The long slender retinal cones of fishes shorten in the light and elongate in the dark. Light-induced cone shortening provides a useful model for stuying nonmuscle contraction because it is linear, slow, and repetitive. Cone cells contain both thin (actin) and thick (myosinlike) filaments oriented parallel to the axis of contraction. This study examines the polarities of the cone's thin filaments and the changes in filament distribution which accompany light-induced contraction, in an attempt to elucidate the structural basis for the cone's contractile process. The proximal half of the cone is fixed to its cellular neighbors in the outer nuclear layer while the distal half is free. Thus, all shortening takes place in a necklike region (the myoid) in the distal half of the cone which extends into the space between the neural retina and the pigmented retinal epithelium. Thin filaments are found throughout the length of the cone, whereas thick filaments occur predominantly in the proximal (axon) regions of both light- and dark-adapted cones. Thus, thick filaments are primarily localized outside the region where shortening takes place. Observations from myosin subfragment-1 binding studies suggest that the cone's thin filaments are organized into two opposing sets. In the distal half of the cone (including the myoid), virtually all filaments have proximally directed arrowheads. In the more proximal regions of the axon, many thin filaments have opposite polarity, their arrowheads being distally directed. Near the synaptic proximal end of the light-adapted (contracted) cone, filaments of opposite polarities occur in approximately equal numbers. Thus, in the cone axon there appear to be two overlapping sets of actin filaments whose opposite polarities correspond to the two actin halves of a muscle sarcomere. In elongated, dark-adapted cones, thick filaments are localized throughout the axon region of the cone. In light, thick filaments accumulate towards the proximal end of the cone. These observations are consistent with a "sliding hypothesis" for cone contraction, in which thick myosinlike filaments produce sliding interdigitation of the two sets of oppositely directed actin filaments in the proximal axon region. Thus, the myoid thin filaments would be essentially reeled into the axon region to produce shortening. The mechanism of re-elongation depends on microtubules, as discussed in the companion paper.

The retinal cones of teleost fish contract at dawn and elongate at dusk. We have previously reported that we can selectively induce detergent-lysed models of cones to undergo either reactivated contraction or reactivated elongation, with rates and morphology comparable to those observed in vivo. Reactivated contraction is ATP dependent, activated by Ca2+, and inhibited by cAMP. In addition, reactivated cone contraction exhibits several properties that suggest that myosin phosphorylation plays a role in mediating Ca2+-activation (Porrello, K., and B. Burnside, 1984, J. Cell Biol., 98:2230-2238). We report here that lysed cone models can be induced to contract in the absence of Ca2+ by incubation with trypsin-digested, unregulated myosin light chain kinase (MLCK) obtained from smooth muscle. This observation provides further evidence that MLCK plays a role in regulating cone contraction. We also report here that lysed cone models can be induced to contract in the absence of Ca2+ by incubation with high concentrations of MgCl2 (10-20 mM). Mg2+-induced reactivated contraction is supported by inosine triphosphate (ITP) just as well as by ATP. Because ITP will not serve as a substrate for MLCK, this finding suggests that Mg2+-activation of contraction does not require myosin phosphorylation. Although Ca2+-induced contraction is completely blocked by cAMP at concentrations less than 10 microM, cAMP has no effect on cone contraction activated by unregulated MLCK or by high Mg2+ in the absence of Ca2+. Because trypsin digestion of MLCK cleaves off not only the Ca2+/calmodulin-binding site but also the site phosphorylated by cAMP-dependent protein kinase, and because Mg2+ activation of cone contraction circumvents MLCK action altogether, both these observations would be expected if cAMP inhibits reactivated cone contraction by catalyzing the phosphorylation of MLCK and thus reducing its affinity for Ca2+, as has been described for smooth muscle. Together our results suggest that in lysed cone models, myosin phosphorylation is sufficient for activating cone contraction, even in the absence of other Ca2+-mediated events, that cAMP inhibition of contraction is mediated by cAMP-dependent phosphorylation of MLCK, and that 10-20 mM Mg2+ can activate actin-myosin interaction to produce contraction in the absence of myosin phosphorylation.

