Search

Vitamin D-binding protein (DBP) is the major transport protein for the vitamin D sterols in a variety of mammalian and avian species. The DBP found in the emydid family of turtles is unique in that it exhibits high-affinity binding of both vitamin D and thyroxine (D/TBP). Sequence analysis has revealed that the emydid D/TBP is homologous to mammalian DBP and shares no homology with mammalian thyroxine-binding globulins. Northern blot analysis was used to examine the spatial profile of D/TBP transcription in the turtle Trachemys scripta. In both adults and hatchlings, two transcripts—the expected 1.5-kb full-length transcript and a second 0.8-kb transcript—were present in nearly all of the tissues examined. In adults, highest expression of the 1.5-kb transcript was seen in the kidney, gonad, and spleen, with lower levels in the liver and lung and no transcripts in skeletal muscle. In hatchlings, the full-length transcript was detected in a variety of tissues at similar levels. Injection of hatchlings with thyroxine increased levels of circulating D/TBP and transcript levels. These data are in marked contrast to observations in mammals in which transcription of DBP is confined predominantly to the liver. Further, the increase in circulating D/TBP associated with increased thyroidal activity may result from a direct or indirect activation of D/TBP transcription by thyroxine.

Thyroid hormone metabolism by 5′-monodeiodinase enzymes (5′MD) was characterized in peripheral tissues of the turtle,Trachemys scripta,and compared with activity measured in the rat. Based on differences in pH dependence, sensitivity to inhibitors, substrate affinity, and cofactor requirements, at least two types of enzyme activities have been identified in the turtle. A 5′MD activity was measured in liver and kidney microsomal fractions that exhibits inhibition by 2n-propyl-6-thiouracil (PTU), a higher affinity for rT3(Km=2 μM) than for T4(Km=6.5 μM), a low cofactor dependence, and a high pH optimum for T4metabolism. The characteristics of this turtle low affinity T4activity correspond to the mammalian type I monodeiodinase. A second type of monodeiodinase (MD) activity that is less sensitive to PTU, has a higher affinity for T4(Km=1 nM), a higher cofactor requirement, and a lower pH optimum was colocalized with the first form. Both turtle MD activities remain active over a range of temperatures, allowing for activity at the preferred body temperature of this species (28 to 37°C compared to the 37°C optimum in the rat). Based on limited comparative data of MD systems from several fish and birds, the turtle most closely resembles avian species. Like birds, turtles possess a mammalian-like type I activity and have colocalized MD forms in the liver. However, the second turtle MD form (MDH) is not comparable to the mammalian or avian MDII-like activity. Analysis of the deiodinase products from both turtle MDs by high-performance liquid chromatography confirmed that the putative turtle MDI produces T3from T4as expected. The MDH produces rT3from T4as does the mammalian type III form, but MDH has a wider tissue distribution (kidney, liver, pancreas, heart, ovary, and brain) and distinct enzyme kinetics. Moreover, MDH activity in the turtle kidney is 100-fold higher than in the liver, indicating that the kidney may play a critical role in the metabolism of thyroid hormones in the turtle; this high renal activity distinguishes the turtle from all other vertebrates studied.

In vitroeffects of purified RanaLH and FSH on cyclic AMP accumulation were examined in ovaries of immature and mature bullfrog, Rana catesbeiana, at different states of the sexual cycle. RanaFSH was active with clear dose dependency only in immature ovaries but was less potent than RanaLH. Mature females did not respond to RanaFSH. RanaLH always increased the cAMP accumulation in mature females but the response was lowest with prespawning and postspawning females.

