Pergamon
PII:
Neuroscience Vol. 79, No. 3, pp. 855–862, 1997 Copyright ? 1997 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/97 $17.00+0.00 S0306-4522(97)00034-1
INDUCIBLE EXPRESSION OF N-METHYL-D-ASPARTATE RECEPTOR, AND DELTA AND MU OPIOID RECEPTOR MESSENGER RNAS AND PROTEIN IN THE NT2-N HUMAN CELL LINE I. W. BECZKOWSKA, K. N. GRACY, V. M. PICKEL and C. E. INTURRISI* Departments of Pharmacology, Neurobiology and Neuroscience, Cornell University Medical College, 1300 York Ave, New York, NY 10021, U.S.A. Abstract––Retinoic acid treatment of NT-era2/cl.D1 (NT2) cells, a human teratocarcinoma cell line, yields 95% pure cultures of terminally differentiated neuronal cells. Concomitant with their terminal differentiation into neurons, NT2 cells are induced by retinoic acid to express neuronal N-methyl--aspartate receptor channels, which are fully functional. We determined the effects of retinoic acid-induced differentiation of NT2 cells on the levels of N-methyl--aspartate, delta opioid and mu opioid receptor messenger RNAs. RNA levels were measured using quantitative solution hybridization assays. The riboprobes were complementary to major portions of the coding regions of the N-methyl--aspartate, delta opioid and mu opioid receptor complementary DNAs. After four weeks of exposure to 10 µM retinoic acid, followed by four weeks of treatment with mitotic inhibitors (1 µM of cytosine arabinoside, 10 µM of fluorodeoxyuridine and 10 µM of uridine) the levels of N-methyl--aspartate receptor messenger RNA in differentiated NT2-N cells increased 10-fold, delta opioid receptor messenger RNA increased three-fold, and mu opioid receptor messenger RNA increased four-fold. Northern blot analysis revealed two transcripts for the N-methyl--aspartate receptor messenger RNA (4.2 and 4.4 kb) and two transcripts for delta opioid receptor messenger RNA (7.0 and 11.0 kb). To determine whether the increases in messenger RNAs were accompanied by an increased synthesis of the respective proteins, we examined the immunoperoxidase localization of N-methyl--aspartate receptor and delta opioid receptor antisera. N-Methyl--aspartate receptor-like immunoreactivity was seen within the cell bodies as well as on the processes of the retinoic acid-differentiated cells. Although delta opioid receptor-like immunoreactivity was detected within the soma of isolated cells prior to retinoic acid treatment, the apparent number of these labelled cells and their ramified processes were markedly enhanced following retinoic acid differentiation. These results demonstrate parallels between the inducible expression of the N-methyl--aspartate and opioid receptor messenger RNAs and proteins during the acquisition of the fully differentiated neuronal phenotype in cultured NT2 cells. Retinoic acid-differentiated NT2 cells express increased levels for the N-methyl--aspartate, delta opioid and mu opioid receptor messenger RNAs, providing the opportunity to study the interactions among these receptor systems in human terminally differentiated neuronal cells in culture. ? 1997 IBRO. Published by Elsevier Science Ltd. Key words: NT2-N cells, delta opioid receptor, mu opioid receptor, N-methyl--aspartate receptor, all-trans retinoic acid, solution hybridization.
All-trans-retinoic acid (RA) treatment of NTera2/ cl.D1 (NT2) cells, a human teratocarcinoma cell line, yields 95% pure cultures of terminally differentiated neuronal cells, designated NT2-N.24 NT2-N cells morphologically resemble primary neuronal cultures from rodents and, like primary neurons, have processes that differentiate into axons and dendrites.2 In addition, they express a variety of neuronal markers including many neuronal cytoskeletal proteins (e.g., MAP1A, MAP1B, MAP2 and tau), secretory *To whom correspondence should be addressed. Abbreviations: DOR, delta opioid receptor; EDTA, ethylenediaminetetra-acetate; LI, -like immunoreactivity; MOR, mu opioid receptor; NMDA, N-methyl-aspartate; NMDAR, NMDA receptor; NT2, Nt-era2/ cl.D1; PBS, phosphate-buffered saline; RA, retinoic acid; SSC, standard saline citrate; TCA, trichloroacetic acid.
