morphologic changes in teleost primary and secondary retinal cells

Morphologic Changes in Teleost Primary andSecondary Retinal Cells Following Brief

Exposure to LightHans-Joachim Wagner* and Ronald H. Douglasf

Light adaptive morphologic changes in the teleost retina, such as movements of rods, cones, andepithelial pigment and spinule formation on horizontal cell terminals, are normally associated withcontinual exposure to light. Depending on a variety of factors these processes are generally completedwithin 30-60 min. In this report we show that a brief exposure to light (1-2 min) preceded andfollowed by darkness is sufficient to elicit these changes in four species of teleost; the trout (Salntogairdneri), the tench (Tinea tinea), the carp (Cyprinus carpio), and the goldfish (Carassius auratus).A brief pulse is as effective in causing cone migration and an increase in the number of spinules ascontinual exposure; however, it is sometimes less effective in causing pigment migration. The pho-tomechanical changes following a brief period of light are always completed more quickly and showgreater species variability than the formation of spinules. These results show that the various mor-phologic manifestations of light adaptation are autonomous processes that need only a short pulseof light to trigger the. whole sequence of events. This is of interest when considering their controlmechanisms and may have consequences for physiologic work involving experiments on dark-adaptedeyes. Invest Ophthalmol Vis Sci 24:24-29, 1983

During retinal light adaptation, four main types ofmorphologic changes can occur. Three of these havebeen observed in most teleosts that have been studied:the shedding of discs from the outer segments of vi-sual cells,1 photomechanical changes of rods, cones,and epithelial pigment,2 and the formation of finger-like projections or "spinules" from horizontal cellterminals associated with cone pedicles.3 The fourthchange, an increase in the number and/or length ofsynaptic ribbons in cone pedicles, occurs in some spe-cies but not in others.4'5

All of these changes are normally observed follow-ing long-term changes in the lighting conditions. De-pending on a variety of factors such as the phase ofthe light/dark cycle and the intensity of the adaptinglight, these movements generally take about 30-60min for completion. Recently, however, Muntz andRichard6 have shown that a brief exposure to lightfollowed by darkness can elicit photomechanicalchanges in dark-adapted rainbow trout. Whereas,

From the Abteilung fur Klinische Morphologie, UniversitatUlm, Neubau Oberer Eselsberg, D-7900 Ulm, Federal Republicof Germany.

Supported by a grant from the DFG* and a Royal Society Eu-ropean fellowship.!

Submitted for publication December 17, 1981.Reprint requests: Hans-Joachim Wagner, Abteilung fur Klin-

ische Morphologie, Universitat Ulm, Neubau Oberer Eselsberg, D-7900 Ulm, Federal Republic of Germany.

0146-0404/83/0100/024/$ 1.15© Association

these effects reflect reactions of the primary retinalelements, ie, receptor cells and pigment epithelialcells, this study reports that such brief exposures (1 -2 min) also affect the adaptational state of secondaryneurons as shown by the morphology of the conehorizontal cell terminals. Spinule formation was ex-amined in four species of teleost (tench, carp, gold-fish, and rainbow trout) and compared to the pho-tomechanical activity.

These observations indicate that whatever the con-trolling mechanisms of the various morphologicchanges are, they are not dependent on a sustainedstimulation, but need only a short pulse of light inorder to trigger the entire process of morphologic lightadaptation. This could have important consequencesfor both past and future physiologic work involvinglight stimulation of dark-adapted retinas.

Materials and Methods

Four species of teleost were used in these experi-ments: tench {Tinea tinea), rainbow trout (Salmogairdneri), carp (Cyprinus carpio), and goldfish (Car-assius auratus). The goldfish were obtained from alocal pet shop, while the other species came fromHeilbronner fisheries, Ulm, Germany.

