Circadian rhythms in Limulus photoreceptors. I. Intracellular studies.

The sensitivity of the lateral eye of the horseshoe crab, Limulus polyphemus, is modulated by efferent optic nerve impulses transmitted from a circadian clock located in the brain (Barlow, R. B., Jr., S. J. Bolanowski, and M. L. Brachman. 1977. Science. 197:86-89). At night, the efferent impulses invade the retinular, eccentric, and pigment cells of every ommatidium, inducing multiple anatomical and physiological changes that combine to increase retinal sensitivity as much as 100,000 times. We developed techniques for recording transmembrane potentials from a single cell in situ for several days to determine what circadian changes in retinal sensitivity originate in the primary phototransducing cell, the retinular cell. We found that the direct efferent input to the photoreceptor cell decreases its noise and increases its response. Noise is decreased by reducing the rate of spontaneous bumps by up to 100%. The response is increased by elevating photon catch (photons absorbed per flash) as much as 30 times, and increasing gain (response per absorbed photon) as much as 40%. The cellular mechanism for reducing the rate of spontaneous quantum bumps is not known. The mechanism for increasing gain appears to be the modulation of ionic conductances in the photoreceptor cell membrane. The mechanism for increasing photon catch is multiple changes in the anatomy of retinal cells. We combine these cellular events in a proposed scheme for the circadian rhythm in the intensity coding of single photoreceptors.

Retinal anatomy of Chorocaris chacei, a deep-sea hydrothermal vent shrimp from the Mid-Atlantic Ridge.

Exploration of deep-sea hydrothermal vents over the past quarter century has revealed that they support unique and diverse biota. Despite the harsh nature of the environment, vents along the Mid-Atlantic Ridge are dominated by large masses of highly motile Bresiliid shrimp. Until 1989, when it was discovered that the vent shrimp Rimicaris exoculata possesses a hypertrophied dorsal eye, many believed that animals populating hydrothermal vents were blind. Chorocaris chacei (originally designated Rimicaris chacei) is a Bresiliid shrimp found at hydrothermal vent fields along the Mid-Atlantic Ridge. Like R. exoculata, C. chacei has a hypertrophied retina that appears to be specialized to detect the very small amount of light emitted from the orifices of black smoker hydrothermal vent chimneys. C. chacei lacks the sophisticated compound eyes common to other decapod crustaceans. Instead, it has a smooth cornea, with no dioptric apparatus, apposed by a tightly packed, massive array of photosensitive membrane. Photoreceptors in the C. chacei retina are segmented into a hypertrophied region that contains the photosensitive membrane and an atrophied cell body that is roughly ten times smaller in volume than the photosensitive segment. The microvillar photosensitive membrane is consistent in structure and ultrastructure with the rhabdoms of decapod and other invertebrate retinas. However, the volume density of photosensitive membrane (> or =60%) exceeds that typically observed in invertebrate retinas. The reflecting pigment cells commonly found in decapod retinas are represented in the form of a matrix of white diffusing cells that exhibit Tyndall scattering and form an axial sheath around the photoreceptors. All photoreceptor screening pigment granules and screening pigment cells are restricted to the region below the photoreceptor nuclei and are thereby removed from the path of incident light. No ultrastructural evidence of rhythmic cycling of photosensitive membrane was observed. The morphological adaptations observed in the C. chacei retina suggest that it is a high-sensitivity photodetector that is of functional significance to the animal.

Retinal anatomy of a new species of bresiliid shrimp from a hydrothermal vent field on the Mid-Atlantic Ridge.

A new species of shrimp (Rimicaris sp.) was recently collected from the Snake Pit hydrothermal vent field on the Mid-Atlantic Ridge. Until the discovery in 1989 that the deep-sea, hydrothermal vent species, Rimicaris exoculata, possessed a hypertrophied dorsal eye, everyone believed that animals recovered from vent environments were blind. Like R. exoculata, Rimicaris sp., a small orange bresiliid shrimp, has an enlarged dorsal eye specialized for detecting light in a very dim environment instead of the expected compound eye. The individual lenses characteristic of a compound eye adapted for imaging have been replaced in Rimicaris sp. by a smooth cornea underlain by a massive array of photosensitive membrane. The number of ommatidia in this species is about the same as in shrimp species that live at the surface; however, the photoreceptors are larger in the deep-sea species and the shape of the photoreceptors is markedly different. The light-sensitive region of the photoreceptor is much larger than those of other shrimp and the rest of the receptor is much smaller than normal. All screening pigment has moved out of the path of incident light to a position below the retina, and the reflecting pigment cells have adapted to form a bright white diffusing screen between and behind the photoreceptors. The ultrastructure of the microvillar array comprising the rhabdom is typical for decapod crustaceans; however, there is a much greater volume density of rhabdom (80% to 85%) than normal. There is no ultrastructural evidence for cyclic rhabdom shedding or renewal. Rimicaris sp. has apparently adapted its visual system to detect the very dim light emitted from the throats of the black smoker chimneys around which it lives.