We tested the effects of low concentrations of 17β-estradiol (E₂) on the in vitro pituitary responsiveness to gonadotropin-releasing hormone (GnRH) in frogs (Rana pipiens). We first showed that the pH indicator phenosulfonpthalein (phenol red. PR) at levels normally used in the culture medium did not significantly affect pituitary basal or GnRH-stimulated gonadotropin secretion in male glands, as may occur in mammals. Incubation of male glands with E₂ at 10 pg/ml for 24 hr in medium lacking PR did not alter basal follicle-stimulating hormone (FSH) or luteinizing hormone (LH) secretion, but significantly attenuated responsiveness of both gonadotropins to GnRH (0.2, 1, and 5 ng/ml); E₂ did not affect responses at the highest dose of GnRH (25 ng/ml). GnRH responsiveness of female pituitaries was also inhibited by E₂ at all doses of GnRH; 1 ng/ml E₂ had greater effects than did 10 pg/ml. The present results demonstrate that E₂ at concentrations commonly found in the circulation of both sexes have a consistent direct inhibitory effect on FSH and LH secretion at the level of the pituitary. Copyright 1993, 1999 Academic Press

To further investigate the action of gonadotropin-releasing hormone (GnRH) on pituitaries of female frogs (Rana pipiens), gonadotropin (Gth) secretion and synthesis was measured following either the in vivoor in vitroexposure to GnRH. Treating frogs with a GnRH-agonist (GnRHa) in vivofor either 2 or 5 days led to an elevation in plasma Gth compared to that of saline-injected controls. In general, the pattern of Gth secretion in vitroparalleled changes observed in their plasma levels. GnRHa-treated frogs also had depressed pituitary content of Gth as compared to controls. Pituitaries from frogs treated with GnRHa for 5 days had higher levels of Gth synthesis than those of controls; however Gth synthesis in females treated with GnRHa for only 2 days was not elevated. After 30 hr of continuous exposure to 100 ng/ml GnRH in vitro, glands from females secreted markedly higher levels of LH and FSH than glands not exposed to the secretagogue. The levels of LH and FSH contained within pituitaries exposed to GnRH were significantly less than those of control glands. Prolonged exposure to GnRH elevated both the secretion of radiolabeled Gth as well as increased the total synthesis of Gth (compared to those of controls). These results indicate that GnRH can stimulate the biosynthesis of Gth in pituitaries of female frogs: however these results differ from those previously obtained with males. It appears that female glands require a much longer duration of GnRH stimulation than males before Gth synthesis is elevated.

Uptake and metabolism of exogenous steroids [testosterone (T), estradiol (E₂), and corticosterone (B)] were examined in anuran larvae. Radioimmunoassay (RIA) of steroids dissolved in water revealed that as much as 0.03 mg of T or E₂ were cleared from water by a single Xenopus laevis tadpole in 24 hr. Studies using RIA in combination with counting of total [³H]steroids suggested that clearance was mainly due to metabolism and not to uptake and storage by the tadpole. Thin-layer chromatography and HPLC analysis confirmed metabolism and revealed that T was metabolized to at least three ether-extractable compounds as well as to water soluble compounds (nonextractable in ether) by nine anuran species, representing eight families. Most species metabolized E₂ to nonextractable compounds, but some produced organic metabolites. In contrast, B was not taken up, cleared, or metabolized by any of the nine species. Studies in X. laevis using [³H]T and radioinert T in the same container confirmed that the decrease in radioimmunoassayable T was due to metabolism as determined by TLC. In a separate study, [³H]T was metabolized more extensively when injected directly into the tadpole than when dissolved in water. Additional studies with X. laevis revealed that steroids were also metabolized in tadpole-conditioned water (water exposed to tadpoles), but the metabolites produced represented <10% of those produced when tadpoles were present. This metabolism was confined to particulate matter from the gut of tadpoles. Metabolites extracted from tadpoles were similar to metabolites found in the water. Copyright 1993, 1999 Academic Press

