1. Excite Keyboard Manual

Abstract

In this study, three somatostatin (SRIF) complementary DNAs (cDNAs) were characterized from goldfish brain. The cDNAs encode three distinct preprosomatostatins (PSS), designated as PSS-I, PSS-II, and PSS-III. The goldfish PSS-I, PSS-II, and PSS-III contain enzymatic cleavage recognition sites, potentially yielding SRIF-14 with sequence identical to mammalian SRIF-14, SRIF-28 with [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14 at its C-terminus, and[ Pro²]SRIF-14, respectively. The brain distribution of the three SRIF messenger RNAs (mRNAs) were differential but overlapping in the telencephalon, hypothalamus and optic tectum-thalamus regions. Seasonal variations in the levels of the three mRNAs were observed, with differential patterns between the three mRNAs and differences between the sexes. However, only the seasonal alteration in the levels of the mRNA encoding PSS-I showed close association with the seasonal variation in brain contents of immunoreactive SRIF-14 and inversely correlated with the seasonal variation in serum GH levels described in the previous studies, suggesting that SRIF-14 is involved in the control of the seasonal variation in serum GH levels. The putative SRIF-14 variant, [Pro²]SRIF-14, inhibited basal GH secretion from in vitro perifused goldfish pituitary fragments, with similar potency to SRIF-14; [Pro²]SRIF-14 also inhibited stimulated GH release from the pituitary fragments, supporting that [Pro²]SRIF-14 is a biologically active form of SRIF in goldfish.

Step 1: Select a Tree. Find a tree that is healthy and clear of obstructions such as bushes, structures or other trees. Select a branch at least 8 inches in diameter, growing roughly parallel to the ground at a height of 10 to 15 feet. A swing attached to a branch higher than 15 feet will result in a greater swinging arc unsafe for children.

SOMATOSTATIN (SRIF or SS) is a polypeptide that was originally isolated from mammalian hypothalamus and characterized as a physiological inhibitor of GH secretion (1). In mammals, SRIF is now known to be a multifunctional peptide widely distributed throughout the central nervous system and peripheral tissues (2, 3). Mammalian SRIF exists as two predominant biologically active forms, SRIF-14 and its NH₂-terminal extension of 14 amino acids, SRIF-28. Both SRIF-14 and SRIF-28 are encoded by a common gene and processed from a single precursor (2–4).

SRIF is also a phylogenetically ancient and multigene family of peptides in vertebrates. SRIF-14 has been identified, with the same amino acid sequence, in representative species of all classes of vertebrate (5, 6). In addition, four molecular variants of SRIF-14,[ Ser¹²]SRIF-14, [Ser⁵]SRIF-14,[ Pro², Met¹³]SRIF-14 and[ Pro²]SRIF-14, have been isolated in nonmammalian vertebrates (7–10). To date, cDNAs for preprosomatostatin-I (PSS-I), which contains SRIF-14 at its C-terminus, have been cloned in several mammalian species (11–14), chicken (15), frog (16), and two teleost species (17–19). The gene for PSS-I has also been characterized in mouse, rat, and human (20–22).

In addition to having PSS-I, teleost fish possess a second SRIF precursor, PSS-II, a molecule that is thought to be processed to a large form of SRIF (SRIF-28 or SRIF-25) with [Tyr⁷, Gly¹⁰]SRIF-14 sequence at its C-terminus. The cDNA sequence for PSS-II has been identified in anglerfish (18) and rainbow trout (23), providing evidence that somatostatins arose from a multigene family. Amino acid sequences for processing products of PSS-II obtained directly from isolates of islet extracts are also known for several teleost species (24–27). Recently, a second SRIF messenger RNA (mRNA) encoding a SRIF-14 variant, [Pro², Met¹³]SRIF-14, was identified in frog brain (9). In addition, a SRIF-related gene termed as cortistatin (CST) has been described in human, rat and mouse, which gives rise to a precursor that contains a tetradecapeptide at its C-terminus with an 11 amino acid homology with SRIF-14 (28–30). This suggests the existence of a multigene family for SRIF in tetrapods.

In teleosts, most of the physiological studies of SRIF have been focused on the regulation of pancreatic hormone and metabolism in salmonids (31, 32) and the regulation of GH secretion and growth in goldfish (33). In goldfish, immunoreactive SRIF-14 has been observed in the brain, with fiber tracts that terminate at the proximal pars distalis of pituitary (34, 35). SRIF-14 is a potent inhibitor of basal and stimulated GH release in goldfish (33). In addition, the concentrations of immunoreactive SRIF-14 in various brain regions vary on a seasonal basis inverse to the seasonal variation in serum GH levels (36). The inhibitory actions of SRIF-14 on GH secretion in vitro or in vivo have also been reported in tilapia, rainbow trout, and several other teleost species (33).

In the present study, three distinct SRIF cDNAs were cloned from goldfish brain RNA or cDNA library. The differential brain distribution and seasonal variation of the three SRIF mRNAs were also examined. In addition, the actions of a cDNA-deduced SRIF-14 variant on pituitary GH secretion were investigated using an in vitro perifusion system.

Materials and Methods

Animals

Goldfish (Carasius auratus) of the common or comet variety with body weight ranging from 30–40 g were purchased from Grassyfork Fisheries (Martinsville, IN) and maintained in 300-liter flow-through aquaria at 17 C under a simulated natural photoperiod of Edmonton, Alberta, Canada. The fish were fed with commercially prepared Unifeed Nu-Way trout ration (United Feeds, Calgary, Canada). Sexually regressed fish (mixed sexes) were used for extraction of RNA and DNA for molecular cloning and tissue distribution studies. For seasonal studies, tissue samples were collected in April [sexually mature fish, with gonadosomatic index (GSI) of 6.5% ± 0.68 for female and 3.5% ± 0.31 for male], July (sexually regressed fish, with GSI of 0.11% ± 0.07 for female and 0.21% ± 0.06 for male) and December (sexually recrudescent fish, with GSI of 4.06% ± 0.91 for female and 2.6% ± 0.33 for male) of the same year. Goldfish were anesthetized with 0.05% tricaine methanesulfonate (Syndel, Vancouver, BC) before tissue collection.

