- Email: [email protected]
The role of cyclic AMP in the control of elasmobranch ocular tapetum lucidum pigment granule migration
~2-6989/88 $3.00 + 0.00 Copyright a 1988 Per~mon Press plc
VisionRes. Vol. 28_ No. 12, pp. 1277-1285, 1988 Primed in Great Britain. .411rights reserved
THE ROLE OF CYCLIC AMP IN THE CONTROL OF ELASMOBRANCH OCULAR TAPETUM LUCIDUM PIGMENT GRANULE MIGRATION ASHLEY
R. HEATH and
HEATHER M. HINDMAN
Department of Biomedical
Abstract-We have studied the effect of adding CAMP to dogfish ocular tapetum lucidum tissue maintained in uifro. Normally, under conditions of dark-adaptation, tapetal pigment granules attain an aggregated state, exposing highly reflective cellular plates (which contain crystalline guanine) to incident illumination. The reflected light from these plates is believed to function in the visual process by increasing photoreceptor photon capture under low light conditions. Upon illumination, the pigment granules attain a dispersed state, occluding the reflective surfaces. We found that this occlusion may be mimicked in dark-adapted tissue in vitro by adding CAMP, or by the use of agents believed to increase intracelhdar CAMP concentrations such as forskolin and IBMX. Additionally, aggregation of pigment in light-adapted tissue transferred to darkness is inhibited by such agents. The results of this and previously published studies indicate that the processes of pigment aggregation and dispersal in ttitio are under the control of the neural retina through the probabie mediation of either hormonal or direct neural communication. Tapetum lucidum
Melanophore
CAMP
Elasmobranch
INTRODUCTION
The tapetum lucidum is an ocular adaptation to low light level environments found in many vertebrate species including both teleost and elasmobranch fish, some reptiles, and mammals (Walls, 1942). A familiar example is the tapetum found in the cat’s eye which gives rise to the spectacular eyeshine observed in direct light at night. Although the function of the tapetum lucidurn, to increase the chance of photon capture by the visual cells, appears to be universal in all species, the cellular mechanism by which the tapetum is implemented varies. A general feature is that the tapetum itself is a highly reflective cellular layer. However, tapeta may be physically located in the cells of the retinal epithelium immediately adjacent to the neural retina, or they may be found outside of the neural retina-retinal epithelial complex towards the scleral side of the eye. In the majority of cases, the tapetum is fixed in the sense that the reflectivity of its surface is not modulated in any fashion to take account of ambient light conditions. In the case of many elasmobranch fishes, however, an ocular tapetum lucidum has evolved outside of the neural retina whose
reflectivity is controlled by the light-dependent, variable migration of melanin pigment granules in specialized accessory melanophores. The elasmobranch tapetum was described in early work by Briicke (ISSS) and Franz (1905). More recently, considerable expe~mental work was carried out by Denton and Nico1(1964), by Nicol (1964, 1965, 1975) and by Kuchnow (1969a, b), Kuchnow and Gilbert (1967) and Kuchnow and Martin (1970). A review article has also appeared dealing with this as well as other aspects of elasmobranch vision (Gruber and Cohen, 1978). To our knowledge, no more recent studies of the tapetum have been published. The earlier authors have given us the following picture of the elasmobranch tapetum lucidum, derived from studies on various species, but in particular from work on the Spiny Dogfish, Squalus acanthias. The tapetum is found immediately adjacent to the choroidal vasculature of the eye. Listing the major cell layers and proceeding from the scleral coat, one finds first the tapetum, then the choroid, then the retinal epithelium and finally the neural retina proper. Unlike most vertebrates, the elasmobranchs possess generally unpigmented retinal epithelia. There is, therefore, no
1277
1278
ASHLEYR. HEATH and HEATHERM. HINDMAK
si~ni~~dnt obstructi~~n to the passage of light through to the tapetum. The tapetum itself consists of two types of cell. T’he reflective cells themselves. or guanophores, contain highly ordered plates of crystalline guanine (Pirie and Simpson, 1946; Denton and Nicol. 1964). Associated with them is a second set of ceils, pigmented melanophores. The nuclei of these cells are found towards the scleral coat. Processes extend from the nuclear region in the direction of the choroid. Once they reach this layer they spread out parallel to it forming a series of terminal feet. In the light microscope, these terminal feet appear to form a continuous sheet across the surface of the vascular layer (see Figs I and 3). In conditions of darkness, or low light levels generally, the pigment granules found in the tapetal melanophores migrate towards the cell nuclei, away from the neural retina. This has the effect of removing pigmentation from the terminal feet and allowing light to reach the g~lanophores. Reflection from the tapetum then occurs. As the light level increases to normal photopic conditions. the pigment movement is reversed until ;I layer of pigment forms in the terminal feet between guanophores and choroid. In this condition. reflection is reduced to minimal levels, This process was shown to take approx. 2 hr to proceed to completion in either direction (Nicol, 1965). The cellular mechanism by which this control of tapetal reflection is accomplished is unknown. In the present paper we begin to address the question of the control of tapetat pigment distribution by reporting the effects of cyclic AMP (cAMP) and related agents on the process in ~,irro.
EXPERIMENTAL
NlETHODS
Adult, male Spiny Dogfish, Squalus acanthias, of 70-80 cm overall length, were obtained from Frenchman’s Bay, Maine, either by netting or by capture on trot-lines. Animals were maintained for periods of up to 5 days in running, fresh, sea water prior to use. They were not fed during this period. All experiments described in this paper were carried out on isolated ocular tissues in citru. The fish was first dark-adapted for a fixed period of 2 hr. Using illumination from a Wratten 2 (red) photographic dark-room lamp, the fish was decapitated and the eyes carefully enucleated. Eyes were then opened,
and the anterior segment and lens discarded. The globe was then hemisected using razor blades. Vitreous humor was carefully removed from the eyecup by repeated application of small wicks constructed from ‘Kimwipes’“. Squafus possesses a very thick vrtreous anti every effort was made to removi:
The tissue pieces, consisting ot mtact sciera, tapeturn, choroid, retinal epithelium and neural retina. were transferred in darkness to small flasks containing 2.5 ml of an elasmobranch Ringer’s solution. This had the following composition (mM): NaCi. 280; KCI, 5; CaClz, 2; MgSO,, 2; KHzPO,, 1.25; NaHCO,, 3% urea, 350 and D-glucose, 10. The Ringer was gassed with 95% 02;5% CO* before use and had a final pH of 7.4. -7.5. The flasks (four in number) were held in a Dubnoff shaking metabolic incubator and the tissue pieces (3-4 per flask) were maintained in darkness unless otherwise stated below. 95% 0,/.5% CO, was continually passed through the flasks at a rate of 1.75-2.0 std cu ft/hr]Rask. In a typical experiment designed to investigate the effect of a particular chemical agent. tissue was incubated for a period Of 2 hr. In other cases, where noted below, tissue was removed from incubatio~l at various time points, In all cases, one flask with its darkadapted tissue pieces was left in darkness for the duration of the experiment with no added effector agents, These tissues are referred to as dark-adapted controls. Experiments were carried out at a water-bath temperature of 20-22 c. Under these incubation conditions chemrcal
1279
Tapetal pigmentgranulemigrationand CAMP agents of interest were added as desired. N6,2’~-Dibutyryladenosine 3’ : S-cyclic monophosphate (db~AMP), nonderivatized CAMP and N?,2’-O-Dibutyrylguanosine-3’: S-cyclic monophosphate (dbcGMP) were added directly to the incubation medium. The phosphodiesterase (PDE) inhibitor 3-isobutyl- l-methyl xanthine (IBMX) and the adenylate cyclase stimulator forskolin were dissolved in dimethylsulfoxide (DMSO) prior to use. The final concentration of DMSO in the Ringer solution was always less than 0.594. DMSO at this concentration was found in control experiments to have no effect on tapetal pigment granule migration. Additionally. some experiments were carried out using the PDE inhibitors caffeine and theophylline. These were also added directly to the incubation medium. All chemicals were obtained from Sigma Chemical Co., St Louis. MO. Light microscopic and data analysis
Experiments were terminated by transferring tissues to cold (4 C) fixative consisting of 2.5% glutaraidehyde, 2% parafo~aldehyde and 17.25% sucrose in 0.16 M sodium cacodylate (Szamier and Ripps, 1983). Postfixation was in 1% osmium tetroxide, 17.25% sucrose in 0.16 M sodium cacodylate at room temperature. Tissues were dehydrated in an ethanol series, embedded in “Polybed 8 12” (Polysciences, Warrington, PA), an Epon substitute, and sectioned at 2 /lrn. Unstained sections were dried on to microscope slides and used directly for photography and data analysis. The sections were photographed using green light and Kodak “Panatomic-X” 35 mM film processed in fullstrength “Microdol-X”. Several photographic prints were prepared from each tissue piece sectioned. In order to quantify the degree of tapetal pigment migration, measurements were made from these prints of the total area of tapetal pigment found in a fixed measuring window of 1500 pm’ placed just above the basement membrane of the retinal epithelium. We accomplished the area measurement using a cursor and Jandel Scientific (Sausahto, CA) “Sigmascan” software running on a portable computer (“Compaq”), interfaced to a Numonies “2210” digitizing pad. To obtain the quantitative data presented here, 750 photographic prints were analyzed in this way. All area data are given in the form of the mean of absolute observed measurements with & 1 SD indicated.