markers (e.g., synaptophysin and chromagranin) and surface markers (e.g., neural cell adhesion molecule and growth-associated protein-43).24 NT2 cells are induced by RA to express neuronal N-methyl-aspartate (NMDA) and non-NMDA glutamate receptors (NMDARs) concomitant with their differentiation into neuron-like cells.32 In addition to susceptibility to glutamate-induced neurotoxicity, the molecular and physiological characteristics of these human glutamate receptors are nearly identical to those of central nervous system neurons, as demonstrated by the polymerase chain reaction and patchclamp recordings.32 RA, in addition to inducing morphological differentiation and changes in ion channels, can produce changes in receptors,1,33 second messenger systems,11,25,29 and NMDAR1 and delta opioid receptor (DOR) mRNA levels in a
855
856
I. W. Beczkowska et al.
neuroblastoma cell line.3 It is not known whether undifferentiated NT2 cells express DOR and mu opioid receptor (MOR) mRNAs, or how RA treatment, resulting in terminal differentiation into human neurons, might affect the expression of these mRNAs. Therefore, to address these questions we determined the levels of DOR, NMDAR1 and MOR mRNAs in undifferentiated NT2 cells and in RAdifferentiated NT2-N cells. In addition, we determined the number and size of the mRNA transcripts for the NMDAR1 and DOR receptors. The localization of DOR and NMDA receptors on RAdifferentiated NT2-N cells was examined by use of immunocytochemistry to further establish whether the heightened expression of the mRNAs was accompanied by an increase in proteins, and if so, the distribution of this labelling in soma and processes. EXPERIMENTAL PROCEDURES
Cell culture NT2 cells were maintained in Dulbecco’s modified Eagle’s medium High Glucose supplemented with 10% fetal calf serum, 100 units/ml penicillin, 100 µg/ml streptomycin and non-essential amino acids. They were grown in monolayer at 37)C, 5% CO2 and 95% humidity. The protocol of Pleasure et al.24 was utilized to induce cell differentiation. In short, semiconfluent cells were treated with 10 µM RA (dissolved in dimethylsulphoxide) twice a week for four weeks. Following RA treatment, the cells were replated 1:6 and two days later mechanically dislodged and replated in fresh Petri dishes. Following the second replate, the cells were maintained in medium supplemented with 1 µM cytosine arabinoside for the first week, and medium supplemented with 1 µM of cytosine arabinoside, 10 µM fluorodeoxyuridine and 10 µM uridine for the first four weeks. RNA extraction Following differentiation, the cells were washed twice with phosphate-buffered saline (PBS) and detached with 0.05% trypsin with 0.02% EDTA. This suspension was then collected in sterile tubes, and centrifuged at 2000 r.p.m. for 5 min. The pellet was resuspended in 2 ml of an RNA extraction buffer containing guanidinium isothiocynate and homogenized.17 Total cellular RNA was extracted with phenol and precipitated with ethanol using a procedure that results in a recovery of RNA that averages 77&7.2% (S.D.).15 20 µm of glycogen (Boehringer–Mannheim, GmbH, Germany) was used during the extraction procedure as a carrier. Poly(A)+ RNA was prepared from 1 mg of total RNA and 0.1 mg of oligo dT cellulose (Sigma) as described by Sambrook.26 Riboprobes A 1058 base pair PstI-SacI DNA fragment containing all of the coding region of the DOR cDNA9 was cloned into a PstI and SacI restricted pGEM-3zf+ plasmid (Promega, Madison, WI) which contained T7 and SP6 promoters.17 A 1074 base 32P-labelled antisense riboprobe was prepared by use of a T7 polymerase transcription system, and an unlabelled 1082 base DOR RNA sense standard was prepared using an SP6 polymerase transcription system.17 The 32Plabelled 18S rRNA riboprobe was prepared by SP6 polymerase transcription of plasmid PS/E.34 The 1413 base pair 32P-labelled NMDAR1 riboprobe, prepared from a plasmid [pGEM-7zf(+)] containing the cDNA sequence for a functional NMDA receptor subunit, NMDAR1, is complementary to a region of the rat
