| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Division of Neuroendocrinology, Institute of Anatomy, University of Zurich, CH-8057 Zurich, Switzerland
Address all correspondence and requests for reprints to: Dr. Manfred Reinecke, Division of Neuroendocrinology, Institute of Anatomy, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland. E-mail: reinecke{at}anatomie.unizh.ch
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Fish are the oldest living vertebrates; they evolved about 500 million yr ago (2). Few IGF-I sequences have been obtained from bony fish, mainly from the small family of salmonids, i.e. Oncorhynchus kisutch (3, 4), O. mykiss (5), and O. tshawytschwa (6). In contrast to most other bony fish, salmonids are tetraploid (7). For this and several other reasons we studied the tilapia (Oreochromis mossambicus). Tilapia belongs to the order of perciformes, which is the most diversified of all bony fish orders as well as the largest order of vertebrates (8). Because tilapia is a mouthbrooder, and thus all developmental stages can be easily obtained, it represents an excellent experimental animal for studies on regulation during early life (9). Previously, the presence of radioimmunoassayable IGF-I in tilapia serum has been shown (10). Furthermore, the gene sequence of GH (GH) has been obtained from Tilapia nilotica (11), and the GH-IGF-I axis probably plays an important role in development of fish (12, 13, 14, 15). Finally, O. mossambicus is euryhaline, and GH might also be involved in its adaption to sea water (16). Therefore, after clarification of essential parameters of the IGF-I system, tilapia may serve as a paradigm to study various IGF-I functions in fish.
In salmonids, IGF-I messenger RNA (mRNA) is expressed not only in liver (15), but also at extrahepatic sites (4, 17, 18, 19, 20). However, compared with mammals fewer organs have been identified to express IGF-I mRNA. Even less information is available on the presence of the IGF-I peptide. A distinct localization of IGF-I immunoreactivity has yet been given only for gastrointestinal tract and endocrine pancreas (21, 22, 23). Thus, the likely cellular sites of IGF-I synthesis are generally unknown in bony fish. Hence, the present study on tilapia aims 1) to the establish the complementary DNA (cDNA) sequence of prepro-IGF-I, 2) to study the expression of IGF-I in various organs, and 3) to identify the likely sites of IGF-I production by the use of immunohistochemical methods.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Reverse transcription-PCR (RT-PCR) for liver and sequence
analysis
Total RNA was extracted from tilapia liver using the Ultraspec
RNA Extraction Kit (ams, Lugano, Switzerland). The RNA concentration
was determined by spectrophotometric absorption at 260 nm. For single
stranded cDNA, 5 µg total RNA were denatured for 3 min at 70 C and
primed with a poly(deoxythymidine)17 primer
(5'-CCTGAATTCTAGAGCTCAT(17)-3'; 30 pmol). For single stranded cDNA
synthesis, the RNA-primer mix was incubated for 1 h at 37 C with
0.5 pmol/µl deoxy-NTPs (Boehringer Mannheim, Mannheim, Germany),
1 x reaction buffer (50 mM KCl, 1.5 mM
MgCl2, 10 mM Tris-HCl), and 10 U AMV-reverse
transcriptase (Appligene, Illkirch, France). RT-PCR was performed in a
50-µl reaction volume [2 µl cDNA, 1 x reaction buffer, 200
µM deoxy-NTPs, 1 µM of each primer, and 1.2
U Taq DNA polymerase (Appligene)].
PCR products were purified using the Wizard PCR DNA Purification Kit (Promega, Madison, WI) and cloned applying the pCR-Script SK+ Cloning Kit (Stratagene, Heidelberg, Germany). For the sequencing reaction (Silver Sequence DNA sequencing system, Promega), plasmid DNA was purified using the Qiagen Plasmid Midi Kit (Qiagen, Basel, Switzerland). IGF-I expression was assessed with primers specific for O. kisutch IGF-I (3), i.e. IGF-IS1 (5'-AGAGAGGCTTTTATTTCAGT-3') and IGF-IS2 (5'-ACTAACCTTGGGTG TTCCTTG-3'). Hot start, touch down RT-PCR was performed with 35 cycles of 1 min at 92 C, 2-min annealing (temperature decreasing from 50 to 40 C), and 3 min at 72 C, followed by two rounds of 3'-rapid amplification of cDNA ends. The first round was performed with 35 cycles of 1 min at 92 C, 2 min at 46 C, and 3 min at 72 C, using IGF-IS1 as gene-specific primer 1 and in an upstream direction the primer 5'-CCTGAATTCTAGAGCTCAT-3' (UDP) fitting on the restriction site of the poly(deoxythymidine)15 primer. For the second round, 1 µl of a dilution of 1:100 of the PCR product of the first round was used as template. Amplification was performed using the primers 5'-CAGGCTATGGCCCCAATGC-3' and UDP with 35 cycles of 1 min at 92 C, 2 min at 56 C, and 3 min at 72 C. The translated 5'-region was amplified using IGF-IS2 as the gene-specific antisense primer and the primer Pre-S TTTGTCGTGCGGAGACCCG in the 5'-untranslated region of salmon IGF-I (6) by RT-PCR with 35 cycles of 1 min at 92 C, 2 min at 53 C, and 3 min at 72 C.