RESULTS
distribution of tapetaf dark-adaptation
pig~lent
in light-
urzd
Light micrographs of the tapetum lucidum from Squalus showing the effects of light- and dark-adaptation on tapetal pigment distribution are presented in Fig. I. Figure 1A shows tissue from the dark-adapted control. Figure IB shows tissue from a companion flask which was exposed to room light for the duration of the 2 hr incubation. In the dark-adapted state, pigment granules may be observed aggregated around the melanophore nuclei. CytopIasmi~ processes extend from the nuclear region towards the choroid. These processes are largely devoid of pigment. In the light-adapted condition, pigment granules move along the length of the processes. In a thoroughly light-adapted preparation as seen in Fig. I B. the pigment moves into the terminal feet of the processes which run parallel to the choroid. In this way, light effectively is prevented from reaching the guanophores. It should be noted that the size and orientation of the tapetal melanophores varies with their location in the eye (Nicol, 1964, 1965). We have quantified the light-adaptation of tapetal tissue previously dark-adapted for the standard period of 2 hr. In these experiments, tissue pieces were transferred to flasks in the usual manner, one flask being covered and acting as the dark-adapted control. Using the red safelight for illumination, care was taken to ensure that the tissue was facing retina-side “up”. Room lights were then turned on, and samples of tissue withdrawn from the other flasks at fixed time points. Pigment area was then measured as a function of time in the light. The results are presented in Fig. 2. It will be seen from the figure that total tapetal pigment area increases with increasing time in the light. maximum pigment dispersion being reached in about 2 hr. Eflhct qf dbcAMP and related agents on tapetai pigment distribution in dark -adapted tissrte
We observed a striking effect upon adding dbcAMP to dark-adapted Squalus tapetai tissue pieces. Although the tissue was maintained in darkness, dbcAMP induced tapetal pigment dispersion characteristic of the light-adapted state. This effect was found to be dose-dependent. In Fig. 3 we present light-micrographs showing the effects on pigment distribution of a series of
1280
ASHLEY R. HEATH and HEATHERM. HINDMAN 6001
Fig. 2. Time course of light”adaptation of the tapetum in &UC&S ncanthim. Tissue was prepared in dim, red light and incubated in room light, as described in Materials and Methods. Each point shows the mean and one standard deviation of the area data. In this and in the following figures the numerals next to each point indicate the number of animals used in each case.
dbcAMP concentrations. The pigment may be seen acquiring an increasingly “light-adapted” distribution with increasing dbcAMP levels. No other agents were added in these experiments; dbcAMP alone was adequate to induce pigment dispersion, and it was not necessary to include phosphodiesterase (PDE) inhibitors as is often the case in such studies. In a series of experiments we have quantified the effect of increasing dbcAMP concentration, and the results may be seen in Fig. 4. Freshly isolated, dark-adapted Squalus tapetal tissue presents to the unaided eye a highly reflective surface. We found that darkening of the surface to a state characteristic of light-adapted tissue could be observed by eye after incubation in OS-l.OmM dbcAMP. (The use of unaided visual inspection of tapetal tissues as a rough check on the state of tapetal pigment was found to be a valuable aid in the conduct of this work. Subsequent histological examination of tissues judged visually always indicated a close correlation between qualitative and quantitative measures of pigment aggregation and dispersion. This method was also used in some cases by earlier workers, e.g. Nicol, 1965.) A maximum pigment area of 400-500 pm* was reached with 5 mM dbcAMP which correlates closely with the maximum area of pigment obtained through the effect of light (Fig. 2). In control experiments, nonderivatized CAMP (5.0 mM), adenosine (2.5 mM), and dbcGMP (5.0 mM) had no observable effect upon dark-adapted tapetal pigment. Other agents were also found to induce pigment dispersion in dark-adapts tissue. Both IBMX and forskolin, a PDE inhibitor and an adenylate cyclase stimulator respectively,
Fig. 4. Quantitative analysis of the dispersive effect ofCAMP on pigment granule movement in Squatus tapetum lucidum in sitro. Tissue was incubated in darkness and dbcAMP was added to the medium at the indicated concentrations. Tissue was prepared and processed and data anatyzed as described in Materials and Methods,
would be expected to elevate endogenous tissue CAMP levels. In confi~ation of this expectation, we observed that both agents produced pigment dispersion patterns similar to that seen from adding exogenous dbcAMP. The dose-response data for IBMX and forskolin have been quantified and the results are presented in Figs 5 and 6. Forskolin is effective in producing pigment dispersion that may be seen with the unaided eye at 10e6M. IBMX is approximately one to two orders of magnitude less effective. Additionally, a few experiments indicated that caffeine at a concentration of 5 mM, and theophylline at a concentration of 1 mM also induce pigment dispersion.
A number of experiments were carried out to examine the effect of dbcAMP on pigment aggregation induced by darkness in lightadapted tissue in vitro. Tissue was prepared in the usual way, except that a period of 2 hr light-adaptation (using standard laboratory Auorescent illumination) was given prior
Fig. 5. Effect of isobuty~methyixanth~ne ff BMX) on tapetal pigment in dark-adapted Squalusacanrhiastissue. Conventions are as given in previous figures.
t-e r Fig. 1. Unstained, light-micrographs showing S9uulrts acanthius tapetum lucidum in dark-adapted (A) and light-adapted (B) states from tissues maintained in O&V as described in the lext. n, nucleus of tapetal melanophore; c, choroid; re, retinal epithelium; r, neural retina. Scale bar, Loom. Arrows indicate positions of tapetal melanophore processes, into which pigment granules migrate on light-adaptation. Note that the size of the tapetal melanophores (including the distance from the tapetal melanophore cell body to the tips of the cytoplasmic processes) and the angle of orientation of the cytoplasmic processes, was observed to vary with location in the eye (see also Nicol, 1964 and 1965). Only the pigment granules move due to the effects of light- or dark-adaptation, the cytoplasmic processes themselves are fixed.
