Thyroxine induces a precocial loss of ultraviolet photo sensitivity in rainbow trout (Oncorhynchus mykiss, teleostei)

Thyroxine induces a precocial loss of ultraviolet photo sensitivity in rainbow trout (Oncorhynchus mykiss, teleostei)
of 10
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Vision Res. Vol. 32, No. 12, pp. 2303-2312, 1992 Printed in Great Britain. All rights reserved 0042-6989/92 $5.00 + 0.00 Copyright 0 1992 Pergamon Press Ltd Thyroxine Induces a Precocial Loss of Ultraviolet Photosensitivity in Rainbow Trout (Oncorhynchus mykiss, Teleostei) HOWARD I. BROWMAN,* CRAIG W. HAWRYSHYN* Received 8 January 1992; in revised form 19 May 1992 Small (<30 g) juvenile rainbow trout (Oncorhyncbus mykiss) possess retinal photoreceptor mechanisms sensitive to ultraviolet (UV), short (S), middle (M) and long (L) wavelengths. During normal development, the sensitivity peak of the UV cone mechanism (360 nm) shifts towards the S-wavelengths (to an intermediate ,, of 390 nm) until, at approx. 60 g, individuals are no longer sensitive in the UV (only a S-wavelength peak at 430 nm remains). This shift in spectral sensitivity is associated with the loss of small accessory corner cones from the retinal photoreceptor cell mosaic. Treating small (~30 g) rainbow trout with thyroid hormone induced a precocial loss of UV photosensitivity and an associated change in the retinal photoreceptor cell mosaic, identical to the events that occur during normal development. UV cones Fish vision Spectral sensitivity Heart-rate conditioning Ontogeny Smoltification Transformation Development INTRODUCTION Small ( < 30 g), juvenile rainbow trout (Oncorhynchus mykiss) possess retinal photoreceptor mechanisms sensi- tive to ultraviolet (UV), short (S), middle (M) and long (L) wavelengths (Hawryshyn, Arnold, Chaisson & Martin, 1989; Hawryshyn, 1992). During normal devel- opment, the sensitivity peak of the UV cone mechanism shifts towards the S-wavelengths until, at approx. 60 g, individuals are no longer sensitive in the IJV (Hawryshyn et al., 1989). This shift in spectral sensitivity is associated with the loss of small accessory corner cones from the retinal photoreceptor cell mosaic (Hawryshyn et al., 1989; this study). Recent evidence indicates that these corner cones contain the zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA   photo- pigment (Hawryshyn et al., 1989; Bowmaker, Thorpe & Douglas, 1991; Loew & Wahl, 1991; Hawryshyn & Harosi, unpublished microspectrophotometric data). Thus, the loss of UV photosensitivity apparently results from the disappearance of UV-sensitive cones from the retina. In the research reported here, we take up the question of what triggers the developmental loss of W photosensitivity in rainbow trout. In many fishes and amphibians, thyroid hormones are associated with the transition from larva to adult (meta- morphosis)(Norris, 1983; Inui & Miwa, 1985; Lam, *Laboratory of Sensory Biology, Department of Biology, University of Victoria, P.O. Box 1700, Victoria, British Columbia, Canada V8W 2Y2. 1985) and with significant morphological, physiological and biochemical changes in the visual system (Evans & Fernald, 1990; Hoskins, 1990). Thyroxine (T4) and its metabolic precursors have also been implicated in another ontogenetic event, the parr to smolt transform- ation in juvenile salmonid fishes. This process of smoltifi- cation usually occurs when the freshwater resident form, Parr, undergoes pronounced biochemical, endocrine, physiological and morphological changes in preparation for the migration to seawater as a smolt (Thorpe, Bern, Saunders & Soivio, 1984; Dickhoff & Sullivan, 1987; Hoar, 1988). Thyroxine appears to regulate the rate of smolt development and treatment with T, induces sev- eral of the morphogenetic changes associated with nor- mal smoltification, including changes in the visual system (Hoar, 1988; Dickhoff, Brown, Sullivan & Bern, 1990). For example, treatment with T, alters A,/A, photopigment ratios in the retinae of some fishes (e.g. Beatty, 1972; Allen, 1977; Tsin & Beatty, 1979). Despite the significant body of work on thyroxine-in- duced changes in scotopic visual pigment ratios, corre- lations with more functional aspects of visual function, such as photopic spectral sensitivity, are rare (though see Muntz & Northmore, 1970; Whitmore & Bowmaker, 1989; Browman & Hawryshyn, 