The influence of a dopaminergic antagonist, metoclopramide (MET), and an agonist, bromocriptine (BROMO), on reproductive status was examined in female frogs, Rana temporaria.MET induced advanced ovulation during hibernation, suggesting dopaminergic inhibition of gonadotropin (LH) release during this period. BROMO did not decrease plasma Lit in intact females in comparison with vehicle (VEH)-treated controls (VEH: 11 ± 6 vs BROMO: 5 ± 4 ng/ml) or in sham-lesioned (SL) females (SL.; 12 ± 5 vs SL + BROMO: 9 ± 8 ng/ml). However, BROMO significantly depressed the rise in plasma LH following lesioning (L) which disconnected the hypothalamus from the median eminence-pituitary complex (L + BROMO: 29 ± 10 vs L: 74 ± 30 ng/ml; P< 0.002). Taken together with previous results of lesion studies, these data point to an important role of dopaminergic inhibition in the regulation of seasonal reproduction in this frog.

Highly specific antisera and reversed-phase HPLC were used to examine the nature and distribution of gonadotropin-releasing hormone (GnRH) in the brains and plasma of three species of Rana frog previously reported to vary in the types and brain distributions of GnRH. The three species examined, Rana pipiens, Rana esculenta, and Rana ridibunda, exhibit differences in total concentration and relative concentrations of GnRH in brain areas, but all three contain the mammalian and chicken-II forms of GnRH (mGnRH and cGnRH-II). Both forms were found in the telencephalon and diencephalon, but mGnRH is consistently the most abundant form in the preoptic-hypothalamic area (e.g., ratios of mGnRH:cGnRH-II > 3:1 in hypothalamus-median eminence), whereas, cGnRH-II is the most abundant in the telencephalon and the only form measured in the cerebellum and medulla. The total content of cGnRH-II in the whole brain is about 1.5-2 times higher than that of mGnRH, due to the larger size of the areas outside the preoptic-hypothalamic area. These general patterns were the same for adults and juveniles. We found no evidence of a third form of GnRH corresponding to salmon GnRH in hypothalamus or other brain areas of R. esculenta as previously reported. These analyses also revealed the presence of both forms of GnRH in plasma draining the hypothalamic area, but the concentration of cGnRH-II is relatively higher than that in the corresponding hypothalamic tissue, suggesting differentially greater release or slower degradation of this form. These results do not support the conclusion that the ranid frogs are highly variable in the nature and distribution of GnRH in the brain, although they suggest that both forms of GnRH are potentially involved in the direct regulation of pituitary gonadotropes. Copyright 1994, 1999 Academic Press

The seasonal gonadal cycle (including gonadal histology, sex steroids, and gonadotropins) was studied in freshly captured common snapping turtles,Chelydra serpentina,from Wisconsin, and the effects of capture stress were evaluated. The ovarian and testicular cycles are shorter than those reported in other freshwater turtles; the cycles commence in mid-May and terminate in early September, immediately after the completion of gonadal growth and maturation. In the female, testosterone (T), 17β-estradiol (E2), and progesterone (Pro) were highly correlated with follicular growth and vitellogenesis. Ovulation in captivity and under natural conditions occurred after mid-May. In captivity, ovulation was a rapid process (24–48 hr); as the follicles descended into the uterine horns there was a significant increase in E2and Pro and eggs were retained in the uterine horns for about 2 weeks before oviposition. In the male, T was sig nificantly correlated with testicular growth and spermiation. Follicle-stimulating hormone (FSH) reached significantly higher levels (P<0.01) in males (8.99±0.38 ng/ml) than in females (2.66±0.22 ng/ml), but luteinizing hormone was undetectable in both sexes. FSH was not correlated with the steroids in either sex. Sex steroids and FSH began to rise before spermiation and vitellogenesis and remained elevated until completion of gonadal growth and maturation. Leydig cells, the main source of plasma androgen in this species, became active shortly after emergence from hibernation and remained steroidogenically active for the rest of the cycle. Sertoli cells became active only after spermatogenesis was under way but also stayed active for the rest of the summer. Courtship and mating behaviors were observed in spring, summer, and fall. The snapping turtle is strictly aquatic with no basking behavior and limited behavioral thermoregulation so there is little daily fluctuation in body temperature. Environmental correlates indicate that the snapping turtle is temperature dependent: recrudesence occurs with a slight increase in water temperature during spring and early summer, while a dramatic drop in gonadal activity accompanies a slight decrease in temperature in fall. Changes in temperature may underlie changes in gonadal activity in the face of relatively stable FSH. Male turtles subjected to captivity and periodic blood sampling show a significant decline in T. The hormonal levels continued to decline whether the turtles are exposed to optimum or extreme temperatures. However, there is more rapid decline in T values in animals with regressed testes (June) than in those with well-developed testes (July). Male and female turtles kept in captivity at different phases of the cycle exhibit different patterns and degrees of response to stress, possibly related to the hormonal levels and the condition of the gonads.

