J Comp Physiol B (2004) 174: 461–470 DOI 10.1007/s00360-004-0432-6
O R I GI N A L P A P E R
M. Krohn Æ J.-P. Hildebrandt
Cross-talk of phosphoinositide- and cyclic nucleotide-dependent signaling pathways in differentiating avian nasal gland cells
Accepted: 28 April 2004 / Published online: 8 June 2004 Springer-Verlag 2004
Abstract In many bird species, the nasal glands secrete excess salt ingested with drinking water or food. In ducks (Anas platyrhynchos), osmotic stress results in adaptive cell proliferation and differentiation in the gland. Using ‘naive’ nasal gland cells isolated from animals that had never ingested excess salt or ‘differentiated’ cells from animals fed with a 1% NaCl solution for 48 h, we investigated the allocation of metabolic energy to salt excretory processes and to other cellular activities. Activation of muscarinic acetylcholine receptors (carbachol) or badrenergic receptors (isoproterenol) in nasal gland cells resulted in a transient peak in metabolic rate followed by an elevated plateau level that was maintained throughout the activation period. Activation of cells using vasoactive intestinal peptide, however, had only marginal effects on metabolic rate. In differentiated cells, sequential stimulation with carbachol and isoproterenol resulted in additive changes in metabolic rate during the plateau phase. Naive cells, however, developed supra-additive plateau levels in metabolic rates indicating cross-talk of both signaling pathways. Using bumetanide, TEA or barium ions to block different components of the ion transport machinery necessary for salt secretion, the relative proportion of energy needed for processes related to ion transport or other cellular processes was determined. While differentiated cells in the activated state allocated virtually all metabolic energy to processes related to salt secretion, naive cells reserved a significant amount of energy for other processes, possibly sustaining cellular signaling and regulating biosynthetic mechanisms related to adaptive growth and differentiation.
Communicated by G. Heldmaier M. Krohn Æ J.-P. Hildebrandt (&) Zoologisches Institut und Museum, Ernst Moritz Arndt-Universita¨t, J.S. Bach-Strasse 11/12, 17489, Greifswald, Germany E-mail:
[email protected] Tel.: +49-03834-864295 Fax: +49-03834-864252
Keywords Metabolic rate Æ Energy allocation Æ Epithelial cells Æ Signal transduction Æ Microphysiometry
Introduction Many bird species have nasal glands that excrete, in less water than imbibed, excess salt ingested with drinking water or food—a process that generates osmotically free water which is retained in the body (Schmidt-Nielsen 1960). In ducks (Anas platyrhynchos) maintained on freshwater, the salt glands remain small and most of the secretory cells in the gland are in a partially differentiated (‘naive’) state. There is no obvious secretion of fluid from the glands. Upon drinking saline (e.g. 1% NaCl solution), secretion of a concentrated fluid rich in NaCl is highly accelerated within a few hours (Bentz et al. 1999). Moreover, prolonged uptake of excess salt results in cell proliferation and cell differentiation in the secretory cells improving salt excretory capacity of the glands (Ernst and Ellis 1969; Hossler 1982; Bentz et al. 1999). Salt secretion, as well as these adaptive processes, depends on an intact nerve supply of the glands, since denervation abolishes both responses to an osmotic load (Peaker and Linzell 1975). Parasympathetic innervation and release of the transmitter acetylcholine seem to be especially essential for glandular functions, as infusion of the muscarinic acetylcholine receptor–antagonist atropine into osmotically stressed animals suppressed salt secretion as well as adaptive proliferation and differentiation in the glands (Peaker and Linzell 1975). However, sympathetic innervation and adrenalin/noradrenalin may also be important, since sympathetic nerve terminals have been found in gland parenchyma (Hasse and Fourman 1970), and b-adrenergic stimulation results in physiological responses in isolated nasal gland cells (Lowy and Ernst 1987; Martin et al. 1994). While activation of muscarinic acetylcholine receptors (mAChRs) in isolated nasal gland cells results in phos-
462