1991). Based upon the action of thyroid hormones on the visual systems of amphibia and fishes, we tested the hypothesis that treating small (< 30 g) rainbow trout (a salmonid) with thyroid hormone would result in changes in their pho- topic spectral sensitivity. Specifically, we examined 2303  1304 HOWARD I BROWMAN and CRAIG W HAWRYSHYh the capacity of T4 to induce a precocial loss of UV photosensitivity and an associated change in the retinal photoreceptor cell mosaic, analogous to the events that occur during normal development. MATERIALS AND METHODS zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Animals and hormone treatment An undomesticated, non-anadromous population of rainbow trout (Badger Lake, British Columbia), raised in outdoor ponds at the Fraser Valley Trout Hatchery in Abbotsford, British Columbia, was used in these experiments. After transport to our holding facility, fish were maintained at 15°C on a 12: 12 hr light:dark photoperiod for at least 8 weeks prior to the first experiments. Fish were fed daily with BioDiet Grower pellets (BioProducts Inc., Warrenton, Ore.) containing fixed proportions of vitamins, nutrients and pigments. All experiments were initiated at least 1 hr after lights on in the holding facility and were terminated at least 1 hr before lights off. Illumination in the holding facility was provided by fluorescent bulbs at an intensity of 33.54 f 14.39 ,uW/cm2 (integrated irradiance, 200-l 100 nm, measured with a Photodyne Inc. radiometer). Experimental animals were held in 5 1 tubs containing 300 pg/l L-thyroxine sodium salt (Sigma Chemical Laboratories, T-2376) dissolved in 0.1 M NaOH. Half of the treatment solution was replaced daily. This manner of hormone treatment has been used to induce photo- pigment changes, and metamorphosis, in fishes (e.g. Cristy, 1974; Miwa & Inui, 1987a, b; Miwa, Tagawa, Inui & Hirano, 1988; Hirata, Kurokura & Kasahara, 1989) and produces a significant elevation in the serum titer of T, in rainbow trout (Allen, 1977). Control fish were handled in an identical manner, but no T4 was added to their water. In order to minimize weight gain during the experiment, fish were fed on a maintenance diet. Handling and maintenance of animals was in accordance with the guidelines set out by the Canadian Council on Animal Care. Heart -rate conditioning experiments Immobilization procedure, fish set -up and optical sys - tern. Fish were anesthetized by immersion in tricaine methane sulfonate (MS-222), at 50 mg/l for 3-5 min and then immobilized with a 0.00012 mg/g body wt intra- muscular injection (total volume injected at several sites) of pancuronium bromide (“Pavulon”, Organon Labora- tories). Immobilization ensured that the stimulus was presented to the same part of the retina throughout the experiment. Test fish were placed in a Plexiglass re- strainer, immersed in a black test tank (80 1) and respired at 400 ml/min with aerated, 15°C water. The fish’s left eye was positioned 5.0 cm from the center of a circular Pyrex window (10 cm in diameter) covered with an Albanene (Kuffler & Esser Co.) screen. The eye was positioned so that the pupillary plane was perpendicular to the stimulus at a 20” roll, the latter so that the stimulus was presented to the ventral retina. A three- channel optical system, optically and vibrationally isolated from the test tank, produced 12 cm d~arneter colored background fields upon which 20 monochro matic stimuli (340--720 nm) were superimposed. All three optical channels contained 250 W tungsten lamps (EHJ Spectra Lamps). The intensity of illumination in thesr channels was controlled by neutral density filters. For each wavelength-intensity combination presented to the test fish, the energy (W cm’) was converted t\, yuantal photon irradiance and corrected for wavelength-depen- dent variations in photon energy. Complete details of the optical system used in these experiments are presented in Hawryshyn and Harosi (1991). Chromatic adaptation was used to isolate the spectral sensitivity of the UV cone mechanism. The illumination used to achieve this effect consisted of a yellow background (550,nm low-pass cut-off filter, Corion), which differen- tially light adapted the M- and L-wavelength sensitive cone mechanisms, and a narrow-band blue background (460 nm narrow band interference filter, Corion) which light adapted the S-wavelength sensitive mechanism. The same background conditions were used during all train- ing sessions and experiments. Fish were allowed a mini- mum of 60 min to adapt to the background conditions before initiation of a training session or experiment. Conditioning protocol. After recovery from the anes- thetic, paralyzed fish were heart-rate conditioned by pairing a 300 msec. 2 3 mA shock (delivered to the caudal peduncle) with monochromatic visual stimuli (Hawryshyn & Beauchamp. 