- Email: [email protected]
A PCR analysis of rhodopsin gene transcription in rat pineal photoreceptor diffentiation
Det'elopmental Brain Research, 69 (1992) 149-152 © 1992 Elsevier Science Publishers B.V. All rights reserved 0165-3806/92/$05.00
149
BRESD 51512
Research Reports
A PCR analysis of rhodopsin gene transcription in rat pineal photoreceptor differentiation Masasuke Araki
a
a n d Shigeru T a k e t a n i
b
a Department of Anatomy, Jichi Medical School, Tochigi 329-04 (Japan) and b Department of Hygiene, Kansai Medical Unirersity, Osaka 570 (Japan) (Accepted 2 June 1992)
Key words: Pineal; Rhodopsin; PCR; Norepinephrine; Culture; Rat
A polymerase chain reaction (PCR) method was used to detect rhodopsin transcripts in rat pineals both in vivo and in vitro. Only very low levels of transcripts were detected in preparations from adult rat pineal tissue, but fairly large amounts were detected in eDNA preparations from cultures of newborn rat pineals. The transcript level was reduced significantly if the cells had been cultured in the presence of 10 ~tM norepinephrine (NE). This concentration of NE had previously been shown to abolish almost completely the rhodopsin immunoreactivity normally seen in such cultures. The present work indicates that part, but not all, of the effect of NE is probably at the level of translation,
INTRODUCTION
The mammalian pineal is considered to be an exclusively endocrine organ which synthesizes and secretes melatonin with a rhythm produced by the suprachiasmatic nucleus and transmitted to the pineal via the superior cervical ganglion9. Several reports, however, have indicated that some molecules characteristic of the retina are also found in the mammalian pineal; examples are S-antigen, rhodopsin kinase, inter-photoreceptor retinoid binding protein (IRBP) and a-transducin 6't°a4'ts. Immunohistochemical studies with antirhodopsin, however, have not revealed any clear-cut positive staining in the rat pineal 4'~6, suggesting the absence of photoreceptor function. Previously, however, we found that numerous rhodopsin-immunoreactive (Rho-l) cells differentiate when pineais from newborn rats are transferred to in vitro conditions 3. Rat pineals removed later than postnatal day 6, however, manifest only a very few Rho-I cells in culture, and norepi.t,~phrine (NE) irreversibly suppresses the differentiation of Rho-i cells in culture of newborn rat pineals 2. In an attempt to learn more about the molecular mechanisms involved in NE suppression of cell differ-
entiation, we have now analyzed the level of rhodopsin gene transcripts by the polymerase chain reaction (PCR) method. The very small sizc of the pineal allowed us to compare in vivo with in vitro transcript levels by this method. It was found that Rho-! cells do synthesize a fair amount of rhodopsin gene transcripts in vitro, and that the intact pineal organ of the mature rat contains only a few transcripts. MATERIALS A N D M E T H O D S
Preparation of tissues and cell culture Adult rats of the Wistar strain were deeply anesthetized with sodium pentobarbital. Retinas and pineals were isolated and spleen was taken as a control tissue. The isolated tissues were rinsed briefly with Hanks' saline solution and stored at -80°C until used. The procedure used for the preparation of pineal cultures has been reported in a previous paper 2. Briefly, pineals were isolated from newborn rats, dissected into small pieces and then treated with collagenase. They were cultured in collagen-coated dishes with Eagle's MEM supplemented with 5% fetal calf serum, 55 ~tg/ml sodium pyruvate and 6 mg/ml glucose. Pineals were cultured for 8 days with or without 10 ~M NE. Polymerase chain reaction Total RNA was isolated from approximately 7x 10s cells or from rat tissues using the guanidine thiocyanate procedure zl. cDNA was synthesized from 2 ~tg of total cellular RNA by Super Script RNAse H reverse transcriptase (Gibco BRL, Gaithersberg, MD), using
Correspondence: M. Araki, Department of Anatomy, Jichi Medical School, Tochigi 329-04, Japan. Fax: (81) (285) 44-8169.