NMDAR1 which codes for one of the two putative glutamate binding sites and three of the four transmembrane regions (TMI, TMII and TMIII) including the TMII sequence that is reported to be part of the ion channel.10 This riboprobe was prepared using an SP6 transcription system and an unlabelled 4272 base (full-length) NMDAR1 sense standard, of which 4213 bases derived from the NMDAR1 cDNA sequence were prepared using an SP6 polymerase transcription system.10 A 32P-labelled riboprobe complementary to the MOR-1 mRNA was transcribed from a pSP73 plasmid containing the HindIII fragment of the rat MOR-1 cDNA.5,6 This fragment includes all of the open reading frame and small regions of the 5* and 3* untranslated sequences of 213 and 38 bases, respectively. The 1388 base unlabelled sense transcript prepared using an SP6 polymerase transcription system served as the standard for MOR-1 mRNA quantitation.5 Solution hybridization The solution hybridization assays are based on the protection of a 32P-labelled riboprobe from ribonuclease digestion when it is hybridized to cRNAs. Riboprobes protected from ribonuclease (RNase) activity are precipitated with trichloroacetic acid (TCA) and quantitated by scintillation counting.10,17,34 For each assay, duplicate aliquots of total RNA extracts were dried in 1.5 ml Eppendorf tubes and resuspended in 30 µl of hybridization buffer (10 mM EDTA, 0.3 M sodium chloride (NaCl) and 0.5% sodium dodecyl sulphate, 10 mM N-Tris[hydroxymethyl]methyl-2-amino-ethane sulphonic acid, pH 7.4) containing a 32P-labelled riboprobe (150,000 d.p.m./30 µl for the DOR, NMDAR1 or MOR riboprobes and 80,000 d.p.m./30 µl for the 18S riboprobe). The hybridization solutions were then covered with two drops of mineral oil and incubated at 75)C for 4 h. Following hybridization 300 µl of digestion buffer (0.3 M NaCl, 5 mM EDTA, 40 µg/ml RNase A and 2 µg/ml RNase T1 in 10 mM Tris–HCl, pH 7.4) was added to each sample. Samples were vortexed, centrifuged and unprotected single strands of RNA were digested at 30)C for 1 h. The ribonuclease reaction was stopped with 1 ml of 5% TCA and 0.75% sodium pyrophosphate, and two drops of 0.5% bovine serum albumin were added to aid precipitation. This solution was mixed gently and the TCA-precipitated radioactive RNA (hybridization signal) was collected onto microfibre filter paper using a 24-place cell harvester (Brandel, Gaithersburg, MD). The filter was washed three times with 5% TCA, dried under an infra-red light and counted by liquid scintillation in 5 ml Hydrofluour scintillation solution (National Diagnostics, Manville, NJ). Comparison was made with a standard calibration curve to convert the d.p.m.s of the hybridization signal to the amount of DOR, MOR and NMDAR1, or total cellular RNA. Since both the DOR and NMDAR1 for NT2-N cells contain multiple transcripts, the mRNA values given below represent equivalents of the correspondence sense transcript used in the calibration curve. Northern blot analysis Samples of Poly(A)+ RNA from NT2-N cells were denatured in 1 M glyoxal/50% (v/v) dimethylsulphoxide for 60 min at 50)C, fractionated by horizontal electrophoresis at room temperature in agarose gel containing 0.01 M sodium phosphate at pH 7.1 (with buffer recirculation), transferred to nitrocellulose filter in the presence of 20#standard saline citrate (1#SSC is 0.15 M NaCl and 0.015 M sodium citrate), and baked in a vacuum oven at 80)C for 4 h.15 Filters were washed three times in 1#SSC at room temperature for 1 h and air-dried before hybridization. Following hybridization, the filters were washed for 1 h in 0.1#SSC at 65)C and then exposed to Kodak X-AR film for three to five days at "70)C.17 The size of each
Retinoic acid induces NMDA, delta and mu receptors
857
Table 1. The increase in N-methyl--aspartate receptor 1, delta opioid receptor and mu opioid receptor messenger RNAs in differentiated NT2-N cells induced by treatment with retinoic acid mRNA NMDAR1 DOR MOR
Untreated1 (pg/µg total RNA)
Retinoic acid1 (pg/µg total RNA)
0.20 (0.01) 0.21 (0.04) 0.08 (0.01)
2.10 (0.02)2 0.60 (0.01)2 0.32 (0.01)2
Triplicate samples of NT2 cells were exposed for four weeks to 10 µM retinoic acid, followed by four weeks of treatment with mitotic inhibitors as described above. The untreated cells served as a control. mRNA levels were measured by solution hybridization. 1 The data are the mean (&S.E.M.) transcripts from RNA extracts obtained from two experiments. 2 Indicates a significant difference (P<0.05) compared to untreated cells.