Morphologically distinct secretory cells in the pituitary pars distalis and pars tuberalis of larval and adult toads (Bufo boreas) immunoreact with antiserum to the beta subunit of human thyroid-stimulating hormone (hTSHβ). The single type of secretory cell in the pars tuberalis has sparse secretion granules uniformly smaller than those of similarly immunoreactive cells in the pars distalis. Thyrotropin immunoactivity appears in pars tuberalis and pars distalis before gonadotropin immunoreactivity during early development. Antisera which distinguish gonadotropes (stained with human and sea turtle LHβ) and thyrotropes (stained with human TSHβ) as separate cell types in the pars distalis of the adult toad immunoreact with the same single type of cell in the pars distalis of the tadpole up through metamorphosis, suggesting the existence of a single pluripotent, glycoprotein-producing precursor cell early in development. Gonadotropin antisera do not react with the pars tuberalis in tadpoles or adults.

Structural homology between the high-affinity thyroxine (T₄)-binding protein (TBP) in the plasma of the turtle, Trachemys scripta, and vitamin D-binding proteins (DBP) of mammals prompted an investigation of plasma vitamin D binding in the turtle. Several lines of evidence indicate that the TBP represents the primary binding protein for 25-OH-cholecalciferol (D₃) in the turtle plasma. D₃-binding protein in whole plasma migrates in the same position as TBP by size-exclusion chromatography and polyacrylamide gel electrophoresis; it is electrophoretically distinct from sex hormone-binding proteins. D₃ binding to purified TBP alone is enhanced (up to sevenfold) in the presence of plasma proteins, including albumin; with this correction, the D₃-binding activity of plasma corresponds to expected TBP titers. Plasma selectively stripped of TBP by affinity chromatography and purified turtle albumin have only trace D₃-binding activity. Variations in physiological state (thryoidal status, age, and sex) previously associated with variable T₄ binding land TBP levels) in T. scripia show correlated variability in plasma D₃ binding. D₃ binding is also highly correlated with T₄ binding (r = 0.81; P &lt; 0.001) for plasma samples taken from 30 adults representing 10 different species of Trachemys. D₃ binding in plasma exhibits a high-affinity site (K_a = 2.3 × 10⁸ M^-1) and a second lower-affinity (K_a = 2 × 10⁶ M^-1), higher-capacity site; capacities are highly variable. Purified TBP has a comparable high affinity (K_a = 2.8 × 10⁸ M^-1), with a capacity close to 1 mol/mol. Binding of T₄ and D₃ are not competitive, indicating separate binding sites for the two ligands; in fact, T₄ tends to enhance the affinity and the capacity for D₃. A single protein in turtle plasma ("TBP/DBP") functions in the transport of two different hormones normally served by two distinct binding proteins (representing different multigene families) in mammals. These results have implications for the mediation of T₄ effects on growth. Copyright 1994, 1999 Academic Press