Reagents and test substances

SRIF-14 (Ala-Gly-Cys-Lys-Asn-Phe-Phe-Trp-Lys-Thr-Phe-Thr-Ser-Cys) was obtained from Sigma Chemical Co. (St. Louis, MO).[ Pro²]SRIF-14 was a gift from Professor H. Kawauchi (Kitasato University, Sanriku, Japan). SRIF peptides were dissolved in HBSS immediately before use. sGnRH ([Trp⁷, Leu⁸]GnRH) was purchased from Peninsula Laboratories, Inc. (Belmont, CA). sGnRH was dissolved in 0.1 m acetic acid. Subsequent dilution of all peptides to appropriate concentrations with culture medium was performed immediately before drug treatment. Dopamine (DA) D1 receptor agonist, SKF 38393 (1-phenyl-2,3,4,5-tetrahydro-(1H)-3-benzazepine-7,8-diol hydrochloride), was purchased from Research Biochemicals International, Inc. (Wayland, MA). SKF 38393 was first dissolved in dimethyl sulfoxide (DMSO), and then diluted to appropriate concentration with culture medium. Final levels of DMSO was less than 0.1%, and did not alter basal GH release from goldfish pituitary fragments (unpublished results). Trizol Reagent, Taq DNA polymerase and SuperScript Preamplification System were purchased from Life Technologies (Gaithersburg, MD). T7QuickPrime Kit was obtained from Pharmacia Biotech (Baie d’Urfe, Québec, Canada). Nybond Nylon membranes and discs, ECL direct nucleic acid labeling and detection system,[α -³²P]deoxy-CTP (dCTP) were obtained from Amersham Life Science (Buckinghamshire, UK). JM109 competent cells and pGEM-T vector system were purchased from Promega Corp. (Madison, WI). Other reagents were of the highest degree of purity available from commercial sources.

Preparation of total RNA and genomic DNA

Total RNA was extracted from goldfish forebrain (telencephalon, including optic nerve and preoptic region, and hypothalamus) using Trizol Reagent, based on the acid guanidinium thiocyanate-phenol-chloroform extraction method (37). The integrity of the RNA was verified by ethidium bromide staining of 28s and 18s ribosomal bands on a denaturing agarose gel. Genomic DNA was isolated from liver tissue of single fish using Trizol Reagent according to the manufacturer’s instruction.

Cloning of three goldfish SRIF cDNAs

For cloning of goldfish PSS-I cDNA, RT-PCR was used to prepare a DNA probe for screening the goldfish brain cDNA library. cDNA was synthesized from 4 μg of total RNA from goldfish forebrain using the SuperScript Preamplification System. For PCR, forward primer SS1-F1 (5′CGGATCCAGTGCGCCCTGGC3′) and reverse primer SS1-R1 (5′GTGAAAGTTTTCCAGAAGAA3′) were designed on the basis of the coding region of the catfish PSS-I cDNA sequence (19). Thirty cycles of PCR amplification were performed with denaturation for 1 min at 94 C, annealing for 1 min at 50 C, extension for 1.5 min at 73 C, and final extension for 10 min at 73 C after the last cycle. Amplification products were separated by agarose gel, and the band of desired size was excised and purified using a Geneclean II kit (Bio 101, La Jolla, CA). The purified DNA fragment (323 bp) was subcloned into pGEM-T vector. The nucleotide sequence analysis indicated that the cloned PCR product (323 bp) contained a partial coding region for a precursor with a partial SRIF-14 sequence at its C-terminus. This DNA fragment was then used as a probe to screen the cDNA library.

A goldfish brain cDNA library (kindly provided by Dr. H. R. Habibi, University of Calgary, Calgary, Alberta, Canada) was constructed using the ZAP-cDNA synthesis kit, including Gigapack II Gold packaging extract (Stratagene, La Jolla, CA). The library was amplified once to a titer of 6 × 10⁹ pfu/ml before being transferred to Nybond-N⁺ discs at a density of 2 × 10⁴ plaques/filter according to the manufacturer’s protocol. The filters were probed with the 323 bp DNA probe prepared by PCR as described above. The probe labeling, hybridization, and signal detection were performed using ECL direct nucleic acid labeling and detection systems based on enhanced chemiluminescence (38). A total of 2.5 × 10⁵ clones were screened, out of which six positives were picked and subjected to secondary screening. Three positive clones were obtained from the secondary screening and subjected to in vivo excision according to the instruction of the ZAP-cDNA synthesis kit. The three clones with cDNA inserts of desired size (approximately 750 bp) were sequenced on an PE Applied Biosystems automated sequencer (373A) according to the manufacturer’s protocol. Sequencing was carried out on both strands using T7 and T3 sequencing primers that flank the insert cDNA.

For cloning of goldfish PSS-II cDNA, RT and rapid amplification of cDNA ends (RACE, 39) were applied. To isolate the 3′ end of cDNA for PSS-II, two degenerate primers, DP1 (5′GCIGGITG(CT)AAGAACTTCTA3′] and DP2 (5′AAGAACTTCTA(CT)TGGAAGGG3′], were designed on the basis of the[ Tyr⁷, Gly¹⁰]SRIF-14 amino acid sequence of anglerfish and rainbow trout PSS-II (18, 23). RT and RACE were carried out as described previously (40, 41). Briefly, total RNA was reverse transcribed to cDNA with dT-adapter primer [dT-AP, 5′GGCCACGCGTCGACTAGTAC(T)₁₇3′] using SuperScript II reverse transcriptase. Two rounds of PCR amplifications were then performed to amplify the 3′ end of cDNA with adapter primer (AP, 5′GGCCACGCGTCGACTAGTAC3′) and DP1, and AP and internal DP2, respectively. After determining the nucleotide sequence of the cDNA 3′ end, new cDNA was synthesized by RT of total RNA with gene specific primer 1 (GSP1, 5′TTAACTTACATTGAGTCAGTTGA3′). After 3′-end tailing of the cDNA with poly(A) using TdT (Life Technologies), the second strand cDNA was synthesized using dT-AP. Two rounds of PCR amplifications were then carried out to amplify the 5′ end of cDNA with AP and GSP1, and AP and internal gene specific primer 2 (GSP2, 5′CAAGCGAGGGCCTCAGCAGG3′), respectively. PCR products were fractionated and the desired bands were purified and subcloned using pGEM-T vector system. DNA sequence analyses were carried out as described above.

To confirm the PSS-II cDNA sequence obtained from RT and RACE, PCR of goldfish genomic DNA was carried out to obtain a partial gene sequence that contained the full coding region for PSS-II. PCR amplification was performed using 100 ng of genomic DNA as template and primers SS2-F1 (5′CGAATCACAGCTACAAAGAGTC3′) and SS2-R1 (5′CAAGCGAGGGCCTCAGCAGG3′) designed from the 5′ and 3′ ends of the isolated PSS-II cDNA sequence. PCR products were purified and subcloned into the pGEM-T vector. DNA sequence analyses were carried out as described above using T7 and M13 reverse sequencing primers and gene specific primers.

For cloning of the third SRIF cDNA, the goldfish brain cDNA library was screened using partial goldfish PSS-II cDNA as probe. A 426-bp DNA probe was prepared by PCR with primer set, SS2-F1, and SS2-R1 specific for the cloned cDNA sequence for PSS-II, and PSS-II cDNA as template. The probe was then used to screen the cDNA library as described above. The probe labeling, hybridization, and signal detection were performed using ECL direct nucleic acid labeling and detection systems. A total of 2 × 10⁵ clones were screened, out of which three positives were picked and subjected to the secondary screening. Two positive clones were obtained from the secondary screening and subjected to in vivo excision and sequence analysis as described above.