150 oligo(dT) 12-18. A (I.05 aliquot of the eDNA product was used for amplification by PCR. The two oligonucleotides prepared for use as PCR amplificathm primers were: 5'-AAGCCGATGAGCAACTTCC-Y, corresponding to nucleotides 515-533 of mouse opsin eDNA; and 5'-TCATCTCCCAGTGGA'fTCTI'-3', complementary to nucleotides i,067-!,086 of that eDNAs. 25 cycles of PCR amplification were c,,rried out, As a= ~aternal control for the amount of PCR reaction product, amplific :tion of ferro-:helatase eDNA, the last-step enzyme of heme biosy~faesis, ',','as carried out by PCR under similar conditions as described above. Tl:e primers used were 5'-TTGATG'VI'AAACATGGGAGG-3', corresponding to nucleotides 211-231 of mouse fcrre~:helatase eDNA; and 5'-TTGGACTGCCAACCAGTCGGTAY, c-',,nplementary to nucleotides 880-901 of the eDNA it.
A
B
:4.3 2.3 :2.0
0.56
Sol~dwrn blot analysis The probe used was a synthetic oligonucleotide, 5'-ATGGTTATCATCATGGTCA-3', corresponding to positions 851-869 of mouse opsin eDNA. It was labeled at the 5' end with [y-32P]ATP (Amersham, 3,000 Ci/mmol), using T4 polynuck:otide kinase. The products of PCR amplification were electrophoresed in a 1.2~ agarose gel and transferred to a nylon membrane (Amersham, Hybond N ÷ ). The filter was prehybridized for 3 h at 48°C in a solution containing 5 x Denhaldt's solution (0.1% bovine serum albumin, 0.1% polyvinyl pyrrolidone, and 0. 1% Ficoll), 5 × SSC (1 × SSC = 0.15 M NaCI, 0.15 M sodium citrate, pH 7.0), 20 mM sodium phosphate (pH 7.0), 0.1% SDS and 40 p,g/ml sonicated salmon sperm DNA. Hybridization with the radiolabelled oligonucleotide (I × 10f' cpm/ml) was carried out at 48°C for 16 h. Stringent washing was performed at 48°C with three changes of 4×SSC containing (I.l~ SDS. The filter was exposed to Kodak X-ray film overnight. S,'cthern blot analy,,.is of ferrochelatase eDNA was performed, using the full-length fragmew, of mouse ferrochelatase eDNA as a probe t i. The intensity of the positive bands was estimated by scanning using an Advantec model DMU-33C densitometer.
RESULT~ it was anticipated that the PCR products amplified with the present primers would correspond to nu. cleotide positions 515-1,086, or approximately 560 base pairs, Fig, IA shows the elcctrophorctic patter, of the PCR products. In the PCR products from the eDNA isolated from neural retinal tissue, there is a major band at the position of mouse opsin eDNA (lane 2). In control experiments using rat spleen eDNA (lane 1) no clear band can be seen at this position. A faint band can be seen in the products from the eDNA from cultured pineal cells, whether or not the cells were cultured in the presence (lane 3) or absence (lane 4) of NE. No band at this position can be clearly seen in the electrophoretic pattern yielded by products from adult pineal tissue eDNA (lane 5). Probe hybridization of the amplified eDNA (Fig. I B) gave a clearer demonstration of the results, In neural retinal tissues a very thick band was found at the position of mouse opsin eDNA (lane 2), Fainter but still obvious bands were seen in the preparations from both normal (lane 4) and NE-treated (lane 3) pineal cultures, Densitometric measurements indicated that the band found in normal cultures was about
Fig. 1. Electrophoretic pattern (A) and probe hybridization analysis (B) of the PCR products. Synthesis of the eDNA samples, amplification by PCR, electrophoresis on 1.2% agarose gel, transfer to a nylon membrane and probe hybridization were as described in Materials and Methods. The sources of the eDNA were: lane I, spleen; lane 2, neural retina; lane 3, pineal cells cultured in the presence of 10/~M norepinephrine; lane 4, cultured pineal cells; and lane 5, adult pineal tissue. Lane 6 contains Hindill-digested fragments of lambda phage used as size markers.