Gonadotropin-releasing hormone (GnRH)-induced desensitization of pituitary luteinizing hormone (LH) secretion in female turtles (Trachemys scripta) was investigated by in vitrosuperfusion of whole or hemipituitaries with the endogenous forms of the peptide. In sub-adults, LH secretion was rapidly attenuated desensitized) by both prolonged and pulsatile treatments of chicken II GnRH (cII-GnRH), but pulsatile GnRH administration was more effective than prolonged administration in stimulating LH release. Prolonged administration of chicken I GnRH (cI-GnRH) also desensitized LH secretion; cI-GnRH was slightly less potent than cII-GnRH. GnRH-induced attenuation of LH secretion was not the result of tissue deterioration or depletion of relesable LH: GnRH responsiveness of desensitized pituitaries was fully restored after 4 hr of in vitrorecovery, and further LH secretion was induced by tetraethylammonium chloride, a membrane depolarizing agent, in the desensitized glands; TRH-stimulated TSH release did not show desensitization in the same glands. Pituitaries from most adult (vitellogenic) females failed to respond to GnRH challenges; however, when responsive, adult pituitaries also desensitized under prolonged GnRH stimulation and were more resistant to GnRH-induced desensitization under pulsatile GnRH treatments. Current results show that adult T. scriptatypically display lower GnRH sensitivity than the subadults; the latter are like mammals, birds, and goldfish in their susceptibility to desensitization by both forms of endogenous GnRH.

The effects of several forms of gonadotropin-releasing hormone (GnRH) on the in vivo secretion of luteinizing hormone (LH) were investigated in the female turtle Trachemys scripta. Serial injections of large vitellogenic turtles with 1-2 µg (ca. 0.15 µg/100 g turtle) of chicken GnRH-I (cGnRH-I) and multiple blood sampling revealed that mature females were unresponsive to GnRRH. In contrast, both forms of avian GnRH, cGnRH-I and cGnRH-II, and an avian GnRH analogue (GnRHa) significantly increased circulating levels of LH in previtellogenic turtles. In these turtles, plasma LH increased within 10 min after the initial injection of 0.5 µg GnRHa or cGnRH-I, but this increase in LH was generally transient and could not be sustained for more than 2 hr even with repeated GnRH injections. Turtles treated with GnRHa on Day 1 showed attenuated pituitary responsiveness to subsequent GnRH challenges on Day 2, whereas turtles treated with both native forms of chicken GnRH remained equally responsive on Day 2. In vitro, GnRHa was more resistant to degradation in plasma than both native forms of GnRH, and the clearance rate of GnRHa in vivo was also slower than that of cGnRH-I. The results show, for the first time, that in vivo GnRH administration stimulates and desensitizes LH secretion in the turtles, but the GnRH responsiveness is related to reproductive status. Furthermore, the results suggest that the desensitizing effects of various forms of GnRH in vivo may be related to their clearance rate. Copyright 1993, 1999 Academic Press

Radioiodinated human FSH (125I-hFSH) was used to study the specificity of the gonadotropin binding sites of various reptilian, avian, and mammalian gonadal tissues by examining competitive interactions with several nonmammalian gonadotropins. All preparations of nonmammalian gonadotropins showed some activity in these radioligand assays, but wide variations in activities were evident depending on the source of the tissue and source of the hormone. Several cases of marked species specificity in binding were apparent: Nonmammalian hormones were relatively inactive with porcine granulosa cells; frog and snake hormones were relatively inactive in turtle tissues; and frog hormones were essentially inactive in avian tissues. FSH-LH specificity of binding determined with hormones of nonmammalian origin differed significantly from that previously described with the aid of mammalian gonadotropins. In particular, when tested with some tissues, three preparations of LH, from the turkey, sea turtle, and frog, were in some cases more active in competing for 125I-hFSH binding than preparations of FSH from the same species; this activity could not be readily accounted for by FSH contamination. These comparative data demonstrate the existence of considerable overlap in the binding characteristics of some species of FSH and LH; the observed differences probably reflect evolutionary changes in both gonadotropin binding sites and in the structure of the gonadotropins.