1281
Fig. 3. Effect of Dibutyryl cychc AMP (dbcAMP) on tapetal pigment distribution in dark-adapted tissue in rirrt>. The tissue was incubated for two hours as described in the text, in the presence of (A) no added dbcAMP (dark-adapted control), (B) 0.25 mM dbcAMP, (C) 0.5 mM dbcAMP, (D) f.0 mM dbcAMP, (E) 2.5 mM dbcAMP and (F) 5.0 mM dbcAMP. Note a progressive dispersal of pigment granules with increasing dbcAMP concentration. As noted in the caption to Fig. I, tapetal melanophore size and orientation varies with location in the eye. Unstained, i~ght-micrographs. Scale bar, IOirm.
Tapetal
pigment
granule
migration
and cAMP
1283
no caffeine and incubated for a further 3.5 hr. These pieces upon visual examination were found to have aggregated pigment, i.e. the effect of caffeine had been reversed. DISCUSSION
Fig. 6. Effect of forskolin on tapetal pigment in darkadapted Squahs acanfhim tissue. Conventions are as given in previous figures.
to sacrifice of the animal. Concentrations of dbcAMP varying from 0.125 to 5.0 mM were tested. The results were judged qualitatively using visual inspection of the tissue pieces to determine the gross state of tapetal pigment dispersion or aggregation. Tissue was scored as aggregated if it presented as dark-adapted with no observable degree ofpigment dispersion. The results are presented:” Table 1. In summary, we observed that dbi 4&l’ F at concentrations greater than 1.OmM’e$rely inhibits the aggregation of tapetal pign$:,observed upon darkadaptation. Rewrsihilit~ of chemkally dispersion in sitra
induced pigment
We do not have data indicating whether or not the effect of dbcAMP %af observed on &_I”.: tapetal tissue irz r*itro is r&ersrble. However, preliminary observations.~~~~ate that the pigment dispersing effect @&ffeine at 5 mM is reversible. Tissue from th?&e animals was incubated in the normal way in darkness in the presence of caffeine for 2 hr. Sample pieces were removed at this point and tapetal pigment was found by visual inspection (see above) to be dispersed. The remaining pieces were then returned to flasks containing fresh medium with ‘Table 1. Inhibition of phore aggregation Concentration of dbcAMP tested (mM) 0.125 0.25 0.50 1.0 2.5 5.0
~~f~f~/~f.~ u~,a~f~ju~
in darkness
Number of animals 3 3 4 3 3 3
tapetal melanoby dbcAMP
Tissue pieces with aggregated pigment: total tissue pieces 9/Y 719 4:11 118 019 O/IO
Note: these objervations are based upon visual scoring of tissue pieces as described in the text. In all cases where tapetal pigment was not aggregated it was observed to he in the dispersed, light-adapted state.
In this paper we have presented evidence that CAMP induces a state of tapetal pigment dispersion in vitro in dark-adapted tissues of Sq~aI~s acanthias, similar to that observed in the normal process of light-adaptation in the absence of exogenously applied effecters. Additionally, tapetal pigment disperses in darkness in the presence of agents that might reasonably be expected to raise the endogenous tissue level of CAMP, such as IBMX and forskolin, through their respective effects on cyclic nucleotide phosphodiesterase and adeny~ate cyclase. The aggregation of pigment that normally accompanies dark-adaptation is also inhibited by CAMP. A number of publications have appeared in the literature that directly link CAMP with the control of pigment granule movement in a variety of melanophore types (Bitensky and Burnstein. 196.5; Abe et al., 1969; Novales and Fujii, 1970; Wakamatsu e? ~1.. 1980; Novales, 1983; Namoto, 1985. 1987; Namoto and Yamada, 1987; Rozdzial and Haimo, 1986a, b; Morishita, 1987). This work has dealt in particular with dermal melanophores of teleosts and amphibians. Recent studies using carefully lysed dermal melanophores from Tilapia have shown that CAMP and ATP are required for melanophore dispersion, but that the absence of CAMP is a prerequisite for aggregation. The dispersion/aggregation cycle in these cells is also directly related to the phosphorylation and dephosphorylation of a 57 kD polypeptide (Rozdzial and Haimo, 1986a. b). Whether or not tapetal pigment granule dispersion in Squa~us is phosphorylationdependent remains to be seen. It has long been known that pigment aggregation in dermal melanophores may be under the control of adrenergic effecters such as epinephrine (Bagnara and Hadley, 1973). and it is supposed that these agents exert their effects through a reduction in intracellular CAMP levels. We currently have no information on the induction of tapetal melanophore pigment aggregation in Squafus by any means other than through the effects of darkness. Normal tapetal pigment aggregation in darkness may be obtained in citrv from previousfy light-adapted tissue under the
1284
ASHLEY R. HEATH and HEATHER
experimental conditions described in this paper. Preliminary experiments carried out by the present authors using light-adapted tissue have shown no effect of epinephrine on the dispersed pigment state. Presumably. in intact lightadapted tissues, sufficient increased endogenous levels of CAMP are maintained to counteract any possible effect of epinephrine. Interestingly~ in at least one class of pigmented cell, the retinal epithelial (RE) cell in teleost fish, CAMP has been demonstrated to have a pigment aggregating effect (Burnside and Nagle, 1983). In vertebrate animals possessing pigmented RE cells extensive reciprocal movements of pigment granules and photoreceptors, referred to as retinomotor movements, occur in response to light- and dark-adaptation. With respect to the retinal epithelium, these movements are analogous to elasmobranch tapetal cell responses to variation in illumination conditions. In teteosts, RE cell pigment granules disperse in the light and aggregate in darkness. In tissues from such animals maintained in the light in the presence of CAMP, pigment aggregation is observed. Thus, this response is directly opposite in direction to that seen in the tapetal cell and in most other types of melanophore. On the other hand, a pigment dispersing effect of CAMP on pigmented RE cells in amphibians was recently described (Astorino et al.. 1987). Although it is possible that we are dealing with a direct effect of CAMP on the tapetal melanophores in our irzvitro experiments, analysis of the cellular mechanisms controlling tapetal pigment movement in Squab is complicated by the results of early experiments carried out by Nicol (1964, 1965) on tapetal tissue from which the neural retina had been carefully removed. He was concerned to determine whether or not the tapetal melanophores themselves act as primary photosensitive effecters, A precedent for such an effector system was already known in the case of the elasmobranch iris constrictor muscfe, which responds directly to illumination (Young, 1933). and melanophores that respond to light, possibly due to variation in endogenous CAMP levels as a result of the action of lightactivated phosphodiesterases, are now wellknown in a number of other animal species (Wakamatsu et ul., 1980; Weber, 1983). HOWever. it was found that following retina removal the tapetal pigment always attained the dispersed state, Dispersal was independent of ambient light conditions and was not affected by a
M.
HINDMAN
variety of physiological treatments including addition of acetylcholine, epinephrine, pituitrin. variations of Ringer ionic composition (e.g. manipulation of calcium, magnesium and potassium levels). alteration of pH balance, and raising hydrostatic pressure. In a few instances, however. Nicol (1965) was able to maintain aggregation of pigment in tapeta from which the retina had been carefully removed and then replaced. although these experin~ents did not give consistent results. It appears that tapetal melanophores in Squalus are not photosensitive primary effecters and that in the dark-adapted state the neural retina itself must exert some kind of inhibitory influence over tapetal pigment dispersion. or must in some fashion actively promote pigment aggregation. The results of our experiments using exogenously applied CAMP indicate that the aggregation of pigment may be related to a decrease in endogenous CAMP as has been suggested for dermal melanophores (see refereqys .abovef. We suggest that the effect of~~~~w,th respect to the retina and tapetum is tolmcrease tissue levels of CAMP, thereby promc$ng” pigment dispersion In our experiments, by adding CAMP in oiwa to dark-adapted tissue, we have presumably bypassed the normal physiological action of the retina. (One might also speculate that CAMP regulates the release or synthesis of some other effector, but weJ ve no evidence on that point at present,) Exa&$ x.,how an increase in CAMP triggered by the ~~~~~se of the neural retina to light is translated in,tc+a cellular response at the level of the tapetu@, lucidum is not known. Although light-activated retinal cyclic nucleotide phosphodiesterases have been known to exist for some time (Bitensky rt ui., 1975) there is to our knowledge no evidence for the existence of light-activated adenylate cyclase systems in the elasmobranch retina. or in any other retina. In addition, any chemical message would have to pass two intervening celiular layers (the retinal epithelium and the choroidal vascuiature) in order to reach the tapetal cells by direct diffusion. Presumptive nerve terminals have been described on tapetal pigment ceils in another species of Dogfish (Must&s mustelus. Kuchnow and Martin. 1970) 50 we cannot exclude the possibility of a neural connection between retina and tapetum. Such an arrangement might give rise to increased tapetal cell. adenylate cyclase activity through the mediation of a neurotransmitter agent. At present, we are searching for evidence for the existence of such
Tapetal pigment granule migration and CAMP
neural connections tapetum lucidum.