three-times as intense as that found with NE-treated cultures. A very faint band could also be seen on probe hybridization in the preparation from adult rat pineal tissue (lane 5). The control preparation from spleen tissue still showed no trace of a band at this position (lane 1), To assess the quantity of the PCR products, eDNA of ferrochelatase, which is the last-step enzyme of heine biosynthesis, and is known to be a house-keeping gone 12, was amplified with the same materials as used in opsin gene amplification. No significant difference was noticed in the intensity of the bands obtained by probe hybridization of amplified ferrochelatase eDNA products from adult pineal tissue, normal pineal culture and NE-treated pineal culture (Fig. 2). Data shown in this study were reproducible with two independent preparations of these materials. DISCUSSION The present study clearly demonstrates the suitability of the PCR method for describing specific gene expression at the transcriptional level during cell differentiation. This method is particularly useful when only small amounts of tissue or cultured cells are available. The results show that cultured rat pineals contain opsin mRNA in small but still obvious amounts, supporting the supposition that the rhodopsin-immunoreactive (Rho-l) cells found in pineal cultures
151
Fig. 2. Southern blot analysis of the PCR products with the ferrochelatase eDNA. Synthesisof the cDNA samples, amplificationby PCR, electrophoresis and transfer to a nylon membranewere similar to those as described in the legend for Fig. 1, except that two oligonucleotides of the ferrochelatase eDNA were used as primers for PCR. Hybridization with the fragment of the ferrochelatase eDNA was performed as described. The sources of the cDNA were the same as shown in the legendfor Fig. 1.
synthesize rhodopsin molecules. It also appeared here that intact pineal tissue from adult rats contains only trace amounts of opsin gene transcripts. When newborn rat pineals are cultured in vitro, numerous Rho-I cells differentiate, but such differentiation is largely suppressed by the presence of 1 # M norepinephrine '~. In the presence of 10 ~M, as used in the present work, the number of Rho-I cells which differentiate is only a few percent of the number found in control cultures. In this previous work, it also appeared that NE caused degeneration of Rho-I cells, although some still seemed normal 2. in contrast to the very few Rho-I cells seen by immunohistochemistry, the rough quantitative estimation used on the hybridization bands (Fig. 1B) suggests that the level of opsin gene transcripts found in cultures treated with 10 # M NE is still about one-third of that in control cultures, although it is very difficult to quantitate the PCR products by reasons deriving from the enormous amplification by the method. Taken together, these results indicate that NE probably affects Rho-I cell differentiation to some extent at the translational level without inducing degeneration of all such cells. In situ hybridization studies may yield more information on the extent of cell degeneration vs. simple inhibition of translation.
In the present study it was shown that adult rat pineals have opsin mRNA although in such very small amounts that it can be detected only after amplification by PCR. Immunohistochemically, no clear-cut positive staining was found in rat pineal 4,~6, although Korf et al. 7 reported such staining in rat and m o u s e 7. No phenotypic evidence was obtained for transgene expression in the pineal, or in any other tissue except the retina, when chimeric gene fusion of rod opsin was used to generate transgenic mice 8. In another study on transgenic mice, it was found that c/s-acting DNA elements required for rhodopsin synthesis are expressed in the retina but not in the pineal ~7. Analogy with some other systems also suggests that such very low levels of opsin gene transcripts as found here in the intact rat pineal do not necessarily mean that synthesis of the protein will occur. For example, lowlevel transcription of the 8-crystallin genes has also been observed in some non-lens tissues of chick embryos without detectable protein synthesis ~. Similarly, low-level transcription of neuron-specific genes is found in adrenal medulla chromaffin cells, but this may only reflect their possible trans-differentiation into neurons under certain conditions ~3. The results of the present study therefore support the view that some neonatal pineal cells have the capacity for rhodopsin synthesis but this is almost completely lost or suppressed in the adult tissue, and the presence of NE is probably an important factor leading to such suppression. Acknowledgements. The authors are deeply grateful to Dr. E.G.
