ASSTRACT Cholinergic receptors controlling iridescence in the iridophores of Loligo opalescens were found to be muscarinic. Selective agonists and antagonists were applied, and responsiveness was measured by a change in iridescence. Carbamulcholine and acetyl-beta-methgl-choline (ABHE) are group A muscarinic agonists that elicited positive responses. Effective group B agonists included carbamul-beta-methyl-choline (CBMC) and oxotremorine. Atropine, scopolamine, and pirenzipine are muscarinic antagonists that effectively hindered iridescence, while tubocurarine, a nicotinic antagonist, did not block iridescence. Agonist groupings are based on their ability to turn over phoshatidglinosito (Phl). The results of these agonistsfantagonists combined with the effect of Li and subsequent addition of myo-inositol, show that Phi turnover mag be involved. Li' selectively inhibits a step in the Phl pathwau. Lif was found to decrease iridescence and addition of muo-inositol seemed to cause its reappearance. Ca seems to be involved, since addition of an agonist in zero Ca' sea water did not result in iridescence. Addition of forskolin and caffeine resulted in a decrease in iridescence, impluing that high cAMP levels inhibit the response. From this data four receptor models were constructed. INTEODUCTION Iridophores, specialized cells in the skin of squid produce structural iridescence bu either diffraction or thin film interference mechaniems (Schäfer, 1938, Brocco and Cloneg, 1980). Two types of iridophore cells are recognized: iridocutes, cells that apparently produce structural colors in a manner similar to a diffraction grating; and reflector cells, which produce structural colors via thin-film interference (Brocco and Cloneg, 1983). In Loliga, reflector cells are located chiefly above the iris, on the ink sac and other internal organs. The iridocutes are found in the skin over the whole body of the squid, and are arranged in donut-shaped splotches in the third dermal lager of the dorsal mantle. Iridescence of iridocytes seems to be caused by the organization of iridosomal platelets of high refractive index located on the periphery of the cell and surrounded by infoldings of the plasmalemma (Brocco and Cloneg, 1980). These platelets are aligned on edge, perpendicular to the integument of the squid (Schäfer, 1938, Kawaguti and Ohgishi, 1962, Hirow, 1972). Iridocutes differ from reflector cells not only in the mechanism bu which they produce structural iridescence, but also in their ability to change their iridescent properties. The neurotransmitter acetulcholine (AChl causes a change in the strength of iridescence (Cooper and Hanlon, 1966), although no one has get successfully identified the innervation of the iridocute dermal lager. This study was conducted to characterize the nature of the receptor responsible for that effect, and the possible signal transduction mechanisms that mag help mediate the structursl changes involved in iridescence. HATERIALS AND METHODS ACh, atropine, tubocurarine, forskolin, caffeine, LiCl, EGTA, HEPES and muo-inositol were obtained from Sigma Chemical. Carbamulcholine, ABMC, CEHE, oxotremorine, pirenzipine, quinuclidinul benzilate (ONB), and scopolamine were the kind gift of Anthong Horielli. Loligo opalescens were collected from Montereg Bag and maintained in round holding tanks 2 m in diameter. Photographs were taken with Kodschrome 400 and Ektachrome 400 using an Olympus UM¬ Camera mounted to an Olumpus upright microscope. The 10k occular lens was combined with either a 40x water immersion Zeiss lens, or a lox American Optical lens, making overall viewing magnification either 400x or 100x. Fiber optic incident lighting was used as the single light source. Skin preparations were obtained by excising patches of the iridophore dermal lauer from the dorsal mantle region and pinning them in a 3 mm diemeter, Sulgard coated dish. The dish was filled with artificial ses water (500 mM Nacl, 12 mMMgi, 10 mac, 10 m KCI, 10 mM HEFES, and enough KOH to bring the pH to 7.8). The sppropriate drugs were diluted in saline and spplied bu exchanging