phoinositide hydrolysis, accumulation of inositol phosphates, and the generation of a sequence of characteristic cytosolic calcium signals (Shuttleworth and Thompson 1989; Hildebrandt and Shuttleworth 1991), the typical response to b-adrenergic activation is the accumulation of cyclic adenosine monophosphate (cAMP). Prolonged cytosolic accumulation of cAMP has been shown previously in avian nasal gland cells in osmotically stressed ducklings (Hildebrandt 1997). It is not known, however, whether adenylyl cyclase is activated primarily by vasoactive intestinal peptide (VIP) co-localized with acetylcholine in the nerve endings of the parasympathetic system or by catecholamines released by sympathetic synapses in the gland. In any case, physiological responses to elevations in cAMP in the secretory cells can be observed (Lowy et al. 1985, 1987; Martin and Shuttleworth 1994a; Martin et al. 1994, Hildebrandt 1997). Each of these signaling pathways affects ion channel activity which mediates salt secretion or modulates secretion rate (Martin and Shuttleworth 1994b; Martin et al. 1994), but may, in addition, have critical roles in triggering the adaptive processes of cell proliferation and cell differentiation (Hildebrandt 1997, 2001). In a previous study (Martin et al. 1994), simultaneous activation of muscarinic and b-adrenergic receptors synergistically stimulated secretion-related energy metabolism (measured as bumetanide-sensitive oxygen-consumption rate). This result was the first indication of a potential cross-talk of these signaling pathways at the cytosolic level. All signal transduction processes in cells are organized in the form of cascades in which certain components may interact on different levels. Since every single biochemical process in the cell needs energy in the form of ATP, we decided to use the rate of carbohydrate metabolism (measured using a microphysiometer) as the parameter to search for potential cross-talk between different signal transduction pathways. In this study, we set out to compare changes in energy metabolism due to activation of either receptor mechanism in partially differentiated (‘naive’) and fully differentiated nasal gland cells and to study the relative portions of metabolism that were related to salt secretion versus other energy-demanding processes. Isolated nasal gland cells either from naive ducklings without any previous experience of osmotic stress or from animals osmotically stressed by replacing their drinking water with 1% NaCl solution for 48 h, were tested for their responses to muscarinic, b-adrenergic or VIPergic activation.
Materials and methods Materials Chemicals for preparing buffered salt solutions for cell isolation and microfluorimetry were obtained from Roth (Karlsruhe, Germany). Minimal essential medium as running buffer for microphysiometry was
obtained from Invitrogen (Paisley, UK). Carbachol, isoproterenol, bumetanide, atropine, propranolol, forskolin, TEA and barium chloride were obtained from Sigma (Deisenhofen, Germany). Mammalian vasoactive intestinal peptide (VIP), which has been shown previously to be highly effective in nasal gland cell receptors (Lowy et al. 1987; Martin and Shuttleworth 1994a) was obtained from Calbiochem (La Jolla, CA, USA). Low-melting agarose was obtained from PeqLab (Erlangen, Germany). Calcium indicator indo-1/AM was bought from Molecular Probes (Eugene, CA, USA).
Experimental animals and nasal gland cell isolation One-day-old ducklings (Anas platyrhynchos) were purchased from a commercial hatchery, placed in cages in a room with a 12:12 h light–dark cycle, and fed chick starter crumbs with drinking water ad libitum. To induce adaptive proliferation and differentiation of secretory cells in the nasal gland, birds were given 1% NaCl solution to drink for 48 h before being used for cell isolation. All experimental animals were 6–15 days old. Ducklings were killed by decapitation, and the nasal glands were rapidly dissected out. The glands were cut into 1-mm cubes. Cell isolation from the tissue cubes was performed using trypsination as described previously (Shuttleworth and Thompson 1989). Isolated cells maintain their biochemical and functional characteristics for several hours after isolation. Their ability to respond to extracellular stimuli with metabolic activation related to salt secretion has been demonstrated several times in different laboratories (Lowy et al. 1987; Hildebrandt and Shuttleworth 1991; Martin and Shuttleworth 1994b).
