Retinal growth hormone in perinatal and adult rats

Journal of Molecular Neuroscience 257 Volume 28, 2006

*Author to whom all correspondence and reprint requests should be addressed. E-mail: [email protected]

Retinal Growth Hormone in Perinatal and Adult Rats

Steve Harvey,* Marie-Laure Baudet, and Esmond J. Sanders

Department of Physiology, University of Alberta, Edmonton, Alberta T6G 2H7, Canada

Received September 17, 2004; Accepted August 18, 2005

Abstract

Growth hormone (GH) mRNA and protein have recently been localized in the neural retina of embryonicchicks, in which exogenous GH promotes cell survival. GH is also expressed in the rat CNS, in which it hasneuroprotective roles, although its presence in the rat neural retina is unknown and is the focus of the presentstudy. GH immunoreactivity, to a 22-kDa protein, was present in extracts of fetal (embryonic day [ED]17) eyesand in extracts from the neural retinas of newborn pups, comparable to GH immunoreactivity in pituitaryextracts. The GH immunoreactivity in the neural retina was widespread but was most intense in large roundedcells in the retinal ganglion cell (RGC) layer and in the optic fiber layer derived from the axons of the RGCs. A693-bp cDNA was also generated by the RT-PCR of RNA extracted from the eyes of ED17 rats and from theneural retinas and eyes of newborn rats, when amplified in the presence of oligonucleotide primers for the ratGH cDNA. Expression of the GH gene in the neural retina was also shown by specific in situ hybridization ofan antisense GH riboprobe to cells in the neural retina, particularly those in the RGC layers of fetal and adultrat eyes. These results demonstrate GH expression in the neural retinas of fetal, newborn, and adult rats, inwhich retinal GH might have neuroprotective roles.

DOI 10.1385/JMN/28:03:257

Index Entries: Growth hormone; neural retina; retinal ganglion cells; rats.

Introduction

Growth hormone (GH) mRNA and protein arepresent in the neural retina of embryonic and neo-natal chicks (Harvey et al., 2001; 2003a; Takeuchi et al., 2001), in which they are primarily localized inretinal ganglion cells (RGCs) (Baudet et al., 2003;Sanders et al., 2003) and in the retinal pigmentedepithelium (RPE) (Takeuchi et al., 2001; Harvey et al., 2003a, 2003b). The presence of GH receptors(GHRs) in these tissues (Harvey et al., 2003a; Baudetet al., 2004) suggests functional autocrine orparacrine roles for retinal GH in retinal developmentor ocular function. The induction of insulin-likegrowth factor-1 (IGF-1) in GH-treated explants ofthe chick retina (Baudet et al., 2003) supports thispossibility, as does the concomitant suppression of

Journal of Molecular NeuroscienceCopyright © 2006 Humana Press Inc.All rights of any nature whatsoever are reserved.ISSN0895-8696/06/28:257–264/$30.00JMN (Online)ISSN 1559-1166DOI 10.1385/JMN/28:03:257

ORIGINAL ARTICLE

caspase-3 and apoptosis-inducing factor (AIF) geneexpression and the attenuation of apoptosis in incu-bated retinal explants (Baudet et al., 2004). RetinalGH might therefore act as a cell-survival factorduring retinal neurogenesis in the chick, which ischaracterized by two developmental waves of apop-tosis in the RGC population (Frade et al., 1997).

GH is similarly known to act as a survival factorin the rat central nervous system (CNS) (Harvey andHull, 2003), in which the GH gene is expressed (Gos-sard et al., 1987; Yoshizato et al., 1998, 1999) and inwhich exogenous GH has neuroprotective actions(Scheepens et al., 1999, 2000, 2001; Ajo et al., 2003).As the neural retina is part of the CNS, GH mightsimilarly be neuroprotective in the rat neural retina,although the presence of GH in this tissue has yet to be determined. The possible presence and

43_jmn28_3_Harvey 3/16/06 6:45 PM Page 257

258 Harvey et al.

Journal of Molecular Neuroscience Volume 28, 2006

localization of GH in the rat neural retina was there-fore assessed in the present study.

Materials and MethodsAnimals and Tissues

All the animals used in these studies wereSprague-Dawley rats (Health Sciences LaboratoryAnimal Services, University of Alberta, Edmonton),which were killed by a sodium pentobarbital over-dose (with approval from the University of Alberta’sCommittee on Animal Policy and Welfare). Wholeeyes or dissected neural retinas were then collectedrapidly at autopsy. Pituitary glands and skeletal(quadricep) muscle from some adult rats were alsocollected for comparison.

Western BlottingTissues were collected into a protease inhibitor

solution (HEPES, MgCI2, EDTA, aprotinin, leu-peptin, and pepstatin in 1% [w/v] PMSF [Baudet et al., 2003]) and homogenized using a Polytronhomogenizer (Brinkman Instruments, Westburg,NY). After centrifugation and protein determination(Bio-Rad Laboratories, Mississauga, Ontario,Canada), the supernatants were stored at –20°C priorto Western analysis by one-dimensional SDS-PAGE.The samples were added to loading buffer (10%[w/v] glycerol, 5% [w/v] 2-β-mercaptoethanol, 2%SDS, 0.001% [w/v] bromophenol blue at pH 6.8) and10% (w/v) DTT and denatured at 100°C for 4 min;the proteins were separated by electrophoresis in 8%gels. After electrophoresis, the proteins were equi-librated in transfer buffer (25 mM Tris, 192 mMglycine 20% [w/v] methanol) and transferred elec-trophoretically (100 V for 1 h at 4°C) to nitrocellu-lose paper (Bio-Rad). After transfer, nonspecificbinding sites were blocked with 5% (w/v) nonfatmilk in Tris-buffered saline (TBS [25 mM Tris HCI,0.5 M NaCI at pH 7.6]), containing Tween-20 (TBST)for 1 h at room temperature. GH immunoreactivitywas detected using a polyclonal antibody raised inrabbits against rat pituitary GH (National HormonePituitary Program, Torrance, CA), diluted 1:1000 inTBS/5% (w/v) nonfat dry milk. After an overnightincubation at room temperature, the membraneswere washed with TBST (3× for 10 min) and thenincubated for 1 h at room temperature with biotiny-lated goat anti-rabbit IgG (Vector Laboratories,Burlingame, CA) at a concentration of 1:200. Anti-body binding was then visualized using ABCreagents (Vector Laboratories) for 1 h at room tem-

perature. After washing, the blots were developedusing an enhanced chemiluminescence detectionsystem (ECL kit, Amersham) and exposed to KodakX-AR film.

ImmunocytochemistryTissues were fixed in freshly prepared

paraformaldehyde (4% [w/v]; Sigma, Mississauga,Ontario, Canada) overnight at 4°C. They were thendehydrated in a graded series of alcohol (50% for15–30 min; 70% for 30–60 min; 95% for 30–120 min)and cleared with Hemo-De (Fisher Scientific,Edmonton, Alberta, Canada) for 30 min. Tissues werethen infiltrated with paraffin wax for 24–48 h at 60°Cunder normal atmospheric pressure. Serial trans-verse (6–8 µM) sections were then taken using amicrotome and mounted onto charged slides (FisherScientific). GH immunoreactivity was determinedusing the same rabbit polyclonal antibody for ratpituitary GH, diluted 1:200 in 4% bovine serum albu-min (BSA) for 1.5 h at room temperature. After incu-bation, slides were washed three times for 10 min in4% BSA in PBS. Sections were then incubated for 1 hat room temperature in goat anti-rabbit IgG conju-gated to FITC at a dilution of 1:200. The labeled sec-tions were then examined using a Zeiss LSM 510confocal microscope equipped with appropriatelasers. Controls, in which the antibodies werereplaced by nonimmune serum, were negative (datanot shown).

RT-PCRTissues were excised in PBS, and RNA was stabi-

lized in RNALater (Ambion, Austin, TX) at 4°C,overnight. Total RNAwas extracted using an RNeasyminikit (Qiagen, Mississauga, Ontario, Canada)according to the manufacturer ’s instructions.Reverse transcription, using approx 3 µg total RNA,was achieved with Thermoscript™ RNase H– reversetranscriptase (Invitrogen, Carlsbad, CA) in a Tech-gene thermal cycler (Techne, Duxford, Cambridge,UK), using an extension temperature of 60°C (60min). mRNAtemplates were then removed by RNaseH (Invitrogen) treatment (20 min at 37°C). Follow-ing reverse transcription, the cDNA products weretransferred to ice and subsequently amplified byPCR, using 1 U of Platinum® Taq DNA polymerase(Invitrogen) per reaction, in the presence of 0.2 Moligonucleotide primers designed to generate 693-bp rat GH cDNA (Seeburg et al., 1977; Beyea et al.,2005) (JAB1, forward primer: 5′-TGG ACA GAT CACTGA GTG GCG-3′; JAB2, reverse primer: 5′-CGC


Retinal GH in Rats 259


AGAGAC ACC AGT GTG TGC-3′). Touchdown PCRwas performed with a starting temperature of 94°C(for 2 min) and then 5 cycles of 94°C (30 s) and 65°C(for 30 s), 72°C (for 1 min), followed by 30 cycles at94°C (30 s), 54.2°C (for 30 s), and 72°C (1 min), witha final extension period at 72°C (for 10 min). Tenmicroliters of PCR products were visualized byethidium bromide in 1.4% agarose gels.

Reverse-transcribed muscle RNA was similarlyamplified with JAB1 and JAB2, to determine thespecificity of retinal GH expression. As a positivecontrol the expression of cyclophilin (a housekeep-ing gene) was also assessed by RT-PCR (Hosford andOlson, 2003), using oligonucleotide primers (rCycf-5′-caccgtgttcttcgacatcac-3′; rCyc r- 5′-ccagtctca-gagctcgaaag-3 ′) designed to generate 114-bpcyclophilin cDNA (Danielson et al., 1988).

In Situ HybridizationIn situ hybridization was performed using full-

length (693 bp) antisense and sense probes (span-ning bases –21 to 672) for rat GH mRNA (Seeburg et al., 1977), as described by Beyea et al. (2005). Theprobes were made from a cDNA moiety amplifiedby RT-PCR from rat pituitary RNA in the presenceof oligonucleotide primers JAB1 and JAB2. The RT-PCR product was cloned into a PCR II–TOPO vector(Invitrogen), which was then linearized with BamH1(for the antisense probe) or Not1 (for the sense probe).Digoxigenin (DIG)-labeled antisense or sense ribo-probes were then synthesized by in vitro transcrip-tion with T7 or SP6 RNA polymerase, respectively,in the presence of DIG RNA labeling mix (RocheDiagnostics, Manheim, Germany), according to themanufacturer’s instructions. Probe concentrationswere determined by dilution and dot-blot analysis.

Tissues were fixed overnight at 4°C in 4%paraformaldehyde in PBS, dehydrated, and embed-ded in paraffin. The 6- to 8-µm tissue sections werethen deparaffinized in Clear-Rite3 (Richard AllenScientific, Kalamazoo, MI), rehydrated through agraded ethanol series, and washed in DEPC-treatedwater and then in DEPC-PBS. The sections were thenincubated in fresh 0.1% DEPC-PBS (2 × 15 min) toremove RNase, treated with proteinase K (10 µg/mL,10 min, 37°C), postfixed in 4% paraformaldehyde(10 min), and incubated in 0.1 M triethanolamine(pH 8.0), containing 0.25% (w/v) acetic anhydride,for 10 min. The sections were then prehybridizedwith 50% formamide in 2X SSC and incubatedovernight in 50% deionized formamide in 5× SSCcontaining 250 µg/mLtRNA, 50 µg/mLheparin, 2%

blocking solution, and sense or antisense DIG-labeled riboprobes (2 µg/mL) that had been previ-ously denatured (5 min, 80°C). The sections werethen sequentially washed in 2× SSC (10 min), 2XSSC/50% formamide (45 min, 45°C), 2× SSC (3 × 15min), 2X SSC (30 min, 37°C), and 0.1× SSC (60 min,60°C). Sections were then incubated (2 × 5 min) in aposthybridization buffer (0.1 M Tris, 0.15 M NaCI,at pH 7.5), and immunohistochemical detection ofthe hybridized probe was achieved by incubation (2h) with an anti-DIG antibody conjugated to alkalinephosphatase (AP [Roche Diagnostics]) in 1% block-ing solution (Roche Diagnostics). The sections werewashed in the same posthybridization buffer (2 × 15min) and equilibrated (5 min) in an additional buffer(0.1 M Tris, 0.1 M NaCI, 0.05 M MgCI2 at pH 9.5), towhich a commercial substrate for AP (BCIP andnitroblue tetrazolium chloride [NBT]; Roche Diag-nostics) was added. The sections were then incu-bated with this substrate until color developmentoccurred. Color development was then stopped bywashing in TE buffer (2 × 5 min). The sections werethen rinsed in PBS (3 × 5 min) and mounted withGel/Mount (Biomeda, Foster City, CA).

ResultsWestern Blotting

As expected, a 22-kDa protein was intenselylabeled by the rat GH antibody in extracts of the ratpituitary gland (Fig. 1). Protein bands of identicalsize were also present in extracts of fetal (ED18) andneonatal (D0) rats (Fig. 1). In each case, GH

Fig. 1. Western blotting of GH-immunoreactive proteinsin the eyes of ED18 rats (lanes 1,2) and in the neural reti-nas of newborn (D0) rats (lanes 3,4), in comparison withthat in a pooled extract of (n = 5) of adult rat pituitary glands(lane 5). The amount of protein loaded on each gel is indi-cated below each lane.


260 Harvey et al.


ImmunocytochemistryGH immunofluorescence was seen throughout

the neural retina of newborn rats (Fig. 2). GH labeling

immunoreactivity was associated with a single pro-tein moiety, with an abundance approx 0.5%–2% ofthat in the pituitary gland (on a weight basis).

Fig. 3. Immunocytochemical staining for GH in the neural retina of adult rats. GH labeling (green) is restricted to thecytoplasm of the large, rounded cells in the RGC layer (white arrows) and to the OFL derived from their axons (repre-sentative of at least three rats).

Fig. 2. Immunocytochemical staining for GH in the neural retina of newborn rats. GH labeling (green) is widespreadthroughout the neural retina but is particularly intense in the large, rounded cells in the RGC layer, in which GH immuno-reactivity is primarily nuclear (red arrowheads) or primarily cytoplasmic (white arrows). GH immunoreactivity is alsopresent in the OFL derived from axons of the RGCs (representative of at least three rats).




was, however, most intense in large rounded cellsthat had the morphological appearance and locationof ganglion cells in the RGC layer (Baudet et al., 2003;Sanders et al., 2003). Two populations of cells wereidentified in this region: one in which the labelingwas primarily nuclear, and another in which theimmunofluorescence was primarily cytoplasmic andfrom which immunoreactive axons formed the opticfiber layer (OFL). Weaker GH immunoreactivity wasalso present in the neural retina of adult rats (Fig. 3),in which GH labeling was primarily in the cytoplasmof the large cells in the RGC layer and in the cytoplasmof the OFL. The number of immunoreactive cells inthe adult neural retina was clearly less than that pre-sent in newborn rats.

RT-PCRIn the presence of oligonucleotide primers JAB1

and JAB2, a cDNAmoiety of 693 bp was, as expected,amplified from reverse-transcribed pituitary RNA.cDNAmoieties of identical size were also amplifiedwith RNA from the eyes of ED17 and newborn ratsand from RNA from the neural retinas of D0 rats(Fig. 4). The specificity of these moieties is indicatedby their absence in negative controls, in which super-script was omitted from the RT-PCR reactions andby the absence of a GH cDNAin reverse-transcribedmuscle mRNA, from which 114-bp cyclophilincDNA was amplified (Fig. 4).

In Situ HybridizationAs expected (data not shown; Beyea et al., 2005),

the antisense GH probe specifically hybridized tosomatotrophs in the rat pituitary gland. Intensehybridization to this probe was also seen in the largeganglion-like cells in the RGC layer of the ED17neural retina (Fig. 5). The hybridization to these cellswas cytoplasmic and also in the axons that comprisethe OFL. DIG labeling was also present, but lessintense, in retinal cells adjacent to the RGC layer (Fig.5). The specificity of hybridization to all of these cellswas demonstrated by the complete absence of DIGlabeling in the presence of the sense probe (Fig. 5E).

Discussion

These results clearly demonstrate the presence ofGH and GH mRNA in the neural retina of perinataland adult rats, as observed in embryonic chicks(Harvey et al., 2001, 2003a; Takeuchi et al., 2001;Baudet et al., 2003; Sanders et al., 2003). AlthoughGH gene expression has been detected previously

in the rat hypothalamus (Yoshizato et al., 1998, 1999),this is the first determination of GH mRNA or pro-tein in this part of the rat CNS.

The RT-PCR amplification of 693-bp cDNA fromretinal RNAwith oligonucleotide primers for rat GHcDNA strongly suggests expression of the GH genein this tissue. This possibility is supported by thehybridization of this cDNA moiety to cells in theneural retina, particularly to the cytoplasm of large,round cells in the RGC layer. This is consistent withstudies in embryonic chicks, in which the full-lengthGH transcript is expressed throughout the neuralretina, but is most intense in the cytoplasm ofimmunologically identified RGCs (Baudet et al.,2003; Harvey et al., 2003a). However, whereas full-length monomer (22-kDa) GH is also present in therat neural retina, monomer GH is barely detectablein the neural retina of embryonic chicks, in contrastto the 22-kDa protein in the chicken pituitary gland(Baudet et al., 2003; Harvey et al., 2003a). Truncated(15- and 16-kDa) proteins account for almost all ofthe GH immunoreactivity in the chick neural retina(Baudet et al., 2003), although these moieties werenot observed in the rat neural retina and are minorcomponents of GH immunoreactivity in the chickenpituitary gland (Aramburo et al., 2001). This suggeststhat transcription and translation of the GH message

Fig. 4. RT-PCR of reverse-transcribed mRNA extractedfrom the neural retina and eyes of newborn (D0) rats andthe eyes of ED17 rats, in comparison with reverse-tran-scribed mRNA from the pituitary glands and skeletal (pec-toralis) muscle of adult rats (using oligonucleotide primersJAB1 and JAB2 designed to generate a 693-bp rat GH cDNA).Control reactions with mRNA are indicated by negativesymbols (–ve). RT-PCR of reverse-transcribed mRNAextracted from muscle using oligonucleotide primersdesigned to generate a 114-bp rat cyclophilin cDNA isshown for comparison. The data are representative of atleast three RT-PCRs using different tissue extracts.


262 Harvey et al.


in the rat neural retina is comparable to that in the rat pituitary gland, whereas transcription of the GH gene (Takeuchi et al., 2001), translation of the GH message, or post-translational processingof the translated protein appears to be tissue-specific in the chick neural retina (Baudet et al., 2003;Harvey et al., 2003a).

Within the rat neural retina, GH immunoreactiv-ity was most intense in cells with the appearanceand location of RGCs in the RGC layer and in theOFL derived from ganglion cells. This is consistentwith the immunocytochemical localization of GH inthe RGCs of embryonic chicks (Baudet et al., 2003;Sanders et al., 2003). Moreover, the intracellularlocalization of GH in the RGCs of the newborn rat

is in both cytoplasmic and nuclear compartments,as in embryonic chicks (Baudet et al., 2003; Harveyet al., 2001, 2003b). The cytoplasmic localization ofGH immunoreactivity likely reflects the synthesisand storage of GH in this compartment, whereas thenuclear localization of GH immunoreactivity likelyreflects the binding of GH to nuclear GHRs (Lobie etal., 1993, 1994; Lincoln et al., 1994; Fraser and Harvey,1994; Harvey and Hull, 2003). It is therefore perti-nent that GHRs have been found throughout theneural retinas of embryonic chicks (Baudet et al.,2004) and fetal rats (Garcia-Aragon et al., 1992).

The GH immunoreactivity in the RGCs of the new-born rat was primarily cytoplasmic in cells adjacentto the OFL(in the inner retina) and primarily nuclear

Fig. 5. In situ hybridization of GH mRNA in the neural retina of newborn rats. (A) Specific hybridization with a 693-bp DIG-labeled BamH1 antisense probe for GH mRNA is shown throughout the neural retina, particularly in large,rounded cells in the RGC layer of the inner retina. Magnification, ×40. (B) Magnification, ×100. (C) Magnification, ×400.(D) Magnification, ×1000. Labeling to a large ganglion cell and its axon (green arrow) is shown. (E) The neural retina isnot labeled in the presence of a Not1 DIG-labeled sense probe. Magnification, ×40. Representative of similar data fromat least three rats.




in cells in the outer RGC layer. This apparent zona-tion was not seen in the retinas of embryonic chicks(Harvey et al., 2001; Baudet et al., 2003), althoughthis may reflect differences in relative maturity anddifferentiation. This apparent zonation might alsoreflect differences in RGC differentiation during themigration of the cells through the neural retina (Snowand Robson, 1995) or spatial differences in RGC mor-phology (Ramon and Yamasaki, 1996). It is of inter-est that this apparent zonation was not present inthe neural retina of the adult rat, in which GHimmunoreactivity was less and purely cytoplasmic.

The presence of GHR immunoreactivity in the ratneural retina (Garcia-Aragon et al., 1992) suggeststhat it is a target site for GH action. Moreover, becausethe blood-retinal barriers are likely to severelyimpede the entry of systemic GH into the neuralretina (James and Cotlier, 1983; Cunha-Vaz, 1997), reti-nal GHRs are likely to be autocrine or paracrine targetsites for GH produced locally. The presence of thesebarriers also suggests that GH immunoreactivityfound in the neural retina is unlikely to be sequesteredfrom peripheral circulation.

Within the rat neural retina, GH binding to retinalGHRs is likely to have functional significance. Thispossibility is supported by the actions of exogenous(monomer) GH in inducing IGF-1 gene expression inchick neural retinal explants (Baudet et al., 2003), inwhich GH inhibits apoptosis and concomitantlyinhibits caspase-3 and AIF gene expression (Baudetet al., 2004). GH might therefore have neuroprotec-tive roles in the rat neural retina, as in the brain(Scheepens et al., 1999, 2000, 2001; Ajo et al., 2003). Itmight therefore be of interest that GH immunoreac-tivity in the neural retina of adult rats appeared to bemuch less than that in the newborn rat. The loss ofRGCs during aging (Robinson et al., 1987; Danias et al., 2004; McKinnon, 2003) might thus reflect anage-related loss of GH neuroprotection. Because RGCdeath in the elderly is a causal factor in the etiologyof glaucoma (McKinnon, 1997, 2003), retinal GH mightprovide a marker for this disease and exogenous GHmight have a novel therapeutic application (Kaushiket al., 2003) in its treatment (Sanders et al., 2003).

In summary, these results demonstrate extrapi-tuitary expression of the GH gene in the neural retina of perinatal and adult rats, in which GHimmunoreactivity is primarily in large cells with themorphological appearance and location of RGCs. Asthis is similar to GH expression in the neural retinasof embryonic chicks, in which GH is physiologicallyinvolved in retinal cell survival, GH might be neuro-

protective for rat RGCs and retinal GH in rats mighthave relevance in the diagnosis and treatment ofglaucoma or other degenerative ocular diseases of aging mammals.

AcknowledgmentsThis work was supported by a grant from the Cana-

dian Institute for Health Research and the Natural Sci-ences and Engineering Research Council of Canada.We thank Dr. A. Parlow for providing the rat GH anti-body. M. L. B. is in receipt of a University of AlbertaFS Chia Studentship and a studentship from theAlberta Heritage Foundation for Medical Research.

ReferencesAjo R., Cacicdeo L., Navarro C., and Sanchez-Franco F.

(2003) Growth hormone action on proliferation and dif-ferentiation of cerebral cortical cells from fetal rat.Endocrinology 144, 1086–1097.

Aramburo C., Carranza M., Reyes M., Luna M., Martinez-Coria H., Berumen L., and Scanes C.G. (2001) Charac-terization of a bioactive 15 kDa fragment produced byproteolytic cleavage of chicken growth hormone.Endocrine 15, 231–240.

Baudet M.- L., Rattray D., Sanders E. J., and Harvey S.(2004) Growth hormone: a survival agent in the devel-oping chick retina. GH IGF Res. 14, 105.

Baudet M.- L., Sanders E. J., and Harvey S. (2003) Retinalgrowth hormone in the chick embryo. Endocrinology144, 5459–5468.

Beyea J. A., Olson D. M., and Harvey S. (2005) Growthhormone expression in the perinatal and postnatal ratlung. Dev. Dynam. 232, 1037–1046.

Cunha-Vaz J. G. (1997) The blood-ocular barriers: past,present and future. Doc. Ophthalmol. 93, 149–157.

Danias J., Lee K. C., Zamora M. F., Chen B., Shen F., Filip-popoulos T., et al. (2004) Quantitative analysis of reti-nal ganglion cell (RGC) loss in aging DBA/2glaucomatous mice: comparison with RGC loss in agingC57/BL6 mice. Invest. Ophthalmol. Vis. Sci.44, 5151–5162.

Danielson P. E., Forss-Petter S., Brow M. A., et al. (1988)pLB15: a cDNA clone of the rat mRNA encodingcyclophilin. DNA 7, 261–267.

Frade J. M., Bovolenta P., Martinez-Morales J. R., ArribasA., Barbas J. A., and Rodriguez-Teba A. (1997) Controlof early cell death by BDNF in the chick retina. Devel-opment 124, 3313–3320.

Fraser R. A. and Harvey S. (1992) Ubiquitous distribution ofgrowth hormone receptors and/or binding proteins inadenohypophyseal tissue. Endocrinology 130, 3593–3600.

Garcia-Aragon J., Lobie P. E., Muscat G. E. O., Gobius K. S.,Norstedt G., and Waters M. J. (1992) Prenatal expres-sion of growth hormone (GH) receptor/binding pro-tein in the rat: A role for GH in embryonic and fetaldevelopment? Development 114, 869–876.


264 Harvey et al.


Robinson S. R., Horsburgh G. M., Dreher B., and McCallM. J. (1987) Changes in the numbers of retinal ganglioncells and optic nerve axons in developing albino rabbit.Brain Res. 432, 161–174.

Sanders E. J., Walter M. A., Parker E., Aramburo C., andHarvey S. (2003) Opticin binds retinal growth hormonein the embryonic vitreous. Invest. Ophthalmol. Vis. Sci.44, 5404–5409.

Scheepens A., Sirimanne E., Beilharz E., Breier B. H., WatersM. J., Gluckman P. D., and Williams C. E. (1999) Alter-ations in the neural growth hormone axis followinghypoxic-ischemic brain injury. Mol. Brain Res. 68, 88–100.

Scheepens A., Sirimanne E. S., Breier B. H., Clark R. G.,Gluckman P. D., and Williams C. E. (2001) Growth hor-mone as a neuronal rescue factor during recovery fromCNS injury. Neuroscience 104, 677–687.