Circadian change in function of Limulus ventral photoreceptors.

Efferent fibers from a central circadian clock innervate photoreceptors along the ventral nerve of Limulus and release octopamine when active. We have recorded ERG-like responses from the ventral eye in vivo over several day periods. We have also used intracellular microelectrodes to study changes in ventral photoreceptor function during exogenous applications of octopamine (the putative efferent neurotransmitter), IBMX (a phosphodiesterase inhibitor), and forskolin (an adenylate cyclase activator): (1) Responses to light measured at night from ventral photoreceptors in vivo are greater in amplitude than those recorded during the day; (2) Octopamine and agents that increase intracellular levels of cAMP in ventral photoreceptors decrease the rate of spontaneous (dark) bumps, increase photoreceptor response to light without changing threshold, and often increase the bump duration; and (3) These changes in function of ventral photoreceptors are similar to those that have been observed in the photoreceptor of the lateral eye during circadian clock activity at night, and in vitro in the presence of those same pharmacological agents.

Theory of the response of the limulus retina to periodic excitation.

Analytical and numerical solutions are given for several problems which arise from a mathematical description of inhibitory interactions in the Limulus retina. The theory developed here takes into account the delay in lateral inhibition and the gradual decay of lateral and self-inhibition. Emphasis is laid on the calculation of responses to excitation fields which are periodic in time and either spatially uniform or of traveling wave type. The analytical solutions given are intended to help experimenters determine the range in which certain linearized equations and reduced measures of excitation are useful for the design and interpretation of experiments. Certain of the numerical solutions obtained describe intrinsically non-linear effects, such as "periodic bursting" under constant excitation.

Theory of delayed lateral inhibition in the compound eye of limulus.

Although the magnitude of lateral inhibition in the retina of Limulus polyphemus depends strongly on the distance between ommatidia, the time of delay tau between the response of one ommatidium and the consequent inhibition of another is independent of the distance between them and is approximately 0.1 sec. Moreover, experimental observations indicate that for intact undamaged retinae in intense spatially uniform illumination, the total inhibitory influence at a given ommatidium can exceed the excitation of that ommatidium. A simple theory combining these two known facts yields the following conclusions: The response of a healthy Limulus retina to an intense, temporally constant, and spatially uniform excitation e (in the usual Hartline-Ratliff units) should be a sustained synchronous oscillation of period 2tau, with the mean response over one complete cycle equal to approximately 1/2e. Under a broad class of circumstances the sustained oscillations should take the form of a succession of "bursts" and "rest periods", each of duration tau. The oscillations can have, however, a fine structure in which there is repeated information about the duration and sequence of any short pulses of excitation which occurred within a time-interval of length tau before the onset of steady excitation. Because it is repeated every 2tau seconds, this information remains available in the retina for interaction with subsequent changes of excitation.

Octopamine modulates photoreceptor function in the Limulus lateral eye.

Activity at night in efferent nerve fibers from a central circadian clock produces changes in photoreceptor function in the lateral compound eye of Limulus: the response to light is increased; membrane potential fluctuations (bumps) occurring in the dark are suppressed; and the duration of bumps occurring both in the dark and under dim illumination is increased (Barlow et al., 1977; Kaplan & Barlow, 1980; Barlow, 1983; Barlow et al., 1985). Efferent nerve terminals release octopamine when activated (Battelle et al., 1982; Battelle & Evans, 1984, 1986); exogenous octopamine in vitro produces some of the changes resulting from efferent nerve activity in vivo (Kass et al., 1988). We report here that the increase in both on-transient and steady-state response to light induced by octopamine in the lateral eye in vitro are concentration dependent with threshold at or below 100 nM, saturation at or above 100 microM, and half-maximal increase in the range 1-10 microM. Octopamine also reduces bump activity in the dark in a concentration-dependent way. Unlike the increase in light response, the dependence of this effect on octopamine concentration is extremely variable from specimen to specimen. The effects of exogenous octopamine on light response and bump activity can sometimes be reversed by removing octopamine from the medium bathing the in vitro preparation. Octopamine also increases bump duration, apparently in a concentration-dependent manner. We have not succeeded in reversing this increase in bump duration. The concentration dependence of changes in photoreceptor response described here agrees qualitatively with the dependence of cAMP levels on octopamine in Limulus photoreceptors (Kaupp et al., 1982), lending further support to the idea that cAMP acts as a second messenger in the circadian control of photoreceptor function. Our results also suggest that the changes induced in the transient and steady-state response to light by both efferent nerve activity and exogenous octopamine have a common origin, which may differ from that responsible for the modulation of bump activity.