Microphysiometry Determination of metabolic rates under steady-state conditions of the cells was performed using a fourchannel Cytosensor Microphysiometer (Molecular Devices, Ismaning, Germany). Experimental cells were enclosed in a capsule cup that was perfused with a weakly buffered running medium (Minimal Essential Medium, Invitrogen 41500-018). Hydrogen ions produced as end products of carbohydrate metabolism and extruded from the cytosol were flushed away with the running medium. During brief stopped-flow periods (40 s every 2 min), protons accumulated within the cell chamber (2 ll total volume) and slightly acidified (maximum 0.01 pH units) the medium around the cells. The resulting pH shifts were measured using a lightaddressable, pH-sensitive sensor chip mounted at the bottom of the cell chamber. The rate of acidification during these regular stopped-flow periods was taken as an indicator of metabolic rate in the experimental cells.
463
Isolated cells were sedimented by centrifugation (1,000·g; 3 min) and resuspended in 0.3% low-melting agarose prepared in running medium at 40 C. Immediately afterwards, 10 ll of this cell suspension was placed on the surface of a porous (3-lm pore diameter) polycarbonate membrane in a capsule cup which was then briefly cooled to 4 C to allow the agarose to solidify. Each capsule cup contained between 500,000 and 800,000 cells. Capsule cups were inserted into the sensor chambers und perfusion of the chambers with running medium was started. During an initial period of 2.5 h, cells were allowed to recover from the isolation procedure and reach steady-state conditions. A valve in the perfusion line of each of the four cell chambers allowed the switch between running media with only a 10-s delay in the arrival of the alternative medium in the cell chamber. Using a standard stimulation protocol within a total experimental time of 120 min, cells were exposed to agonists or antagonists of plasma membrane receptors or compounds affecting transport activity of cells or intracellular signaling pathways. Carbachol (1 lmol/l) was used as a muscarinic agonist. This concentration of carbachol activates nasal gland cells submaximally and most likely corresponds with the physiological situation of nasal gland cells in osmotically stressed animals in vivo (Shuttleworth and Hildebrandt 1999). A submaximal concentration (10 lmol/l) of isoproterenol was used to activate b-adrenergic receptors. Mammalian VIP (20 nmol/l), which has been shown previously to activate avian plasma membrane receptors specific for this peptide hormone (Lowy et al. 1987; Gerstberger et al. 1988; Martin and Shuttleworth 1994a), was used instead of avian VIP, which is not commercially available. Bumetanide (10 lmol/l) was used to block the activity of the Na+/K+/2Cl co-transporter, which is an essential component of the secondary active chloride secretion mechanism in nasal gland cells (Lowy et al. 1989). Inhibition of this transporter immediately stops salt secretion. As a consequence, pumping of sodium and potassium ions by the Na+/K+-ATPase ceases, which, in turn, reduces the need for aerobic ATP generation. Any residual metabolic activity seen in the presence of bumetanide must therefore be due to biochemical processes independent of salt secretion. Forskolin (1 lmol/l) was used to directly activate adenylyl cyclase and to elevate the cAMP concentration in cells without receptor occupation. Perfusion of cell chambers with pure running medium terminated drug actions on the cells within 5–10 min, so that use of receptor antagonists could be avoided. Cells in one of the four cell chambers were left unstimulated during each experiment and treated with agonists at the end of the experiment to control for vitality and responsiveness of the cells. Experimental data were normalized to the means of metabolic rate data during the last hour before activation of cells.
Fluorimetric measurements of cytosolic calcium concentrations Cytosolic calcium concentrations in nasal gland cells were measured as described previously (Shuttleworth and Thompson 1989). Briefly, freshly isolated nasal gland cells were loaded with the calcium indicator dye indo-1 and resuspended in HEPES-buffered saline in the cuvette of a Spex Fluoromax-3 spectrofluorimeter (Jobin Yvon, Longiumeau, France). The excitation wavelength was set to 350 nm and emission was recorded at 400 and 495 nm at 2 s intervals. Calibration and calculation of calcium concentrations were performed as described (Grynkiewicz et al. 1985). Termination of carbachol- or isoproterenol-mediated cell stimulation was achieved by adding the muscarinic antagonist atropine (100 lmol/l) or the adrenergic antagonist propranolol (50 lmol/l), respectively.
Statistics Normalized metabolic-rate data of different cell preparations were used to calculate means and standard deviations (SD). Statistical analysis of means at certain points during the plateau phases in the continued presence of agonists was performed using either Student’s t-test or one-way ANOVA (GraphPad InStat-Software, v. 3.06). Statistical significance was claimed at P