Prior to the experimental light exposure, the trout(15-20 cm), tench (6-8 cm) and carp (15-40 cm)were transferred from the fish farm, where they had

for Research in Vision and Ophthalmology

24

Downloaded From: http://iovs.arvojournals.org/pdfaccess.ashx?url=/data/journals/iovs/933338/ on 04/08/2018

No. 1 MORPHOLOGIC RETINAL CHANGES FOLLOWING BRIEF LIGHT EXPOSURE / Wagner and Douglas 25

Salmo gairdneri

Fig. 1. Epithelial pigmentand cone indices and thespinule/ribbon ratio of (a)Salmo gairdneri, (b) Car-assius auratus, (c) Tineatinea, and (d) Cyprinus car-pio following a brief (1-2min) exposure to light. Thecircles and solid line( • • ) represent thecones, the squares and thedotted line (• • ) thepigment, and the stars andthe dashed line (* • )the spinule/ribbon ratio. Thepigment and cone indicesand s/r of normally light-adapted individuals aregiven by the symbols in thecircles (© ® ®) and those offish exposed to 1 min of redlight at 10-min intervals for40-50 min by the symbolsin the squares (E H H). Thebar under the ordinate rep-resents the duration of thelight exposure. As two fishwere killed at each interval,and both eyes of an individ-ual were examined, all reti-nomotor indices and spi-nule/ribbon ratios shown arethe means of the averagevalues from four retinas.The vertical bars representthe standard deviations ofthese four averages (in somecases only the positive ornegative deviation has beenshown so as to avoid con-fusion with other lines).

C.I. PI.Carassius auratus

C.L#

0 20 40minutes after flash

Tinea tinea

07

03

as 05

'03/

0 20 40 60minutes after flash

Cyprinus caroio

25

15

0 5

Jr"

r15



60

been exposed to the natural changes in illuminationonly, to the laboratory, where they were put in dark-ness. The goldfish (5-6 cm) were treated in a similarmanner, except that prior to being put in darknessthey had been in an artificial light/dark cycle (lightphase 06.00-18.00). Following two hours dark ad-aptation, fish were sampled at around noon, imme-diately prior to a brief (1 min except for tench, whichwas two min) exposure from a tungsten iodide lamp(average of 2.7 X 103 lx at the water surface) and atintervals thereafter (see Fig. 1 for exact samplingtimes) (Table 1). Fish were killed at a standardizedtime such as noon, controlled for any possible vari-

ation of indices due to endogenous circadian rhythms.All manipulations were carried out using only a dimred flashlight. At every sampling interval, two fishwere killed by decapitation and both eyes of an in-dividual enucleated and hemisected. Subsequent fix-ation and embedding were carried out as describedpreviously.3

To determine the relative positions of the conesand epithelial pigment, 1-2 /*m transverse sectionswere stained with Richardson's stain7 and examinedunder a light microscope. The position of the epithe-lial pigment is given as a pigment index, which is thedistance from Bruch's membrane to the outermost


26 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / January 1983 Vol. 24

Table 1. Relative effectiveness of a brief (1-2 min)exposure to light in causing an increase in thespinule/ribbon ratio compared to continuouslight. Tinea tinea received 2 min of light,the others 1 min

Species

Tinea tineaCyprinus carpioCarassius auratusSalmo gairdneri

Normal change inspinule/ribbonratio on light

adaption

1.121.992.282.67

Change inspinule/ribbonratio followingbrief exposure

1.161.711.741.68

Percent

104867663

projection of pigment granules expressed as a per-centage of the distance from Bruch's membrane tothe external limiting membrane (elm). Similarly, thecone index is the distance from the gap between thecone ellipsoid and outer segment to the elm again asa percentage of the total thickness. For the cone indi-ces only the principal member of double cones weremeasured. In each retina at least 50 cones and 10pigment positions were measured, and these wereaveraged to give the indices for that retina. All mea-surements were made in the vicinity of the optic nerveso as to control for any regional variation in photo-mechanical changes.

Spinules were counted for at least 50 cone pedicles

per eye in ultrathin transverse sections on an electronmicroscope at a magnification of 30,000X and ex-pressed as a fraction of the number of synaptic rib-bons, as described in Wagner3 (see Table 2). In somecases, where ribbons were very short (possibly due todark adaptation, see Wagner4'5) synaptic ridges werecounted instead. For convenience, we generally referto ribbons, however.

In order to obtain light-adapted baseline values forthe retinal indices and spinule/ribbon ratios, the eyesof two tench, two trout, and two carp sampled at thefish farm during the day were examined. For goldfishbaselines fish sampled during the light phase of a lab-oratory light/dark cycle were used. These baselinevalues are shown in Figure 1 by the points within thecircles.