Tissue expression
Total RNA of O. mossambicus organs was extracted, and
first strand cDNA was synthesized as described above. The primers
5'-CAGGCTATGGCCCCAATGC-3' and IGF-IS2 were used in a hot start, touch
down PCR run with 35 cycles and decreasing annealing temperatures (55
to 45 C). The PCR products were separated on 2% agarose gel and
stained with ethidium bromide. Southern blotting was performed on
amplification products. DNA was transferred on positively charged nylon
membrane (Boehringer Mannheim) overnight. The blot was prehybridized
for 30 min in DIG EasyHyb (Boehringer Mannheim) and hybridized
for 2 h at 68 C with the addition of 4 µl digoxygenin
(DIG)-labeled RNA probe. The DIG-labeled RNA probe spanned the region
between the primers IGF-IS1 and IGF-IS2 of tilapia IGF-I and used the
DIG RNA Labeling Kit (Boehringer Mannheim). The hybridized probe was
detected with the DIG DNA detection kit (Boehringer Mannheim) and was
visualized by chemiluminesence reaction on x-ray photographs.
Immunohistochemical protocol
To reduce unspecific binding, sections were treated with PBS
containing 2% BSA and 2% normal goat serum and processed for
immunofluorescence. Three antisera (K37, 116, and 118) raised in
rabbits against human IGF-I were used diluted at 1:300 (22, 24).
Sections were incubated with one of the IGF-I antisera for 12 h at
4 C. After repetitive washing in PBS (pH 7.4), the primary antisera
were detected using biotinylated goat antirabbit IgG (Bioscience
Products, Emmenbrucke, Switzerland; 1:100) for 30 min at room
temperature. Thereafter, sections were washed in PBS and incubated with
streptavidin-fluorescein-isothiocyanate (FITC; Bioscience Products;
1:50) for 30 min at room temperature. For analysis of the coexistence
of the classical islet hormones with IGF-I in the endocrine pancreas
(24) and IGF-I and tyrosine hydroxylase (TH) in the head kidney (25),
single sections were incubated for double immunofluorescence. Rabbit
IGF-I antisera and nonrabbit antisera against the islet hormones and TH
were applied consecutively and detected by appropriate secondary
antisera and labeling with FITC or Texas Red, respectively, as
described previously (22, 23, 24, 25).
Specificity controls
The specificity of the reactions obtained was tested using the
following controls: 1) replacement of the primary antiserum by
nonimmune serum, 2) preabsorption of the IGF-I antisera with
recombinant human IGF-I, IGF-II, and bovine insulin (40 µg; 400 µg
peptide/ml diluted antiserum). As positive controls, sections of rat
pancreas (24) were processed. The different IGF-I antisera used
probably reacted with identical cells as was found by analysis of
consecutive sections. Preabsorption of the antisera with 40 µg
recombinant human IGF-I/ml completely blocked immunoreactions.
IGF-I-immunoreactions were not affected by preabsorption with IGF-II
and insulin at concentrations as high as 400 µg/ml. Likewise,
reactions obtained with the antisera against the islet hormones and TH
were extinguished only by preabsorption with the corresponding
antigen.
Photomicrographs were taken with a Zeiss Axiophot (Zeiss, Zurich, Switzerland). The fluorochromes were visualized with fluorescence modules for FITC and for Texas Red (24).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
500 and
200 bp) were detected, but the PCR product of the expected size
(200 bp) occurred in all organs studied (Fig. 2A). By Southern blotting of the DNA and
hybridization with an IGF-I specific RNA probe, the approximately
200-bp product was verified. (Fig. 2B).
|
, A and B). In all portions
of the gastrointestinal tract, IGF-I-IR endocrine cells (Fig. 3C)
occurred. The distribution and frequency of the gastrointestinal
IGF-I-IR cells varied largely among individuals. Whereas some
individuals showed scarce small accumulations of IGF-I-IR cells, others
exhibited frequent big clusters of IGF-I-IR cells. Furthermore,
depending on the individual, IGF-I-IR cell clusters occurred in
stomach, small or large intestine, or several portions of the
gastrointestinal tract. The gastrointestinal IGF-I-IR cells did not
react with any of the antisera against the classical islet hormones.