Binding affinities (K0) for thyroxine (T4) by blood plasma and purified plasma proteins from two turtles, the slider (Trachemys scripta) and snapper (Chelydra serpentina), were compared with those of a human using equilibrium dialysis. The purified T4binding protein (TBP) from T. scriptahad a high affinity that was lower by about fivefold than that of human TBG (3.2 × 109vs 1.7 × 1010M-1). The affinity of T4binding by TBP was similar to that determined for whole plasma from this species. T.scriptaplasma stripped of TBP by affinity chromatography showed a greatly reduced Ka(2.8 × 105M-1) similar to that of albumin purified from the same species (3.6 × 105M-1), as well as a marked increase in free T419.7%) as compared to whole turtle plasma (0.13%). The T4binding of snapping turtle plasma (Ka= 5 × 105and free T4= 2.7%) was similar to that of the TBP-stripped plasma and albumin. Binding affinities for the two turtle and human albumins were similar. Because of low concentrations of albumin in the turtle blood (estimated at ca. 10 mg/ml based on binding), TBP probably accounts for a greater proportion of T4binding in the slider turtle than does TBG in the human; e.g., when plasma T4and TBP are elevated in T. scripta, >98% of bound T4would be associated with TBP. When both TBP and T4are depressed (e.g., as in hatchlings), free T4may actually be higher than in conditions when total T4is elevated. Turtles like the snapper presumably rely almost wholly on albumin for T4transport. Calculations based on these data indicate that temporal changes or interspecific differences in total T4levels among reptiles likely misrepresent variations in free T4.

Thyroxine (T₄) binding proteins in turtle plasma were compared with those of various mammalian and nonmammalian (fish, birds) species using gel filtration chromatography (Bio-Gel). High-affinity T₄ binding was observed in all mammals studied, in the chicken, and in one species of turtle, Trachemys scripta. This high affinity T₄ binding protein (TBP) appears to be the major component of T₄ transport in the turtle; selective removal of this TBP greatly diminished binding activity in T. scripta plasma. In all other nonmammalian species studied (including other turtles), T₄ binding appeared to be associated primarily with relatively low affinity components such as albumin. Albumin concentrations in turtle (T. scripta, and Chelydra serpentina) plasmas were examined by measurement of dry weight, protein assay (Bradford), radioimmunoassay, and electrophoretic densitometry. While C. serpentina may have total protein concentrations approaching that of human plasma, albumin content in both turtles was estimated at about 3-4 mg/ml, less than 10% that of human plasma albumin; larger molecular weight components that do not bind T₄ comprise a greater proportion of turtle plasma protein. Additionally, purified T. scripta albumin exhibits lower T₄ binding than does human albumin; binding to turtle albumin is also reduced by fatty acids. Taken together, these results indicate a lesser role for albumin in plasma T₄ binding in T. scripta than in humans, and, thus, a larger role for the T₄ binding protein, TBP. Other turtles lacking a TBP may rely wholly on low affinity albumin binding that also has a lower capacity than that in mammals. Copyright 1993, 1999 Academic Press