McGeer (The Universityof British Columbia) for her careful reading of the manuscript. This work was supported partly by a Grant-in-Aid from the Ministryof Education, Scienceand Culture, Japan and by a grant provided by the lchiro Kanehara Foundation. REFERENCES 1 Agata, K., Yasuda, K. and Okada, T.S., Gene coding for a lens-specific protein, 8-crystallin, is transcribed in non-lens tissues of chick embryos, Dec'. Biol., 100 (1983) 222-226. 2 Araki, M., Ceikdar mechanismfor norepinephrine suppressionof pineal photoreceptor..like cell differentiation in rat pinc~dlcultures, Deu. Biol., 149 (1992)440-447. 3 Araki, M. and Tokunaga, F., Norepinephrinc suppresses bo~h photoreceptor and neuron.like properties expressed by cultured rat pineal glands, Cell Differ. Dev., 31 (1990) 129-135. 4 Araki, M., Watanabe, K., Tokunaga, F. and Nonaka, T., Phenotypic expression of photoreceptor and endocrine properties by cultured pineal cells of the newborn rat, Cell Differ. Def., 25 (1988) 155-164. 5 Baehr, W., Falk, J.D., Bugra, K., Triantafyllos, J.T. and McGinhis, J.F., Isolation and analysisof the mouse opsin gene, FEBS Lett., 238 (1988) 253-256. 6 Donoso, L.A., Merryman, C.F., Edelberg, H.E., Naids, R. and Kalsow, C., S-Antigen in the developingretirta and pineal gland: a monoclonal antibody study, Invest. Ophthalmol. Vis. Sci., 26 (1985) 561-567.
152 7 Kerr, H.-W., Foster, R.G., Ekstr6m, P. and Schalken, J.J., Opsin-like immunoreaction in the retinae and pineal organs of four mammalian species, Cell Tissue Res., 242 (1985) 045-648. 8 Lcm, J., Applebury, M.L., Falk, J.D., Flannery, J.G. and Simon, M.I., Tissue-specific and developmental regulation of rod opsin chimeric genes in transgenic mice, Neuron, 6 (1991) 201-210. 9 Reiter, R.J., The mammalian pineal gland: structure and function, Am. J. Anat., 162 (1981) 287-313. 10 Somers, R.L. and Klein, D.C., Rhodopsin kinase activity in the mammalian pineal gland and other tissues, Science, 226 (1984) 182-184. I i Taketani, S, Nakahashi, Y., Osumi, T. and Tokunaga, R., Molecular cloning, sequencing and expression of mouse ferrochelatase, J. Biol. Chem., 265 (1990) 19377-19380. 12 Taketani, S., inazawa, J., Nakahashi, Y., Abe, I". and Tokunaga, R., Structure of the human ferrochelatase gene: exon/intron gene organization and location of the gene to chromosome 18, Fur. J. Biochem., 205 (1992) 217-.9.22. 13 Vandenberg, D.J., Wuenschell, C.W., Mori, N. and Anderson,
D.J., Chromatin structure as a molecular marker of cell li:~eage and developmental potential in neural crest-derived chromaffin cells, Neuron, 3 (1989) 507-518. 14 van Veen, T., Ostholm, T., Gierschik, P., Spiegel, A., Somers, R., Korf, H.W. and Klein, D.C., a-Transducin immunoreactivity in retinae and sensory pineal organs of adult vertebrates, Prec. Natl. Acad. Sci. USA, 83 (1986) 912-916. 15 van Veen, T., Katial, A., Shinohara, T., Barrett, D.J., Wiggert, B., Chader, G.J., and Nickerson, J.M., Retinal photoreceptor neurons and pinealocytes accumulate mRNA for inter-photoreceptor retinoid binding protein (IRBP), FEBS Lett., 208 (1986) 133-137. lb Vigh-Teichman, 1. and Vigh, B., Electron microscopic localization of immunoreactive opsin in the pineal organ. In P.J. O'Brien and D.C. Klein (Eds.), Pineal and Retina Relationships, Academic Press, Orlando, FL, 1986, pp. 401-413. 17 Zack, DJ., Bennett, J., Wang, Y., Davenport, C., Klaunberg, B., Gearhart, J. and Nathans, J., Unusual topography of bovine rhodopsin promoter-lacZ fusion gene expression in transgenic mouse retinas, Neuron, 6 (1991) 187-199.