Analysis of gills revealed two cellular sites of IGF-I
immunoreactivity. In the gill filaments, IGF immunoreactivity
occurred in some chondrocytes (Fig. 3D), which in part seemed to
divide. In addition, several cells in the gill epithelium, probably
chloride cells, exhibited IGF-immunoreactivity (Fig. 3E). In the
kidney, numerous IGF-I-IR cells were located in the epithelium of
proximal tubules (Fig. 3F). In the head kidney, IGF-I
immunoreactivity was detected in interrenal cells, as
identified by their specific location and their failure to react with
antisera against TH (Fig. 4, A and B). In
the ovary, IGF-I immunoreactivity was occasionally located in granulosa
cells around developing oocytes (Fig. 5A)
and constantly around mature (Fig. 5B) oocytes. In the male
gonad, IGF-I-IR Sertoli cells and spermatocytes were present (Fig. 5C).
IGF-I-IR neurons were detected at all levels of the brain. In all
individuals studied, a varying number of Purkinje perikarya and/or
dendrites (Fig. 5D) in the cerebellum exhibited IGF-I
immunoreactivity. However, both frequency and distribution
patterns of IGF-I-IR neurons in other areas (Fig. 5E) showed
pronounced variation among the different individuals. In the retina,
IGF-I immunoreactivity occurred in receptor and bipolar neurons and in
axons of the ganglionic cells (Fig. 5F).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
), however, are remarkably high compared
with those of salmonids, S. aurata, and humans, indicating a
particularly conservative phylogeny. This result is in accordance to
the main physiological impact of IGF-I A and B domains as established
in mammals (1).
|
In contrast to the clear expression of IGF-I mRNA in tilapia liver, no IGF-I-IR liver cells could be identified for certain. Similarly, only low or no IGF-I immunoreactivity was localized in rat liver, but pretreatment with colchicine significantly increased cytoplasmic IGF-I immunoreactivity (28). Therefore, in tilapia liver, IGF-I is probably rapidly released into the circulation after synthesis, as has been presumed for mammals (28).
IGF-I mRNA was expressed in both portions of the tilapia gastroentero-pancreatic system. The expression of IGF-I mRNA in pancreas and the correlated presence of IGF-I-IR in islet cells agree with previous immunohistochemical localizations of IGF-I in islet cells of other teleosts (21, 22, 23) and the detection of IGF-I mRNA in salmonid principal islet (19). The early appearance of IGF-I in islets during fish development (23) suggests important biological functions of islet-derived IGF-I. On the one hand, IGF-I derived from the tilapia somatostatin cells may be a paracrine regulator of insulin secretion from the B cells, as is likely in mammals (29). On the other hand, considerable amounts of sulfation activity have been determined in eel pancreas that were unaffected by hypophysectomy (30). In goby, 35SO4 incorporation in cartilage was significantly reduced after isletectomy, but hepatic GH binding was unchanged (12, 31). Based on these experiments, it has been assumed that insulin plays the crucial role and interacts with GH in the regulation of hepatic IGF-I production in teleosts (12, 31). Although mammalian proinsulin stimulated 35SO4 incorporation in salmonid cartilage (32), the presence of IGF-I in teleost islets may indicate that the observed effects (12, 30, 31) are directly caused by islet-derived IGF-I. Thus, islet-derived IGF-I may be involved in fish growth-promoting processes as an endocrine hormone.
The presence of IGF-I at the gene and peptide levels in tilapia gut is consistent with the localization of IGF-I immunoreactivity in entero-endocrine cells of other bony fish (21, 23) and the determination of IGF-I mRNA in salmonid intestine (20, 33). Although in tilapia islets, IGF-I immunoreactivity occurred in somatostatin-IR cells as found in other bony fish (22, 23), the IGF-I-IR entero-endocrine cells did not react with any antiserum against the classical islet hormones. Furthermore, in contrast to the IGF-I-IR islet cells, the intestinal IGF-I-IR cells showed remarkable interindividual variation in frequency and distribution. The physiological meaning of these observations is as yet unclear. However, it is tempting to speculate that IGF-I released from mucosal cells may exert paracrine effects in response to varying local demands in contrast to the above-suggested, mainly endocrine effects of islet-derived IGF-I. This hypothesis is supported by studies in the pig, in which the amounts of IGF-I immunoreactivity and type 1 receptors coincide with villous growth and maturation (34). As in addition to gill and kidney, gut is a major osmoregulatory organ in fish (12), an involvement of the IGF-IR mucosal cells in osmoregulation is also conceivable.