in the Squalus acanthias
Aeknoi~ledgements-This work was supported by grants from the National Eye Institute (EY 06536) the Pennsylvania Lion’s Sight Conservation and Eye Research Foundation, Breastplate Laboratories of Houston, TX and by a fellowship to ARH from the Lucille P. Markey Trust. We wish to thank the Director and staff of the Mount Desert Island Biological Laboratory for their efforts in ensuring the successful completion of this work. We thank also Richard Stegen for providing technical assistance. REFERENCES Abe K., Robison G. A., Liddle G. W., Butcher R. W., Nicholson W. E. and Baird C. E. (1969) Role of cyclic AMP in mediating the effects of MSH, norepinephrine, and melatonin on frog skin color. Endocrinology 85, 674-682. Astorino A.. Flannery J. G., Marc R. and Basinger S. F. ( 1987) Effect of adrenergic agonists and cyclic nucleotides on pigment granule migration in the RPE of Runa pip&s eyecups in r&w. Invest. Ophthal. risuai Sri. (SuppI.) t8, 256. Bagnara J. T. and Hadley M. E. [ 1973) Chromutophores and Color Chunge. Prentice-Hall, Englewood Cliffs, N.J. Bitensky M. W. and Burnstein S. R. (I965) Effects of cyclic adenosine monophosphate and melanocyte stimulating hormone on frog skin in vitro. Nature, Lond. 208, 1X--1284.
Bitensky M. W., Miki N.. Keirns .I. J., Baraban J. A., Wheeler M. A.. Lacy J. and Marcus F. R. (1975) Activation of photoreceptor disc membrane phosphodiesterase by light and ATP: cyclic GMP as a modulator of sense receptor function. Adz.. cyclic Nucleotide Rex 5, 2 13.-240.
Briicke E. (1845) Anatomische Untersuchungen fiber die ~ogennanten leuchtenden Augen bei den Wirbelthieren. .4&i. Anat. Physiol, Lpz. Jahrg. (IS45), 387-406. Burnside B. and Nagle B. (1983) Retinomotor movements of photoreceptors and retinal pigment epithelium: mechanisms and regulation. In Progress in Refinai Research (Edited by Osborne N. and Chader G.), pp. 67-109. Pergamon Press, Oxford. Denton E. J. and Nicol J. A. C. (1964) The chorioidal tapeta of some cartilaginous fishes (Chondrichthyes). J. mar. Rio/. A.?,~.. i’.R
44, 219~.258.
Franz V. (1905) Zur Anatomie, Histologie und funktionellen Gestaltung des Selachierauges. Jena. 2. Naruw. Bd. 40, 697 840. Gruher S. H. and Cohen J. L. (1978) Visual system of the eiasmobranchs: state of the art 1960-1975. In Sensory Biology o/‘Sharks, Skatpv. and Rays (Edited by Hodgson E. and Mathewson R.). pp. I I-105. Office of Navai Research.
Arlington,
VA.
Kuchnow K. P. (1969a) Threshold for the elasmobranch tapetal-pigment response. Vision Res. 9, lS7- 194. Kuchnow, K. P. (1969b) Tapetal pigment response of elasmobranchs. Vision Res. 9, 8499854. Kuchnow K. P. and Gilbert P. W. (1967) Preliminary in viva studies on pupillary and tapetal pigment responses in the
128.5
lemon shark, Negaprion brevirostris. In Sharks, Skates and Rays (Edited by Gilbert P. W., Mathewson R. F. and Rail D. P.), pp. 465-477. Johns Hopkins Press, Baltimore, MD. Kuchnow K. P. and Martin R. (1970) Pigment migration in the tapetum lucidum of the elasmobranch eye: evidence for a nervous mechanism. Vision Res. 10, 825-828. Morishita F. (1987) Responses of the melanophores of the medaka, Oryzius lutipes, to adrenergic drugs: evidence for involvement of c(?adrenergic receptors mediating melanin aggregation. Camp. Biochem. Physiol. SSC, 6974. Namoto S. (1985) Effects of adenylate cyclase activating agents on pi~ent-aggregating action of lithium ions to fish melanophores. J. Sci. Hiroshima Univ. 32, 105-I 16. Namoto S. (I 987) A subtype of adenosine receptors mediating pigment dispersion in leucophores of the medaka: evidence for an A,-receptor. Camp. Biochem. Physiol. SSC, 75-81.
Namoto S. and Yamada K. (1987) Effects of forskolin, isoproterenol and lithium ions on leucophores of a teleost, Oryzius Iutipes: evidence for involvement of adenylate cyclase in pigment-dispersing response. Camp. Biochem. Physiol, 86C, 9 I-95. Nicol J. A. C. (1964) Reflectivity of the chorioidal tapeta of selachians. J. Fish Res. Bd. Can. 21, lOS9-11~. Nicol J. A. C. (1965) Migration of chorioidal tapetal pigment in the spur dog, Sguutus ucanthius. J. mar. biol. Ass., U.K. 45, 405427.
Nicol J. A. C. (1975) Studies on the eyes of fishes: structure and ultrastructure. In Vision in Fishes: New Approaches in Research (Edited by AIi M. A.), pp. 5799607. Plenum Press, New York, N.Y. Novales R. R. (1983) Cellular aspects of hdrmonally controlled pigment translocations within chromatophores of poikilothermic vertebrates. Am. Zool. 23, 5599568. Novales R. R. and Fujii R. (1970) A melanin-dispersing effect of cyclic adenosine monophosphate on Fundulus meianophores. J. Ceil Physiol. 75, 133-136. Pirie A. and Simpson D. M. (1946) Preparation of a fluorescent substrate from the eye of the dogfish (~~ua~us acanthjas). Biochem. J. 40, l&20. Rozdzial M. M. and Haimo L. T. (lPS6a) Bidirectional pigment granule movements of melanophores are regulated by protein phosphorylation and dephosphorylation. Cell 47, 1061-.1070. Rozdzial M. M. and Haimo L. T. (1986b) Reactivated melanophore motility: differential regulation and nucleotide requirements of bidirectional pigment granule transport. J. Cell Biol. 103, 2755-2764. Szamier R. B. and Ripps H. (1983) The visual cells of the skate retina: structure, histochemistry, and disc-shedding properties. J. Camp. Neurol. 215, 51-62. Wakamatsu Y., Kawamura S. and Yoshizawa T. (1980) Light-induced pigment aggregation in cuttured fish melanophores: spectral sensitivity and inhibitory effects of theophylfine and cyclic adenosine-3’,5’-monophosphate. J. Cell Sci. 41, 6574. Walls G. L. (1942) The Vertebrate Eye and its Adaptive Radiation. The Cranbrook Press. Bloomfield Hills, MI. Weber W. (1983) Photosensitivity of chromatophores. Am. Zool. 23, 495-506. Young J. Z. (1933) Comparative studies on the physiology of the iris. I. Selachians. Proc. R. Sot. 8. 112, 228-241.