149
BRESD 51512
Research Reports
A PCR analysis of rhodopsin gene transcription in rat pineal photoreceptor differentiation Masasuke Araki
a
a n d Shigeru T a k e t a n i
b
a Department of Anatomy, Jichi Medical School, Tochigi 329-04 (Japan) and b Department of Hygiene, Kansai Medical Unirersity, Osaka 570 (Japan) (Accepted 2 June 1992)
Key words: Pineal; Rhodopsin; PCR; Norepinephrine; Culture; Rat
A polymerase chain reaction (PCR) method was used to detect rhodopsin transcripts in rat pineals both in vivo and in vitro. Only very low levels of transcripts were detected in preparations from adult rat pineal tissue, but fairly large amounts were detected in eDNA preparations from cultures of newborn rat pineals. The transcript level was reduced significantly if the cells had been cultured in the presence of 10 ~tM norepinephrine (NE). This concentration of NE had previously been shown to abolish almost completely the rhodopsin immunoreactivity normally seen in such cultures. The present work indicates that part, but not all, of the effect of NE is probably at the level of translation,
INTRODUCTION
The mammalian pineal is considered to be an exclusively endocrine organ which synthesizes and secretes melatonin with a rhythm produced by the suprachiasmatic nucleus and transmitted to the pineal via the superior cervical ganglion9. Several reports, however, have indicated that some molecules characteristic of the retina are also found in the mammalian pineal; examples are S-antigen, rhodopsin kinase, inter-photoreceptor retinoid binding protein (IRBP) and a-transducin 6't°a4'ts. Immunohistochemical studies with antirhodopsin, however, have not revealed any clear-cut positive staining in the rat pineal 4'~6, suggesting the absence of photoreceptor function. Previously, however, we found that numerous rhodopsin-immunoreactive (Rho-l) cells differentiate when pineais from newborn rats are transferred to in vitro conditions 3. Rat pineals removed later than postnatal day 6, however, manifest only a very few Rho-I cells in culture, and norepi.t,~phrine (NE) irreversibly suppresses the differentiation of Rho-i cells in culture of newborn rat pineals 2. In an attempt to learn more about the molecular mechanisms involved in NE suppression of cell differ-
entiation, we have now analyzed the level of rhodopsin gene transcripts by the polymerase chain reaction (PCR) method. The very small sizc of the pineal allowed us to compare in vivo with in vitro transcript levels by this method. It was found that Rho-! cells do synthesize a fair amount of rhodopsin gene transcripts in vitro, and that the intact pineal organ of the mature rat contains only a few transcripts. MATERIALS A N D M E T H O D S
Preparation of tissues and cell culture Adult rats of the Wistar strain were deeply anesthetized with sodium pentobarbital. Retinas and pineals were isolated and spleen was taken as a control tissue. The isolated tissues were rinsed briefly with Hanks' saline solution and stored at -80°C until used. The procedure used for the preparation of pineal cultures has been reported in a previous paper 2. Briefly, pineals were isolated from newborn rats, dissected into small pieces and then treated with collagenase. They were cultured in collagen-coated dishes with Eagle's MEM supplemented with 5% fetal calf serum, 55 ~tg/ml sodium pyruvate and 6 mg/ml glucose. Pineals were cultured for 8 days with or without 10 ~M NE. Polymerase chain reaction Total RNA was isolated from approximately 7x 10s cells or from rat tissues using the guanidine thiocyanate procedure zl. cDNA was synthesized from 2 ~tg of total cellular RNA by Super Script RNAse H reverse transcriptase (Gibco BRL, Gaithersberg, MD), using
Correspondence: M. Araki, Department of Anatomy, Jichi Medical School, Tochigi 329-04, Japan. Fax: (81) (285) 44-8169.