The expression of IGF-I mRNA in tilapia gill and kidney is consistent with similar results obtained in salmonids (17, 20, 33) and, partly, Sparus aurata (26). GH and IGF-I significantly improved the ability of rainbow trout to maintain serum osmolarity and sodium levels after transfer to seawater; the effect of IGF-I is greater than that of GH (13). The levels of IGF-I mRNA in salmonid gill and kidney notably increased after injection of GH; transfer to seawater did not alter the IGF-I mRNA levels in liver, but did so in gill and kidney (17). Based on these experiments, the hypothesis has been raised that in gill and kidney, IGF-I may function as an autocrine/paracrine factor in osmoregulation (17). The IGF-IR chloride cells of the gill epithelium shown in the present study are probably the production and release sites for the assumed paracrine effects of IGF-I on osmoregulation in teleost gill. IGF-I seems to exert its osmoregulatory effects directly, as in in vitro preparations of coho salmon gill, recombinant bovine IGF-I stimulated Na+,K+-adenosine triphosphatase (35). Nothing is yet known about the possible effects of IGF-I on teleost kidney function. However, our data demonstrating cells of the proximal tubule as the likely sites of IGF-I production indicate that IGF-I in teleost kidney may be involved in several parameters of renal function, as is likely in mammals (36).
A subpopulation of probably proliferating chondrocytes in the gill filaments also showed IGF-I immunoreactivity. For mammals, the presence of IGF-I in chondrocytes is still under discussion. The expression of IGF-II mRNA has been reported for murine chondrocytes at any stage of development, whereas no IGF-I mRNA could be detected (37). In contrast, IGF-I immunoreactivity has been localized in proliferating rat chondrocytes (28), which also express IGF-I mRNA (38). Although no data were obtained on IGF-II in tilapia cartilage, our results strongly suggest that the observed effects of GH on fish cartilage that are probably mediated by IGF-I (12, 13, 30, 31) under physiological conditions are caused by IGF-I derived from local chondrocytes.
No IGF-I transcripts could be detected in salmonid head kidney (17). The present study, in contrast, demonstrates the expression of IGF-I mRNA in tilapia head kidney and identifies interrenal cells as the likely source of IGF-I. Poor and mostly negative evidence has been presented on IGF-I in mammalian adrenal cortex (28, 39, 40). Thus, the presence of IGF-I in interrenal cells may be peculiar to bony fish. An involvement of locally produced IGF-I in the regulation of steroid production has been postulated for mammalian gonads (41). Although the present study gives the first evidence for the presence of IGF-I in fish interrenal cells, we suggest that IGF-I derived from tilapia interrenal cells may be involved in the local regulation of steroidogenesis.
The expression of IGF-I mRNA in tilapia brain is in agreement with similar results obtained in salmonids (18, 20, 33) and catfish (42) and by the identification of IGF-I receptors in the brain of Cottus scorpius (43). Remarkably few immunohistochemical localizations of IGF-I have been presented in the central nervous system (CNS). These deal with protochordates and mammals and, thus, indicate that IGF-I in the CNS has a long phylogenetic history. In protochordates, neurons throughout the CNS contained IGF-I immunoreactivity (44). In the developing rat cerebellum, IGF-I immunoreactivity was localized in neurons and glial cells (45), and in adult rat brain, IGF-I-IR neurons, especially Purkinje cells, were observed after colchicine treatment (28). In correspondence, tilapia also exhibited IGF-I-immunoreactivity most pronounced in the cerebellum, but additional IGF-IR neurons were present throughout the brain. The overall and in part varying distribution pattern resembles the expression of IGF-I mRNA at all levels of rat brain (46, 47) and, thus, argues against a neurotransmitter or modulator function of IGF-I. Rather, IGF-I in adult tilapia brain may support the survival of neurons and glial cells, as has been assumed for mammals (47).
The present study also shows that IGF-I mRNA is expressed in tilapia eye and localizes IGF-I immunoreactivity in retinal receptors and neurons. Recently, expression of IGF-I mRNA has been reported for rat (48) and the cichlid fish, Haplochromis burtoni (49). The distribution pattern of IGF-I immunoreactivity in tilapia retina resembles that of IGF-I mRNA in both species. IGF-I in neuronal cells of the adult tilapia retina may serve as a maintenance factor, whereas for the receptor cells, an additional involvement of IGF-I in regeneration is conceivable (49).
In both female and male gonads of tilapia, IGF-I mRNA expression was achieved. The presence of IGF-I mRNA in ovary (18, 20, 33) and testis (20, 50) is supported by results obtained in salmonids. In the ovary, our peptide localization revealed that IGF-I immunoreactivity occurs in granulosa cells of developing and mature oocytes, similar to that described in mammals (28). A season-dependent IGF-I binding that reached its maximum at the end of prespawning has been observed in carp ovaries (51), and IGF-I has been shown to induce final oocyte maturation in red seabream in vitro (52). Thus, in fish, an intraovarian IGF-I system may be involved in the autocrine/paracrine regulation of follicular growth and oocyte maturation. In isolated salmonid spermatogenic and Sertoli cells, the expression of IGF-I has been shown (50). These observations are consistent with those obtained in tilapia, where IGF-I immunoreactivity also occurred in Sertoli cells and spermatocytes. As IGF-I receptors also occur in salmonid testis (50), future research may show that local IGF-I plays an important physiological role in several parameters of fish testicular function, as is likely in mammals (41).
In summary, the presence of IGF-I transcripts in many organs of tilapia supports and extends results obtained in salmonids indicating that extrahepatic IGF-I systems are present at the phylogenetic level of bony fish. In combination with the physiological data available, the present identification of parenchymal cells as potential sites of IGF-I production indicates that local IGF-I systems are involved in the autocrine/paracrine regulation of organ-specific functions in fish. Therefore, both the liver-derived, i.e. endocrine, and autocrine/paracrine organ-specific actions of IGF-I as established in mammals seem to have a long phylogenetic history. This is also stressed by the very conservative evolution of the functionally significant A and B domains.
|
Footnotes |
|---|
Received March 10, 1997.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  N. Shved, G. Berishvili, H. D'Cotta, J.-F. Baroiller, H. Segner, E. Eppler, and M. Reinecke Ethinylestradiol differentially interferes with IGF-I in liver and extrahepatic sites during development of male and female bony fish J. Endocrinol., December 1, 2007; 195(3): 513 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. H. Evans, P. M. Piermarini, and K. P. Choe The Multifunctional Fish Gill: Dominant Site of Gas Exchange, Osmoregulation, Acid-Base Regulation, and Excretion of Nitrogenous Waste Physiol Rev, January 1, 2005; 85(1): 97 - 177. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. D. McCormick Endocrine Control of Osmoregulation in Teleost Fish Integr. Comp. Biol., August 1, 2001; 41(4): 781 - 794. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Reinecke, A. C. Schmid, B. Heyberger-Meyer, E. B. Hunziker, and J. Zapf Effect of Growth Hormone and Insulin-Like Growth Factor I (IGF-I) on the Expression of IGF-I Messenger Ribonucleic Acid and Peptide in Rat Tibial Growth Plate and Articular Chondrocytes in Vivo Endocrinology, August 1, 2000; 141(8): 2847 - 2853. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Fruchtman, L. Jackson, and R. Borski Insulin-Like Growth Factor I Disparately Regulates Prolactin and Growth Hormone Synthesis and Secretion: Studies Using the Teleost Pituitary Model Endocrinology, August 1, 2000; 141(8): 2886 - 2894. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Perrot, E. B. Moiseeva, Y. Gozes, S. J. Chan, and B. Funkenstein Insulin-Like Growth Factor Receptors and Their Ligands in Gonads of a Hermaphroditic Species, the Gilthead Seabream (Sparus aurata): Expression and Cellular Localization Biol Reprod, July 1, 2000; 63(1): 229 - 241. [Abstract] [Full Text]
|
|
|
|
|
|
M. R. Nader, T. Miura, N. Ando, C. Miura, and K. Yamauchi Recombinant Human Insulin-Like Growth Factor I Stimulates All Stages of 11-Ketotestosterone-Induced Spermatogenesis in the Japanese Eel, Anguilla japonica, In Vitro Biol Reprod, October 1, 1999; 61(4): 944 - 947. [Abstract] [Full Text]
|
|
|
|
|
|
P. Melamed, G. Gur, H. Rosenfeld, A. Elizur, and Z. Yaron Possible Interactions between Gonadotrophs and Somatotrophs in the Pituitary of Tilapia: Apparent Roles for Insulin-Like Growth Factor I and Estradiol Endocrinology, March 1, 1999; 140(3): 1183 - 1191. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