VisionRes. Vol. 28_ No. 12, pp. 1277-1285, 1988 Primed in Great Britain. .411rights reserved
THE ROLE OF CYCLIC AMP IN THE CONTROL OF ELASMOBRANCH OCULAR TAPETUM LUCIDUM PIGMENT GRANULE MIGRATION ASHLEY
R. HEATH and
HEATHER M. HINDMAN
Department of Biomedical
Abstract-We have studied the effect of adding CAMP to dogfish ocular tapetum lucidum tissue maintained in uifro. Normally, under conditions of dark-adaptation, tapetal pigment granules attain an aggregated state, exposing highly reflective cellular plates (which contain crystalline guanine) to incident illumination. The reflected light from these plates is believed to function in the visual process by increasing photoreceptor photon capture under low light conditions. Upon illumination, the pigment granules attain a dispersed state, occluding the reflective surfaces. We found that this occlusion may be mimicked in dark-adapted tissue in vitro by adding CAMP, or by the use of agents believed to increase intracelhdar CAMP concentrations such as forskolin and IBMX. Additionally, aggregation of pigment in light-adapted tissue transferred to darkness is inhibited by such agents. The results of this and previously published studies indicate that the processes of pigment aggregation and dispersal in ttitio are under the control of the neural retina through the probabie mediation of either hormonal or direct neural communication. Tapetum lucidum
Melanophore
CAMP
Elasmobranch
INTRODUCTION
The tapetum lucidum is an ocular adaptation to low light level environments found in many vertebrate species including both teleost and elasmobranch fish, some reptiles, and mammals (Walls, 1942). A familiar example is the tapetum found in the cat’s eye which gives rise to the spectacular eyeshine observed in direct light at night. Although the function of the tapetum lucidurn, to increase the chance of photon capture by the visual cells, appears to be universal in all species, the cellular mechanism by which the tapetum is implemented varies. A general feature is that the tapetum itself is a highly reflective cellular layer. However, tapeta may be physically located in the cells of the retinal epithelium immediately adjacent to the neural retina, or they may be found outside of the neural retina-retinal epithelial complex towards the scleral side of the eye. In the majority of cases, the tapetum is fixed in the sense that the reflectivity of its surface is not modulated in any fashion to take account of ambient light conditions. In the case of many elasmobranch fishes, however, an ocular tapetum lucidum has evolved outside of the neural retina whose
reflectivity is controlled by the light-dependent, variable migration of melanin pigment granules in specialized accessory melanophores. The elasmobranch tapetum was described in early work by Briicke (ISSS) and Franz (1905). More recently, considerable expe~mental work was carried out by Denton and Nico1(1964), by Nicol (1964, 1965, 1975) and by Kuchnow (1969a, b), Kuchnow and Gilbert (1967) and Kuchnow and Martin (1970). A review article has also appeared dealing with this as well as other aspects of elasmobranch vision (Gruber and Cohen, 1978). To our knowledge, no more recent studies of the tapetum have been published. The earlier authors have given us the following picture of the elasmobranch tapetum lucidum, derived from studies on various species, but in particular from work on the Spiny Dogfish, Squalus acanthias. The tapetum is found immediately adjacent to the choroidal vasculature of the eye. Listing the major cell layers and proceeding from the scleral coat, one finds first the tapetum, then the choroid, then the retinal epithelium and finally the neural retina proper. Unlike most vertebrates, the elasmobranchs possess generally unpigmented retinal epithelia. There is, therefore, no
1277
1278
ASHLEYR. HEATH and HEATHERM. HINDMAK
si~ni~~dnt obstructi~~n to the passage of light through to the tapetum. The tapetum itself consists of two types of cell. T’he reflective cells themselves. or guanophores, contain highly ordered plates of crystalline guanine (Pirie and Simpson, 1946; Denton and Nicol. 1964). Associated with them is a second set of ceils, pigmented melanophores. The nuclei of these cells are found towards the scleral coat. Processes extend from the nuclear region in the direction of the choroid. Once they reach this layer they spread out parallel to it forming a series of terminal feet. In the light microscope, these terminal feet appear to form a continuous sheet across the surface of the vascular layer (see Figs I and 3). In conditions of darkness, or low light levels generally, the pigment granules found in the tapetal melanophores migrate towards the cell nuclei, away from the neural retina. This has the effect of removing pigmentation from the terminal feet and allowing light to reach the g~lanophores. Reflection from the tapetum then occurs. As the light level increases to normal photopic conditions. the pigment movement is reversed until ;I layer of pigment forms in the terminal feet between guanophores and choroid. In this condition. reflection is reduced to minimal levels, This process was shown to take approx. 2 hr to proceed to completion in either direction (Nicol, 1965). The cellular mechanism by which this control of tapetal reflection is accomplished is unknown. In the present paper we begin to address the question of the control of tapetat pigment distribution by reporting the effects of cyclic AMP (cAMP) and related agents on the process in ~,irro.
EXPERIMENTAL
NlETHODS
Adult, male Spiny Dogfish, Squalus acanthias, of 70-80 cm overall length, were obtained from Frenchman’s Bay, Maine, either by netting or by capture on trot-lines. Animals were maintained for periods of up to 5 days in running, fresh, sea water prior to use. They were not fed during this period. All experiments described in this paper were carried out on isolated ocular tissues in citru. The fish was first dark-adapted for a fixed period of 2 hr. Using illumination from a Wratten 2 (red) photographic dark-room lamp, the fish was decapitated and the eyes carefully enucleated. Eyes were then opened,
and the anterior segment and lens discarded. The globe was then hemisected using razor blades. Vitreous humor was carefully removed from the eyecup by repeated application of small wicks constructed from ‘Kimwipes’“. Squafus possesses a very thick vrtreous anti every effort was made to removi:
The tissue pieces, consisting ot mtact sciera, tapeturn, choroid, retinal epithelium and neural retina. were transferred in darkness to small flasks containing 2.5 ml of an elasmobranch Ringer’s solution. This had the following composition (mM): NaCi. 280; KCI, 5; CaClz, 2; MgSO,, 2; KHzPO,, 1.25; NaHCO,, 3% urea, 350 and D-glucose, 10. The Ringer was gassed with 95% 02;5% CO* before use and had a final pH of 7.4. -7.5. The flasks (four in number) were held in a Dubnoff shaking metabolic incubator and the tissue pieces (3-4 per flask) were maintained in darkness unless otherwise stated below. 95% 0,/.5% CO, was continually passed through the flasks at a rate of 1.75-2.0 std cu ft/hr]Rask. In a typical experiment designed to investigate the effect of a particular chemical agent. tissue was incubated for a period Of 2 hr. In other cases, where noted below, tissue was removed from incubatio~l at various time points, In all cases, one flask with its darkadapted tissue pieces was left in darkness for the duration of the experiment with no added effector agents, These tissues are referred to as dark-adapted controls. Experiments were carried out at a water-bath temperature of 20-22 c. Under these incubation conditions chemrcal
1279
Tapetal pigmentgranulemigrationand CAMP agents of interest were added as desired. N6,2’~-Dibutyryladenosine 3’ : S-cyclic monophosphate (db~AMP), nonderivatized CAMP and N?,2’-O-Dibutyrylguanosine-3’: S-cyclic monophosphate (dbcGMP) were added directly to the incubation medium. The phosphodiesterase (PDE) inhibitor 3-isobutyl- l-methyl xanthine (IBMX) and the adenylate cyclase stimulator forskolin were dissolved in dimethylsulfoxide (DMSO) prior to use. The final concentration of DMSO in the Ringer solution was always less than 0.594. DMSO at this concentration was found in control experiments to have no effect on tapetal pigment granule migration. Additionally. some experiments were carried out using the PDE inhibitors caffeine and theophylline. These were also added directly to the incubation medium. All chemicals were obtained from Sigma Chemical Co., St Louis. MO. Light microscopic and data analysis
Experiments were terminated by transferring tissues to cold (4 C) fixative consisting of 2.5% glutaraidehyde, 2% parafo~aldehyde and 17.25% sucrose in 0.16 M sodium cacodylate (Szamier and Ripps, 1983). Postfixation was in 1% osmium tetroxide, 17.25% sucrose in 0.16 M sodium cacodylate at room temperature. Tissues were dehydrated in an ethanol series, embedded in “Polybed 8 12” (Polysciences, Warrington, PA), an Epon substitute, and sectioned at 2 /lrn. Unstained sections were dried on to microscope slides and used directly for photography and data analysis. The sections were photographed using green light and Kodak “Panatomic-X” 35 mM film processed in fullstrength “Microdol-X”. Several photographic prints were prepared from each tissue piece sectioned. In order to quantify the degree of tapetal pigment migration, measurements were made from these prints of the total area of tapetal pigment found in a fixed measuring window of 1500 pm’ placed just above the basement membrane of the retinal epithelium. We accomplished the area measurement using a cursor and Jandel Scientific (Sausahto, CA) “Sigmascan” software running on a portable computer (“Compaq”), interfaced to a Numonies “2210” digitizing pad. To obtain the quantitative data presented here, 750 photographic prints were analyzed in this way. All area data are given in the form of the mean of absolute observed measurements with & 1 SD indicated.