In cells of the teleost retinal pigment epithelium (RPE), melanin granules disperse into the RPE cell's long apical projections in response to light onset, and aggregate toward the base of the RPE cell in response to dark onset. The RPE cells possess numerous microtubules, which in the apical projections are aligned longitudinally. Nocodazole studies have shown that pigment granule aggregation is microtubule-dependent (Troutt, L. L., and B. Burnside, 1988b Exp. Eye Res. In press.). To investigate further the mechanism of microtubule participation in RPE pigment granule aggregation, we have used the tubulin hook method to assess the polarity of microtubules in the apical projections of teleost RPE cells. We report here that virtually all microtubules in the RPE apical projections are uniformly oriented with plus ends toward the cell body and minus ends toward the projection tips. This orientation is opposite that found for microtubules of dermal melanophores, neurons, and most other cell types.

We have been using lysed cell models of teleost retinal cones to examine the mechanism of contraction in nonmuscle cells. We have previously reported that dark-adapted retinas can be lysed with the detergent Brij-58 to obtain cone motile models that undergo Ca++- and adenosine triphosphate (ATP)-dependent reactivated contraction. In this report we further dissect the roles of ATP and Ca++ in activation of contraction and force production by (a) characterizing the Ca++ and nucleotide requirements in more detail, (b) by analyzing the effects of inosine triphosphate (ITP) and the ATP analog ATP gamma S and (c) by testing effects of cyclic adenosine monophosphate (cAMP) on reactivated cone contraction. Exposing lysed cone models to differing free Ca++ concentrations produced reactivated contraction at rates proportional to the free Ca++ concentration between 3.16 X 10(-8) and 10(-6) M. A role for calmodulin (CaM) in this Ca++ regulation was suggested by the inhibition of reactivated contraction by the calmodulin inhibitors trifluoperazine and calmidazolium ( R24571 ). The results of analysis of nucleotide requirements in lysed cone models were consistent with those of smooth muscle studies suggesting a role for myosin phosphorylation in Ca++ regulation of contraction. ATP gamma S and ITP are particularly interesting in that ATP gamma S, on the one hand, can be used by kinases to phosphorylate proteins (e.g., myosin light chains) but resists cleavage by phosphatases or adenosine triphosphatases (ATPases), e.g., myosin ATPase. ITP, on the other hand, can be used by myosin ATPase but does not support Ca++/calmodulin mediated phosphorylation of myosin light chains by myosin light chain kinase. Thus, these nucleotides provide an opportunity to distinguish between the kinase and myosin ATPase requirements for ATP. When individual nucleotides were tested with cone motile models, the nucleotide requirement was highly specific for ATP; not only ITP and ATP gamma S, but also guanosine triphosphate, cytosine triphosphate, adenylyl-imidodiphosphate (AMPPNP) failed to support reactivated contraction when substituted for ATP throughout the incubation. However, if lysed cones were initially incubated with ATP gamma S and then subsequently incubated with ITP, the cones contracted to an extent that was comparable to that observed with ATP. As observed in skinned smooth muscle, adding cAMP to contraction medium strongly inhibited contraction in lysed cone models.

