Richard H. White

Polarized Light Helps Monarch Butterflies Navigate

During their spectacular migratory journey in the fall, North American monarch butterflies (Danaus plexippus) use a time-compensated sun compass to help them navigate to their overwintering sites in central Mexico. One feature of the sun compass mechanism not fully explored in monarchs is the sunlight-dependent parameters used to navigate. We now provide data suggesting that the angle of polarized skylight (the e-vector) is a relevant orientation parameter. By placing butterflies in a flight simulator outdoors and using a linear polarizing filter, we show that manipulating the e-vector alters predictably the direction of oriented flight. Butterflies studied in either the morning or afternoon showed similar responses to filter rotation. Monarch butterflies possess the anatomical structure needed for polarized skylight detection, as rhabdoms in the dorsalmost row of photoreceptor cells in monarch eye show the organization characteristic of polarized-light receptors. The existence of polarized-light detection could allow migrants to accurately navigate under a variety of atmospheric conditions and reveals a critical input pathway into the sun compass mechanism.

Not all butterfly eyes are created equal: Rhodopsin absorption spectra, molecular identification, and localization of ultraviolet-, blue-, and green-sensitive rhodopsin-encoding mRNAs in the retina of Vanessa cardui

Surveys of spectral sensitivities, visual pigment spectra, and opsin gene sequences have indicated that all butterfly eyes contain ultraviolet‐, blue‐, and green‐sensitive rhodopsins. Some species also contain a fourth or fifth type, related in amino acid sequence to green‐sensitive insect rhodopsins, but red shifted in absorbance. By combining electron microscopy, epi‐microspectrophotometry, and polymerase chain reaction cloning, we found that the compound eye of Vanessa cardui has the typical ultrastructural features of the butterfly retina but contains only the three common insect rhodopsins. We estimated lambda‐max values and relative densities of the rhodopsins in the Vanessa retina (0.72, P530; 0.12, P470; and 0.15, P360) from microspectrophotometric measurements and calculations based on a computational model of reflectance spectra. We isolated three opsin‐encoding cDNA fragments that were identified with P530, P470, and P360 by homology to the well‐characterized insect rhodopsin families. The retinal mosaic was mapped by opsin mRNA in situ hybridization and found to contain three kinds of ommatidia with respect to their patterns of short wavelength rhodopsin expression. In some ommatidia, P360 or P470 was expressed in R1 and R2 opposed receptor cells; in others, one cell expressed P360, whereas its complement expressed P470. P530 was expressed in the other seven cells of all ommatidia. P470‐expressing cells were abundant in the ventral retina but nearly absent dorsally. Our results indicated that there are major differences between the color vision systems of nymphalid and papilionid butterflies: the nymphalid Vanessa has a simpler, trichromatic, system than do the tetrachromatic papilionids that have been studied. J. Comp. Neurol. 458:334–349, 2003.

The retina of Manduca sexta: rhodopsin expression, the mosaic of green-, blue- and UV-sensitive photoreceptors, and regional specialization

SUMMARY Spectral sensitivities of individual photoreceptors in the compound eye of Manduca sexta were verified by immunocytochemistry, and the retinal mosaic was mapped, using polyclonal antisera raised against amino-terminal sequences of three identified rhodopsins: P520, P450 and P357. Retinulae are composed of a small proximal cell and seven or eight elongate cells extending across the retina. In each retinula, one or two elongate dv cells oriented in the dorsal-ventral axis of the retinal lattice express either P450 or P357. Six elongate ap and ob cells in the anterior-posterior and oblique axes express P520. The small proximal pr cell also appears to express P520. The retinal mosaic is regionalized into three distinct domains: ventral and dorsal domains that divide the main retina, and a large dorsal rim area. The immunocytochemical data provide a high-resolution map of the Manduca retina that confirms and refines earlier low-resolution ERG spectral sensitivity measurements. The dorsal and ventral domains, separated at a well-defined equatorial border, are distinguished by differences in the proportion of blue-sensitive dv cells: these cells dominate the ventral retina but are less abundant in the dorsal retina. Green-sensitive ap and ob receptors are uniformly distributed across the dorsal and ventral domains, and UV-sensitive dv cells are fairly uniformly distributed because many retinulae in the dorsal domain contain only one dv cell. Similarly, dorsal rim retinulae contain only the ventral member of the dv pair of receptors, two-thirds of which express P357. Otherwise, dorsal rim receptors express none of the three sequenced Manduca opsins; they must express rhodopsins that have yet to be cloned.