RESULTS
distribution of tapetaf dark-adaptation
pig~lent
in light-
urzd
Light micrographs of the tapetum lucidum from Squalus showing the effects of light- and dark-adaptation on tapetal pigment distribution are presented in Fig. I. Figure 1A shows tissue from the dark-adapted control. Figure IB shows tissue from a companion flask which was exposed to room light for the duration of the 2 hr incubation. In the dark-adapted state, pigment granules may be observed aggregated around the melanophore nuclei. CytopIasmi~ processes extend from the nuclear region towards the choroid. These processes are largely devoid of pigment. In the light-adapted condition, pigment granules move along the length of the processes. In a thoroughly light-adapted preparation as seen in Fig. I B. the pigment moves into the terminal feet of the processes which run parallel to the choroid. In this way, light effectively is prevented from reaching the guanophores. It should be noted that the size and orientation of the tapetal melanophores varies with their location in the eye (Nicol, 1964, 1965). We have quantified the light-adaptation of tapetal tissue previously dark-adapted for the standard period of 2 hr. In these experiments, tissue pieces were transferred to flasks in the usual manner, one flask being covered and acting as the dark-adapted control. Using the red safelight for illumination, care was taken to ensure that the tissue was facing retina-side “up”. Room lights were then turned on, and samples of tissue withdrawn from the other flasks at fixed time points. Pigment area was then measured as a function of time in the light. The results are presented in Fig. 2. It will be seen from the figure that total tapetal pigment area increases with increasing time in the light. maximum pigment dispersion being reached in about 2 hr. Eflhct qf dbcAMP and related agents on tapetai pigment distribution in dark -adapted tissrte
We observed a striking effect upon adding dbcAMP to dark-adapted Squalus tapetai tissue pieces. Although the tissue was maintained in darkness, dbcAMP induced tapetal pigment dispersion characteristic of the light-adapted state. This effect was found to be dose-dependent. In Fig. 3 we present light-micrographs showing the effects on pigment distribution of a series of
1280
ASHLEY R. HEATH and HEATHERM. HINDMAN 6001
Fig. 2. Time course of light”adaptation of the tapetum in &UC&S ncanthim. Tissue was prepared in dim, red light and incubated in room light, as described in Materials and Methods. Each point shows the mean and one standard deviation of the area data. In this and in the following figures the numerals next to each point indicate the number of animals used in each case.
dbcAMP concentrations. The pigment may be seen acquiring an increasingly “light-adapted” distribution with increasing dbcAMP levels. No other agents were added in these experiments; dbcAMP alone was adequate to induce pigment dispersion, and it was not necessary to include phosphodiesterase (PDE) inhibitors as is often the case in such studies. In a series of experiments we have quantified the effect of increasing dbcAMP concentration, and the results may be seen in Fig. 4. Freshly isolated, dark-adapted Squalus tapetal tissue presents to the unaided eye a highly reflective surface. We found that darkening of the surface to a state characteristic of light-adapted tissue could be observed by eye after incubation in OS-l.OmM dbcAMP. (The use of unaided visual inspection of tapetal tissues as a rough check on the state of tapetal pigment was found to be a valuable aid in the conduct of this work. Subsequent histological examination of tissues judged visually always indicated a close correlation between qualitative and quantitative measures of pigment aggregation and dispersion. This method was also used in some cases by earlier workers, e.g. Nicol, 1965.) A maximum pigment area of 400-500 pm* was reached with 5 mM dbcAMP which correlates closely with the maximum area of pigment obtained through the effect of light (Fig. 2). In control experiments, nonderivatized CAMP (5.0 mM), adenosine (2.5 mM), and dbcGMP (5.0 mM) had no observable effect upon dark-adapted tapetal pigment. Other agents were also found to induce pigment dispersion in dark-adapts tissue. Both IBMX and forskolin, a PDE inhibitor and an adenylate cyclase stimulator respectively,
Fig. 4. Quantitative analysis of the dispersive effect ofCAMP on pigment granule movement in Squatus tapetum lucidum in sitro. Tissue was incubated in darkness and dbcAMP was added to the medium at the indicated concentrations. Tissue was prepared and processed and data anatyzed as described in Materials and Methods,
would be expected to elevate endogenous tissue CAMP levels. In confi~ation of this expectation, we observed that both agents produced pigment dispersion patterns similar to that seen from adding exogenous dbcAMP. The dose-response data for IBMX and forskolin have been quantified and the results are presented in Figs 5 and 6. Forskolin is effective in producing pigment dispersion that may be seen with the unaided eye at 10e6M. IBMX is approximately one to two orders of magnitude less effective. Additionally, a few experiments indicated that caffeine at a concentration of 5 mM, and theophylline at a concentration of 1 mM also induce pigment dispersion.
A number of experiments were carried out to examine the effect of dbcAMP on pigment aggregation induced by darkness in lightadapted tissue in vitro. Tissue was prepared in the usual way, except that a period of 2 hr light-adaptation (using standard laboratory Auorescent illumination) was given prior
Fig. 5. Effect of isobuty~methyixanth~ne ff BMX) on tapetal pigment in dark-adapted Squalusacanrhiastissue. Conventions are as given in previous figures.
t-e r Fig. 1. Unstained, light-micrographs showing S9uulrts acanthius tapetum lucidum in dark-adapted (A) and light-adapted (B) states from tissues maintained in O&V as described in the lext. n, nucleus of tapetal melanophore; c, choroid; re, retinal epithelium; r, neural retina. Scale bar, Loom. Arrows indicate positions of tapetal melanophore processes, into which pigment granules migrate on light-adaptation. Note that the size of the tapetal melanophores (including the distance from the tapetal melanophore cell body to the tips of the cytoplasmic processes) and the angle of orientation of the cytoplasmic processes, was observed to vary with location in the eye (see also Nicol, 1964 and 1965). Only the pigment granules move due to the effects of light- or dark-adaptation, the cytoplasmic processes themselves are fixed.
1281
Fig. 3. Effect of Dibutyryl cychc AMP (dbcAMP) on tapetal pigment distribution in dark-adapted tissue in rirrt>. The tissue was incubated for two hours as described in the text, in the presence of (A) no added dbcAMP (dark-adapted control), (B) 0.25 mM dbcAMP, (C) 0.5 mM dbcAMP, (D) f.0 mM dbcAMP, (E) 2.5 mM dbcAMP and (F) 5.0 mM dbcAMP. Note a progressive dispersal of pigment granules with increasing dbcAMP concentration. As noted in the caption to Fig. I, tapetal melanophore size and orientation varies with location in the eye. Unstained, i~ght-micrographs. Scale bar, IOirm.