Brain distribution of three goldfish somotostatin mRNAs

Tissues of five discrete goldfish brain areas, olfactory bulbs and tracts, telencephalon (including optic nerve and preoptic region), hypothalamus, optic tectum-thalamus, and posterior brain (including cerebellum, medulla and spinal cord), and pituitary were freshly excised and homogenized for extraction of total RNA using Trizol Reagent as described above. About 10 μg of total RNA from individual tissues were fractionated by electrophoresis in a denaturing agarose gel (1.5%) with formaldehyde and blotted onto Nybond nylon membrane by capillary transfer. The cDNAs for goldfish PSS-I, PSS-II, and PSS-III were labeled with [α-³²P]dCTP using T7QuickPrime Kit. Hybridization was performed using methods described by Church and Gilbert (42). In brief, the membranes were prehybridized in hybridization solution (0.5 m NaHPO₄, 7% SDS, 1 mm EDTA, and 1% BSA) for at least 1 h. The hybridization solution was then changed, and the labeled probe was added. After hybridization overnight at 65 C, the membranes were washed three times with washing solution (0.04 m NaHPO₄, 1 mm EDTA, and 1% SDS) and exposed to a phosphoimaging screen for 72 h. To serve as an internal control, the membrane was stripped and reprobed with a[α -³²P]dCTP-labeled partial cDNA for goldfishα -tubulin (kind gift from Dr. H. R. Habibi, University of Calgary, Calgary, Alberta, Canada).

The brain distribution of the three goldfish somatostatin mRNAs was further confirmed by RT-PCR followed by Southern blot analysis. First strand cDNA was prepared from total RNA using the SuperScript Preamplification System and used as templates for PCR using specific primers for the three goldfish SRIF mRNAs. The primer sets are: SS1-F2 (5′GCGTATCCAGTGCGCACTGGC3′) and SS1-R2 (5′GTGAAAGTTTTCCAGAAGAA3′) for PSS-I mRNA, SS2-F1 (5′CGAATCACAGCTACAAAGAGTC3′) and SS2-R1 (5′CAAGCGAGGGCCTGAGCAGG3′) for PSS-II, and SS3-F1 (5′GGAGCTACAAGACTTCAAC3′) and SS3-R1 (5′CTGTGTCAGAGTAAGTCCACG3′) for PSS-III. PCR condition was denaturation 1 min at 94 C, annealing 1 min at 51−55 C and extension 1 min at 73 C for a total of 30 cycles, and a final extension of 10 min at 73 C. For internal control, RT-PCR was performed at the same time using a primer set (Actin-F: CTACTGGTATTGTGATGGACTCCG; Actin-R: TCCAGACAGAGTATTTGCGCTCAG) for goldfish β-actin. Twenty microliters of each PCR reaction were fractionated on 1.5% agarose gel and transferred onto a Nybond nylon membrane by capillary transfer. The membranes were then hybridized with cDNA probes for one of the three goldfish SRIF mRNAs, using ECL direct nucleic acid labeling and detection systems.

Seasonal variation of three goldfish somatostatin mRNAs

Tissue samples of forebrain regions of 20 goldfish were collected on April 3, July 31, and December 21 of the same year, respectively, and stored at −80 C until total RNA was extracted. The mRNA levels for each PSS form in the total RNA samples were analyzed using Northern blot as described in the previous section. The hybridization signals were scanned using phosphorimaging (Molecular Dynamics, Inc., Sunnyvale, CA) and quantitated by ImageQuant software (Molecular Dynamics, Inc.). The mRNA levels for each PSS form were expressed as a ratio of PSS mRNA toα -tubulin mRNA. These ratios were then normalized as a percentage of mRNA levels from male fish collected in April.

Column perifusion of goldfish pituitary fragments

To examine the actions of the deduced SRIF-14 variant,[ Pro²]SRIF, on pituitary GH secretion, an in vitro column perifusion system for goldfish pituitary fragments was used (43). In brief, goldfish (sexually recrudescent) pituitary fragments (0.2 mm³) from 30 pituitaries were divided over 8 columns. The pituitary fragments were perifused overnight (15–18 h) with M199 (Life Technologies) containing 56 U/ml Nystatin (Sigma Chemical Co.), at a flow rate of 5 ml/h. Two hours before the experiment, the perifusion medium was switched to HBSS containing 25 mm HEPES and 0.1% BSA (Sigma Chemical Co.), and the flow rate increased to 15 ml/h. After this period of continuous perifusion, GH release from the pituitary fragments remained relatively stable in the absence of stimulation. Test substances were then applied from a drug reservior to the perifusion column through a three-way stopcock. Perifusate samples were collected in 5-min fractions and stored frozen at −25 C. GH contents in these samples were assayed using a RIA previously validated for goldfish GH (43).

For the dose-response study, goldfish pituitary fragments were administered with 2 min pulses of increasing doses of[ Pro²]SRIF-14 or SRIF-14 from 0.01 nm to 100 nm at intervals of 30 min. To examine the actions of[ Pro²]SRIF-14 on stimulated GH release from pituitary fragments, the pituitary fragments were first exposed to a 2-min pulse of 1 μm SKF 38393 or a 5-min pulse of 100 nm sGnRH. Sixty min after the pulse of sGnRH or SKF 38393, the fragments were exposed to 100 nm [Pro²]SRIF-14 or medium (control) for 60 min. During the exposure to SRIF or medium, the fragments were also exposed to the second pulse of 100 nm sGnRH (5 min) or 1 μm SKF 38393 (2 min). A third pulse of 100 nm sGnRH (5 min) or 1 μm SKF 38393 (2 min) was administered to the fragments 60 min after exposure to SRIF or medium. The interval between each of the three pulses of 100 nm sGnRH (5 min) or 1 μm SKF 38393 (2 min) was 90 min.

Data transformation and statistics

GH data from each column were expressed as a percentage of the mean GH contents of the first six fractions (30 min) collected at the beginning of perifusion before any drug treatment commenced (referred to as % pretreatment) (43, 44). This data transformation allows pooling of GH data from separate columns of the same experiment. GH responses were quantified by calculating the accumulated net change in GH release during the 30-min period (6 fractions) after a pulse of drug treatment (i.e. a net change in the area under the curve). Dose-response curves for [Pro²]SRIF-14 and SRIF-14 were analyzed with the ALLFIT computer program to obtain the respective ED₅₀ values (45). Data for GH release quantitated by RIA and SRIF mRNA levels quantitated by Northern blot analysis were subjected to statistical analyses using ANOVA followed by Fisher’s least significant difference (LSD) test. Differences were considered significant at P < 0.05.