As a control for any possible effect of the red torchon photomechanical movements and spinule forma-tion, two individuals from each species were darkadapted in the same manner as experimental animals.However, instead of receiving a 1-min white light ex-posure, they were exposed to the red torch for 1 min.This red light exposure was then repeated at 10-minintervals for 40-50 min, after which the fish were killedand the retinas processed as described above. Thesecontrol values are shown in Figure 1 by the pointswithin the squares and in all cases are not significantlydifferent from dark-adapted control values.

Table 2. Numbers and ratios of cone pedicles, synaptic ribbons, and spinules in the retinas of fish exposedto various lighting conditions

(a) Salmo gairdneriControl light adaptedMaximum light adaptation after

brief exposureControl dark adapted before

brief exposureRed light control

(b) Carassius auratusControl light adaptedMaximum light adaptation after



(c) Tinea tineaControl light adaptedMaximum light adaptation after



(d) Cyprinus carpioControl light adaptedMaximum light adaptation after



Number ofpedicles

202

197

218322

203

216

245222

216

210

210250

200

212

200207

Number ofspinules

1792

1468

280502

1017

916

147366

1064

780

406291

1192

928

310290

Number ofribbons

576

691

6411088

407

468

659486

592

416

562717

428

370

386290

Spinule/ribbonratio

3.11

2.12

0.440.46

2.50

1.96

0.220.75

1.80

1.88

0.720.41

2.79

2.51

0.801.00

Spin ule/pedicleratio

8.87

7.45

1.281.56

5.01

4.24

0.601.65

4.93

3.71

1.931.16

5.96

4.38

1.551.40

Ribbon/pedicleratio

2.85

3.51

2.943.38

2.00

2.17

2.692.19

2.74

1.98

2.682.87

2.14

1.75

1.931.40



Table 3. Relative effectivenesscompared to continuous light.

Species

Tinea tineaCarassius auratusSalmo gairdneri

of a brief (1-2 min) exposure to light in causing double coneTinea tinea received 2 min of light, the others 1 min

Change in cone index duringnormal light adaptation

0.300.38

Change in cone indexfollowing brief exposure

0.290.36

migration

Percent

9795

ResultsThe positions of the epithelial pigment and double

cones as well as the ratio of spinules to ridges (s/r)for all four species immediately prior to and followinga brief light exposure are shown in Figure 1. In allcases a brief exposure to light caused some movementof the cones and epithelial pigment toward their light-adapted positions, and an increase in the number ofspinules relative to the number of synaptic ribbons.Photomechanical changes for the carp are not shownas the results obtained were too variable.

Spinules/Ribbons

In all four species a brief exposure to light causedan increase in s/r to approximately the levels foundin normally light-adapted animals (Figure 1, Table1). In trout, carp, and goldfish, the maximum numberof spinules was reached in 50 min, while in tench thelight-adapted level was reached after only 20.

Table 2 shows the number of cone pedicles, spi-nules, and synaptic ribbons in the retinas of controland experimental fish. It can be seen that the ratioof spinules/pedicles, like the ratio of spinules/ribbons,is significantly higher in light-adapted retinas than indark-adapted ones. The ratio of ribbons/pedicles,however, shows no variation with the state of adap-tation. Thus, it is safe to assume that the increase inthe spinule/ribbon ratio following a brief exposure towhite light is caused by an increase in the numberof spinules and not by a decrease in the number ofsynaptic ribbons.

Double Cones

In trout and goldfish these cones migrate to typicallight-adapted positions following a 1-min exposureto light (Figure 1, Table 3). For trout the maximum

cone index was reached within 10 min after the lightexposure, and the cones stayed in this light-adaptedposition for the rest of the experiment (60 min). Ingoldfish the maximum was reached after 20 min, butthe cones then started to move back toward theirdark-adapted positions. Tench cones must be consid-ered separately as photomechanical movements inthis species are unusual. As described elsewhere8 thecones do not undergo significant migration in eitherthe light or dark. They always lie in close proximityto the elm. A brief exposure to light therefore has noeffect on tench cones.