Tapetal
pigment
granule
migration
and cAMP
1283
no caffeine and incubated for a further 3.5 hr. These pieces upon visual examination were found to have aggregated pigment, i.e. the effect of caffeine had been reversed. DISCUSSION
Fig. 6. Effect of forskolin on tapetal pigment in darkadapted Squahs acanfhim tissue. Conventions are as given in previous figures.
to sacrifice of the animal. Concentrations of dbcAMP varying from 0.125 to 5.0 mM were tested. The results were judged qualitatively using visual inspection of the tissue pieces to determine the gross state of tapetal pigment dispersion or aggregation. Tissue was scored as aggregated if it presented as dark-adapted with no observable degree ofpigment dispersion. The results are presented:” Table 1. In summary, we observed that dbi 4&l’ F at concentrations greater than 1.OmM’e$rely inhibits the aggregation of tapetal pign$:,observed upon darkadaptation. Rewrsihilit~ of chemkally dispersion in sitra
induced pigment
We do not have data indicating whether or not the effect of dbcAMP %af observed on &_I”.: tapetal tissue irz r*itro is r&ersrble. However, preliminary observations.~~~~ate that the pigment dispersing effect @&ffeine at 5 mM is reversible. Tissue from th?&e animals was incubated in the normal way in darkness in the presence of caffeine for 2 hr. Sample pieces were removed at this point and tapetal pigment was found by visual inspection (see above) to be dispersed. The remaining pieces were then returned to flasks containing fresh medium with ‘Table 1. Inhibition of phore aggregation Concentration of dbcAMP tested (mM) 0.125 0.25 0.50 1.0 2.5 5.0
~~f~f~/~f.~ u~,a~f~ju~
in darkness
Number of animals 3 3 4 3 3 3
tapetal melanoby dbcAMP
Tissue pieces with aggregated pigment: total tissue pieces 9/Y 719 4:11 118 019 O/IO
Note: these objervations are based upon visual scoring of tissue pieces as described in the text. In all cases where tapetal pigment was not aggregated it was observed to he in the dispersed, light-adapted state.
In this paper we have presented evidence that CAMP induces a state of tapetal pigment dispersion in vitro in dark-adapted tissues of Sq~aI~s acanthias, similar to that observed in the normal process of light-adaptation in the absence of exogenously applied effecters. Additionally, tapetal pigment disperses in darkness in the presence of agents that might reasonably be expected to raise the endogenous tissue level of CAMP, such as IBMX and forskolin, through their respective effects on cyclic nucleotide phosphodiesterase and adeny~ate cyclase. The aggregation of pigment that normally accompanies dark-adaptation is also inhibited by CAMP. A number of publications have appeared in the literature that directly link CAMP with the control of pigment granule movement in a variety of melanophore types (Bitensky and Burnstein. 196.5; Abe et al., 1969; Novales and Fujii, 1970; Wakamatsu e? ~1.. 1980; Novales, 1983; Namoto, 1985. 1987; Namoto and Yamada, 1987; Rozdzial and Haimo, 1986a, b; Morishita, 1987). This work has dealt in particular with dermal melanophores of teleosts and amphibians. Recent studies using carefully lysed dermal melanophores from Tilapia have shown that CAMP and ATP are required for melanophore dispersion, but that the absence of CAMP is a prerequisite for aggregation. The dispersion/aggregation cycle in these cells is also directly related to the phosphorylation and dephosphorylation of a 57 kD polypeptide (Rozdzial and Haimo, 1986a. b). Whether or not tapetal pigment granule dispersion in Squa~us is phosphorylationdependent remains to be seen. It has long been known that pigment aggregation in dermal melanophores may be under the control of adrenergic effecters such as epinephrine (Bagnara and Hadley, 1973). and it is supposed that these agents exert their effects through a reduction in intracellular CAMP levels. We currently have no information on the induction of tapetal melanophore pigment aggregation in Squafus by any means other than through the effects of darkness. Normal tapetal pigment aggregation in darkness may be obtained in citrv from previousfy light-adapted tissue under the
1284
ASHLEY R. HEATH and HEATHER
experimental conditions described in this paper. Preliminary experiments carried out by the present authors using light-adapted tissue have shown no effect of epinephrine on the dispersed pigment state. Presumably. in intact lightadapted tissues, sufficient increased endogenous levels of CAMP are maintained to counteract any possible effect of epinephrine. Interestingly~ in at least one class of pigmented cell, the retinal epithelial (RE) cell in teleost fish, CAMP has been demonstrated to have a pigment aggregating effect (Burnside and Nagle, 1983). In vertebrate animals possessing pigmented RE cells extensive reciprocal movements of pigment granules and photoreceptors, referred to as retinomotor movements, occur in response to light- and dark-adaptation. With respect to the retinal epithelium, these movements are analogous to elasmobranch tapetal cell responses to variation in illumination conditions. In teteosts, RE cell pigment granules disperse in the light and aggregate in darkness. In tissues from such animals maintained in the light in the presence of CAMP, pigment aggregation is observed. Thus, this response is directly opposite in direction to that seen in the tapetal cell and in most other types of melanophore. On the other hand, a pigment dispersing effect of CAMP on pigmented RE cells in amphibians was recently described (Astorino et al.. 1987). Although it is possible that we are dealing with a direct effect of CAMP on the tapetal melanophores in our irzvitro experiments, analysis of the cellular mechanisms controlling tapetal pigment movement in Squab is complicated by the results of early experiments carried out by Nicol (1964, 1965) on tapetal tissue from which the neural retina had been carefully removed. He was concerned to determine whether or not the tapetal melanophores themselves act as primary photosensitive effecters, A precedent for such an effector system was already known in the case of the elasmobranch iris constrictor muscfe, which responds directly to illumination (Young, 1933). and melanophores that respond to light, possibly due to variation in endogenous CAMP levels as a result of the action of lightactivated phosphodiesterases, are now wellknown in a number of other animal species (Wakamatsu et ul., 1980; Weber, 1983). HOWever. it was found that following retina removal the tapetal pigment always attained the dispersed state, Dispersal was independent of ambient light conditions and was not affected by a
M.
HINDMAN
variety of physiological treatments including addition of acetylcholine, epinephrine, pituitrin. variations of Ringer ionic composition (e.g. manipulation of calcium, magnesium and potassium levels). alteration of pH balance, and raising hydrostatic pressure. In a few instances, however. Nicol (1965) was able to maintain aggregation of pigment in tapeta from which the retina had been carefully removed and then replaced. although these experin~ents did not give consistent results. It appears that tapetal melanophores in Squalus are not photosensitive primary effecters and that in the dark-adapted state the neural retina itself must exert some kind of inhibitory influence over tapetal pigment dispersion. or must in some fashion actively promote pigment aggregation. The results of our experiments using exogenously applied CAMP indicate that the aggregation of pigment may be related to a decrease in endogenous CAMP as has been suggested for dermal melanophores (see refereqys .abovef. We suggest that the effect of~~~~w,th respect to the retina and tapetum is tolmcrease tissue levels of CAMP, thereby promc$ng” pigment dispersion In our experiments, by adding CAMP in oiwa to dark-adapted tissue, we have presumably bypassed the normal physiological action of the retina. (One might also speculate that CAMP regulates the release or synthesis of some other effector, but weJ ve no evidence on that point at present,) Exa&$ x.,how an increase in CAMP triggered by the ~~~~~se of the neural retina to light is translated in,tc+a cellular response at the level of the tapetu@, lucidum is not known. Although light-activated retinal cyclic nucleotide phosphodiesterases have been known to exist for some time (Bitensky rt ui., 1975) there is to our knowledge no evidence for the existence of light-activated adenylate cyclase systems in the elasmobranch retina. or in any other retina. In addition, any chemical message would have to pass two intervening celiular layers (the retinal epithelium and the choroidal vascuiature) in order to reach the tapetal cells by direct diffusion. Presumptive nerve terminals have been described on tapetal pigment ceils in another species of Dogfish (Must&s mustelus. Kuchnow and Martin. 1970) 50 we cannot exclude the possibility of a neural connection between retina and tapetum. Such an arrangement might give rise to increased tapetal cell. adenylate cyclase activity through the mediation of a neurotransmitter agent. At present, we are searching for evidence for the existence of such
Tapetal pigment granule migration and CAMP
neural connections tapetum lucidum.