Retinal rod photoreceptors of teleost fish elongate in the light and shorten in the dark. Rod cell elongation and shortening are both mediated by actin-dependent mechanisms that occur in the inner segment myoid and ellipsoid. The intracellular signaling pathways by which light and dark regulate the actin cytoskeleton in the inner segment are unknown. To investigate the intracellular signals that regulate teleost rod motility, we have been using mechanically isolated rod inner/outer segments (RIS-ROS) obtained from the retinas of green sunfish, Lepomis cyanellus. In culture, RIS-ROS retain the ability to elongate in response to light; myoids elongate 15-20 microns in length during 45 min of light culture. A pharmacological approach was taken to investigate the role of cyclic nucleotides, cyclic nucleotide-dependent kinases, and protein phosphatases in the regulation of RIS-ROS motility. Millimolar concentrations of cAMP and cGMP analogues were both found to inhibit light-induced myoid elongation and two cyclic nucleotide analogues, SpCAMPS and 8BrcGMP, promoted myoid shortening after RIS-ROS had elongated in response to light. The cyclic nucleotide-dependent kinase inhibitor, H8, mimicked light by promoting myoid elongation in the dark. The effects of H8 were dose dependent, with maximal elongation occurring at concentrations of approximately 100 microM. Similar to the effects of cyclic nucleotide analogues, the phosphatase inhibitor, okadaic acid (0.1-10 microM), inhibited light-induced elongation and promoted shortening. The results presented here suggest that RIS-ROS motility is regulated by protein phosphorylation: phosphorylation in the dark by cyclic nucleotide-dependent protein kinases promotes rod shortening, while dephosphorylation in the light promotes rod elongation.

Agents which elevate cyclic AMP (cAMP) cause teleost retinal rods to contract. We have characterized this cAMP effect and have evaluated the role of the cytoskeleton in cyclic nucleotide-induced contraction, using actin and microtubule inhibitors. The necklike myoid region of the rod contracts in the dark and elongates in the light. If long, light-adapted rods are cultured with cAMP analogs and IBMX, rods contract to their short dark-adapted position. Cyclic nucleotide-induced rod contraction occurs in constant light, requires a phosphodiesterase inhibitor, and is specific to cAMP (db cyclic GMP, 8-bromocyclic GMP, 5'AMP, and adenosine have no effect on rod myoid length). Cyclic AMP effects on rod length are consistent with observations from several species that cAMP levels are higher in dark-adapted than in light-adapted retinas. Since rod myoids contain paraxially aligned actin filaments and microtubules, we have used the motility inhibitors cytochalasin D and cold and nocodazole to investigate the roles of these cytoskeletal elements in rod contraction. Cyclic nucleotide-induced contraction is not inhibited when myoid microtubules are disrupted with cold and nocodazole treatments, but contraction is blocked if myoid actin filaments are disrupted with cytochalasin D. Thus, we conclude that actin filaments, but not microtubules, are required for rod contraction. We propose that rod contraction in vivo is triggered by a rise of cytoplasmic cAMP at onset of darkness and that this contraction is mediated by an actin-dependent mechanism.

Teleost retinal rods elongate when exposed to light. Elongation is mediated by a narrow necklike region called the myoid. In the cichlid Sarotherodon mossambicus, the rod inner segment (composed of the myoid with adjacent ellipsoid) increases in length from 12 micrometers in the dark to 41 micrometers in the light. Long light-adapted myoids contain longitudinally oriented microtubules and bundles of parallel 60-A filaments that we have identified as actin by their ability to bind myosin subfragment 1. In short dark-adapted myoids, only microtubules are recognizable. Colchicine experiments reveal that light-induced rod elongation can occur in the absence of myoid microtubules. Intraocular injections of colchicine at concentrations that disrupt virtually all rod myoid microtubules do not block rod elongation. However, rod elongation is blocked by intraocular injections of cytochalasin B or cytochalasin D. The hierarchy of effectiveness of these drugs is consistent with their effectiveness in inhibiting actin assembly and in disrupting other actin-dependent motile processes. On the basis of ultrastructural observations and the results of these inhibitor studies, we propose that the forces responsible for rod elongation are dependent not on microtubules but on actin filament assembly.

We have examined the effects of changes in extracellular ionic composition on cone and retinal pigment epithelium (RPE) retinomotor movements in cultured isolated teleost retinas. In vivo, the myoid portion of teleost cones contracts in the light and elongates in the dark; RPE pigment disperses in the light and aggregates in the dark. In vitro, cones of dark-adapted (DA) retinas cultured in constant darkness contracted spontaneously to their light-adapted (LA) positions if the culture medium contained greater than or equal to 10(-3)M Cao++. DA cones retained their long DA positions in a medium containing less than or equal to 10(-6)M Cao++. Low [Ca++]o (10(-5)-10(-7)M) also permitted darkness to induce cone elongation and RPE pigment aggregation. Light produced cone contraction even in the absence of Cao++, but the extent of contraction was reduced if [Ca++]o was less than 10(-3) M. Thus, full contraction appeared to require the presence of external Ca++. High [K+]o (greater than or equal to 27 mM) inhibited both light-induced and light-independent Ca++-induced cone contraction. However, low [Na+]o (3.5 mM) in the presence of less than or equal to 10(-6)M Cao++ did not mimic light onset by inducing cone contraction in the dark. High [K+]o also promoted dark-adaptive cone and RPE movements in LA retinas cultured in the light. All results obtained in high [K+]o were similar to those observed when DA or LA retinas were exposed to treatments that elevate cytoplasmic cyclic 3',5'-adenosine monophosphate (cAMP) content.

As a test of the hypothesis that cyclic nucleotides play a role in the regulation of retinomotor movements and disc shedding in the photoreceptor-pigment epithelial complex, we have used an in vitro eyecup preparation that sustains both disc shedding and cone retinomotor movements, Eyecups were prepared in white light from animals in which both shedding and cone movement had been blocked by 4 d of constant-light treatment. In eyecups incubated for 3 h in light, disc shedding was negligible and cones remained in the light-adapted (contracted) position. In eyecups incubated in darkness, however, a massive shedding response (dominated by rod photoreceptors) was induced, and at the same time cone photoreceptors elongated to their dark-adapted position. In eyecups incubated in light dbcAMP promoted cone elongation and thus mimicked darkness; the dbcAMP effect was potentiated by the phosphodiesterase inhibitors papaverine and 3-isobutylmethylxanthine. In eyecups incubated in darkness, on the other hand, both phosphodiesterase inhibitors and dbcAMP reduced the phagosome content of the pigment epithelium. The effects of dbcAMP on the cone elongation and rod shedding appear to be specific in that dbcGMP, adenosine, and adenosine 5'-monophosphate had no significant effect. Our results suggest that cAMP plays a role in the regulation of both retinomotor movements and disc shedding.

Teleost retinal cones elongate in the dark and contract in the light. In isolated retinas of the green sunfish Lepomis cyanellus, cone myoids undergo microtubule-dependent elongation from 5 to 45 micron. We have previously shown that cone contraction can be reactivated in motile models of cones lysed with Brij-58. Reactivated contraction is both actin and ATP dependent, activated by calcium, and inhibited by cAMP. We report here that we have obtained reactivated cone elongation in lysed models prepared by the same procedures. Reactivated elongation is ATP dependent, activated by cAMP, and inhibited by calcium. The rate of reactivated elongation is proportional to the cAMP concentration between 10 microM and 0.5 mM, but is constant between 10 microM and 1.0 mM Mg-ATP. No elongation occurs if cAMP or Mg-ATP concentration is less than or equal to 5 microM. Mg-ATP is required for both cAMP-dependent and cAMP-independent processes, suggesting that Mg-ATP is required both for a regulatory process entailing cAMP-dependent phosphorylation and for a force-producing process. Free calcium concentrations greater than or equal to 10(-7) reduce the elongation rate by 78% or more, completely inhibiting elongation at 10(-5) M. This inhibition is not due to competition from calcium-activated contraction. Cytochalasin D blocks reactivated contraction, but does not abolish calcium inhibition of reactivated elongation. Thus calcium directly affects the elongation mechanism. Calcium inhibition is calmodulin dependent. The calmodulin inhibitor trifluoperazine abolishes calcium inhibition of elongation. Furthermore, calcium blocks elongation only if present during the lysis step; subsequent calcium addition has no effect. However, if calcium plus exogenous calmodulin are subsequently added, elongation is again inhibited. Thus calcium inhibition appears to require a soluble calmodulin which is lost shortly after lysis.