Epithelial Pigment

In all species the brief light exposure caused somepigment movement, but unlike cones, the extent ofmigration differed from species to species (Figure 1,Table 4). In two cases, tench and goldfish, it wasenough to cause the pigment to move to totally light-adapted positions in 15 and 20 min respectively. Thepigment, like the cones, of the goldfish started tomigrate back toward a dark-adapted position after 20min and was nearly totally dark-adapted after the 60min of the experiment. In trout, however, the pig-ment only migrated to 63% of its normal extent, andstayed in a semiadapted position for most of the ex-periment. It also showed a tendency towards return-ing to the dark-adapted position towards the end ofthe experiment.

In summary our results show that: (1) A brief ex-posure to light is essentially as effective, in terms offinal levels reached, in causing cone migration andan increase in the number of spinules as continualexposure to light. It is, however, sometimes less ef-fective in causing pigment migration. (2) Photome-chanical movements following brief exposure to lightare always completed more quickly than the increase

Table 4. Relative effectivenesscompared to continuous light.

Species

Tinea tineaCarassius auratusSalmo gairdneri

of a brief (1-2 min) exposure to light in causing epithelialTinea tinea received 2 min of light, the others 1 min

Change in pigment index duringnormal light adaptation

0.570.440.43

Change in pigment indexfollowing brief exposure

0.480.370.27

pigment migration

Percent

848463


28 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / January 1983 Vol. 24

in the number of spinules. (3) Species differences inphotomechanical movements are more marked thanfor the increase in the number of spinules.

Discussion

In invertebrates there are several examples showingthat a brief exposure to light, followed by a return todarkness, is sufficient to elicit morphologic changesnormally associated with light adaptation.910 For ver-tebrate photoreceptors both cone migration6 as wellas disc shedding11 have been observed following aflash of light in trout and frogs respectively. Our re-sults show that a brief exposure to light cannot onlyaffect the primary elements within the retina but canalso lead to morphologic changes in the neural, ie,secondary, part of the retina. Due to the few speciesstudied, however, it is not possible to generalize thesefindings to all vertebrate classes.

As an anatomical "marker," we have used smallprocesses that originate from the cone horizontal cellterminals in the first synaptic layer. HI and H2 cellprocesses that invaginate the cone pedicles give riseto secondary finger-like spinules following light ad-aptation.3 As these spinules are largely absent duringdark adaptation, their formation clearly reflects thestate of adaptation of these secondary neurons. Al-though the function of these structures is not fullyunderstood, their absence or presence is associatedwith important physiologic changes in the feedbackmechanisms of the outer plexiform layer12 and in thechromatic receptive field organization of ganglioncells.13 Therefore, the formation of such spinules fol-lowing a brief exposure to light can be taken as anindication of a change in the adaptive state of thewhole neural retina.

Spinule formation following a brief exposure tolight was observed to occur more slowly than bothpigment and cone migration. This is not surprisingin view of the fact that their formation must dependon stimulation of the visual cells. Furthermore, it isinteresting to note that spinule formation peaks at 50min after the light exposure in the three species ex-amined for this length of time. Although our exper-imental design does not make it possible to obtainsufficient evidence to confirm the existence of sucha turning point back towards dark adaptation, thesimilarity between the three species is strongly indic-ative of it. In all cases, except perhaps the trout, briefexposure to light was as effective in causing spinuleformation as continued exposure.

Whereas there seems to be little interspecific dif-ference in spinule formation, the reaction of the conesand epithelial pigment exhibits a marked variabilitybetween the species. Although in all cases the conestook up a position of maximal light adaptation, this

was only so for the epithelial pigment of two species.Also, although in trout the cones stayed light adaptedfor the duration of the experiment (60 min), goldfishcones and pigment started to return to a dark-adaptedposition.

Species differences in the reaction of the cones andepithelial pigment are also known to occur when fishare exposed to other unnatural lighting conditions.For instance, some species when put in darkness forextended periods of time showed normal photome-chanical movements as if the changes in illuminationwere taking place.14 AH1516 and Wagner and Ali,5

however, failed to find any endogenous rhythm ofpigment or cone movement in several species of sal-monid, while Douglas17 observed two peaks of light-adaptive migration occurring around about the timethat would have been dawn and dusk when rainbowtrout were kept in continual darkness. That speciesdifferences in response to brief exposures to light oc-cur is, therefore, not surprising, although their sig-nificance remains unclear.