Wavelength Discrimination and the Role of Ultraviolet Vision in the Feeding Behavior of Hawkmoths1

Nocturnal Sphingidae (hawkmoths or sphinx moths) are important pollinators in tropical forests. Hawkmoth flowers are typically white to the human eye. As the retinas of hawkmoths contain ultraviolet-sensitive photoreceptors, flower patterns reflecting ultraviolet wavelengths (that are not visible to humans) might be significant to sphingid feeding behavior. The flowers of ten hawkmoth-pollinated species were examined with an ultraviolet sensitive video system in Monteverde, Costa Rica. All were found to lack ultraviolet reflectance. A common hawkmoth species, Manduca sexta, whose range extends to Costa Rica was then used in laboratory free choice experiments to determine which wavelengths elicited proboscis extension, probing and drinking of sugar water. When offered a choice between artificial flowers or backlighted filters, Manduca strongly preferred to feed at those reflecting or transmitting only wavelengths longer than 400 nm, avoiding those that also included ultraviolet wavelengths. That is, feeding behavior was best elicited by stimuli that mimicked the reflectance of typical hawkmoth flowers. Feeding behavior must be primarily activated by either the green- or violet-sensitive mechanisms (or both) of the hawkmoth visual system, while concurrent activation of the ultraviolet-sensitive mechanism interferes with it.

Turnover of Photoreceptor Membrane in the Larval Mosquito Ocellus: Rhabdomeric Coated Vesicles and Organelles of the Vacuolar System

We have presented evidence in previous papers showing that the degradation and renewal of photoreceptor membrane in larval mosquito ocelli is regulated by light. Mosquito rhabdomeres grow larger in darkness and smaller in light as the amount of membrane increases and decreases. The kinetics of rhabdomere growth and diminution indicate that these changes arise from variations in the rate of membrane turnover (1).

Influence of carotenoid deficiency on visual sensitivity, visual pigment and P-face particles of photoreceptor membrane in the mothManduca sexta

Summary1.Several larval diets (Table 1) were developed for rearing the tobacco hornworm mothManduca sexta in an effort to control the synthesis of adult visual pigments (generically, ‘rhodopsins’) through the availability of their chromophore, retinaldehyde or, more likely, 3-hydroxyretinaldehyde.2.Rhodopsin was measured in difference spectra from detergent extracts of adult retinas. Opsin was identified and measured on SDS gels after electrophoretic separation of retinal proteins reduced with cyanoborohydride to convert rhodopsin to fluorescent N-retinyl opsin. The density of P-face particles in photoreceptor membranes was measured in freeze-fracture preparations. Visual sensitivity of compound eyes was measured from the electroretinogram (ERG).3.One diet containing corn meal and soy flour, rich sources of potential carotenoid precursors of the chromophore, producedfortified animals with the highest level of rhodopsin: 60 pM/retina. The addition of spinach leaves to the fortified diet did not increase the amount of rhodopsin. A second diet containing wheat germ producedintermediate moths with about 25% of the visual pigment of the fortified group. A third diet containing potato starch and lacking all sources of carotenoids except for a small amount of yeast produceddeprived animals whose visual pigment could not be measured but must have been less than 0.6 pM/retina (Fig. 1B).4.A band at 35–38 kDa on SDS gels prepared from cyanoborohydride-reduced extracts of fortified retinas was identified as n-retinyl opsin from its intense fluorescence. The fluorescence of the band was less intense in preparations from intermediate retinas. No fluorescence was detected in preparations of deprived retinas. However, this relatively insensitive assay would not allow detection of rhodopsin levels less than 6 pM/retina. When the gels were stained for protein, the density of the 35 kDa band from intermediate and deprived retinas was about 45% and 6%, respectively, of that from fortified retinas. Thus the relative density of the band from preparations of deprived retinas is about 6 times greater than the estimated maximum amount of rhodopsin present in extracts. Either there is excess opsin in the deprived retinas, or another minor protein runs at the same position on the gel as opsin (Fig. 2).5.P-face particle densities of rhabdomeric membrane ranged from 104/μm2 in the fortified animals to 4×103/μm2 in intermediate animals to 5×102/μm2 in deprived moths (Figs. 3, 4 and Table 2).6.The sensitivity of the intermediate and deprived animals averaged 55% and 0.06%, respectively, of that of the fortified animals (Fig. 1 A). Measurement of the ERG proved to be the simplest and most sensitive method for measuring visual impairment. If sensitivity remains linear with rhodopsin content at low concentrations, deprived retinas contain about 0.04 pM of rhodopsin.7.Visual sensitivity increased by 10 to 40-fold following the addition ofβ-carotene or xanthophyll to the deprived diet. Addition of either retinol or retinal did not significantly increase sensitivity (Fig. 1A).