1985). Heart beat was mon- itored with a Grass P-15 amplifier and integrated into a microcomputer system that measured interbeat intervals during prestimulus (10 set) and stimulus (3 set) periods. A criterion response was defined when a single stimulus period interbeat interval was at least 1.05 times longer than the mean pre-stimulus period interbeat interval. Variability (standard deviation of the mean pre-stimulus interbeat interval) in the fish’s heartbeat was measured during each pre-stimulus period so that the robustness of the criterion level could be evaluated throughout the experiment. A criterion of 1.05 was only used when the absolute value of the standard deviation around the mean pre-stimulus interbeat interval was 2% or less of that value. During conditioning a bright 620 nm stimulus was presented every 60 set until the fish gave five consecutive criterion responses. Thereafter, and in order to achieve appropriate levels of generalization in the response, a series of stimuli of different wavelengths and intensities were presented until two consecutive criterion responses were obtained at each combination. At the end of each conditioning or experimental session three blanks were presented to assure that responses were related to the light stimulus and not to extraneous factors such as sound or vibration. To negate the stimulus path’s shutter as a potential cue, a second shutter, positioned nearby, opened and closed at random intervals. Threshold determination and experimental design. Threshold determinations began 3060 min after com- pletion of a training session. Thresholds were determined by presenting a subthreshold stimulus intensity followed  PRECOCIAL LOSS OF UV PHOTOSENSITIVITY 2305 TABLE 1. Weight, length, age (days from hatching to experiment, +4) and duration of controls and thyroxine treatments for rainbow trout used in spectral sensitivity experiments Age Weight Length Fish No. Exoerimen t cdavs wst hatch) w zyxwvutsrqponmlkjihgfedcbaZYXWV mm) 169 Pretreatment 251 14.7 114 169 23 day thyroxine treatment 278* 16.5 117 169 36 day thyroxine treatment 291 16.6 121 170 Pre-treatment 223 12.3 109 170 22 day thyroxine treatment 277 16.8 116 170 36 day thyroxine treatment 291 17.7 119 172 Pre-treatment 241 13.0 107 172 21 day thyroxine treatment 276 17.6 114 181 Pre-treatment 262 10.1 109 181 22 day control 298 10.9 111 181 36 day control 312 12.5 116 181 36 day thyroxine treatment 375 23.2 135 182 Pre-treatment 278 15.5 119 182 22 day control 299 15.3 123 182 33 day control 314 19.0 129 I82 23 day thyroxine treatment 362 32.0 149 186 Pre-treatment 309 11.3 111 186 23 day thyroxine treatment 363 23.7 135 188 Pre-treatment 272 10.1 104 I88 23 day control 300 8.1 104 188 35 day control 313 9.0 104 188 23 day thyroxine treatment 364 17.4 118 190 Pre-treatment 318 21.3 132 190 23 day control 367 21.7 145 197 Pre-treatment 325 16.1 122 197 23 day control 368 14.6 119 204 Pre-treatment 384 15.1 118 204 39 day thyroxine treatment 435 28.8 129 205 Pre-treatment 385 17.0 123 205 45 day thyroxine treatment 443 22.9 135 *Thyroxine exposure was usually initiated approx. 1 week after the pre-treatment experiment by increasing intensity steps (0.1 log unit every 30 see) until the test fish gave two consecutive criterion responses. Threshold intensity was taken as the first intensity level to which the fish responded. To minimize the incidence of false positives, the threshold intensity had to be preceeded by at least three negative responses (i.e. three 0.1 log-unit intensity increments). Detection thresholds for mono- chromatic stimuli from 340 to 720 nm, presented in semi-random order so that the same cone mechanism was not tested several times in succession, were determined. Since fish survive the experiments, spectral sensitivity curves were obtained from the same individuals prior to initiation of T, treatment (or control) and after approx. 