The relation of a dual binding protein, involved in the transport of both thyroxine (T₄) and vitamin D₃, in the blood of emydid turtles to blood proteins in other vertebrates was examined immunologically. Binding studies with 25-OH-[³H]cholecalciferol (D₃) confirmed the presence of a D₃-binding protein (DBP) in the plasma of a wide variety of chelonian species representing both major suborders, as well as other species of reptiles, amphibians, birds, and mammals. Analysis by polyacrylamide gel electrophoresis (PAGE) showed that these binding proteins exhibited variable electrophoretic mobilities and in some species multiple DBPs were observed. Western blot analysis of these gels and additional tests with SDSPAGE showed that an antiserum against the dual T₄/D₃-binding protein (TBP/DBP) from the turtle Trachemys scripta cross-reacted with the DBPs of diverse turtles, one crocodilian (Alligator), and birds; however, there was little or no cross-reaction with plasma from squamate reptiles (three snakes, four lizards), one crocodilian (Osteolaemus), amphibians (two anurans, one urodele), or mammals (six eutherians, one metatherian). The cross-reacting proteins (DBPs) all exhibit a similar M_r (≈60 K). Adsorption of plasma by TBP/DBP-affinity chromatography confirmed this phylogenetic pattern of cross-reactions between the anti-TBP/DBP serum and functional DBPs in chicken, turtle, and alligator blood (affinity adsorption effectively eliminated D₃ binding); there was only weak cross-reaction with DBP in a lizard and the second crocodilian (50% reduction in D₃ binding) and none with DBP of snakes or mammals (D₃ binding was unaffected). Results suggest that the plasma protein (TBP/DBP) that exhibits dual, high-affinity binding of T₄ and D₃ in one turtle family evolved from the more "primitive" vitamin D-binding protein of stem reptiles; thus, the high-affinity T₄-binding site on this molecule is probably a derived characteristic of DBP in the Emydidae. Copyright 1994, 1999 Academic Press

The influence of androgen on pituitary sensitivity to gonadotropin-releasing hormone (GnRH) was investigated in juvenile female bullfrogs. Newly metamorphosed bullfrogs were treated in vivo for 7 days or their pituitaries were treated in vitro for 24 hr with 5α-dihydrotestosterone (DHT). Pituitary sensitivity to GnRH was assessed by incubating glands with 100 ng/ml GnRH. The secretion of luteinizing hormone (LH) and follicle-stimulating hormone (FSH) was quantified by separate radioimmunoassays. Gth biosynthesis was quantified using immunoprecipitation to measure the incorporation of [³⁵S]methionine. Prior exposure to DHT, in vitro or in vivo, significantly elevated the GnRH-induced secretion of LH and FSH. However, DHT had a differential influence on the two gonadotropins; both the basal and the GnRH-induced secretion of LH was enhanced, whereas only the GnRH-induced secretion of FSH was elevated. DHT did not significantly alter Gth biosynthesis. Treating older frogs (5 months postmetamorphic) in vivo with DHT (for 7 days) combined with a GnRH agonist (GnRHa) for 3 days enhanced the GnRH-stimulated Gth biosynthesis compared to treatment with either DHT or GnRHa alone. Therefore, while DHT may act on the pituitary to enhance gonadotropin secretion in response to GnRH, this action does not result in a concomitant increase in Gth biosynthesis. Juvenile female bullfrogs may require an increase in both GnRH and DHT in order to stimulate Gth biosynthesis. Copyright 1994, 1999 Academic Press

Vitamin D binding protein (DBP) was isolated from the plasma of the snapping turtle,Chelydra serpentina(mixed sex adult), and compared with the previously characterized dual function binding plasma protein that binds both thyroxine (T4) and vitamin D (tsTDBP) in an emydid turtle,Trachemys scripta.Purification ofChelydra serpentinaDBP (csDBP) was accomplished by ion exchange chromatography, preparative polyacrylamide gel electrophoresis, and reverse-phase high-performance liquid chromatography. Estimates of increased purification (ca. 71-fold), recovery (ca. 1.5%), and corresponding plasma concentration (ca. 0.6 mg/ml) are confounded by interference with other proteins. A protein was identified that showed the same high affinity for 25-OH-cholecalciferol (vitamin D3) as thetsTDBP, and the two exhibited similar heat resistances in binding. ThecsDBP andtsTDBP had similar molecular weights by SDS–PAGE (ca. 58 kDa), showed immunological cross reactivity, and differed by only three residues, representing conservative substitutions, in the 30 NH2-terminal amino acid sequence. A slightly lesser homology (up to 89% similarity based on conservative substitutions) was seen with three mammalian DBPs. However, unliketsTDBP, the DBP from the snapper did not bind T4. These data support the view that the DBP of emydid turtles secondarily evolved a second T4binding site, and this is probably independent of the D3binding domain.