Results

Three goldfish SRIF cDNAs

Screening of a goldfish brain cDNA library with a probe for the coding region of partial PSS-I cDNA obtained six positive clones. After secondary screening, three clones were characterized by nucleotide sequence analysis. Sequence analysis confirmed their identity as PSS-I cDNA clones because the deduced protein sequence contains SRIF-14 at its C-terminus. All three clones contained the identical nucleotide sequence of PSS-I cDNA, only differing in the length of poly(A) tails at their 3′-end. The 749-nucleotide sequence of goldfish PSS-I cDNA is shown in Fig. 1A (GenBank accession number U40754), along with the deduced amino acid sequence of the PSS-I. The cDNA comprises 79 bases of 5′-untranslated region, 345 bases of open reading frame, and a long 325 base 3′-untranslated region, which contains a polyadenylation signal (AATAAA) and a poly(A) tail. The nucleotide sequence of the open reading frame predicts a goldfish PSS-I of 114 amino acid residues. The PSS-I consists of a putative signal peptide sequence of 24 amino acids from position +1 to 24. The signal peptide has features common to other signal peptides, notably a region of hydrophobic residues preceded by a positively charged residue (Arg at +5) near the N terminus (46). The precursor also contains a number of putative enzyme cleavage recognition sites (47). These include an Arg monobasic site at position +88, potentially yielding a 26-amino acid large form of SRIF (SRIF-26), and an Arg-Lys dibasic site at position +99 to 100, potentially yielding SRIF-14 with sequence identical to mammalian SRIF-14 (1).

The cDNA and deduced amino acid sequences of the three goldfish preprosomatostatins (PSS), PSS-I (A), PSS-II (B), and PSS-III (C). The DNA sequences of the 5′- and 3′-untranslated region are shown as lower case letters, while coding regions are shown as upper case letters. The putative signal peptide sequences are shown as bold upper case letters. Potential monobasic and dibasic enzymatic cleavage recognition sites are indicated in boxes. The amino acid sequences for somatostatin-14 (SRIF-14) or its potential variants are underlined at the C-terminus of each precursor. The polyadenylation signals (AATAAA or ATTAAA) in the 3′-untranslated regions are shown as upper case letters.

The cDNA and deduced amino acid sequences of the three goldfish preprosomatostatins (PSS), PSS-I (A), PSS-II (B), and PSS-III (C). The DNA sequences of the 5′- and 3′-untranslated region are shown as lower case letters, while coding regions are shown as upper case letters. The putative signal peptide sequences are shown as bold upper case letters. Potential monobasic and dibasic enzymatic cleavage recognition sites are indicated in boxes. The amino acid sequences for somatostatin-14 (SRIF-14) or its potential variants are underlined at the C-terminus of each precursor. The polyadenylation signals (AATAAA or ATTAAA) in the 3′-untranslated regions are shown as upper case letters.

RT and RACE were performed to clone goldfish PSS-II cDNA. 3′-RACE produced a distinct band of approximately 200 bp size. Sequence analysis of this product indicates that this partial cDNA contained the 3′-untranslated region and coding region for the C-terminal portion of the [Tyr⁷, Gly¹⁰]SRIF-14 sequence. Analysis of the 5′-RACE products revealed a distinct band of about 450 bp. Subcloning and sequence analysis showed that this 5′-RACE product contained nucleotide sequences for the 5′-untranslated region and an open reading frame for the second goldfish SRIF precursor (PSS-II), which contains [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14 at its C-terminus. The nucleotide sequences of the overlapping portions were identical between the partial cDNAs obtained by 3′-RACE and 5′-RACE. The same PCR amplification and sequence analyses for 3′-RACE and 5′-RACE were repeated to confirm that the sequences did not include any errors caused by misincorporation. Additional confirmation of the complete nucleotide sequence of the goldfish PSS-II cDNA was achieved by performing PCR amplification using specific primers positioned at the most 5′ and 3′ end of the cDNA and sequence analysis of the cDNA. The complete nucleotide sequences and the deduced amino acid sequence of the PSS-II cDNA, size 589 bp, is shown in Fig. 1B (GenBank accession number U60262). The cDNA comprises 57 bp 5′-untranslated region, an open reading frame (363 bp), and a 169 bp 3′-untranslated region, including two polyadenylation signals (AATAAA) and a poly(A) tail. The open reading frame encodes a 120-amino acid precursor, which contains a putative 26-amino acid signal peptide sequence from position +1 to 26. The precursor contains an Arg monobasic site at position +92, potentially yielding a large form of SRIF (SRIF-28) with 28 amino acids, and an Arg-Lys dibasic site at position +105 to +106, potentially yielding [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14. The [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14 is likely a variant of[ Tyr⁷, Gly¹⁰]SRIF-14, which was identified at the C-terminus of PSS-II in anglerfish (18) and rainbow trout (23). To confirm the PSS-II cDNA sequence derived from RT and RACE, PCR of genomic DNA was performed to obtain a partial gene sequence for PSS-II. Sequence analysis of the PCR product confirmed the cDNA sequence. This 1269-bp partial gene sequence contains an internal intron of 841 bp and two partial exons for a reading frame of 363 bp encoding the full-length of PSS-II (Data not shown, GenBank accession number AF025686).

Screening of the goldfish brain cDNA library using partial PSS-II cDNA as probe identified two positive clones. After secondary screening, two positive clones were characterized by nucleotide sequence analysis. The identity of the sequence indicates a novel cDNA, as the deduced peptide sequence contains a SRIF-14 variant with a proline substitution at position 2 of SRIF-14. The 694 bp of goldfish PSS-III cDNA (GenBank accession number U72656) is shown in Fig. 1C, along with the deduced amino acid sequence of the precursor (PSS-III). The cDNA is comprised of a 104 bp 5′-untranslated region, 336 bp of open reading frame, and a 254 bp 3′-untranslated region, which contains a modified polyadenylation signal (ATTAAA) and a poly (A) tail. The deduced 111-amino acid precursor consists of a putative signal peptide sequence of 19 amino acids from position +1 to 19. The precursor contains an Arg-Lys dibasic site at position +96 and +97, potentially processing into [Pro²]SRIF-14. In addition, the precursor contains two Arg monobasic sites at +87 and +82, potentially generating a 24-amino acid peptide and a 29-amino acid peptide, respectively.

The nucleotide sequence similarity of the precursor coding regions of the three goldfish SRIF cDNAs are 53%, 22%, and 15% between PSS-I and PSS-II cDNA, PSS-I and PSS-III cDNA, and PSS-II, and PSS-III cDNA, respectively. The amino acid sequence similarity between three goldfish PSSs are 38%, 32%, and 18% between PSS-I and PSS-II, PSS-I and PSS-III, and PSS-II and PSS-III, respectively. Table 1 summarizes the percentage amino acid sequence similarity of the known PSS between goldfish and other veterbrate species. The multiple sequence alignment was performed using Clustal W program at the online service website of European Bioinformatics Institute.