3 and 5 weeks of exposure (or control) (Table 1). Pre-treat- ment spectral sensitivity curves were obtained for a total of eleven fish. Thyroxine treatment usually began several days after the pre-treatment spectral sensitivity curve was obtained. Before calculating mean spectral sensitivity curves, absolute values of log photon sensitivity were normalized for inter-individual variability. zyxwvutsrqponmlkjihgfedcbaZYXWVUTSRQPONMLKJIHGFEDCBA Histological procedures Twelve fully light-adapted rainbow trout were sac- rificed by spinal section; eyes were enu~leated immedi- ately. Five fish were controls and five had been treated with T, (as above) (Table 2). In addition, one small (HC12) and one large (HC13) fish, of the same chrono- logical age, were also sacrificed (Table 2). Eyecups were immersed in primary fixative (2.5% glutaraldehyde, 1 parafo~aldehyde, 3% sucrose, in 0.06 M phosphate buffer) where they remained, refriger- ated, for 12-24 hr. Prior to immersion, the cornea, iris, lens and some of the vitreal fluid were removed to improve penetration of the primary fixative. Retinae were then dissected into four pie-shaped wedges repre- senting dorsal, ventral, nasal and temporal regions. Small notches were cut along the edge of each wedge as orientation markers. The retina and pigmented epi- thelium were separated from the sclera and each pie- wedge of retina was thereafter processed individually.  2306 HOWARD I. BROWMAN and CRAIG W. HAWRYSHYN TABLE 2. Weight, length, age (days from hatching to experiment, _t4) and duration of comtol~ and thyroxine treatments for rainbow trout used in histological experiments Fish No. Experiment HI 42 day thyroxine treatment H2 42 day thyroxine treatment H3 42 day thyroxine treatment H4 42 day thyroxine treatment H9 69 day thyroxine treatment As Weight (days post hatch) (g) 381 20.0 384 13.0 3x3 5.5 3x5 I?..0 446 14.0 Length (mm) I30 1’0 I?(, 120 II‘; HCl HC2 HC3 HClO HCll 42 day control 42 day control 42 day control 67 day control 69 day control 390 ?I .O 12s 391 18.0 1’5 392 20.5 13s 450 31.5 135 445 28.5 140 HC12 Small fish 661 16.9 120 HC13 Large fish 653 75.4 193 After primary fixation, the tissue was washed in buffer (0.06 M phosphate buffer, 3% sucrose, pH 7.3) post- fixed in 1% OsO,, rinsed, dehydrated in a graded ethanol series, infiltrated and embedded in Polybed 812 follow- ing the methodology of Ali and Anctil (1976). Tissue from the central ventral retina, the area to which stimuli were presented in the spectral sensitivity experiments, were sectioned tangentially (1 pm thick) to the base of the cone outer segments. Sections were stained with Richardson’s stain for light microscope examination. RESULTS Spectral sensitivity Small rainbow trout (< 30 g) possess four cone mech- anisms: UV-, S-, M- and L-wavelength sensitive (Hawryshyn et al., 1989). The ,I,,, of these cone mechan- isms are 360, 430, 535 and 620 nm respectively. During normal development there is a shift in the A,,,,, of the UV mechanism from 360 nm in individuals smaller than approx. 30g to 390 nm in 60 g fish. Fish weighing 70-90 g are no longer UV sensitive (see Figs 4 and 5 in Hawryshyn et al., 1989). All pretreatment and control fish in the current exper- iments exhibited sensitivity peaks at UV and S wave- lengths [Fig. l(A)]. M and L mechanisms were also present, although their sensitivity was depressed by the adapting background (Fig. 1). The same pattern was observed for all individual control fish [Fig. l(B)]. The UV-cone mechanism sensitivity points were most effec- tively fitted with a 360 nm ,I,,,,, visual pigment absorption curve, corrected for ocular media absorption (Fig. 1). This is consistent with microspectrophotometric esti- mates of a 370-380 nm a,, for the UV-sensitive cone pigment in rainbow trout (Hawryshyn & Harosi, unpub- lished observations) and with estimates for UV-sensitive cones in other species (Harosi, 1985; Hawryshyn & Beauchamp, 1985; Whitmore & Bowmaker, 1989; Bowmaker et al., 1991; Hawryshyn & Harosi, 1991). Spectral sensitivity points in the neighboring S-wave- length region were most effectively fitted with a 430 nm &lax visual pigment absorption curve, corrected for ocular media absorption (Fig. 1). This is consistent with observations on the S-wavelength sensitive cones of brown trout Salmo trutta) (Bowmaker & Kunz, 1987). There was no significant change in the mean nor within-individual spectral sensitivity curves of the con- trol fish during these experiments [Fig. l(A. B)]. Both UV- and S-wavelength sensitive cone mechanisms re- mained as described above. The mean spectral sensitivity curves for T,-treated fish exhibited a shift in the Itimax f the UV mechanism towards S-wavelengths until, after approx. 