150 oligo(dT) 12-18. A (I.05 aliquot of the eDNA product was used for amplification by PCR. The two oligonucleotides prepared for use as PCR amplificathm primers were: 5'-AAGCCGATGAGCAACTTCC-Y, corresponding to nucleotides 515-533 of mouse opsin eDNA; and 5'-TCATCTCCCAGTGGA'fTCTI'-3', complementary to nucleotides i,067-!,086 of that eDNAs. 25 cycles of PCR amplification were c,,rried out, As a= ~aternal control for the amount of PCR reaction product, amplific :tion of ferro-:helatase eDNA, the last-step enzyme of heme biosy~faesis, ',','as carried out by PCR under similar conditions as described above. Tl:e primers used were 5'-TTGATG'VI'AAACATGGGAGG-3', corresponding to nucleotides 211-231 of mouse fcrre~:helatase eDNA; and 5'-TTGGACTGCCAACCAGTCGGTAY, c-',,nplementary to nucleotides 880-901 of the eDNA it.
A
B
:4.3 2.3 :2.0
0.56
Sol~dwrn blot analysis The probe used was a synthetic oligonucleotide, 5'-ATGGTTATCATCATGGTCA-3', corresponding to positions 851-869 of mouse opsin eDNA. It was labeled at the 5' end with [y-32P]ATP (Amersham, 3,000 Ci/mmol), using T4 polynuck:otide kinase. The products of PCR amplification were electrophoresed in a 1.2~ agarose gel and transferred to a nylon membrane (Amersham, Hybond N ÷ ). The filter was prehybridized for 3 h at 48°C in a solution containing 5 x Denhaldt's solution (0.1% bovine serum albumin, 0.1% polyvinyl pyrrolidone, and 0. 1% Ficoll), 5 × SSC (1 × SSC = 0.15 M NaCI, 0.15 M sodium citrate, pH 7.0), 20 mM sodium phosphate (pH 7.0), 0.1% SDS and 40 p,g/ml sonicated salmon sperm DNA. Hybridization with the radiolabelled oligonucleotide (I × 10f' cpm/ml) was carried out at 48°C for 16 h. Stringent washing was performed at 48°C with three changes of 4×SSC containing (I.l~ SDS. The filter was exposed to Kodak X-ray film overnight. S,'cthern blot analy,,.is of ferrochelatase eDNA was performed, using the full-length fragmew, of mouse ferrochelatase eDNA as a probe t i. The intensity of the positive bands was estimated by scanning using an Advantec model DMU-33C densitometer.
RESULT~ it was anticipated that the PCR products amplified with the present primers would correspond to nu. cleotide positions 515-1,086, or approximately 560 base pairs, Fig, IA shows the elcctrophorctic patter, of the PCR products. In the PCR products from the eDNA isolated from neural retinal tissue, there is a major band at the position of mouse opsin eDNA (lane 2). In control experiments using rat spleen eDNA (lane 1) no clear band can be seen at this position. A faint band can be seen in the products from the eDNA from cultured pineal cells, whether or not the cells were cultured in the presence (lane 3) or absence (lane 4) of NE. No band at this position can be clearly seen in the electrophoretic pattern yielded by products from adult pineal tissue eDNA (lane 5). Probe hybridization of the amplified eDNA (Fig. I B) gave a clearer demonstration of the results, In neural retinal tissues a very thick band was found at the position of mouse opsin eDNA (lane 2), Fainter but still obvious bands were seen in the preparations from both normal (lane 4) and NE-treated (lane 3) pineal cultures, Densitometric measurements indicated that the band found in normal cultures was about
Fig. 1. Electrophoretic pattern (A) and probe hybridization analysis (B) of the PCR products. Synthesis of the eDNA samples, amplification by PCR, electrophoresis on 1.2% agarose gel, transfer to a nylon membrane and probe hybridization were as described in Materials and Methods. The sources of the eDNA were: lane I, spleen; lane 2, neural retina; lane 3, pineal cells cultured in the presence of 10/~M norepinephrine; lane 4, cultured pineal cells; and lane 5, adult pineal tissue. Lane 6 contains Hindill-digested fragments of lambda phage used as size markers.
three-times as intense as that found with NE-treated cultures. A very faint band could also be seen on probe hybridization in the preparation from adult rat pineal tissue (lane 5). The control preparation from spleen tissue still showed no trace of a band at this position (lane 1), To assess the quantity of the PCR products, eDNA of ferrochelatase, which is the last-step enzyme of heine biosynthesis, and is known to be a house-keeping gone 12, was amplified with the same materials as used in opsin gene amplification. No significant difference was noticed in the intensity of the bands obtained by probe hybridization of amplified ferrochelatase eDNA products from adult pineal tissue, normal pineal culture and NE-treated pineal culture (Fig. 2). Data shown in this study were reproducible with two independent preparations of these materials. DISCUSSION The present study clearly demonstrates the suitability of the PCR method for describing specific gene expression at the transcriptional level during cell differentiation. This method is particularly useful when only small amounts of tissue or cultured cells are available. The results show that cultured rat pineals contain opsin mRNA in small but still obvious amounts, supporting the supposition that the rhodopsin-immunoreactive (Rho-l) cells found in pineal cultures
151
Fig. 2. Southern blot analysis of the PCR products with the ferrochelatase eDNA. Synthesisof the cDNA samples, amplificationby PCR, electrophoresis and transfer to a nylon membranewere similar to those as described in the legend for Fig. 1, except that two oligonucleotides of the ferrochelatase eDNA were used as primers for PCR. Hybridization with the fragment of the ferrochelatase eDNA was performed as described. The sources of the cDNA were the same as shown in the legendfor Fig. 1.
synthesize rhodopsin molecules. It also appeared here that intact pineal tissue from adult rats contains only trace amounts of opsin gene transcripts. When newborn rat pineals are cultured in vitro, numerous Rho-I cells differentiate, but such differentiation is largely suppressed by the presence of 1 # M norepinephrine '~. In the presence of 10 ~M, as used in the present work, the number of Rho-I cells which differentiate is only a few percent of the number found in control cultures. In this previous work, it also appeared that NE caused degeneration of Rho-I cells, although some still seemed normal 2. in contrast to the very few Rho-I cells seen by immunohistochemistry, the rough quantitative estimation used on the hybridization bands (Fig. 1B) suggests that the level of opsin gene transcripts found in cultures treated with 10 # M NE is still about one-third of that in control cultures, although it is very difficult to quantitate the PCR products by reasons deriving from the enormous amplification by the method. Taken together, these results indicate that NE probably affects Rho-I cell differentiation to some extent at the translational level without inducing degeneration of all such cells. In situ hybridization studies may yield more information on the extent of cell degeneration vs. simple inhibition of translation.
In the present study it was shown that adult rat pineals have opsin mRNA although in such very small amounts that it can be detected only after amplification by PCR. Immunohistochemically, no clear-cut positive staining was found in rat pineal 4,~6, although Korf et al. 7 reported such staining in rat and m o u s e 7. No phenotypic evidence was obtained for transgene expression in the pineal, or in any other tissue except the retina, when chimeric gene fusion of rod opsin was used to generate transgenic mice 8. In another study on transgenic mice, it was found that c/s-acting DNA elements required for rhodopsin synthesis are expressed in the retina but not in the pineal ~7. Analogy with some other systems also suggests that such very low levels of opsin gene transcripts as found here in the intact rat pineal do not necessarily mean that synthesis of the protein will occur. For example, lowlevel transcription of the 8-crystallin genes has also been observed in some non-lens tissues of chick embryos without detectable protein synthesis ~. Similarly, low-level transcription of neuron-specific genes is found in adrenal medulla chromaffin cells, but this may only reflect their possible trans-differentiation into neurons under certain conditions ~3. The results of the present study therefore support the view that some neonatal pineal cells have the capacity for rhodopsin synthesis but this is almost completely lost or suppressed in the adult tissue, and the presence of NE is probably an important factor leading to such suppression. Acknowledgements. The authors are deeply grateful to Dr. E.G.