hybridization band was estimated by use of an unlabelled RNA marker. The developed films were then scanned with a SILVER SCANNERII using the Adobe Photoshop software and the image was printed on photographic paper. The presence of multiple bands for both mRNAs was confirmed by densitometry analysis of the computer scans using the IP Lab Gel software. Immunocytochemistry The cells were processed for immunoperoxidase localization of either polyclonal DOR antiserum (p34) which was prepared against a peptide consisting of amino acids 34 to 47 of the cloned DOR7 or a commercially available (Chemicon International) polyclonal NMDAR1 antiserum corresponding to the C-terminus of the rat NMDA receptor subunit. Following RA treatment, NT2-N cells were plated in 1 ml chamber slides and treated with mitotic inhibitors for four weeks as described above. The undifferentiated NT2 cells were plated in 1 ml chamber slides and allowed to grow for 48 h before fixation. The cells were washed with 0.1 M PBS, fixed with 3% acrolein (Polysciences) and 2% paraformaldehyde for 5 min, and incubated for 48 h at 4)C with normal primary or preadsorbed control antiserum against p34 for DOR and a no antiserum control for NMDAR1 receptor. The concentration of the antibody for detection of NMDAR1 receptor ranged from 1:500 to 1:2000, and for detection of DOR ranged from 1:500 to 1:1000. After the removal of the antisera, the NT2 and NT2-N cells were washed and placed for 30 min each in (i) for DOR: biotinylated goat anti-guinea-pig, and for NMDAR1: biotinylated goat anti-rabbit IgG (1:400, Amersham) in 0.1% bovine serum albumin, and (ii) peroxidase–avidin complex, in order to identify the bound immunoglobins. Peroxidase avidin–biotin complex (ABC, Elite Kit) for detection of primary antisera was obtained from Vector Laboratories, Burlingame, CA. The peroxidase was visualized by a 6 min reaction with 3,3*diaminobenzidine (0.22 mg/ml) and 0.01% hydrogen peroxide. All incubations and washes between each step were carried out in 0.1 M Tris–saline at room temperature. Statistics Data were analysed using a one-way analysis of variance with subsequent pair comparisons made using the Student Newman–Keuls test at the 0.05 level of significance. RESULTS
After four weeks of exposure to a 10 µM dose of RA followed by treatment with mitotic inhibitors (1 µM of cytosine arabinoside, 10 µM of fluorodeoxyuridine and 10 µM of uridine), there was a significant increase in the level of each of the three
Fig. 1. Northern blot analysis of the N-methyl--aspartate (NMDAR1) mRNA from retinoic acid (RA)-differentiated NT2-N cells. Poly(A)+ RNA from the RA-differentiated NT2-N cells (5 µg in lane a) and rat liver RNA extract (20 µg, lane b) were separated on a 1% agarose gel, transferred onto a nitrocellulose filter and hybridized with the NMDAR1 riboprobe. The sizes of the NMDAR1 hybridization bands were estimated by use of unlabelled RNA markers. ‘‘0’’ represents the origin of the electrophoresis.
mRNAs of interest as measured by solution hybridization. A 10-fold increase in the level of NMDAR1 mRNA was the largest effect of RA-induced differentiation. This was followed by a four-fold increase in the level of MOR mRNA and a three-fold increase in the level of DOR mRNA (Table 1). A Northern blot analysis with the 32P-labelled NMDAR1 riboprobe (Fig. 1) revealed two hybridization bands (approximately 4.2 and 4.4 kb) that were present in the poly(A)+ fraction from NT2-N cells (lane a), but not present in rat liver (lane b). The densitometry analysis with the IP Lab Gel program confirmed that the most intense peak corresponded to the 4.4 kb band, compared to the less intense 4.2 kb band (ratio 14:1). Northern blot analysis with the 32P-labelled DOR riboprobe (Fig. 2) revealed two
858
I. W. Beczkowska et al.
were also extensively labelled using the DOR antiserum at a dilution of 1:500 (Fig. 3f). At this dilution, in the absence of RA, DOR-LI was detected in only a few isolated NT2 cells (Fig. 3d). Neither the differentiated nor the undifferentiated cells showed peroxidase labelling when the DOR antiserum was preadsorbed with the cognate peptide (Fig. 3e). DISCUSSION
Fig. 2. Northern blot analysis of the delta opioid receptor (DOR) mRNA from the retinoic acid (RA)-differentiated NT2-N cells. Poly(A)+ RNA from the RA-differentiated NT2-N cells (10 µg in lane a) and rat liver RNA extract (20 µg, lane b) were separated on the 1.2% agarose gel, transferred onto a nitrocellulose filter and hybridized with the DOR riboprobe. The sizes of the DOR hybridization bands were estimated by use of unlabelled RNA markers. ‘‘0’’ represents the origin of the electrophoresis.