5 weeks of exposure, UV sensitivity was lost [Fig. 2(A)]. All within- individual spectral sensitivity curves exhibited the same trend [Fig. 2(B)]. After 3 weeks of exposure to T,, the UV-cone sensi- tivity points could no longer be fitted with a 360 nm I.,;,, ocular media corrected visual pigment absorption curve. These points were most effectively fitted with a 390 nm i,,, visual pigment absorption curve, corrected for ocular media absorption (Fig. 2). After 5 weeks of exposure to T,, there was no longer a sensitivity peak in the UV. These fish possessed only S-, M- and L-sensitive cone mechanisms. The S-sensitive cone mechanism points for these fish were most effec- tively fitted with a 430 nm A,.,,,, isual pigment absorption curve, corrected for ocular media absorption (Fig. 2). There was no pronounced lateral movement of the sensitivity points for the M- and L-wavelength sensitive cone mechanisms (Fig. 2). However, the relative sensi- tivity of the L-wavelength cone mechanism increased during the progressive loss of the UV sensitivity peak (Fig. 2). Visual pigment absorption curves The visual pigment absorption curves fitted to our data were generated by an eighth-order polynomial template for vertebrate cone visual pigments (Bernard, 1987; Gary D. Bernard, personal communication). The nomogram’s main absorption peak was generated from absorption spectrum data for chicken iodopsin (Wald, Brown & Smith, 1955). The long-wavelength tail was generated from suction-pipette data for vertebrate cones (Nunn, Schnapf & Baylor, 1984). These two data sets  PRECOCIAL LOSS OF UV PI-IOTOSENSITIVITY 2307 zyxwvutsrqpon  A) I n= 11 I l 5 weeks control (13.50 f 5.07 g) l 3 weeks control (15.23 f 7.51 g) * Pre -treatment (14.23 f 3.34 g) I I I I 1 300 400 500 600 Wavelength (nm I 700 800 .A A l * l Fish 188: 36 days control (9.0 g) l Fish 188 : 23 days control (8.1 g) b Fish 188 : before (70.1 g 1 A I I I I I 300 400 500 600 Wavelength nm) 700 800 FIGURE I. (A) Mean spectral sensitivity curves for rainbow trout used as controls obtained, from the same individuals, after 3 and 5 weeks of treatment. (B) Spectral sensitivity curves from a single control fish after 3 and 5 weeks of treatment. A yellow background was used to “isolate” the UV sensitive cone mechanism in all experiments. The 360 and 430 nm I,, visual pigment absorption curves were compared with the spectral sensitivity of all fish. Note: (i) visual pigment absorption curves (corrected for ocular media absorption) are represented by solid lines; (ii) spectral sensitivity curves were arbitrarily arranged on the ordinate; (iii) 1 division on the ordinate = 1 log unit. were pooled and plotted as relative absorbance vs &,,,/A, following MacNichol (1986). The pooled template data were then log-transformed and best-fitted to an eighth-order polynomial function. This function pro- duces absorption curves for cone visual pigments that are much tighter approximations than those calculated from extensions of the Ebrey and Honig (1977) nomograms. As an example, we plotted the microspectro- photometrically determined absorption curve for the UV photopigment of carp, Cyprinus carpio (Hawryshyn & Harosi, 1991) and superimposed upon it visual pigment absorption curves for a pigment with the same Amax generated from both the Ebrey and Honig nomogram and the Bernard eighth-order polynomial function; the latter clearly produces a better fit (Fig. 3). Histology of the ventral retina The ventral retina of the small fish (HC 12) contained a regular square mosaic of cones, consisting of a single cone surrounded by four double cones [Fig. 4(A)]. Small single cones were found at each corner of this square mosaic. Cone cells in the retinae of all five control fish ex- hibited the same mosaic pattern [Fig. 4(C)]. Although of the same chronological age as small fish HC 12, the ventral retina of the large fish (HC13) contained no small single cones at the corners of the square mosaic [Fig. 4(B)].
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks