concentration (20 ng/ml each), although to a far lesser extent compared with TSH. As shown in Figure 2A, both I_o and I_i were 20.0 ± 7.5 pQ/s for ACh alone and 21.3 ± 8.9 for ACh/ FSH (P = 0.99; n = 6) and 28.1 ± 5.4 versus 42.0 ± 9.7 pQ/s (P < 0.05; n = 8), respectively. Similarly, I_o was 28.8 ± 4.2 pQ/s for ACh alone and 35.0 ± 4.0 after treatment with LH (20 ng/ml; P = 0.40, n = 9), and I_i was 19.6 ± 2.9 versus 26.1 ± 2.2 pQ/s (P = 0.08; n = 10). In addition, thyrotropin-releasing hormone had no effect on either baseline or triggered currents at concentrations up to 100 µM (data not shown; n = 12). These results indicate that the present TSH effect on tracheal gland is specific to its -subunit that determines the hormonal specificity.

View larger version (78K): [in this window] [in a new window]   Figure 3.   Representative original recordings showing effects of: (A) LH in comparison with TSH in the same cell; (B) isoproterenol; and (C ) mixture of IBMX and cpt-cAMP. LH is one of the pituitary hormones structurally related to TSH. LH (and FSH) showed a weak cross-reactivity with TSH (see also Figure 2). TSH acts on thyrocytes with cAMP as a major second messenger in releasing thyroid hormones. However, maneuvers anticipating a rise in cAMP failed to mimic the TSH effect on tracheal gland.

TSH is known to increase the cAMP level and PKA activity in thyrocytes (30), and forskolin or membrane-permeable analogues of cAMP mimic many actions of TSH, including the release of thyroid hormones. To test the involvement of cAMP in the present signaling pathway in tracheal gland, we examined the effects of isoproterenol, a -adrenergic agonist, or a cocktail of 3-isobutyl-1-methylxanthine (IBMX) (a cyclic nucleotide phosphodiesterase inhibitor; 10³ M) and 8-(4-chlorophenylthio) (cpt)-cAMP (a membrane-permeable cAMP analogue; 10⁴ M) on the ACh-induced electric responses. As shown in Figures 3B and 3C, however, neither of the procedures that intended to raise the level of cAMP could mimic the potentiating effect of TSH, but both showed an opposite inhibitory effect on the ACh-induced responses. These results indicate possibilities that TSH made use of a signaling cascade other than cAMP-PKA via TSH receptors on tracheal gland cells, or otherwise, that the present TSH effect was not mediated by the thyroid-type TSH-R.

In an attempt to identify TSH receptors on tracheal gland acinar cells, immunohistochemical and reverse transcription (RT)-PCR experiments were performed using human tracheal specimens with thyroid gland as a positive control. As shown in Figure 4A, in contrast to the dense positive label on thyroid follicular epithelium, we observed little immunostaining on submucosal gland or on surrounding tissues including the epithelium. Similarly, the cDNA segment of TSH-R was not detected in submucosal gland acinar cells at all, whereas the thyrocytes expressed a clear band corresponding to TSH-R cDNA, as expected (Figure 4B).

View larger version (141K): [in this window] [in a new window]   View larger version (68K): [in this window] [in a new window]   Figure 4.   (A) Immunoalkaline phosphatase stain of human submucosal gland (c and d) using mouse mAb against human TSH-R, with thyroid gland (a and b) as a positive control. The sections in the upper panels (a and c) were with mAb against TSH-R, whereas those in the lower panels were with control IgG2a. All sections were counterstained weakly with hematoxylin. In contrast to the positive red stain on thyroid follicular epithelium (a), no significant label was detected in submucosal gland (c). (Bar: 50 µm). (B) cDNA segments of TSH-R (upper panel) and -actin (lower panel) amplified by RT-PCR from human thyroid gland and tracheal submucosal gland. Thyroid gland showed a clear band of TSH-R cDNA, whereas submucosal gland did not. The 631 1,017-bp DNA segments are expected to be amplified from TSH-R cDNA and -actin cDNA, respectively.