hybridization bands, a diffuse band at 11.0 kb and the main band of approximately 7.0 kb that was present in the poly (A)+ extracts from NT2-N cells (lane a), but not present in the rat liver (lane b). Densitometry analysis revealed three peaks: a small peak corresponding to the diffuse 11.0 kb band, a large peak corresponding to the 7.0 kb band and a small third peak corresponding to a 28S RNA band (ratio 2:6:1). A Northern blot analysis with the 32 P-labelled MOR riboprobe was attempted, but it did not yield detectable signal, presumably due to the low abundance and the exceptionally large size (14 kb) of the MOR-1 mRNA.5 Morphological differentiation of NT2 cells treated with RA was observed in each experimental plate and was characterized by cell clustering and the appearance of numerous axon- and dendrite-like processes (Fig. 3). NMDAR1-like immunoreactivity (LI) was prominently seen in the RA-differentiated NT2-N cells, even with the lowest concentration of the NMDAR1 antibody (1:2000, Fig. 3c). This peroxidase immunoreactivity was not seen when differentiated cells were processed for immunocytochemistry with omission of the NMDAR1 antiserum (1:1000, Fig. 3b). Additionally, no immunoperoxidase labelling for the NMDAR1 receptor was detected in undifferentiated NT2 cells even with the highest concentration of the NMDAR1 antibody (1:500, Fig. 3a). RA-differentiated NT2-N cells and processes
These results demonstrate that a long duration of differentiation of NT2 teratocarcinoma cells with 10 µM of RA produced a dramatic increase in the level of the expression of NMDAR1, DOR and MOR mRNAs. These include the two transcripts for NMDAR1 mRNA: 4.2 and 4.4 kb, and the two transcripts for DOR mRNA: 7.0 and 11.0 kb. Additionally, we have shown that the increase in the levels of these mRNAs is accompanied by a dramatic morphological differentiation of the NT2 cells and enhanced immunolabelling for both NMDAR1 and DOR protein. Retinoic acid induces NT2 cell differentiation and the expression of N-methyl--aspartate receptor 1 mRNA and protein The dramatic increase in the level of NMDAR1 mRNA and the detection of positive NMDAR1-LI only in NT2-N cells is consistent with the induction of the expression of functional NMDAR1 receptors. These results support an earlier study showing an absence of functional NMDA receptors in NT2 cells prior to RA treatment, and the induction of expression of the functional neuronal NMDA and nonNMDA glutamate receptors concomitant with their differentiation into neuron-like cells.32 Based on the ultrastructural studies in several brain regions, we expect that the differentiated processes immunolabelled for NMDAR1 include both dendrites and axons.12,14,22 The labelling would also suggest that the functional receptor may be present on the plasma membrane of the cell soma. Alternatively, the intense somatic NMDAR1 immunolabelling may indicate sites of synthesis or membrane recycling. The present observation of multiple transcripts of NMDAR1 revealed by Northern blot analysis of NT2-N extracts probed with NMDAR1 riboprobe can be explained by the existence of multiple isoforms of the NMDA receptor generated by alternative splicing.30 These splice variants may exist across species, including the human.23 Northern blot analysis of the rat NMDAR1 mRNA suggests that splice variants NMDAR1-1a, -1b, -2a and -2b, ranging in size from 4.3 to 4.5 kb, co-migrate in a gel, as do splice variants NMDAR1-3a, -3b, -4a and -4b, ranging in size from 3.9 to 4.1 kb (M. Brodsky, personal communication). The size of the NMDA mRNA transcripts in NT2-N cell extracts are identical with human NMDA mRNA10 and may reflect the
Retinoic acid induces NMDA, delta and mu receptors
859
Fig. 3. Immunoperoxidase staining of the N-methyl--aspartate (NMDAR1) and delta opioid receptor (DOR) in NT2 and retinoic acid (RA)-differentiated NT2-N cells. NT2 and NT2-N cells were processed for immunoperoxidase localization of polyclonal NMDAR1 antiserum following no treatment (a), no antiserum (negative control, b) and differentiation with RA (c). Immunostaining was observed within the cytoplasm of both soma and processes in RA-differentiated NT2-N cells but not in undifferentiated NT2 cells. Note the absence of staining when antiserum is omitted (b). NT2 and NT2-N cells were processed for immunoperoxidase localization of the polyclonal DOR antiserum following no treatment (d), a condition in which the antiserum was adsorbed by a specific DOR peptide (e) and differentiation with RA (f). Immunostaining was observed within the cytoplasm but not on the processes of the NT2-N cells and weak immunostaining restricted to few cells was present in NT2 undifferentiated cells. Magnification #400.