Photoreceptor cells dissociated from the compound lateral eye of the horseshoe crab, Limulus polyphemus, I: Structure and ultrastructure.

Isolated photoreceptors are desirable for whole-cell and patch-clamp studies of functional properties of visual processes that cannot be clearly analyzed when the photoreceptors are coupled. The retina of the compound lateral eye of the horseshoe crab, Limulus polyphemus, was dissociated into individual retinular cells using an enzyme pretreatment consisting of collagenase, papain, and trypsin, and a two-stage mechanical dissociation. These photoreceptors are functionally viable in an organ culture medium for up to 1 week and possess naked arhabdomeral and rhabdomeral segment membranes which are easily accessible for whole-cell recordings. A dissection technique was also developed whereby the retinal epidermis and neural plexus, as well as the second-order eccentric cells, could be separated from the ommatidia of the compound lateral eye in one simple step, providing viable isolated ommatidia attached to the cornea. The enzyme pretreatment used for dissociating the retina was then used to remove the individual ommatidia from the corneal cones. Hoffman modulation contrast microscopy was used to develop a reliable method for sorting and collecting viable isolated retinular cells for morphological and electrophysiological studies. Morphological analysis using light microscopy and scanning and transmission electron microscopy revealed that isolated retinular cells are morphologically nearly identical to retinular cells in situ. Isolated retinular cells possess a normal rhabdomere with no apparent loss of microvillar membrane as a result of the isolation process. Ommatidia can presently be isolated with up to six retinular cells possessing essentially normal structure and ultrastructure including thick rays of rhabdom. Isolated ommatidia possess naked A-segment membranes which are also well suited for whole-cell recording techniques.

Photoreceptor cells dissociated from the compound lateral eye of the horseshoe crab, Limulus polyphemus, II: Function.

A combination of enzymatic digestions and mechanical disruption was used to isolate photoreceptor cells from the compound lateral eye of the horseshoe crab, Limulus polyphemus. The cells were maintained in a culture medium and tested for function using whole-cell and cell-attached patch configurations of the gigaseal technique. The cells dissociated from the eye generated spontaneous voltage and current bumps in the dark, and depolarized in a graded fashion to increasing intensities of light over several decades, producing responses similar to those of cells in vivo. Currents evoked during voltage clamp were similar to those in ventral photoreceptor cells of Limulus, although transient currents in the dark- and light-activated currents were smaller in isolated lateral eye cells, perhaps because of the slow speed and spatial nonuniformity of the clamp in these large cells. In addition to isolated cells, dissociation of the compound eye produced small clusters of cells and isolated ommatidia which were also tested for function. Comparison of the electrical characteristics of isolated cells with those of cells in small clusters and in their ommatidial matrix suggests that the electrical junctions normally connecting photoreceptor cells within an ommatidium are functional in the latter groups, but not in isolated cells. Cell-attached patches of rhabdomeral membrane of isolated cells contained light-activated channels, resembling those observed in ventral photoreceptor cells, but no voltage-activated channels. Similar patches of arhabdomeral membrane contained voltage-activated channels, but no light-activated channels. We conclude that this preparation is suitable for studies of processes involved in generating the light response in invertebrate photoreceptor cells.

Efferent neurotransmission of circadian rhythms in Limulus lateral eye. II. Intracellular recordings in vitro.

We investigated efferent neurotransmission in the Limulus lateral eye by studying the action of pharmacological agents on responses of photoreceptor cells in vitro. We recorded transmembrane potentials from single cells in slices of retina that were excised during the day and maintained for several days in a culture medium. Potentials recorded in the absence of pharmacological agents resemble those recorded from cells in vivo during the day. Octopamine, a putative efferent neurotransmitter, induced changes in photoreceptor potentials that mimicked in part those generated at night by a circadian clock located in the brain. Specifically, octopamine (100 to 500 microM) decreased the frequency of occurrence of quantum bumps in the dark and increased the amplitude of photoreceptor responses to intermediate and high light intensities. Similar actions were produced by naphazoline (25 to 100 microM, potent agonist of octopamine), forskolin (8 to 400 microM, activator of adenylate cyclase), IBMX (1 mM, inhibitor of phosphodiesterase), and 8-bromo-cAMP (500 microM, analogue of cAMP). 8-bromo-cGMP (500 microM, analogue of cGMP) decreased the rate of spontaneous quantum bumps only. Our results support the hypothesis that (1) octopamine is an efferent neurotransmitter of circadian rhythms in the Limulus eye and that (2) it activates adenylate cyclase to increase levels of the second messenger, cAMP, in photoreceptor cells. Circadian changes in photoreceptor responses to moderate intensities may be a specific action of cAMP, since cGMP has no effect. Circadian changes in the rate of spontaneous quantum bumps may involve a less specific intermediate, since both cAMP and cGMP reduce bump rate. Characteristics of the retinal slice preparation precluded a detailed study of the effects of pharmacological agents on retinal morphology.

