Published ahead of print on July 10, 2003, doi:10.1165/rcmb.2003-0177OC
© 2004 American Thoracic Society DOI: 10.1165/rcmb.2003-0177OC Ionotropic Glutamate Receptors in Lungs and AirwaysMolecular Basis for Glutamate ToxicityV.A. Medical Center, Northport; and Department of Medicine, SUNY Health Sciences Center, Stony Brook, New York Address correspondence to: Sami I. Said, M.D., Pulmonary and Critical Care Medicine, T-17 025 Health Sciences Center, SUNY at Stony Brook, Stony Brook, NY 11794–8172. E-mail: ssaid{at}notes.cc.sunysb.edu
We earlier showed that the ionotropic glutamate receptor agonist N-methyl D-aspartate (NMDA) induces excitotoxic pulmonary edema, and that endogenous activation of NMDA receptors (NMDAR) could mediate lung injury caused by oxidative stress. In this study, we searched for evidence of NMDAR expression in the rat lung and in the alveolar macrophage (AM) cell line NR8383, and for possible regulation of receptor expression by NMDA. The presence of mRNA for NMDAR 1 and the four known NMDAR 2 subtypes (A, B, C, and D) was examined by reverse transcriptase–polymerase chain reaction using isoform-specific primers. NMDAR 1 was expressed in all lung regions examined (peripheral, midlung, and mainstem), as well as in trachea and the AMs. Expression of NMDAR 2A and 2B subtypes was not detected, whereas NMDAR 2C was present only in peripheral and mid-lung samples. NMDAR 2D was the dominant subtype expressed in the peripheral, gas-exchange zone of lung and in alveolar macrophages, and this expression was upregulated in lungs treated with NMDA. Western blot confirmed the presence of NMDAR 1 protein in all lung regions and in AMs. These findings provide a molecular-biological basis for the excitotoxic actions of glutamate in rat lungs and airways, and raise the question of a possible physiologic role for NMDAR in lung and airway function.
Abbreviations: alveolar macrophage, AM • central nervous system, CNS • glyceraldehyde phosphodehydrogenase, G3PDH • N-methyl D-aspartate, NMDA • NMDA receptor, NMDAR • pulmonary artery, PA • phosphate-buffered saline, PBS • polymerase chain reaction, PCR • polyvinylidene fluoride, PVDF • radioimmunoprecipitation assay buffer, RIPA • reverse transcription (or transcriptase), RT • Tris-borate-EDTA, TBE • Tris-buffered saline, TBS • Tris-buffered saline with 0.1% Tween, TBST • vasoactive intestinal peptide, VIP
The amino acids glutamate and aspartate, abundantly present in the mammalian central nervous system (CNS), are major excitatory neurotransmitters. Acting on glutamate receptors, they play important physiologic roles (1). But glutamate can also be lethal to neurons. This lethal action is exerted in part through activation of ionotropic N-methyl D-aspartate (NMDA) receptors (NMDAR), resulting in an influx of intracellular calcium, which triggers a series of toxic events ultimately leading to cell death (2–4).