the solution in the beth. The zero Ca solution was prepared by substituting Ng for Ca and adding 2 mM EGTA. Preparations were placed under the microscope, and a single iridophore donut was observed. A baseline level of iridescence was obtained bu adjusting the light to produce a maximum amount of iridescence (incident light made an angle of approximately 30° with the horizontal). A photograph was then taken to objectively evaluate this level of iridescence. Iridescence was judged by three factors: 1) Area: the number of cells that were iridescent in the donut vs those that were not, 2) Intensity: intensity of iridescence, and 3) Color: the color of the light seems to be related to the strength of the response. White corresponds to completely off, Blues and Greens represent a semi-iridescent stage, and Reds, Golds, and Oranges indicate a more fullu iridescent form of the iridophore. The various drugs were added topicallu to the bath, and changes in levels of iridescence were noted. Photographs were taken at times of 1, 5, and 10 minutes after addition of each chemical. Photographs were taken at later times depending on the response to the drug. RESULTS Characterization of the Receptor Various cholinergic agonists and antagonists were tested at a final concentration of .1 mM. Agonists were judged to induce iridescence if application to the bath resulted in an increase in iridescence. Theg were judged ineffective if addition to the bath caused no change, and subsequent washing followed by application of ACh resulted in a positive change in iridescence. Antagonists were considered ineffective if the combined application of the antagonist and ACh resulted in a change in iridescence that could not be enhanced by subsequent washing and application of ACh alone. The antagonists were judged to be successful blockers if iridescence could be enhanced by washing and addition of ACh. See Table 1 Cholinergic receptors can be divided into two tupes: muscarinig and niocotinic. The non-specific cholinergic agonist carbamylcholine uielded a positive result, as would be expected in either a muscarinic or nicotinic system. Specific muscarinic agonists, ABMC, CBME, and oxotremorine, gave positive responses to varging degrees. Host of the specific muscarinic antagonists that were tested, atropine, scopolamine, and pirenzipine effectively competed with the agonist and hindered iridescence. From these results, it appears that a muscarinic system is present in the iridocytes of Loligo. Muscarinic agonists have been divided into two groups depending on their effect on Phl turnover (Fisher et al, 1983). Group A agonists trigger Phl turnover to a large degree, while Group B agonists trigger it to a somewhat lesser extent. Oftentimes, the degree of response to the various muscarinic agonists can indicate whether or not Phi turnover is involved in the system. Solely on the basis of agonist binding experiments, the results of this study are too preliminaru to assess the role of Phl turnover in iridescence. Oxotremorine, a Group B agonist, did elicit a less dramatic response than ang of the Group A agonists that were applied, but CBMC, also a Group B agonist, gave just as strong a response. Other methods of determining Phl involvement were required. Effect of Li', myo-inositol The cation, Li', provides one of the few ways to selectivelu inhibit the cycle of Phl turnover, thereby providing a method for testing the role of Phl involvement (Drummond et al, 1986). LiCl (lOmM) was added to a bath containing a semi-iridescent preparation. After 30 minutes, no change in the level of iridescence was observed. After 45 minutes, iridescence began to fade, and after an hour, there was a substantial decrease in iridescence. 