Percentage amino acid sequence similarity of the known preprosomatostatins (PSS) between goldfish and other vertebrate species

Differential brain distribution of three golfish SRIF mRNAs

To examine the brain distribution of the three goldfish SRIF mRNAs, total RNA prepared from pituitary and five different brain areas, including olfactory bulbs and tracts, telencephalon, hypothalamus, optic tectum-thalamus, and posterior brain region, was subjected to Northern blot analysis (Fig. 2A). Hybridization signals to[α -³²P]-labeled probe for PSS-I mRNA were found in the pituitary and in all brain areas except the olfactory bulbs. Hybridization signals to the probe for PSS-II mRNA were only found in the hypothalamus and optic tectum-thalamus. Hybridization signals to the probe for PSS-III mRNA were detected in the olfactory bulbs, telencephalon, hypothalamus and optic tectum-thalamus. The α-tubulin transcript as an internal control was observed in all of the tissues examined (Fig. 2A).

Brain distribution of the three goldfish somatostatin mRNAs as revealed by Northern blot analysis (A) and RT-PCR followed by Southern blot analysis (B). Total RNA was prepared from five brain regions, including olfactory bulb and tract (1 ), telencephalon (2 ), hypothalamus (3 ), optic tectum-thalamus (4 ) and posterior brain region (5 ), and pituitary (6 ). The RNA (10 μg) was fractionated on denaturing agarose gel and transblotted onto Hybond nylon membranes. The membrane was hybridized with one of the goldfish somatostatin cDNA probes labeled with [α-³²P]dCTP and exposed to phosphoimaging screen. Northern blot of goldfish α-tubulin was also performed as internal control. For RT-PCR, cDNA was prepared from the total RNA samples from the five brain regions as described above. PCR was performed using the specific primer sets for each of the three goldfish somatostatin mRNAs. RT-PCR for goldfish β-actin was also performed as internal control. PCR reaction was fractionated and transblotted onto nylon membrane. The membrane was hybridized with cDNA probe and detected using ECL nucleic acid labeling and detection system.

Brain distribution of the three goldfish somatostatin mRNAs as revealed by Northern blot analysis (A) and RT-PCR followed by Southern blot analysis (B). Total RNA was prepared from five brain regions, including olfactory bulb and tract (1 ), telencephalon (2 ), hypothalamus (3 ), optic tectum-thalamus (4 ) and posterior brain region (5 ), and pituitary (6 ). The RNA (10 μg) was fractionated on denaturing agarose gel and transblotted onto Hybond nylon membranes. The membrane was hybridized with one of the goldfish somatostatin cDNA probes labeled with [α-³²P]dCTP and exposed to phosphoimaging screen. Northern blot of goldfish α-tubulin was also performed as internal control. For RT-PCR, cDNA was prepared from the total RNA samples from the five brain regions as described above. PCR was performed using the specific primer sets for each of the three goldfish somatostatin mRNAs. RT-PCR for goldfish β-actin was also performed as internal control. PCR reaction was fractionated and transblotted onto nylon membrane. The membrane was hybridized with cDNA probe and detected using ECL nucleic acid labeling and detection system.

The differential brain distribution of the three goldfish SRIF mRNAs was confirmed using RT-PCR followed by Southern blot analysis (Fig. 2B). The primer set specific for goldfish PSS-I mRNA amplified a PCR product of 323 bp and was detected with Southern blot in the telencephalon, hypothalamus, optic tectum-thalamus and posterior brain. The primer set specific for goldfish PSS-II mRNA amplified a PCR product of 426 bp and was detected with Southern blot in the hypothalamus and optic tectum-thalamus as well as in telencephalon, where PSS-I mRNA was not detected by Northern blot analysis. The primer set specific for goldfish PSS-III mRNA amplified a PCR product of 531 bp and was detected with Southern blot in the olfactory bulbs, telencephalon, hypothalamus, optic tectum-thalamus area as well as in posterior brain, where PSS-III mRNA was not detected by Northern blot analysis. The primer set specific for β-actin mRNA amplified a PCR product of 600 bp in all of the tissues examined, verifying the quality and integrity of the cDNA samples. There was no PCR product detected in the negative control (PCR without DNA template). No cross hybridization was seen for each of the three probes in Northern blot and Southern blot analyses.

Seasonal variation of three goldfish SRIF mRNAs

To examine the seasonal variation of the three SRIF mRNAs, total RNA prepared from forebrain tissue of male or female goldfish collected in April (sexually mature), July (sexually regressed) or December (sexually recrudescent) was used to quantitate mRNA levels for each SRIF mRNA using Northern blot analysis.

Seasonal changes in PSS-I mRNA levels were observed for female but not male fish (Fig. 3A). Females in December and April exhibited significantly higher levels of PSS-I mRNA than that in July, but there were no significant differences in PSS-I mRNA levels between female fish collected in April and in December. Male fish showed no significant differences in PSS-I mRNA levels between April, July, and December. No sexual differences in mRNA levels for PSS-I were seen in July or December (Fig. 3A). In April, however, female fish had significantly higher levels of PSS-I mRNA levels than males.

Seasonal variation of the three goldfish somatostatin mRNAs. Total RNA was prepared from the goldfish forebrain tissues collected in April, July, and December, respectively, and subjected to Northern blot analysis as described in Fig. 2. Hybridization signals were detected using phosphoimaging and quantitated using ImageQuant program. The somatostatin mRNA levels were expressed as a ratio between somatostatin mRNA and α-tubulin mRNA (internal control) and then normalized as a percentage of somatostatin mRNA levels from male fish collected in April. Data are mean ± sem (n = 6). Statistical analysis was performed using ANOVA followed by Fisher’s LSD test. Significant differences at P < 0.05 between groups are designated by different lower case letters above the bars.

Seasonal variation of the three goldfish somatostatin mRNAs. Total RNA was prepared from the goldfish forebrain tissues collected in April, July, and December, respectively, and subjected to Northern blot analysis as described in Fig. 2. Hybridization signals were detected using phosphoimaging and quantitated using ImageQuant program. The somatostatin mRNA levels were expressed as a ratio between somatostatin mRNA and α-tubulin mRNA (internal control) and then normalized as a percentage of somatostatin mRNA levels from male fish collected in April. Data are mean ± sem (n = 6). Statistical analysis was performed using ANOVA followed by Fisher’s LSD test. Significant differences at P < 0.05 between groups are designated by different lower case letters above the bars.

The pattern of seasonal alteration in PSS-II mRNA levels was similar for both female and male fish, with significantly higher levels occurring in July compared with April and December (Fig. 3B). No significant differences in PSS-II mRNA levels were found in both sexes between April and December. Significant sexual differences in PSS-II mRNA levels were observed in April, with higher expression of PSS-II in females (Fig. 3B).