McGeer (The Universityof British Columbia) for her careful reading of the manuscript. This work was supported partly by a Grant-in-Aid from the Ministryof Education, Scienceand Culture, Japan and by a grant provided by the lchiro Kanehara Foundation. REFERENCES 1 Agata, K., Yasuda, K. and Okada, T.S., Gene coding for a lens-specific protein, 8-crystallin, is transcribed in non-lens tissues of chick embryos, Dec'. Biol., 100 (1983) 222-226. 2 Araki, M., Ceikdar mechanismfor norepinephrine suppressionof pineal photoreceptor..like cell differentiation in rat pinc~dlcultures, Deu. Biol., 149 (1992)440-447. 3 Araki, M. and Tokunaga, F., Norepinephrinc suppresses bo~h photoreceptor and neuron.like properties expressed by cultured rat pineal glands, Cell Differ. Dev., 31 (1990) 129-135. 4 Araki, M., Watanabe, K., Tokunaga, F. and Nonaka, T., Phenotypic expression of photoreceptor and endocrine properties by cultured pineal cells of the newborn rat, Cell Differ. Def., 25 (1988) 155-164. 5 Baehr, W., Falk, J.D., Bugra, K., Triantafyllos, J.T. and McGinhis, J.F., Isolation and analysisof the mouse opsin gene, FEBS Lett., 238 (1988) 253-256. 6 Donoso, L.A., Merryman, C.F., Edelberg, H.E., Naids, R. and Kalsow, C., S-Antigen in the developingretirta and pineal gland: a monoclonal antibody study, Invest. Ophthalmol. Vis. Sci., 26 (1985) 561-567.
152 7 Kerr, H.-W., Foster, R.G., Ekstr6m, P. and Schalken, J.J., Opsin-like immunoreaction in the retinae and pineal organs of four mammalian species, Cell Tissue Res., 242 (1985) 045-648. 8 Lcm, J., Applebury, M.L., Falk, J.D., Flannery, J.G. and Simon, M.I., Tissue-specific and developmental regulation of rod opsin chimeric genes in transgenic mice, Neuron, 6 (1991) 201-210. 9 Reiter, R.J., The mammalian pineal gland: structure and function, Am. J. Anat., 162 (1981) 287-313. 10 Somers, R.L. and Klein, D.C., Rhodopsin kinase activity in the mammalian pineal gland and other tissues, Science, 226 (1984) 182-184. I i Taketani, S, Nakahashi, Y., Osumi, T. and Tokunaga, R., Molecular cloning, sequencing and expression of mouse ferrochelatase, J. Biol. Chem., 265 (1990) 19377-19380. 12 Taketani, S., inazawa, J., Nakahashi, Y., Abe, I". and Tokunaga, R., Structure of the human ferrochelatase gene: exon/intron gene organization and location of the gene to chromosome 18, Fur. J. Biochem., 205 (1992) 217-.9.22. 13 Vandenberg, D.J., Wuenschell, C.W., Mori, N. and Anderson,
D.J., Chromatin structure as a molecular marker of cell li:~eage and developmental potential in neural crest-derived chromaffin cells, Neuron, 3 (1989) 507-518. 14 van Veen, T., Ostholm, T., Gierschik, P., Spiegel, A., Somers, R., Korf, H.W. and Klein, D.C., a-Transducin immunoreactivity in retinae and sensory pineal organs of adult vertebrates, Prec. Natl. Acad. Sci. USA, 83 (1986) 912-916. 15 van Veen, T., Katial, A., Shinohara, T., Barrett, D.J., Wiggert, B., Chader, G.J., and Nickerson, J.M., Retinal photoreceptor neurons and pinealocytes accumulate mRNA for inter-photoreceptor retinoid binding protein (IRBP), FEBS Lett., 208 (1986) 133-137. lb Vigh-Teichman, 1. and Vigh, B., Electron microscopic localization of immunoreactive opsin in the pineal organ. In P.J. O'Brien and D.C. Klein (Eds.), Pineal and Retina Relationships, Academic Press, Orlando, FL, 1986, pp. 401-413. 17 Zack, DJ., Bennett, J., Wang, Y., Davenport, C., Klaunberg, B., Gearhart, J. and Nathans, J., Unusual topography of bovine rhodopsin promoter-lacZ fusion gene expression in transgenic mouse retinas, Neuron, 6 (1991) 187-199.