10 m myo-inositol was then added to the bath, and iridescence returned within 30 seconds to a minute. This procedure was repeated twice. The first repeat showed no fading of virtually full iridescence after an hour, and addition of myo-inositol did not increase it. The second repeat resulted in fading of a semi-iridescent preparation after 30 minutes, but addition of myo-inositol did not have ang effect. Effect of 2ero Ca Ses Water There is some evidence that Ca' is required for iridescence (Cooper and Hanlon, 1986), and muscarin has been shown to affect Ca' fluxes in some sustems (Parod et al, 1980). After bathing the iridophore in zero Ca' sea water for 50 minutes, addition of ACh produced no change in iridescence. The prep was then washed with regular artificial ses water, and subsequent addition of the agonist resulted in a significant increase in iridescence. This procedure was repeated on a different preparation and the results were similar, with slightlg less iridescence after washing and application of ALh. Effect of Forskolin, Caffeine Muscarinic receptor mechanisms, in some cases, have been shown to act by decreasing the sunthesis of cAMP (Birdsall et al, 1980). Forskolin (.1 mt1) and caffeine were added to a bath containing a semi-iridescent preparation. After 5 minutes, iridescence decreased dramaticallu. After 30 minutes, iridescence came back slightly but was still far from the original level. Addition of carbamylcholine to the bath at this stage resulted in a sharp increase in iridescence. This procedure wos repeated twice on different preparations, a fully iridescent preparation and another semi-iridescent preparation, each time resulting in a virtually complete loss of iridescence, that was reversed bu application ofImM carbamglcholine. DISCUSSION The iridescence of iridocutes seems to be controlled by a muscarinic-tupe receptor system that involves Phi turnover, a rise in intracellular Ca' levels, and a decrease in cAMP levels. The responses to a number of muscarinic agonists and the block of the response by Specific muscarinic antagonists indicates that a muscarinic receptor is involved. Muscarinic receptors have been proposed to exist in two subtipes, MI and M2, each able to express more than one conformational state (Fisher and Bartus, 1985). pirenzipine specificallu blocks the Mi subtupe (Fisher and Bartus, 1985), and on the basis of this result, it seems that iridocytes possess at least the MI tupe of muscarinic receptor. The distinction between the two muscarinic receptor subtupes based on the effects on Phl turnover is still vaque (Fisher and Bartus. 1985). Comparisons of the effectiveness of Group A and B agonists also gield inconclusive evidence whether Phl turnover is involved in the induction of iridescence. A role for Phl turnover is suggested in iridescence on the basis of the experiment involving Li and myo-inositol. Addition of Li' to cells has been shown to induce a sharp decrease in myo-inositol concentrations, while causing a drastic increase in myo-inositol 1-phosphate levels, apparently by inhibiting the enzyme myo-inositol 1-phosphatase (Allison et al, 1976). Ph breskdown products that are known secondarg messengers, IP- and DAG occur well before the myo-inositol 1-phosphatase step in the pathwad, and Li' action has been hupothesized to occur by depleting the Fhi pool of mgo-inositol, therebg rendering it unable to regenerate Phl, and placing the turnover mechaniem into check (Berridge et al, 1982). The cell can either absort myo-inositol through the plasma mernbrane or sunthesize it from glucose (Drummond et al, 1986). The fading of iridescence upon addition of small concentrations of Li' can be correlated with a depletion of myo-inositol from the Phl pool, since addition of myo-inositol seems to restore iridescence, at least to some degree, The preparation that did not fade after an hour perhaps had an inordinatelu high Phl pool that was not get depleted. The Fhl breakdown product, IFz, may act by releasing Cat from intracellular stores. Roger Hanlon and Kag Cooper have been investigating the role of Ca“ in iridocyte iridescence of L. Brevis. Addition of the Ca ionophore 423187 to a bath solution containing Ca" caused an increase in iridescence. Verapamil, a Ca channel blocker, inhibited agonist-induced iridescence to a great degree. Addition of the agonist ACh to a zero Ca bath solution resulted in a brief increase in iridescence that could not be maintained (Hanlon and Cooper, 