Sex Differences in Sensory and Motor Branches of the Pudendal Nerve of the Rat

The morphology of the pudendal nerve was quantified in adult male and female rats. The sensory branch of the pudendal nerve was about three times as large in cross section in males as in females, and the motor branch was about five times as large. Electron microscopy was used to determine the ultrastructural bases of these gross size differences. Differences that were found included greater packing density of both myelinated and unmyelinated axons in females, larger myelinated and unmyelinated axons in males, larger myelin sheaths of sensory axons in males, more numerous myelinated axons in both branches of males, and more numerous unmyelinated axons in the sensory branch of males. There was also some indication that myelinated sensory axons were more likely to branch in the dorsal clitoral nerve of females than in the homologous nerve of males. Morphological differences in the structure of pudendal axons, their associated Schwann cells, and the extracellular matrix as well as differences in sensory and motor axonal number all have potential implications for the sexual differentiation of the central nervous system and behavior.

Adult stemmata of the butterfly Vanessa cardui express UV and green opsin mRNAs

Adult stemmata are distinctive insect photoreceptors located on the posterior surfaces of the optic lobes. They originate as larval eyes that migrate inward during metamorphosis. We used a combination of light microscopy and in situ hybridization to examine their anatomical organization in the butterfly Vanessa cardui and to test for the presence of visual pigments, the light sensitive components of the visual transduction pathway. The bilateral cluster of six internal stemmata is located near the ventral edge of the lamina. They retain the dark screening pigment and overlying crystalline cones of the larval stemmata. We found two opsin mRNAs expressed in the stemmata that are also expressed, respectively, in UV-sensitive and green-sensitive photoreceptor cells in the compound eye. A third mRNA that is expressed in blue-sensitive photoreceptor cells of the compound eye was not expressed in the stemmata. Our results reinforce the idea that the adult stemmata are not merely developmental remnants of larval eyes, but remain functional, possibly as components of the circadian input channel.

Carotenoid replacement inDrosophila: freeze-fracture electron microscopy

SummaryBecause of the consequent lack of photopigment chromophore, carotenoid/retinoid (vitamin A) deprivation during the larval period ofDrosophila leads to decreased rhodopsin in adult photoreceptors. Decreased density of P-face particles in photoreceptor membrane (rhabdomeric microvilli) is a prominent ultrastructural feature of this rhodopsin deficiency. When adults are fed carotenoid, the rhabdomeric P-face particle density-which reflects the concentration of rhodopsin-increases halfway to the replete control level during the first 12 hours, and is fully restored by 2 days. Based on freeze-fracture replicas, there is a continuity of membrane between rhabdomeric microvilli and the parent retinula cell. That confluence is relevant to turnover of photoreceptive membrane. Microvillar and retinula cell P-face particle densities covary. The relevance of the demonstration of rapid recovery from chromophore depletion is discussed in relation to hypotheses that the chromophore and/or related retinoids regulate opsin gene transcription, and/or post-translational processing and deployment from the endoplasmic reticulum to the rhabdomere.