Mechanism Underlying the TSH-Mediated Potentiation of Triggered Ionic Currents in Submucosal Gland

TSH highly sensitized the tracheal gland to neurotransmitters without any visible action by itself. It is unlikely that this effect was mediated by the thyroid-type TSH-R, because cAMP, a major second messenger of TSH in releasing thyroid hormones, had an opposite inhibitory effect on tracheal gland (Figure 3) and, moreover, TSH-R protein and its messenger RNA (mRNA) appeared to be absent in this cell type (Figure 4). This reminded us of the fact that TSH acts on other receptors for factors such as EGF (1), TGF- (2), and adrenergic agents (31), resulting in the upregulation of cell growth, endothelin secretion, or hormone release in thyrocytes. It has been well documented that TSH increases EGF binding to its receptor, activating the receptor's intrinsic tyrosine kinase (1). Interestingly enough, this activation of tyrosine kinase can even occur in FRTL-5 cells, which lack the high-affinity cooperative EGF receptor binding sites (32). Therefore, we examined the effects of protein-tyrosine kinase inhibitors on the present TSH action. We used two different inhibitors, genistein and herbimycin A, which are known as reversible (33) and irreversible (34) tyrosine kinase inhibitors, respectively. As shown in Figure 5A, genistein (25 µg/ml; n = 7) reversibly abolished the potentiation activity of TSH. In like manner, TSH exhibited no potentiation on the ACh-evoked electric response when the cells were treated with herbimycin A (0.5 µg/ml; n = 3, Figure 5B). Neither inhibitor affected the baseline currents or the occurrence of the ACh-induced response (Figures 5B and 5D). The estimates of the ACh currents potentiated by TSH were restored to the pre-TSH level in the presence of genistein (Figure 6).

View larger version (76K): [in this window] [in a new window]   Figure 5.   Representative original recordings showing the effects of tyrosine kinase inhibitors (A and B) and a known tyrosine kinase activator, IFN- (C and D). (A) Genistein reversibly abolished the TSH-mediated potentiation in the ionic currents without affecting the occurrence of the ACh response. (B) Cells incubated with herbimycin A, another tyrosine kinase inhibitor, for 5 h before the experiments normally responded to ACh but they completely lost the ability to respond to TSH. (C and D) IFN- mimicked TSH in both the augmentation and the sensitivity to genistein.

View larger version (23K): [in this window] [in a new window]   Figure 6.   Data summarizing the effect of genistein (25 µg/ml) on TSH (20 ng/ml)- enhanced ACh response (n = 7). The estimates of the ACh currents potentiated significantly by TSH were restored to the pre-TSH level in the presence of genistein (GST). *P < 0.05.

IFN- has been reported to act via protein-tyrosine kinase in various cells, including a human intestinal epithelial cell line that release SC (13). Hence, we tested the effect of IFN- on tracheal gland as an activator of protein-tyrosine kinase. As shown in Figure 5C, IFN- (> 400 U/ml) mimicked the TSH-mediated augmentation of the ACh-induced electric response in submucosal gland acinar cells (n = 4). This augmentation by IFN- was abolished completely in the presence of genistein (Figure 5D; n = 6).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Articles concerning the physiologic properties of freshly isolated submucosal gland are quite limited in number (22- 25, 35, 36), and most studies investigating the submucosal gland electrophysiology made use of cells in culture (21, 37, 38). This is probably because the number of cells available is anatomically very small, and because small experimental animals including rat, mouse, rabbit, and goose, have few tracheobronchial glands, if any (39). The submucosal gland bioelectric properties in fresh- and cultivated-cell preparations are similar in terms of the presence of [Ca²⁺]_i-activated Cl secretion in response to cholinergic agents. However, conflicting observations have been reported concerning the responses to -adrenergic agonists. For instance, the -agonist isoproterenol induced a large short-circuit current across a cultured sheet of human tracheobronchial gland (21), whereas freshly isolated human and feline tracheal gland acinar cells failed to respond to isoproterenol in a whole-cell patch-clamp study (23), despite the fact that the fresh cells secreted mucus glycoproteins in response to -agonists (36). However, the electrophysiologic properties of fresh tracheal gland are quite similar to those of other freshly isolated exocrine glands, including lacrimal (27), salivary (40), and exocrine pancreas (29) when investigated by patch-clamp methodology. Moreover, an in vivo observation indicated that cholinergic agents were much more potent stimulators of gland secretion than were adrenergic agonists, when estimated using hillock formations of a powdered tantalum layer coating the airway surface (41). Therefore, we used ACh and NE as cholinergic and -adrenergic agonists, respectively, both of which activate Ca²⁺- induced ionic currents in freshly isolated submucosal gland preparations (23, 24).