1986). The zero Ca sea water experiment executed in this study implied that Ca" was required in the external medium to induce iridescence. However. the 50 minute bathing period might be ample time for intracellular stores to become depleted; therefore, Hanlon's experiment with zerg Car sea water mag indeed support the theory of releasing Ca from intracellular stores. Agonist binding resulted in a brief increase in iridescence, perhaps associated with this intracellular Ca release. Internal Ca pumps would then have removed Ca' from the cell where it would be chelated in the external medium. cAMP levels may also have an important modulating effect on the iridescence of squid iridocutes. Huscarinic agonist binding has been shown, in some cases, to decrease sunthesis of CAMP (Birdsall et al, 1983). The results of the forskolin and caffeine experiment may indicate that cAMP levels cannot be too high if the iridocute is to maintain iridescence. This does not necessarily imply that agonist binding in iridocytes decresses cAMP levels, but it does seem to show that high levels of intracellular CAMF produce some kind of inhibitorg effect on iridescence. In general, muscarinic agonists affect cAMP levels by linking to the G, protein that inhibits adenglate cyclase, therebg decressing sunthesis of CAMP. A useful follow up experiment would then be to test the effect of cholers and pertussis tokins on iridocyte iridescence, since theg are known to affect cAMP sunthesis by acting on the G protein. Consideration of these aspects of intracellular mediation of the receptor signal, has led to four receptor models that attempt to link the receptor to Phl turnover and to inhibition of cAMP sunthesis. (See Figures 1 to 4.) Theg assume that the Car rise is under control of IPz, one of the products in the Phl cycle. EECEPTOR MODEL 1: The receptor is linked directlg to Phospholipase C, the enzyme involved in the initial stages of Fhl turnover. Activation of this enzyme sets this pathwag into motion. The products of Fhl turnover then either stimulate the G; protein to inhibit adenglate cuclase and decrease overall cAMP synthesis, or theg inhibit adenglate cyclase directly. IP causes Ca- release to turn on iridescence. EECEETOP MODEL 2: Two receptors are used: one directig linked to Fhospholipase C, and the other to the G, protein. The two receptors may have different binding affinities, affecting each process to varging degrees. There is some evidence that muscarinic receptors mag each interconvert to high and low affinity forms, adding another aspect to requlation and allowing more flexibilty in the system. FECEPTOR HODEL3 There is one receptor directlg linked to both Phospholipase C and the G;. Agonist binding would cause a joint effect, triggering both aspects of the system at once. PECEPTOR HODEL 4: Both Fhospholipsse C and the G, protein are under control of a single receptor, but the receptor is linked to only one of them at ang given time. The receptor interconverts between high and low affinitu forms and floats in the membrane of the cell, showing preference for one integral membrane protein or the other depending on the receptor's state. These models deal only with the earliest parts of the signal. It seems likelu that the external signal is transduced into an enzymatic process. The muscarinic signal mag utilize cyclic quanine phosphate products to either inhibit cAMP production or stimulate a specific enzume. DAG, one of the products of Phl turnover, effects change through protein kinase C, allowing for phosphorglation of various enzumes, and switching them to an active state. Ca effects usuallg occur through some sort of Ca* binding protein, like calmodulin. No one has conclusively proven how the external signal elicits an ultrastructural change in the iridosomal platelets. Hicrotubule and microfilament mechaniems have been proposed, but further experimentation is needed. In conclusion, muscarinic receptors are involved in the iridescence of iridocytes, and they appear to act through the cycle of Phl turnover, rise in internal Ca' levels, and by hindering the synthesis of CAMP. would like to thank Stuart Thompson for his support and quidance, Anthonu Morielli for his time and drugs in the characterization of the receptor, Chris Patton for his technical expertise, Bruce Hopkins for providing all the squid I could possibly kill, and mang hearty thanks to Roger Hanlon for his helpful advice and the use of his unpublished data. Lastlu, I would like to thank the entire class of Sio 175H for one helluva good time. REFERENCES Allison, J. H., Blisner, M. E., Holland, W. H., Hipps, P. P. and W. R. Sherman. 