Three forms of glutamate receptors have been identified: NMDA, Studies in the CNS have determined that the NMDAR consists of tetrameric, heteromeric assemblies of two subunits: NMDAR 1 and NMDAR 2, which have different physiologic and pharmacologic properties. These subunits, respectively, contain the glycine and glutamate recognition sites. The NMDAR 2 subfamily consists of four individual subunits, NMDAR 2 A, B, C, and D (5–7), that are differentially expressed within the brain. The presence of NMDA-type glutamate receptors has been reported in non-neuronal tissues, including pancreatic ß cells, the male lower urogenital tract, kidney, lymphocytes, megakaryocytes, and cerebral microvasculature (8–16). The first indication that NMDA may play a role in lung physiology and disease was our observation that high concentrations (1 mM) of NMDA in the perfusate of isolated, ventilated rat lungs induced acute high-permeability edema that could be prevented by the noncompetitive NMDAR antagonist MK-801 (17). We later reported that inhibitors of NMDAR greatly attenuated oxidant lung injury caused by paraquat or xanthine oxidase, suggesting a role for endogenous activation of NMDA receptors in mediating some forms of acute lung injury (18). In addition, we have made three observations that suggest a link between NMDAR activation and airway hyperresponsiveness, a cardinal feature of bronchial asthma (19). First, NMDA (10-8–10-4 M), applied to perifused tracheal segments of guinea pig, increased resting muscle tone and enhanced the contractile response to acetylcholine (10-5–10-4 M) or methacholine (10-7–10-5 M). Second, in whole guinea-pig lungs perfused via the trachea, NMDA (1–2 mM) increased airway perfusion pressure, and the increase was totally abolished by MK-801 (dizocilpine, 10 µM). Third, dizocilpine greatly attenuated the increase in airway pressure elicited by capsaicin (10-8–10-7 M), a vanilloid that provokes acute inflammation and reproduces some of the main abnormalities of bronchial asthma. Together, these findings suggested that NMDAR activation might be an important, though still unrecognized, mechanism of acute lung injury and of the airway inflammation and hyperreactivity of bronchial asthma (20). We were, therefore, prompted to search for direct evidence for ionotropic glutamate receptors in lung and airways. Our goals in this study were to: (i) seek direct evidence for the expression of NMDAR in rat lung and alveolar macrophages; (ii) identify the subtypes of these receptors and localize them to various lung regions; and (iii) examine their possible upregulation by the receptor agonist NMDA.
Isolated, Perfused Lung Preparation Male Sprague-Dawley Rats (300–350 g; Taconic, Germantown, NY) were anesthetized with sodium pentobarbital (100 mg/kg, intraperitoneally). Following tracheal intubation, the lungs were mechanically ventilated with a humidified gas mixture of 95% O2–5% CO2, using a rodent ventilator (Model 640; Harvard Apparatus, South Natick, MA), at 60 cycles/min and a tidal volume of 5 ml/kg. The lungs were perfused in situ with Mg2+-free Krebs solution (117 mM NaCl, 5.4 mM KCl, 25 mM NaHCO3, 1 mM NaH2PO4, 2.5 mM CaCl2) supplemented with 4% bovine serum albumin (Krebs-BSA), through a cannula inserted into the pulmonary artery (PA) via the right ventricle. The perfusate was collected via a cannula placed in the left atrium. Perfusion, by a pulsatile circulation pump (Model 1405; Harvard Apparatus), was started at a rate of 8 ml/min; the rate was adjusted to reach an initial mean PA pressure of 5–8 cm H2O, then was held constant. Initially, the lungs were perfused in an open, nonrecirculating mode for 10 min to washout residual blood, after which perfusion was switched to the recirculating mode, with 100 ml Krebs-BSA. Peak airway and mean PA pressures were continuously monitored by pressure transducers (Statham P23A; Statham, Hato Rey, PR) attached to the tracheal cannula and to a polyethylene catheter in the main PA, respectively, and were recorded (2600S recorder; Gould, Cleveland, OH). Left atrial pressure was maintained at 2 cm H2O by placing the tip of the outflow cannula 2 cm above the level of the atrium. After stabilization of airway and perfusion pressures, samples were taken from the peripheral areas containing alveoli and microvessels, as well as from central lung, containing mainstem bronchi, and from midlung, containing a mix of peripheral tissue and small airways. All tissue samples were immediately frozen in liquid nitrogen and stored at –70°C. In some experiments, the effect of NMDA on NMDAR expression was examined as follows. After stabilization of airway and PA pressures, 10 mM L-arginine was added to the perfusate (17), and 10 min later peripheral lung samples (control) were taken. The sampling area was ligated and 1 mM NMDA was added to the perfusate. After perfusion for 1 h, peripheral lung samples were again taken and immediately frozen in liquid nitrogen. Peripheral lung samples were also checked for the wet/dry weight ratio as a measure of the presence and severity of pulmonary edema. One lung was also lavaged with 3 ml saline for measurement of bronchoalveolar lavage fluid protein content, as an index of protein leakage due to alveolar-microvascular injury (17).