presence of NMDAR1 and NMDAR2 splice variants in this cell line. Additional evidence for the existence of these splice variants in the NT2-N cells comes from Western blot analysis.21 Retinoic acid increases delta opioid receptor mRNA and immunoreactivity in differentiated NT2-N cells In contrast to the NMDAR1 system, there was a detectable level of DOR mRNA and DOR-LI in the NT2 cells prior to RA treatment. The increase in the level of DOR mRNA and immunolabelling for DOR seen following RA treatment in the present study most likely reflects the induction of a transcription factor and protein expression in the differentiated
NT2-N cells with neuronal properties. However, the fact that both DOR mRNA and protein were present in undifferentiated NT2 cells suggests the possibility that the increase in DOR expression (measured as pg/µg of total RNA) seen in the NT2-N cells might be due, at least in part, to selection by mitotic inhibitors. If this was the case, then only cells expressing DOR prior to differentiation would survive, resulting in a higher DOR signal per µg of total RNA. However, at the present time, we are unable to test this possibility. The apparent number of labelled cells and processes was dramatically increased following treatment with retinoic acid. This suggests that DOR may serve a function in signalling cellular changes in undifferentiated NT2 cells that lack the axon and
860
I. W. Beczkowska et al.
dendritic type processes characteristic of neurons. In contrast, the more extensive labelling of soma and processes in differentiated NT2-N cells indicates a receptor distribution and, presumably, a function comparable to that seen in neurons within the central nervous system. These are likely to include a prominent localization to axons and dendrites as shown by electron microscopy immunocytochemistry using the same DOR antiserum in spinal cord and limbic cortex.7,31 This localization would suggest a prominent involvement of DOR ligands in both the presynaptic release and postsynaptic responses of differentiated NT2-N cells. Additionally, however, we observed a far more prominent labelling of NT2-N soma than has been reported in sections of neural tissue.13 While this could indicate a potentially greater role for somatic receptor in NT2-N cells, it is most likely attributed to greater thickness of whole mounts versus sections of neurons. The NT2-N RNA extracts probed with the DOR riboprobe revealed two transcripts represented by a main band at 7 kb and a diffuse band at approximately 11 kb. A multiple band pattern in the NG108-15 neuroblastoma cells is thought to result from differential Poly(A+) addition sites.19 In this Northern blot analysis (Fig. 2), we observed some radioactive material in the size range of the 28S RNA that was not completely removed by stringent buffer washing. This material appeared in extracts of rat liver RNA as well as in the NT2-N Poly (A)+ extracts. However, since rat liver does not express the DOR gene and rat liver RNA does not protect the DOR probe in the solution hybridization assay, we did not consider this material to represent a true hybridization band. Also, it is possible that there was some carryover in the Poly(A)+ extracts that resulted in the presence of residual 28S RNA. It is not possible at this time to determine whether these transcripts represent human DOR mRNA. Northern blot analysis of DOR mRNA from SH-SY5Y human neuroblastoma cells indicated the presence of a transcript of 9 kb, but no signal was found in human brain samples, presumably due to low levels of the DOR mRNA transcripts, compared with SH-SY5Y cells.28 A Northern blot analysis of mouse and rat brain RNA extracts revealed a single ribonucleaseresistant hybridization band of 10 kb in length in mouse brain, while the rat brain contained a prominent 12 kb band and a diffuse 4.5 kb band.18 The size of the main DOR hybridization band in the NT2-N cells is closer in size to the postulated human DOR mRNA transcript than the size of DOR mRNA in rodents. Mu opioid receptor mRNA is increased by retinoic acid treatment RA-induced increase in the level of MOR mRNA in the present study was much lower than NMDAR1 and lower than DOR mRNA. The almost undetect-
able level of MOR mRNA in the undifferentiated NT2 cells and the insignificant increase in RAdifferentiated NT2-N cells may indicate it as a nonfunctional receptor. This exceptionally low level of MOR mRNA in the NT2-N cells and the large size of the MOR transcript might have contributed to the failure to define MOR transcripts by Northern blot analysis. It would be of value to attempt to establish the cellular localization of MOR protein expression in NT2-N cells following RA treatment, although the very low level of MOR mRNA suggests that the level of protein expression might be undetectable. Although it is clear that NT2-N cells have functional NMDA receptors,24,32 we have not yet determined whether the mu and delta opioid receptors induced in this cell line are functional in terms of ligand binding and signal transduction. Possible mechanisms of the N-methyl--aspartate receptor 1, mu and delta opioid receptor mRNA induction by retinoic acid in NT2-N cells The long duration of the exposure of NT2 cells to RA resulting in either induction of NMDAR1 mRNA or increase in the DOR or MOR mRNAs suggests that RA is affecting transcription by an indirect mechanism which requires activation of some gene or genes other than the DOR, MOR or NMDAR1 genes. The likely candidates would be the homeobox (Hox) genes. Tissue and cell culture studies have revealed that RA can induce various Hox gene expression and that this activation is time dependent (e.g., see Refs 4, 8 and 20). It has been shown that in the NT2 cell line RA sequentially activates genes of the human Hox 1, Hox 2, Hox 3 and Hox 5 clusters;27 however, the mechanisms by which RA activates these genes are not fully understood. In a hybrid neuroblastoma cell line, NG10815, six day treatment with 10 µM of RA resulted in an enhanced expression of both DOR and NMDAR1 mRNAs, suggesting that RA-activated Hox-target elements in the DOR gene may be present in this cell line. CONCLUSIONS
We have identified a cell line in which there is the concurrent induction of the expression of NMDAR1 receptors and opioid receptors during the acquisition of neuronal processes. The fact that RAdifferentiated NT2-N cells express functional NMDA receptors in addition to opioid receptors may allow these cells to be used in vitro to study the interactions among these three receptors.16 Acknowledgements—We thank Dr Samuel Pleasure for his generous gift of NT2 cells and Dr Jose L. Walewski for his assistance in densitometry analysis. This research was supported in part by NIDA Grant DA01457 and DA05130 (CEI), NIDA Training Grant DA07274 (IWB). CEI is a recipient of a Research Scientist Award from NIDA (DA00198).