A major finding in this study was the striking augmentation by TSH of the triggered ionic currents in tracheal gland. A significant potentiation was recognized at a TSH concentration close to the plasma level in healthy humans (0.1 to 1.5 ng/ml). Moreover, even the cells that appeared silent in the presence of neurotransmitters acquired a responsiveness by TSH, which might explain the excessive requirement of agonists in many in vitro experimental situations.

TSH is a heterodimeric glycoprotein hormone (28,000 mol. wt.) composed of an -subunit which it shares with FSH and LH, and a unique -subunit that confers hormonal specificity (42). Because of this structural peculiarity, a cross reactivity among these anterior pituitary hormones at high concentrations has been well documented (43). Hence we tested the effect of FSH or LH on submucosal gland. These hormones showed only a very weak potentiation at the maximal concentration used, and only the effect of FSH on I_i was statistically significant (Figures 2 and 3A). These results suggested that the present TSH effect could be ascribed to its hormonal activity that is assigned to the -subunit of TSH.

Although the TSH-specific action on tracheal gland suggested the presence of TSH-R on the acinar cells, we could not detect it by either immunohistochemistry or RT-PCR (Figure 4). These findings confirmed the results of an earlier study that investigated the binding of ¹²⁵I-labeled bovine TSH to human thyroid, testicular, fat, adrenal, liver, kidney, pancreas, and lung cell membranes (3). The investigators found that the first four tissues had comparable high-affinity constant values, but the rest of the tissues, including lung, lacked high-affinity sites. In the present study, we validated their finding at the protein and mRNA levels. Moreover, increases in cellular cAMP by either isoproterenol or the IBMX/cpt-cAMP mixture failed to mimic the TSH effect, which further supports the idea that the present potentiation was not mediated by the thyroid-type TSH-R. However, the binding study does not preclude the possibility of the presence of a presumed non- thyroid type TSH-R.

The present TSH action on tracheal gland is likely to be mediated by a signaling pathway involving tyrosine kinase(s) because specific inhibitors of tyrosine kinase totally abolished the TSH effect without affecting the occurrence and continuance of the neurotransmitter-induced responses (Figures 5A and 5B). Moreover, IFN-, a known tyrosine kinase activator, mimicked the TSH action (Figures 5C and 5D). Recently, two groups independently published observations regarding the upregulated transcytosis of SC, an Ig transporter, by IFN- across a cultured sheet of Calu-3 cells (12, 15) which appears to resemble the tracheobronchial gland acinar cells (16). Because TSH has been shown to release IFN- from lymphocytes in vivo and in vitro (9), and because TSH enhances Ig production, probably from plasma cells (8), TSH can be positioned directly upstream of the Ig secretion as well as fluid secretion from the gland.