Alveolar Macrophages
Reverse Transcriptase–Polymerase Chain Reaction
Total RNA was isolated from frozen lung samples (
Western Blot
Immunofluorescence
Selective Expression of Receptor Subtypes by RT-PCR in Lung and Airways As expected, mRNA for NMDAR 1 and all four NMDAR 2 subtypes was clearly evident in rat brain samples taken from cerebral cortex (Figure 1). NMDAR 1 was moderately expressed in trachea and major airways (mainstem), and weakly detectable in peripheral and midlung samples. mRNA for NMDAR 2A and 2B receptor subtypes was totally absent in all lung regions examined, as well as trachea. NMDAR 2C was observed in peripheral and midlung regions, and absent in mainstem and trachea, whereas NMDAR 2D was notably expressed in peripheral and midlung samples, and weakly expressed in major airways and trachea. No signal was detected using samples that had not been reverse-transcribed before PCR (-RT), indicating that bands, when present, were not due to genomic amplification (Figure 1).
Detection of NMDAR 1 by Western Blot Analysis The presence of NMDAR 1 in the lung was confirmed by Western blot, using an antibody directed against the C2 region of the receptor present in four of the eight possible splice variants. The predominant band in brain, as expected, was detected at  109 kD (Figure 2A). In nonperfused peripheral lung, two bands were noted, at 115 and 138 kD. The higher molecular weight of bands present in lung compared with brain may be due to differences in glycosylation or phosphorylation of the receptor protein, or to the presence of a variant of the receptor subunit, as noted in the male lower urogenital tract (11). In samples from lungs previously perfused with Krebs-BSA, the 138-kD band was absent, suggesting that it was attributable to blood elements (Figure 2A). NMDAR 1 protein was detected as the 115-kD band in peripheral, midlung and mainstem samples from perfused lung, as well as in the alveolar macrophage cell line NR833 (Figure 2B).
Receptor Expression by AM Cells from the rat alveolar cell line NR8383 expresed NMDAR 1 by RT-PCR (Figure 3A), and strongly and selectively expressed NMDAR 2D as the only NMDAR 2 receptor subtype (Figure 3B). We confirmed the same finding in rat alveolar macrophage in primary culture. The presence of NMDAR 1 was confirmed by both Western blotting (Figure 2B) and immunofluorescence, as shown in Figure 4. Staining was consistently detected in all cells in the field (Figure 4A, immunofluorescence image and corresponding brightfield image, Figure 4B), and was not observed in cells exposed to secondary antibody only (Figure 4C, immunofluorescence image and corresponding brightfield image, Figure 4D). Immunofluorescent staining for NMDAR 1 appeared to be predominantly localized to the cell surface, possibly the cell membrane.
Upregulation of NMDAR 2D by NMDA RT-PCR analysis of peripheral lung samples taken before and after 60 min perfusion with 1 mM NMDA, a procedure shown to induce acute high-permeability edema (17), revealed marked upregulation of NMDAR 2D expression (Figure 5). No effects were noted on ß-actin (Figure 5) or G3PDH expression (data not shown).