Seasonal alterations in PSS-III mRNA levels were observed in both females and males (Fig. 3C). Both female and male fish in July had the highest levels of PSS-III mRNA. In addition, female fish had significantly higher levels of PSS-III mRNA in April compared with that in December. In contrast, male fish in December had significantly higher PSS-III expression compared with that in April, but there was no difference in PSS-III mRNA levels between males in July and in December (Fig. 3C). Significant sexual differences in mRNA levels for PSS-III were observed in both April and December (Fig. 3C). In December, males expressed significantly higher PSS-III than females. In contrast, in April females expressed higher PSS-III than males.

Actions of [Pro2]SRIF-14 on goldfish pituitary GH secretion

Direct actions of [Pro²]SRIF-14 on GH release were examined in goldfish pituitary fragments under column perifusion. Both SRIF-14 and [Pro²]SRIF-14 (0.01–100 nm) inhibited basal GH secretion from goldfish pituitary fragments in a dose-dependent manner, with similar potency between the GH release response to SRIF-14 and to [Pro²]SRIF-14 (Fig. 4). The ED₅₀ values of[ Pro²]SRIF-14 and SRIF-14 were 2.33 ± 2.17 nm and 3.56 ± 2.43 nm, respectively. GH secretion after each pulse rapidly returned to baseline levels, with increasing duration of the inhibition with increasing concentrations of SRIF peptides.

Dose-dependent inhibiting effect of SRIF-14 and[ Pro²]SRIF-14 on GH secretion from perifused goldfish pituitary fragments. Two-min pulses of increasing concentrations of SRIF-14 or [Pro²]SRIF-14 were administered at 30-min intervals. GH responses were quantitated as the accumulated net change in GH release after each pulse of SRIF peptide treatment. Data showed mean ± sem (n = 4). ED₅₀ values analyzed using the ALLFIT program for the GH-inhibiting effect of SRIF-14 and [Pro²]SRIF-14 were 2.33 ± 2.17 nm and 3.56 ± 2.43 nm, respectively.

Dose-dependent inhibiting effect of SRIF-14 and[ Pro²]SRIF-14 on GH secretion from perifused goldfish pituitary fragments. Two-min pulses of increasing concentrations of SRIF-14 or [Pro²]SRIF-14 were administered at 30-min intervals. GH responses were quantitated as the accumulated net change in GH release after each pulse of SRIF peptide treatment. Data showed mean ± sem (n = 4). ED₅₀ values analyzed using the ALLFIT program for the GH-inhibiting effect of SRIF-14 and [Pro²]SRIF-14 were 2.33 ± 2.17 nm and 3.56 ± 2.43 nm, respectively.

[Pro²]SRIF-14 also inhibited DA D1 receptor agonist-stimulated GH release from perifused goldfish pituitary fragments (Fig. 5A). Two min pulses of SKF 38393, adminstered at 90-min intervals, significantly stimulated GH release. Exposure of the fragments to [Pro²]SRIF-14 decreased basal GH secretion and completely blocked SKF 38393-induced (the second pulse) GH secretion. After removal of[ Pro²]SRIF-14, GH levels returned to near the initial baseline levels. However, subsequent exposure of the fragments to the third SKF 38393 pulse did not stimulate GH secretion from the fragments.

Effects of [Pro²]SRIF-14 on dopamine D1 receptor agonist SKF 38393-stimulated (A) and sGnRH-stimulated (B) GH secretion from perifused goldfish pituitary fragments. Three 2-min pulses of 100 nm sGnRH or 5-min pulses of 1μ m SKF 38393 were administered at 90-min intervals. The pituitary fragments were exposed to [Pro²]SRIF-14 or medium (control) for 1 h when the second pulse of drug treatment was supplied in the middle of this exposure. The upper panels of (A) and (B) show the GH release profile expressed as percentage of basal levels in columns treated with[ Pro²]SRIF-14. The lower panels of (A) and (B) show the GH release responses to the three pulses of the drug treatment in the control columns and in the columns treated with[ Pro²]SRIF-14. GH responses were quantitated as the accumulated net change in GH release after each pulse of drug treatment. Data showed mean ± sem (n = 4–6). Statistical analysis was performed using ANOVA followed by Fisher’s LSD test. Significant differences at P < 0.05 between groups are designated by different lower case letters above the bars.

Effects of [Pro²]SRIF-14 on dopamine D1 receptor agonist SKF 38393-stimulated (A) and sGnRH-stimulated (B) GH secretion from perifused goldfish pituitary fragments. Three 2-min pulses of 100 nm sGnRH or 5-min pulses of 1μ m SKF 38393 were administered at 90-min intervals. The pituitary fragments were exposed to [Pro²]SRIF-14 or medium (control) for 1 h when the second pulse of drug treatment was supplied in the middle of this exposure. The upper panels of (A) and (B) show the GH release profile expressed as percentage of basal levels in columns treated with[ Pro²]SRIF-14. The lower panels of (A) and (B) show the GH release responses to the three pulses of the drug treatment in the control columns and in the columns treated with[ Pro²]SRIF-14. GH responses were quantitated as the accumulated net change in GH release after each pulse of drug treatment. Data showed mean ± sem (n = 4–6). Statistical analysis was performed using ANOVA followed by Fisher’s LSD test. Significant differences at P < 0.05 between groups are designated by different lower case letters above the bars.

The actions of [Pro²]SRIF-14 on sGnRH-stimulated GH secretion was also examined (Fig. 5B). Five-minute pulses of sGnRH, given at 90-min intervals, resulted in stimulation of GH release from the pituitary fragments. Exposure of the fragments to[ Pro²]SRIF-14 decreased basal GH secretion and completely blocked sGnRH-induced (the second pulse) GH secretion. Subsequent exposure of the fragments to the third sGnRH pulse resulted in the stimulation of GH secretion, which was not significantly different from the GH response to the third sGnRH pulse in the control columns (without exposure to [Pro²]SRIF-14).

Discussion

In the present study, three distinct SRIF cDNAs were cloned from goldfish brain. The three cDNAs encode three goldfish SRIF precursors designated as PSS-I, PSS-II, and PSS-III, respectively. The deduced amino acid sequence similarities between three goldfish PSSs are from 18–38%, indicating high divergence between three goldfish SRIF precursors. Notably, goldfish PSS-III showed 32% and 27% similarity to frog [Pro², Met¹³]SRIF-14 precursor (9) and mammalian CST precursor (28–30), respectively. A phylogenetic tree was produced by the amino acid sequence alignment of all known PSSs and CST precursors using Clustal W program (data not shown). This phylogenetic tree showed that goldfish PSS-I was grouped with mammalian, chicken, frog and catfish PSS-I in a separate clade, and that goldfish PSS-II was grouped with rainbow trout and anglerfish PSS-II in a clade. However, goldfish PSS-III was grouped with frog[ Pro², Met¹³]SRIF-14 precursor and mammalian CST precursors in a branch, along with catfish SRIF-22 precursor. Goldfish [Pro²]SRIF-14 precursor (PSS-III) shows little structural similarity with either PSS-I or PSS-II outside the C-terminal region. Similar results were observed between frog[ Pro², Met¹³]SRIF-14 precursor and frog PSS-I and the known fish PSS-IIs (16). Together, this suggests that goldfish PSS-III represents a third gene for SRIF in goldfish and is phylogenetically associated with the frog [Pro², Met¹³]SRIF-14 and mammalian CST genes.