In trout the pigment was found to migrate only toabout half its maximum extent, whereas the conesalways migrated maximally. This decreased ability ofthe pigment to migrate again corresponds well to whatis known about rhythms of photomechanical move-ment when fish are kept in darkness over extendedperiods of time. In several such experiments there isan absence of, or decreased, pigment migration whilethe cones show a clear endogenous pattern of move-ment.818"20 Therefore, it seems that the epithelial pig-ment is more dependent on the external light stimulusthan the cones, possibly indicating differences in con-trolling mechanisms of cones and epithelial pigment.

The fact that only a brief pulse of light is necessaryto initiate morphologic changes associated with lightadaptation is of interest when considering the mech-anisms that control these events. These biochemicalprocesses are at present only incompletely under-stood. Cyclic adenosine 3':5' monophosphate(cAMP),21 melatonin,22"30 and pituitary hor-mones31"33 have all been implicated as having an ef-fect on photomechanical changes. Formation anddegradation of spinules, on the other hand, may berelated to the transmitter flux released by the cones.It is not clear, however, what other controlling mech-anisms are involved in the sequence of events leadingto light adaptation. What is clear, in light of our ev-idence, is that morphologic light adaptation is an au-tonomous process that only needs a brief pulse oflight in order to trigger all subsequent events that leadto structural light adaptation of the retina.

A note of caution may be added in conclusion. Inview of the fact that both the receptor and the neu-ronal part of the fish retina can be light adapted bya brief light exposure, great care should be taken in



all physiologic experiments on dark-adapted speci-mens to avoid any illumination. In all cases of doubtmorphologic controls should be carried out, as evenflashes as short as 0.0024 seconds, which is similarin duration to the stimulus used in physiologic ex-periments, can cause photomechanical changes inotherwise dark-adapted trout.6 These results couldalso cast doubt on some of the previous physiologicwork in vision research requiring light stimulation ofdark-adapted animals. Although most work has beencarried out near thresholds, which require only lowintensity stimulation, it is possible that some "dark-adapted" retinas were actually not as dark-adaptedas supposed.

Key words: retina, cones, epithelial pigment, horizontalcells, adaptation, photomechanical movements, spinules.

Acknowledgments

The authors would like to thank Professor W. R. A.Muntz for useful comments on the manuscript and Mrs.Verena Frosch and Mrs. Heather Douglas for technical as-sistance.

References1. O'Day WT and Young RW: Rhythmic daily shedding of outer

segment membranes by visual cells in the goldfish. J Cell Biol76:593, 1978.

2. Ali MA: Retinomotor responses. In Vision in Fishes; NewApproaches in Research, Ali MA, editor. 1974 NATO Ad-vanced Study Institute. New York, Plenum Press, 1975, pp.313-356.

3. Wagner HJ: Light-dependent plasticity of the morphology ofhorizontal cell terminals in cone pedicles of fish retinas. JNeurocytol 9:573, 1980.

4. Wagner HJ: Quantitative changes of synaptic ribbons in thecone pedicles of Nannacara: light dependent or governed bya circadian rhythm? In Vision in Fishes; New Approaches inResearch, Ali MA, editor. 1974 NATO Advanced Study In-stitute. New York, Plenum Press, 1975, pp. 679-686.

5. Wagner HJ and Ali MA: Cone synaptic ribbons and retino-motor changes in the brook trout, Salvelinus fontinalis (Sal-monidae, Teleostei), under various experimental conditions.Can J Zool 55:1684, 1977.

6. Muntz WRA and Richard DS: Photomechanical movementsin the trout retina following brief flashes of light. Vision Res,22:529, 1982.

7. Richardson KC, Jarett L, and Finke EH: Embedding in epoxyresins for ultrathin sectioning in electron microscopy. StainTechnol 35:313, 1960.

8. Douglas RH and Wagner HJ: Endogenous patterns of pho-tomechanical movements and their relation to activity rhythmsin teleosts. Cell Tissue Res 226:133, 1982.

9. Daw NW and Pearlman AL: Pigment migration and adapta-tion in the eye of the squid, Loligo pealei. J Gen Physiol 63:22,1974.

10. Olivo RF and Larsen ME: Brief exposure to light initiatesscreening pigment migration in retinula cells of the crayfish,Procambarus. J Comp Physiol [A] 125:91, 1978.