Our results demonstrate the expression of a variety of NMDA glutamate receptor subtypes in the lung and airways. With the use of RT-PCR and Western blotting as well, our data permit these conclusions: (i) NMDAR 1 mRNA was moderately expressed, and its protein was easily detectable by Western blot, in all regions of the lung and in airways. (ii) NMDAR 2D mRNA was predominantly expressed in the peripheral, gas-exchange regions of the lung. (iii) The same receptor subtype, as well as NMDAR 2C, were expressed in medium-sized and larger airways. (iv) The complementary receptor subunits were often not expressed to the same degree in the same part of the lung. (v) All of these receptor subtypes are probably localized in nonneuronal elements, because neurons are not found beyond the conducting airways in rat lungs (24). Two related observations, reported separately, support and complement these findings: (i) The demonstration by receptor autoradiography with radio-labeled MK-801 as the ligand for the NMDA receptor, of binding sites in rat peripheral lung including alveolar walls, as well as in bronchial smooth muscle and bronchial epithelium (25); and (ii) Immunocytochemical evidence that NMDAR 2B subtype is expressed in neurons supplying the upper airways of rat lung (26). The latter study revealed that in some of these neurons, the NMDAR subtype was co-localized with vasoactive intestinal peptide (VIP) and neuronal nitric oxide synthase (20, 26). This finding is of special pathophysiologic significance, because we have shown that excitotoxic lung injury is mediated by excessive generation of nitric oxide, and is markedly attenuated by VIP (17). Although the NMDAR subtypes have been characterized as to their functional roles in the CNS and their relationship to neuronal excitotoxicity in neurological disorders (27), little or nothing is known about the physiologic and pathophysiologic significance of these receptor in the respiratory system. We have already referred to the implications, based on experimental evidence, that NMDA can induce excitotoxicity in the lung, in the form of an acute, high-permeability pulmonary edema (17), and that glutamate may be linked to the pathogenesis of bronchial asthma and airway inflammation (19). The expression of NMDAR subtypes reported here provides the first molecular-biological basis for the observed effects of glutamate and NMDA on the lungs and airways. The differential localization of these receptor subtypes is consistent with the view that NMDAR 2D may be the receptor subtype most closely associated with acute excitotoxic lung injury, and with NMDAR, may also be involved in the mediation of glutamate-enhanced airway reactivity and airway inflammation. The possible link between NMDAR 2D and excitotoxic lung injury is supported by (i) the selective expression of this particular receptor subtype of NMDAR 2 in the alveolar macrophage, one of the principal cells participating in the lung inflammatory response; and (ii) the upregulation of this receptor subtype by NMDA itself, a positive feedback mechanism by which NMDA toxicity can be perpetuated and amplified. Injured cells, including neurons and neutrophils (28), release glutamate at high (millimolar) concentrations, which becomes a further source of injury. These tentative conclusions, however, must be tempered by the fact that our knowledge of NMDA receptors in non-neuronal tissues and cell lines (11, 29) is relatively limited compared with what is already known about these receptors in the brain (27). Specifically, it is recognized that: (i) combinations of the glycine-binding NMDAR 1 subunit and the glutamate-binding NMDAR 2 subunit are required for full functional activity of NMDAR in mammalian neural cells; and (ii) combinations of different receptor subtypes give rise to receptors with different properties. Transfection of cloned NMDA receptors into cell lines has clarified important relationships between the makeup of the receptors and their ability to induce toxicity and cell death. In the CNS, NMDAR 2A– or 2B– containing receptors give rise to greater levels of toxicity than NMDAR 2C– and 2D–containing receptors (27). As to the potential physiologic significance of NMDA receptors in the lung, one can speculate about a possible contribution to noncholinergic regulation of smooth muscle tone in the airway and pulmonary vessels. Our own observations in airway smooth muscle (19) suggest that glutamate receptors may contribute to constrictor tone, whereas other authors have reported both a vasoconstrictor (11) and a vasodilator (13, 30) influence. The latter would result in large measure from stimulation of nitric oxide synthesis by NMDA (12, 13). In conclusion, the expression of NMDA glutamate receptors in the lungs and airways provides a solid foundation for the view that signaling via these receptors may be a factor in the pathogenesis of two major clinical disorders: acute lung injury simulating the acute respiratory distress syndrome, and acute bronchial asthma. Recognition of the link between these diseases and specific NMDA receptor subtypes could lead to the introduction of novel, more successful therapeutic strategies. The findings also provide an impetus to the exploration of a potentially important but virtually unexplored neuroregulatory system.
This work was supported by research funds from the Department of Veterans Affairs, and from NIH Grants HL-30450 and HL-70212 from the National Heart, Lung and Blood Institute. Received in original form May 5, 2003 Received in final form June 29, 2003
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
-amino-3-hydroxy-5-methyl-4-isoxazole-propionate (AMPA), and kainate. The NMDAR has been the one most strongly implicated in excitotoxic as well as neuroexcitatory events. Permeable to Ca2+ and uniquely blocked by Mg2+, it plays a key role in brain development, learning and memory, and synaptic plasticity (
