SUMMARYRainbow trout undergo natural cone degeneration and thus are interesting models for examining mechanisms of neural degeneration. They have ultraviolet-sensitive (UVS) cones that are lost over most of the retina during development; only a small functional population remains in the dorsal retina. How this spatial distribution of UVS cones is maintained is unclear. Thyroxine (T4) induces UVS cone loss, and local thyroid hormone regulation was hypothesized to control UVS cone distribution. Thyroid hormone receptor alpha (TRα), thyroid hormone receptor beta (TRβ) and Type 2 deiodinase (D2) regulate thyroid hormone exposure to target cells. Regional retinal expression of these genes was investigated during exogenous T4 treatment and natural smoltification of rainbow trout. Each retina from dark-adapted parr, T4-treated parr and natural smolts was divided into four quadrants, and total RNA was isolated. Quantitative real-time RT-PCR analysis demonstrated that all retinal quadrants had increased accumulation of TRβ transcripts 2 days post-T4 treatment, corresponding to initiation of SWS1 opsin downregulation. Smolts exhibited decreased accumulation of TRα and TRβ transcripts in all quadrants, but this effect was most pronounced in the dorso-temporal (DT) retinal quadrant where UVS cones persist. By contrast, in 2 day T4-treated parr, the DT quadrant showed increased expression of TRα and TRβ. Furthermore, D2 transcripts decreased in the DT quadrant of T4-treated parr but increased in the DT quadrant of smolts. These results suggest that T4 upregulates TRβ expression to initiate SWS1 opsin downregulation, while TRα and TRβ downregulation occurs to prevent natural loss of UVS cones from the DT retina.

The ability of developing rainbow trout Oncorhynchus mykiss embryos to compensate for elevated oocyte triiodothyronine (T 3 ) content and whether elevation of oocyte T 3 content within a physiologically meaningful range affects growth rates of the embryo or the expression of genes encoding for thyroid hormone receptors a (TRa) and b (TRb) were examined. Oocytes were immersed in ovarian fluid alone (control) or T 3 -enriched ovarian fluid prior to fertilization and water hardening, to induce a dose-dependant increase in oocyte T 3 content of c. 3 (control), c. 30 (LT 3 ) or c. 110 ng egg À1 (HT 3 ). To examine the interaction of embryo somatic growth with altered thyroid state more effectively, the embryos were reared at two ambient temperatures (8Á5 and 5Á5 C ) to induce different growth rates. A significant decline in whole embryo T 3 content was measured in the T 3 -treatment groups reared at both water temperatures by 3 weeks post-fertilization (dpf), and may have reflected the action of outer ring monodeiodinase, which was present in microsomes prepared from embryos 23 dpf. Whole embryo T 3 levels in the HT 3 group, however, remained higher than controls until phase 2 of development [the onset of endogenous thyroid hormone (TH) release]. This suggested that the embryos exerted some control over their response to exogenous TH, but that there was a limit to the level of control exerted by the embryonic tissues. Reverse transcriptase-polymerase chain reaction (RT-PCR) revealed the presence of mRNA encoding for the two TR isoforms as early as 26 dpf, and quantitative real-time RT-PCR (qPCR) was used to examine the effect of elevated oocyte T 3 content on the expression of these TR genes in embryos raised at 8Á5 and 5Á5 C, and sampled at similar developmental stages prior to the onset of embryonic TH synthesis, to ensure that the oocyte T 3 was the only source of TH exposure to the embryo. There was a suppression of the TRa gene expression in the control 5Á5 C group relative to the control 8Á5 C group. In addition, both TRa and TRb mRNA accumulation was lower, relative to the controls, in the LT 3 treatment group reared at 8Á5 C suggesting a suppressive effect of the lower level of T 3 treatment on the TR gene expression. Conversely, there were no differences from controls in the HT 3 treatment group, possibly indicating that this level of exposure overrides the downregulating capacity of the embryo. Similar patterns were seen for TRa and TRb mRNA accumulation in embryos reared at 5Á5 C, but because of the temperature suppressed level of TRa mRNA in the controls, significant affects of the LT 3 treatment were only found for TRb. There were no measurable effects of T 3 treatment on oocyte fertility or embryo somatic growth for either temperature treatment group, nor was somatic growth hormone content (measured only in the 8Á5 C treatment group) apparently related to in ovo T 3 levels. The results suggest †Author to whom correspondence should be addressed.

The development of the thyroid tissue in rainbow trout (Oncorhynchus mykiss) embryos between 20 and 88 days postfertilization (dpf) was studied using traditional histological methods in combination with immunostaining using antibodies raised against L-thyroxine (T4) and triiodo-L-thyronine (T3). In 20 dpf embryos, the thyroid primordium appeared as a simple tube aligned with the dorsal surface of the ventral aorta. The tubular primordium becomes progressively bifurcated at each end and follicles are formed by pinching off balls of cells from the tips of the bifurcations. The secretory activity of the thyroid tissue appears first as a synthesis phase, evident at 30 dpf, characterized by T4 and T3 immunostaining present only in the follicle lumen. A later hormone release or secretory phase was first evident for T4 immunostaining at 43 dpf and for T3 immunostaining at 46 dpf and was characterized by the presence of immunostaining both within the follicle lumen and also in the cytoplasm of some (but not all) thyrocytes. This stage was also coincident with a marked increase in the number of pituitary thyrotrophs. The proportion of embryos exhibiting immunostained thyrocytes increased progressively from 43 to 52 dpf, at which stage all embryos had T4- and T3-immunostained thyrocytes.

No abstract

No abstract