Efferent Innervation to Limulus Eyes In Vivo Phosphorylates a 122 kD Protein.

Efferent fibers innervate all of the eyes of the horseshoe crab, Limulus polyphemus. Driven by a circadian clock located in the central nervous system, the activity of the fibers at night is responsible for anatomical, biochemical, and physiological changes in the eyes, which increase their ability to detect and respond to light. We showed previously that octopamine, a putative efferent neurotransmitter, stimulates the phosphorylation of a 122 kD protein in in vitro preparations of both ventral and lateral eyes by means of a cAMP-dependent mechanism. We now report that phosphorylation of the 122 kD protein in the lateral eye is enhanced in vivo: (1) at night, in correlation with efferent nerve input activated by the circadian clock; and (2) during the day, in response to electrical stimulation of efferent axons. We show further that the 122 kD protein is enriched in, and may be restricted to, tissues that contain photoreceptors. We postulate that this protein is involved in the efferent-stimulated increase in retinal sensitivity.

The morphology of the dorsal eye of the hydrothermal vent shrimp, Rimicaris exoculata.

The bresiliid shrimp, Rimicaris exoculata, lives in large masses on the sides of hydrothermal vent chimneys at two sites on the Mid-Atlantic Ridge. Although essentially no daylight penetrates to depths of 3500 m, very dim light is emitted from the hydrothermal vents themselves. To exploit this light, R. exoculata has evolved a modified compound eye on its dorsal surface that occupies about 0.5% of the animal's body volume. The eye's morphology suggests that it is extremely sensitive to light. The cornea of the dorsal eye is smooth with no dioptric apparatus. The retina consists of two wing-shaped lobes that are fused across the midline anteriorly. The rhabdomeral segments of the 7000 ommatidia form a compact layer of photosensitive membrane with an entrance aperture of more than 26 mm2. Within this layer, the volume density of rhabdom is more than 70%. Below the rhabdomeral segments, a thick layer of white diffusing cells scatters light upward into the photoreceptors. The arhabdomeral segments of the five to seven photoreceptors of each ommatidium are mere strands of cytoplasm that expand to accommodate the photoreceptor nuclei. The rhabdom is comprised of well-organized arrays of microvilli, each with a cytoskeletal core. The rhabdomeral segment cytoplasm contains mitochondria, but little else. The perikaryon contains a band of mitochondria, but has only small amounts of endoplasmic reticulum. There is no ultrastructural indication of photosensitive membrane cycling in these photoreceptors. Vestigial screening pigment cells and screening pigment granules within the photoreceptors are both restricted to the inner surface of the layer of the white diffusing cells. Below the retina, photoreceptor axons converge in a fanshaped array to enter the dorsal surface of the brain. The eye's size and structure are consistent with a role for vision in shrimp living at abyssal hydrothermal vents.

Sulfide as a Chemical Stimulus for Deep-Sea Hydrothermal Vent Shrimp.

Organisms dependent on deep-sea hydrothermal vents for their existence face extinction when their vents expire, unless they can establish populations on neighboring vents or on new vent sites. Propagules, including larvae and motile adults, are readily dispersed broadly by seafloor currents, but how they recognize active hydrothermal sites is problematical. Compelling evidence that vent organisms can find and colonize hydrothermal sites has been provided by a series of observations on the East Pacific Rise (1). New hydrothermal vents created there following a volcanic eruption on the seafloor in March 1991 were colonized by sessile invertebrates in less than one year. On the Mid-Atlantic Ridge, shrimp that normally cluster on sulfide surfaces have been observed to swim directly back to the surfaces when displaced from them. How do vent animals locate new or existing vents? Passive transport by currents (2) or active swimming without guidance by some physical cue is not likely to result in success (3). Chemicals present in hydrothermal fluids have been proposed as attractants. We provide the first evidence of a chemosensory response in a vent invertebrate to sulfides, which are prevalent in vent fluids and provide the energy,for chemosynthetic primary production at vents.