Retinoic acid induces NMDA, delta and mu receptors
861
REFERENCES
1. 2. 3. 4. 5. 6. 7.
8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31.
Adem A., Mattsson M. E., Nordberg A. and Pahlman S. (1987) Muscarinic receptors in human SH-SY5Y neuroblastoma cell line: regulation by phorbol ester and retinoic acid-induced differentiation. Brain Res. 430, 235–242. Andrews P. W. (1984) Retinoic acid induces neuronal differentiation of a cloned human embryonal carcinoma cell line in vitro. Devl Biol. 103, 285–293. Beczkowska I. W., Buck J. and Inturrisi C. E. (1996) Retinoic acid-induced increase in delta-opioid receptor and N-methyl--aspartate receptor mRNA levels in neuroblastoma # glioma (NG108-15) cells. Brain Res. Bull. 39, 193–199. Breier G., Bucan M., Francke U., Colberg-Poley A. M. and Gruss P. (1986) Sequential expression of murine homeo box genes during F9 EC cell differentiation. Eur. molec. Biol. Org. J. 5, 2209–2215. Brodsky M., Elliott K., Hynansky A., Jenab S. and Inturrisi C. E. (1995) Quantitation of ä-opioid receptor (MOR-1) mRNA in selected regions of the rat CNS. NeuroReport 6, 725–729. Chen Y., Mestek A., Liu J., Hurley J. A. and Yu L. (1993) Molecular cloning and functional expression of a µ-opioid receptor from rat brain. Molec. Pharmac. 44, 8–12. Cheng P. Y., Svingos A. L., Wang H., Clarke C. L., Jenab S., Beczkowska I. W., Inturrisi C. E. and Pickel V. M. (1995) Ultrastructural immunolabeling shows prominent presynaptic vesicular localization of ä-opioid receptor within both enkephalin and non-enkephalin containing axon terminals in superficial layers of the rat cervical spinal cord. J. Neurosci. 15, 5976–5988. Colberg-Poley A. M., Voss S. D., Chowdury K. and Gruss P. (1985) Structural analysis of murine genes containing homeo box sequences and their expression in embryonal carcinoma cells. Nature 314, 713–718. Evans C. J., Keith D. E. Jr., Morrison H., Magendzo K. and Edwards R. H. (1992) Cloning of a delta opioid receptor by functional expression. Science 258, 1952–1955. Franklin S. O., Elliott K., Zhu Y. S., Wahlestedt C. and Inturrisi C. E. (1993) Quantitation of NMDA receptor (NMDAR1) mRNA levels in the adult and developing rat CNS. Molec. Brain Res. 19, 93–100. Girard P. R. and Kuo J. E. (1990) Protein kinase C and its 80-kilodalton substrate protein in neuroblastoma cell neurite outgrowth. J. Neurochem. 54, 300–306. Gracy K. N. and Pickel V. M. (1995) Comparative ultrastructural localization of the NMDAR1 glutamate receptor in the rat basolateral amygdala and bed nucleus of the stria terminalis. J. comp. Neurol. 362, 71–85. Hazum E., Chang K. J. and Cuatrecasas P. (1979) Opiate (enkephalin) receptors of neuroblastoma cells: occurrence in clusters on the cell surface. Science 206, 1077–1079. Huntley G. W., Vickers J. C., Janssen W., Brose N., Heinemann S. F. and Morrison J. H. (1994) Distribution and synaptic localization of immunocytochemically identified NMDA receptor subunit proteins in sensory-motor and visual cortices of monkey and human. J. Neurosci. 14, 3603–3619. Inturrisi C. E., Branch A. D., Robertson H. D., Howells R. D., Franklin S. O., Shapiro J. R., Calvano S. E. and Yoburn B. C. (1988) Glucocorticoid regulation of enkephalins in cultured rat adrenal medulla. Molec. Endocrinol. 