In some experiments, we used a maximal concentration of ACh (10⁶ M) to stimulate the acinar cells. However, we found no potentiation by TSH (either with concentrations 2 or 20 ng/ml) of the maximally stimulated outward and inward currents. This fact gave an important insight into the mechanism underlying the present TSH-mediated potentiation. That is, TSH affected neither the already activated (fully opened) channel populations nor the increase in the number of channel proteins on the plasma membrane. Presumably, TSH might influence the Ca²⁺-mobilization pathway or a mechanism sensitizing the channels to cellular Ca²⁺, resulting in a leftward shift of the concentration-response curves. When considering the cellular mechanism of the present TSH action, reports describing another endocrine-exocrine interaction may be of help. The islet hormone insulin has been reported to markedly potentiate the effect of ACh on rat pancreatic enzyme secretion without influencing the baseline secretion by itself (44). The highly specific tyrosine kinase inhibitor genistein abolished the potentiation by insulin without affecting the ACh-induced amylase secretion and Ca²⁺ rise (45), which was very similar to the present TSH action on submucosal gland. Taking into consideration the fact that the insulin receptor is a member of the receptor tyrosine kinase superfamily, TSH may also act on some tyrosine kinases in tracheal gland acinar cells. In this respect, it may be of importance that TSH has an ability to activate the EGF receptor's intrinsic tyrosine kinase in cultured thyrocytes (1, 32), resulting in the increased affinity of EGF to its receptor. The EGF receptor is another member of the receptor tyrosine kinase superfamily (46) and is known to be present on airway submucosal gland (47). We have no data at present on how the activated tyrosine kinase potentiates the neurotransmitter-induced ionic currents. One possibility is that tyrosine phosphorylation of phospholipase C- (48) may catalyze the breakdown of phosphatidyl inositol bisphosphate into inositol 1,4,5-trisphosphate (InsP₃) and diacylglycerol, and the former releases intracellular Ca²⁺, resulting in the potentiation of the Ca²⁺-activated currents. Although InsP₃ plays an important role in ACh-mediated current activation in submucosal gland (23), this possibility is remote because TSH exhibited no effect on the baseline currents. An increased level of [Ca²⁺]_i by neurotransmitters may be a prerequisite for the present TSH action on submucosal gland.

We showed that TSH and IFN- markedly potentiated the electrolyte secretion from the tracheal submucosal gland in the presence of physiologic neurotransmitters. This possibly serves as a nonspecific airway protection. Assuming that TSH can be released locally, as has been reported in intestinal mucosa or lymphocytes, TSH may also contribute to the hyperplastic or hypertrophic remodeling of submucosal glands in chronic inflammatory airway diseases by enhancing the affinity of EGF to its receptors on the gland. In addition, TSH has been known to increase antibody production and to increase plasma IFN-, which may facilitate specific mucosal immunity via the submucosal gland in association with upregulated SC turnover. Taken together with the short plasma half-life of about 30 min, TSH may be one of the important regulators of airway mucosal immunity (see Figure 7).

View larger version (55K): [in this window] [in a new window]   Figure 7.   A drawing delineating the proposed pathways of thyrotropin (TSH) in regulating the airway defense mechanisms. The left half of the figure depicts the TSH action on the nonspecific defense mechanism, which we showed in the present investigation. TSH markedly potentiated the neurotransmitter-mediated electrolyte secretion, resulting in an upregulated fluid secretion that washed away the inhaled substances. The right half of the figure depicts a hypothetical enrollment of TSH in the specific defense. TSH is known to bind to lymphocytes, raise the plasma level of IFN-, and enhance the production of Igs by promoting the differentiation and proliferation of plasma cells. IFN- then upregulates the turnover of SC, a transporter of Igs synthesized in glandular cells. In inflammatory or infected airways, plasma cells have been found to localize in close proximity of submucosal glands. The Igs released from the plasma cells bind in turn to SC at basolateral aspect of the acinar cells. The secretory Igs (sIgs) thus formed are transported through the cytoplasm and released into the lumen at the apical side of the cells. M3: muscarinic M3-receptor; : -adrenergic receptor; Th: helper T lymphocyte. Numbers in parentheses are reference numbers.