Scheepens A., Williams C. E., Breier B. H., Guan J., andGluckman P. D. (2000) A role for the somatotropic axisin neural development, injury and disease. J. Pediatr.Endocrinol. Metab. 13, 1483–1491.

Seeburg P. H., Shine J., Martial J. A., Baxter J. D., and Good-man H. M. (1977) Nucleotide sequence and amplifica-tion in bacteria of structural gene for rat growthhormone. Nature 270, 486–494.

Snow R. L. and Robson J. A. (1995) Migration and differ-entiation of neurons in the retina and optic tectum ofthe chick. Exp. Neurol. 134, 13–24.

Takeuchi S., Haneda M., Tshigawara K., and Takahashi S.(2001) Identification of a novel GH isoform: a possiblelink between GH and melanocortin systems in thedeveloping chicken eye. Endocrinology 142, 5158–5166.

Yoshizato H., Fujikawa T., Shibata M., Tanada M., andNakashima K. (1999) Stimulation of growth hormonegene expression in the pituitary and brain by Panaxginseng C.A. MEYER. Endocr. J. Suppl. S85–S88.

Yoshizato H., Fujikawa T., Soya H., Tanaka M., andNakashima K. (1998) The growth hormone (GH) geneis expressed in the lateral hypothalamus: Enhancementby GH-releasing hormone and repression by restraintstress. Endocrinology 139, 2545–2551.

Gossard F., Dihl F., Pelletier G., Dubois P. M., and MorelG. (1897) In situ hybridization to rat brain and pituitarygland of growth hormone cDNA. Neurosci. Lett. 79,251–256.

Harvey S. and Hull K. (2003) Neural growth hormone: anupdate. J. Mol. Neurosci. 20, 1–14.

Harvey S., Johnson C. D. M., and Sanders E. J. (2001) Growthhormone in neural tissues of the chick embryo. J. Endocrinol. 169, 87–498.

Harvey S., Kakebeeke M., and Sanders E. J. (2003a) Growthhormone localization in the neural retina and retinalpigmented epithelium of embryonic chicks. J. Mol. Neu-rosci. 22, 139–145.

Harvey S., Kakebeeke M., Murphy A. E., and Sanders E. J.(2003b) Growth hormone in the nervous system:autocrine or paracrine roles in retinal function? Can. J.Physiol. Pharmacol. 81, 371–384.

Hosford G. E. and Olson D. M. (2003) Effects of hypoxiaon VEGF, its receptors, and HIF-2α in the newborn ratlung. Am J. Physiol. 288, L161–L168.

James C. R. and Cotlier E. (1983) Fate of insulin in theretina: an autoradiographic study. Br. J. Ophthalmol. 67,80–88.

Kaushik S., Pandav S. S., and Ram J. (2003) Neuroprotec-tion in glaucoma. J. Postgrad. Med. 49, 90–95.

Lincoln D. T., El-Hifnawi E., Sinowatz F., and Waters M. J.(1994) Immunohistochemical localization of growthhormone receptor binding protein in the mammaliancerebellum. Ann. Anat. 176, 419–427.

Lobie P. E., Garcia-Aragon J., Lincoln D. T., Barnard R.,Wilcox J. N., and Waters M. J. (1993) Localization andontogeny of growth hormone receptor gene expressionin the central nervous system. Dev. Brain Res. 74, 225–233.

Lobie P. E., Wood T. J., Chen C. M., Waters M. J., and NorstedtC. (1994) Nuclear translocation and anchorage of thegrowth hormone receptor. J. Biol. Chem.269, 31,735–31,746.

McKinnon S. J. (1997) Glaucoma, apoptosis, and neuro-protection. Curr. Opin. Ophthalmol. 8, 28–37.

McKinnon S. J. (2003) Glaucoma: ocular Alzheimer’s dis-ease? Front. Biosci. 1, 1140–1156.


Documents

Retinal growth hormone in perinatal and adult rats