1976. Biachem. Biophus. Res. Commun 71, 664-670. Berridge, M.J., Downes, C. P. and M. R. Hanleg. 1982. Biachem 206, 587-595 Birdsall, M. J.M., Berrie, C.P., Burgen A. S. V., and E. C. Hulme. 1980. in Feceptors for Neuratrensmitters and Feptide Hormenes (Pepeu, G., Kuhar, M. J., and J. S. Enna, eds.), pp. 107-116, Raven Press, Hew Tork. Brocco, S.L. and R. A. Cloneg. 1980. Reflector Cells in the skin of Octopus dofleim. Cell Tissue Res. 205: 167-136. Cloneu, R. A. and S. L. Brocco. 1983. Chromatophore organs, reflector cells, iridocutes and leucophores in Cephalopods. Amer. 2ool. 23. 581-592. K. H. and R.T. Hanlon. 1986. Correlation of iridescence with Cooper changes in iridophore platelet ultrastructure in the squid Lolliguncula brevis. J Exg. Bicl Commin. Drummond, A. H., Joels, L. A. and P. H. Hughes. 1986. The interaction of lithium ions with inositol lipid signalling sustems. Biochem. Soc. Transact. 15:32-35. Fisher, S. K., Klinger, P. D and B. W. Agranoff. 1983. Muscarinic agonist binding and phosholipid turnover in brain. J. Biol. Chem. 258, 7350-7363 Fisher, S. K. and R. T. Bartus. 1985. Regional differences in the coupling of muscarinic receptors to inositol phospholipid hudrolisis in quinea pig brain. J. Neurochem. 45, 1085-1095. Hanlon, R. T. 1982. The functional organization of chromatophores and iridescent cells in the bodg patterning in the body patterning of Loligo Flei (Cephalopoda. Huopsida). Halacologia 23(1: S9-119. Kawaguti, S. and S. Ohgishi. 1962. Electron microscopic study of the iridophores of a cuddlefish, Segia esculente. Biol. J. Okamaga University 8: 115-129. Mirow, S. 1972. Skin color in the squids Laliga Fealiiand Laliga Cralescens. II. Iridophores. 2. Zellforsch. 125: 176-190. Parod, R. J., Leslie, B. A. and J. W. Putneg. 1980. Am d Physiol 239, 699-105 Schäfer, W. 1938. Bau Entwicklung und Farbenentstehung bei den Flitterzellen von Sepis officinslis. 2. Zellforsch. 27: 222-245. EIGURE LEGEND Eigure 1: Single link model. Single receptor linked to Phospholipase C. Phospholipase C activation indirectly decreases cAMP levels. Eigure 2: Dual receptor model. Two receptors, each linked to either Phosholipase Cor the G, inhibitorg protein. Eigure 3: Double link model. Single receptor directlg linked to both Phosholipase C and the G; inhibitorg protein. Eigure 4: Floating receptor model. Single receptor links to either Fhosholipase C or the G, inhibitorg protein, depending on the receptors conformational state. Plate 1: Lateral view of an iridophore donut in the off position. Some Blues and Greens are visible. Plate 2: 30 seconds after addition of carbarnglcholine. The donut shows semi-iridescence. More Blues and Greens are visible. Elate 3: Five minutes after addition of carbamglcholine. The donut now shows a ruch greater degree of iridescence. Many more iridocutes are on, and they show predominantly Reds and Oranges. Some Blues and Greens are still visible. ABLE IRIDESCENCE RESPONSE TO VARIOUS AGONISTS/ANTAGONISTS IRIDESCENCE EESC HUSCAPINIC AGONIST GROUP *** cetyl Cholne *** Carbamyl Cholne *** Acetyl eta ethyl holne (O Groue 8 *** Carbamyl Beta Methyl holne () Oxotremorne ANTAGONISTS HUSCARINIC trone Scopolamne nucyl enate (INB) Pirenzipire NICCTINIC fubcurare * : INDUCED IRIDEECENCE - : BLOCKED IRIDESCENCE : NO ESFECT aost a aso RECEPTOR MODEL AGONIST —— Gi PHOSPHOLIPASE C ADENYLATE CYCLASE — dire link - - indire link IGURE RECEPTOR MODEL 2 AGONIST PHOSPHOLIPASE C FIGURE 2 AGONIST RECEPTOR MODEL 3 AGONIST PHOSPHOLIPASE C Gi GURE 3 RECEPTOR MOOEL 4 AGONIST PHOSPHOLIPASEC P floating link FIGURE 4