Posttranslational processing of PSS-I has been well studied in mammals and some fish species (6). In mammals, processing of PSS-I by converting enzymes is tissue specific. In brain, pancreas, and stomach, PSS-I is primarily hydrolyzed to yield SRIF-14 by cleavage at Arg-Lys dibasic sites, whereas in the intestine, PSS-I is mostly processed to SRIF-28 by cleavage at an upstream Arg monobasic site (48, 49). In nonmammalian vertebrates, SRIF-14 is the only PSS-I product identified in extracts of the pancreas of all species of birds, reptiles, and amphibias studied to date (6) and in islet tissue from some fish species (24–26, 50, 51). Therefore, according to the processing of PSS-I in other species, goldfish PSS-I is likely processed at the Arg-Lys dibasic site to generate SRIF-14, identical to the mammalian SRIF-14 (1). However, it is also possible that goldfish PSS-I can be processed at monobasic Arg site to yield SRIF-26, which has only three residue substitutions compared with bowfin SRIF-26 (52). The SRIF-26 is the most abundant SRIF component in extracts of the islet organ in bowfin, and SRIF-14 is present as a minor component in this species (52). Similarly, catfish PSS-I contains a monobasic Arg site, which could generate a SRIF-26 with an identical sequence to the proposed goldfish SRIF-26 (19).

Processing of PSS-II into SRIF-25 or SRIF-28 has been demonstrated indirectly by isolates of islet or intestine extracts from several fish species (24–27). In anglerfish, both PSS-I and PSS-II contain dibasic and monobasic cleavage sites that could be cleaved respectively to generate SRIF-14 and SRIF-28 (17, 18). However, it has been shown that anglerfish SRIF-14 and SRIF-28 are two separate and independent products of anglerfish PSS-I and II, respectively, although both precursors are also cleaved at corresponding mono- and dibasic sites to some minor extent (50, 53–55), and both PSSs are expressed in different cells of the Brockmann body (56, 57). On this basis, it is suggested that goldfish PSS-II is likely processed at the monobasic Arg site to yield SRIF-28, which contains [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14 at its C-terminus. The goldfish SRIF-28 deduced from the cloned cDNA has 7 amino acids different from anglerfish SRIF-28 (18). Recently, a SRIF-28 peptide was isolated from goldfish intestine (27), with 5 amino acids different from the goldfish SRIF-28 deduced from cDNA in this study. These results suggest that there are at least two forms of SRIF-28 in goldfish, which may be differentially expressed in specific tissues.

Goldfish PSS-III deduced from the third cDNA sequence contains dibasic Arg-Lys site to generate [Pro²]SRIF-14 at its C-terminus.[ Pro²]SRIF-14 has been identified in Russian sturgeon, a modern representative of the primitive bony fish (10). Sturgeons are one of the extant representatives of chondrosteans that gave rise to holosteans and teleosts. In addition, sturgeons exhibit dramatic similarities with amphibians; in particular they possess a functional hypothalamic-hypophysial blood portal system similar to the tetrapods, a system that is vestigial or absent in teleosts (58). A SRIF-14 variant [Pro², Met¹³]SRIF-14 was identified and subsequently its cDNA was cloned from an amphibian species (9, 16). Recently, cDNA encoding a SRIF-related peptide, CST, has been characterized in several mammalian species (28–30). Preprocortistatin contains 112 amino acids with the sequence of a lysine-extended SRIF-14-related peptide at its C-terminus. CST-14 binds to mammalian SRIF receptors (sst) and inhibits stimulated-cAMP production likely through sst (28, 30). In addition, [Pro²]SRIF-14,[ Pro², Met¹³]SRIF-14 and rat CST-14 all contain Gly-Pro substitution at position 2 (position 1 for CST), although the functional implication of this substitution is unclear. Given the presence of [Pro²]SRIF-14 in both sturgeon and goldfish and the close phylogenetic relationship between goldfish[ Pro²]SRIF-14 precursor (PSS-III) and [Pro², Met¹³]SRIF-14 and CST precursors, the[ Pro²]SRIF-14 gene may be the origin for the[ Pro², Met¹³]SRIF-14 and CST genes and represent an evolutionary branch of the SRIF family.

In the present study, [Pro²]SRIF-14 peptide inhibited basal GH release from goldfish pituitary fragments in a dose-dependent manner, with similar potency to SRIF-14. In addition,[ Pro²]SRIF-14 inhibited sGnRH- or DA D1 receptor agonist-stimulated GH release from goldfish pituitary fragments; sGnRH and dopamine are two major stimulators of pituitary GH secretion in goldfish (43, 44). These results suggest that[ Pro²]SRIF-14 is a biologically active form of SRIF in goldfish. In frogs, [Pro², Met¹³]SRIF-14 is more potent than SRIF-14 in competing with the radioligand in receptor binding studies, and both [Pro², Met¹³]SRIF-14 and SRIF-14 inhibit GH secretion from the frog pituitary (59).

In goldfish, immunoreactive SRIF (SRIF-ir) has been demonstrated in the brain using antiserum recognizing equally well mammalian SRIF-14 and SRIF-28 (34, 35). SRIF-ir neurons were widely distributed in the telencephalon, diencephalon, including the preoptic region, ventrobasal hypothalamus, epithalamus, midbrain, reticular formation and spinal cord (35). In the present study, distribution of the three SRIF mRNAs were consistently observed in the telencephalon, hypothalamus, and optic tectum-thalamus. However, the distribution of the three SRIF mRNAs were differential in the other brain regions. PSS-I mRNA was also detected in the posterior brain area by both RT-PCR and Northern blot analysis, consistent with the previous observation of SRIF peptide localization using immunocytochemistry. PSS-II mRNA was restricted in the forebrain and midbrain areas. PSS-III mRNA was also detected in posterior brain by RT-PCR but not by Northern blot analysis, suggesting lower expression levels of PSS-III mRNA in this brain area. In addition, PSS-III mRNA was detected in the olfactory bulbs by both RT-PCR and Northern blot analysis, whereas mRNAs for PSS-I and PSS-II were not detected in this brain area. Interestingly, goldfish PSS-I mRNA, which encodes SRIF-14, was also detected in the pituitary using Northern blot analysis. The presence of SRIF mRNA in the pituitary suggests a paracrine function of SRIF-14.