in the Squalus acanthias
Aeknoi~ledgements-This work was supported by grants from the National Eye Institute (EY 06536) the Pennsylvania Lion’s Sight Conservation and Eye Research Foundation, Breastplate Laboratories of Houston, TX and by a fellowship to ARH from the Lucille P. Markey Trust. We wish to thank the Director and staff of the Mount Desert Island Biological Laboratory for their efforts in ensuring the successful completion of this work. We thank also Richard Stegen for providing technical assistance. REFERENCES Abe K., Robison G. A., Liddle G. W., Butcher R. W., Nicholson W. E. and Baird C. E. (1969) Role of cyclic AMP in mediating the effects of MSH, norepinephrine, and melatonin on frog skin color. Endocrinology 85, 674-682. Astorino A.. Flannery J. G., Marc R. and Basinger S. F. ( 1987) Effect of adrenergic agonists and cyclic nucleotides on pigment granule migration in the RPE of Runa pip&s eyecups in r&w. Invest. Ophthal. risuai Sri. (SuppI.) t8, 256. Bagnara J. T. and Hadley M. E. [ 1973) Chromutophores and Color Chunge. Prentice-Hall, Englewood Cliffs, N.J. Bitensky M. W. and Burnstein S. R. (I965) Effects of cyclic adenosine monophosphate and melanocyte stimulating hormone on frog skin in vitro. Nature, Lond. 208, 1X--1284.
Bitensky M. W., Miki N.. Keirns .I. J., Baraban J. A., Wheeler M. A.. Lacy J. and Marcus F. R. (1975) Activation of photoreceptor disc membrane phosphodiesterase by light and ATP: cyclic GMP as a modulator of sense receptor function. Adz.. cyclic Nucleotide Rex 5, 2 13.-240.
Briicke E. (1845) Anatomische Untersuchungen fiber die ~ogennanten leuchtenden Augen bei den Wirbelthieren. .4&i. Anat. Physiol, Lpz. Jahrg. (IS45), 387-406. Burnside B. and Nagle B. (1983) Retinomotor movements of photoreceptors and retinal pigment epithelium: mechanisms and regulation. In Progress in Refinai Research (Edited by Osborne N. and Chader G.), pp. 67-109. Pergamon Press, Oxford. Denton E. J. and Nicol J. A. C. (1964) The chorioidal tapeta of some cartilaginous fishes (Chondrichthyes). J. mar. Rio/. A.?,~.. i’.R
44, 219~.258.
Franz V. (1905) Zur Anatomie, Histologie und funktionellen Gestaltung des Selachierauges. Jena. 2. Naruw. Bd. 40, 697 840. Gruher S. H. and Cohen J. L. (1978) Visual system of the eiasmobranchs: state of the art 1960-1975. In Sensory Biology o/‘Sharks, Skatpv. and Rays (Edited by Hodgson E. and Mathewson R.). pp. I I-105. Office of Navai Research.
Arlington,
VA.
Kuchnow K. P. (1969a) Threshold for the elasmobranch tapetal-pigment response. Vision Res. 9, lS7- 194. Kuchnow, K. P. (1969b) Tapetal pigment response of elasmobranchs. Vision Res. 9, 8499854. Kuchnow K. P. and Gilbert P. W. (1967) Preliminary in viva studies on pupillary and tapetal pigment responses in the
128.5
lemon shark, Negaprion brevirostris. In Sharks, Skates and Rays (Edited by Gilbert P. W., Mathewson R. F. and Rail D. P.), pp. 465-477. Johns Hopkins Press, Baltimore, MD. Kuchnow K. P. and Martin R. (1970) Pigment migration in the tapetum lucidum of the elasmobranch eye: evidence for a nervous mechanism. Vision Res. 10, 825-828. Morishita F. (1987) Responses of the melanophores of the medaka, Oryzius lutipes, to adrenergic drugs: evidence for involvement of c(?adrenergic receptors mediating melanin aggregation. Camp. Biochem. Physiol. SSC, 6974. Namoto S. (1985) Effects of adenylate cyclase activating agents on pi~ent-aggregating action of lithium ions to fish melanophores. J. Sci. Hiroshima Univ. 32, 105-I 16. Namoto S. (I 987) A subtype of adenosine receptors mediating pigment dispersion in leucophores of the medaka: evidence for an A,-receptor. Camp. Biochem. Physiol. SSC, 75-81.
Namoto S. and Yamada K. (1987) Effects of forskolin, isoproterenol and lithium ions on leucophores of a teleost, Oryzius Iutipes: evidence for involvement of adenylate cyclase in pigment-dispersing response. Camp. Biochem. Physiol, 86C, 9 I-95. Nicol J. A. C. (1964) Reflectivity of the chorioidal tapeta of selachians. J. Fish Res. Bd. Can. 21, lOS9-11~. Nicol J. A. C. (1965) Migration of chorioidal tapetal pigment in the spur dog, Sguutus ucanthius. J. mar. biol. Ass., U.K. 45, 405427.
Nicol J. A. C. (1975) Studies on the eyes of fishes: structure and ultrastructure. In Vision in Fishes: New Approaches in Research (Edited by AIi M. A.), pp. 5799607. Plenum Press, New York, N.Y. Novales R. R. (1983) Cellular aspects of hdrmonally controlled pigment translocations within chromatophores of poikilothermic vertebrates. Am. Zool. 23, 5599568. Novales R. R. and Fujii R. (1970) A melanin-dispersing effect of cyclic adenosine monophosphate on Fundulus meianophores. J. Ceil Physiol. 75, 133-136. Pirie A. and Simpson D. M. (1946) Preparation of a fluorescent substrate from the eye of the dogfish (~~ua~us acanthjas). Biochem. J. 40, l&20. Rozdzial M. M. and Haimo L. T. (lPS6a) Bidirectional pigment granule movements of melanophores are regulated by protein phosphorylation and dephosphorylation. Cell 47, 1061-.1070. Rozdzial M. M. and Haimo L. T. (1986b) Reactivated melanophore motility: differential regulation and nucleotide requirements of bidirectional pigment granule transport. J. Cell Biol. 103, 2755-2764. Szamier R. B. and Ripps H. (1983) The visual cells of the skate retina: structure, histochemistry, and disc-shedding properties. J. Camp. Neurol. 215, 51-62. Wakamatsu Y., Kawamura S. and Yoshizawa T. (1980) Light-induced pigment aggregation in cuttured fish melanophores: spectral sensitivity and inhibitory effects of theophylfine and cyclic adenosine-3’,5’-monophosphate. J. Cell Sci. 41, 6574. Walls G. L. (1942) The Vertebrate Eye and its Adaptive Radiation. The Cranbrook Press. Bloomfield Hills, MI. Weber W. (1983) Photosensitivity of chromatophores. Am. Zool. 23, 495-506. Young J. Z. (1933) Comparative studies on the physiology of the iris. I. Selachians. Proc. R. Sot. 8. 112, 228-241.