11. Basinger SF and Matthes MT: The effect of long-term constant

light on the frog pigment epithelium. Vision Res 20:1143,1980.

12. Wagner HJ and Weiler R: Untersuchungen zur synaptischenPlastizitat von Horizontalzellen mit Hilfe von intrazellularerAbleitung und HRP Injektion. Verh Anat Ges, in press.

13. Raynauld JP, Laviolette JR, and Wagner HJ: Goldfish retina:a correlate between cone activity and morphology of the hor-izontal cell in cone pedicles. Science 204:1436, 1979.

14. John KR and Kaminester LH: Further studies on retinomotorrhythms in the teleost Astyanax mexicanus. Physiol Zool42:60, 1969.

15. Ali MA: The ocular structure, retinomotor and photobehav-ioural responses of juvenile Pacific salmon. Can J Zool 37:965,1959.

16. Ali MA: Histophysiological studies on the juvenile Atlanticsalmon (Salmo satar) retina. II. Responses to light intensities,wavelengths, temperatures, and continuous light or dark. CanJ Zool 39:511, 1961.

17. Douglas RH: An endogenous crepuscular rhythm of rainbowtrout (Salmo gairdneri) photomechanical movements. J ExpBiol 96:377, 1982.

18. Arey LB and Mundt GH: A persistent diurnal rhythm in visualcones. (Abstract) Anat Rec 79(suppl):41, 1941.

19. Welsh JH and Osborn CM: Diurnal changes in the retina ofthe catfish, Ameiurus nebulosus. J Comp Neurol 66:349, 1937.

20. Olla BL and Marchioni WW: Rhythmic movements of conesin the retina of bluefish, Pomatomus saltatrix, held in constantdarkness. Biol Bull 135:530, 1968.

21. Evans M and Burnside B: Cyclic nucleotide effects on teleostcone elongation. (Abstract) J Cell Biol 83(No 2, pt 2):344a,1979.

22. Cheze G: L'epiphyse des poissons Teleosteens—Techniquesd'approche. Recherches morphologiques et experimentales.Ph. D., Toulouse, 1974.

23. Cheze G: Recherche qualitative de derives indoliques dansl'epithelium pigmentaire retinien de la truite (Salmo irideus).Vision Res 18:1587, 1978.

24. Cheze G and Ali MA: Role de l'epiphyse dans la migrationdu pigment epithelial retinien lors d'une stimulation photiqueunilateral chez Salmo irideus. Rev Can Biol 35:69, 1976.

25. Cheze G, Peyraud C and Serfaty A: Etude des effets del'epiphysectomie sur la dynamique pigmentaire retinienned'un teleosteen: la truite, Salmo irideus L. C R Soc Biol (Paris)165:1985, 1971.

26. Kraus-Ruppert R and Lembeck F: Die Wirkung von Mela-tonin auf die Pigmentzellen der Retina von Froschen. PfliigersArch Ges Physiol 284:160, 1965.

27. Pang SF and Yew DT: Pigment aggregation by melatonin inthe retinal pigment epithelium and choroid of guinea-pigs,Cavia porcellus. Experientia 35:231, 1979.

28. Quay WB: Retinal and pineal hydroxyindole-O-methyl trans-ferase activity in vertebrates. Life Sci 4:983, 1965.

29. Quay WB: Comparative physiology of serotonin and mela-tonin. Adv Pharmacol 6A:283, 1968.

30. Quay WB and McLeod RW: Melatonin and photic stimulationof cone contraction in the retina of larval Xenopus laevis.(Abstract) Anat Rec 160:491, 1968.

31. Ali MA and Pickford GE: The effects of hypophysectomy onthe retinomotor responses of the killifish, Fundulus heierocli-tus. Rev Can Biol 38:139, 1979.

32. Flight WFG: Morphological and functional comparison be-tween the retina and the pineal organ of lower vertebrates.Prog Brain Res 52:131, 1979.

33. Vilter V: Regulation sympathico-hypophysaire de migrationpigmentaires dans la retine de l'anguille (Anguilla vulgaris L.)Arch Physiol Biol (Paris) 16 (suppl):6, 1942.


morphologic changes in teleost primary and secondary retinal cells

Documents