In a previous study, seasonal variations in circulating levels of GH in the goldfish have been observed with highest mean daily serum GH levels in June and the lowest levels in November (60). The seasonal variation in body growth rates correlates in part with the seasonal cycle in serum GH levels (60). The brain and pituitary content of immunoreactive SRIF-14 measured by RIA using antiserum against SRIF-14 was shown to vary on a seasonal basis inverse to the seasonal variation in the levels of serum GH in goldfish (36), with highest levels of immunoreactive SRIF-14 in telencephalon, hypothalamus, and thalamus-midbrain at November and the lowest levels of immunoreactive SRIF in the same brain regions at June and July. These results suggest that the seasonal variations in GH reflect changes in the intensity of the SRIF inhibitory tone. In the present study, seasonal variations of the levels of three SRIF mRNAs were observed, with differential patterns and sexual differences. Goldfish PSS-II and PSS-III mRNAs levels are highest for both males and females in late July in sexually regressed fish, and lower in April in sexually mature fish. Therefore, the seasonal variations in PSS-II and PSS-III mRNA levels are not closely correlated with the seasonal variations in GH levels and immunoreactive SRIF-14 levels from the previous studies (36, 60). However, the seasonal variations in PSS-I mRNA in female fish were closely correlated with the seasonal variations in GH levels and immunoreactive SRIF levels from the previous studies (36, 60), with higher PSS-I mRNA levels in sexually mature fish in April and in sexually recrudescent fish in December and lower levels in sexually regressed fish in July. These results suggest that PSS-I mRNA, which encodes SRIF-14, is potentially involved in the control of seasonal variations in serum GH levels. On the other hand, the present results do not demonstrate a significant seasonal variation in PSS-I mRNA in the forebrain of male fish, suggesting that SRIF-14 is not involved directly in controlling the seasonal GH cycle in male fish. These differences between males and females in brain PSS-I mRNA expression patterns implicate sex steroids in the modulation of PSS-I mRNA expression in goldfish. Indeed, our recent studies (unpublished observations) have shown that brain PSS-I mRNA levels are affected by estradiol but not by testosterone.

In summary, three SRIF cDNAs were characterized from goldfish brain. The cDNAs encode three distinct PSSs, which potentially yield SRIF-14, SRIF-28 with [Glu¹, Tyr⁷, Gly¹⁰]SRIF-14 at its C-terminus, and[ Pro²]SRIF-14, respectively. The brain distribution of the three SRIF mRNAs were differential but overlapping in the telencephalon, hypothalamus and optic tectum-thalamus regions. Seasonal variations in the levels of the three mRNAs were observed, with differential patterns between the three mRNAs and differences between sexes. The putative SRIF-14 variant, [Pro²]SRIF-14, inhibited basal goldfish pituitary GH secretion, with similar potency to SRIF-14; [Pro²]SRIF-14 also inhibited stimulated GH release, supporting that [Pro²]SRIF-14 is a biologically active form of SRIF in goldfish.

Acknowledgments

We acknowledge Dr. H. R. Habibi for providing goldfish brain cDNA library and goldfish α-tubulin clone. We thank Carol Nahorniak, Gary Ritzel, Eugene Chomey, and Pierre Peyon for their assistance.

, , , , , , . :–, . :–. :–, . :–, , . :–. In: , , (eds) . , , pp –, , , , . :–. : –, , . :–, , , , . :–, , . :–, , , , , , , , , , . :–, , , , , , . :–, . :–, , , , , , . , , , , , , , . :–, , , , , . :–, , , , . :–, , , , , , . :–, , , . :–, . :–, , , . :, , , . :–, , , . :–, , , , . :–, ,

Excite Keyboard Manual

, , , , . :–, , , . :–, , , , , , , , , . :–, , , , , , . :–, , , , , , , , . :–, . In: , , (eds) . , , pp –. :–, . :–, , , . :–, , . :–, , . :–, . :–, , , , . :–, , . :–, . :–, . :–, . :–, , , . :–, , . :–, , . :–, . :–, , . :–, , , , , , , . :–, . :–, , , . :–, . :–, , . :–, . :–, , . :–, , . :–, , , , . :–, , , , . :–, . In: , (eds) . , , vol :–, , , , , , , , , , . :–, . :–

Drawing is a skill you can master at any age. When you are ready, you can learn the basics of drawing by taking a free online drawing class. The websites all offer helpful instruction for beginning artists, and many of them offer classes at intermediate or advanced levels. When you use the web as your art instructor, you can log on to learn whenever you please.

Kline Creative

The free online drawing lessons at Kline Creative website are designed for beginners of any age, from young children to adults. The site offers instructional videos on a range of drawing subjects. The videos are designed to give the beginner core skills to enhance any art medium you choose to use.

ArtyFactory

The ArtyFactory Art Lessons Gallery offers free online art lessons that include basic drawing classes for pencil, ink and colored pencil. For visitors who want to expand their knowledge of art, the site also offers an Art Appreciation Gallery and a Design Lessons Gallery.

YouTube.com

Don't overlook YouTube when you are searching for free online drawing classes. YouTube is a treasure trove of videos on the subject. Just enter a search term such as 'drawing lessons' and choose from the enormous selection of videos on the topic. You may need to filter the list to see the topics of most interest to you, such as 'drawing animals' or 'drawing figures.'

DrawingCoach.com

Visit DrawingCoach.com for free drawing classes that skip the heavy theory and help students get started drawing immediately. Have fun learning how to draw portraits, cartoons, caricatures, and tattoos. All the lessons include step-by-step instructions and examples. Some lessons also include video tutorials.

DrawSpace

DrawSpace offers free and paid drawing lessons. This free collection of online drawing classes contains dozens of illustrated lessons for beginning, intermediate and advanced artists. Learn how to set up a studio, create line drawings, shade correctly and cartoon. Some of the free classes are:

• Introduction to Drawing
• Drawing From Line to Life: Beginner and Intermediate
• Introduction to Contour Drawing
• Drawing a Symmetrical Design
• Drawing With Colored Pencils

Academy of Art University

This high-quality video class from the Academy of Art University entitled 'How to Draw a Head' teaches you how to draw a head from a photo or from memory. The instruction focuses on facial proportion, expression and sketching basics

Toad Hollow Studio

Check out these free online drawing lessons at Toad Hollow Studio for instruction at all skill levels. Beginning lessons include line drawing, contour drawing, and shading. The lessons are available in text and video formats and are all free to the user. Also available is information on art theory and various drawing techniques.

How to Draw It

The How to Draw It website offers a simple approach to drawing animals and people. The animal tutorials are super easy to do, while the people lessons a bit more advanced. All are free to site visitors and make instant progress in your drawing skills possible.

How to Draw Cartoons Online!

If drawing cartoons is your thing, this site offers plenty of free instruction on the topic. The site covers categories like '80s style cartoons, video game characters like Pacman, and Mr. Spock and Darth Vader.

Free Online Art Classes

This site covers a wide range of art classes, but there are several free drawing tutorials for online learners, including:

• Learn Basic Drawing
• Draw With Pen and Ink
• Learn Colored Pencils

Some of the classes are downloadable and some are in video form.