2, 633–640. Inturrisi C. E. (1994) NMDA receptors, nitric oxide and opioid tolerance. Regul. Pept. 54, 129–130. Jenab S. and Inturrisi C. E. (1994) Ethanol and naloxone differentially upregulate delta opioid receptor gene expression in neuroblastoma hybrid (NG108-15) cells. Molec. Brain Res. 27, 95–102. Jenab S., Kest B., Franklin S. O. and Inturrisi C. E. (1995) Quantitative distribution of the delta opioid receptor mRNA in the mouse and rat CNS. Life Sci. 56, 2343–2355. Magendzo I., Merickel A., Keith D. Jr. and Evans C. J. (1993) Analysis of multiple RNA transcripts encoding a ä-opioid receptor. Soc. Neurosci. Abstr. 19, 72. Mavilio F., Simeone A., Boncinelli E. and Andrews P. W. (1988) Activation of four homeobox gene clusters in human embryonal carcinoma cells induced to differentiate by retinoic acid. Differentiation 37, 73–79. McGonigle P., Lu L., Wolfe B. B. and Munir M. (1995) Human NT2-N neurons express the NR1 and NR2B subunits of the NMDA receptor. Soc. Neurosci. Abstr. 2, 1836. Petralia R. S., Yokotani N. and Wenthold R. J. (1994) Light and electron microscope distribution of the NMDA receptor subunit NMDAR1 in the rat nervous system using a selective anti-peptide antibody. J. Neurosci. 14, 667–696. Plannels-Cases R., Sun W., Ferrer-Montiel A. V. and Montal M. (1993) Molecular cloning, functional expression, and pharmacological characterization of an N-methyl--aspartate receptor subunit from human brain. Proc. natn. Acad. Sci. U.S.A. 90, 5057–5061. Pleasure S. J., Page C. and Lee M.-Y. V. (1992) Pure, postmitotic, polarized human neurons derived from NT-era 2 cells provide a system for expressing exogenous proteins in terminally differentiated neurons. J. Neurosci. 12, 1802–1815. Ponzoni M. and Lanciotti M. (1990) Retinoic acid rapidly decreases phosphatidylinositol turnover during neuroblastoma cell differentiation. J. Neurochem. 54, 540–546. Sambrook J., Fritsch E. F. and Maniatis T. (1989) Molecular Cloning, A Laboratory Manual, pp. 7.26-7.29. Cold Spring Harbor Laboratory Press, New York. Scott M. P. (1992) Vertebrate homeobox gene nomenclature. Cell 71, 551–553. Simonin F., Befort K., Gaveriaux Ruff C., Matthes H., Nappey V., Lannes B., Micheletti G. and Kieffer B. (1994) The human ä-opioid receptor: genomic organization, cDNA cloning, functional expression, and distribution in human brain. Molec. Pharmac. 46, 1015–1021. Slack R. S. and Proulx P. (1990) Effects of retinoic acid and staurosporine on the protein kinase C activity and the morphology of two related human neuroblastoma cell lines. Biochim. biophys. Acta 1053, 89–96. Sugihara H., Moriyoshi K., Ishii T., Masu M. and Nakanishi S. (1992) Structure and properties of seven isoforms of the NMDA receptor generated by alternative splicing. Biochem. biophys. Res. Commun. 185, 826–832. Svingos A. L., Cheng P. Y., Clarke C. L. and Pickel V. M. (1995) Ultrastructural localization of delta opioid receptor and Met-enkephalin immunoreactivity in rat insular cortex. Brain Res. 700, 25–39.
862
I. W. Beczkowska et al.
32.
Younkin D. P., Tang C.-M., Hardy M., Reddy U. R., Shi Q.-Y., Pleasure S. J., Lee V. M.-Y. and Pleasure D. (1993) Inducible expression of neuronal glutamate receptor channels in the NT2 human cell line. Proc. natn. Acad. Sci. U.S.A. 90, 2174–2178. 33. Yu V. C. and Sadee W. (1988) Efficacy and tolerance of narcotic analgesics at the mu opioid receptor in differentiated neuroblastoma cells. J. Pharmac. exp. Ther. 245, 350–355. 34. Zhu Y. S., Branch A. D., Robertson H. D., Huang T. H., Franklin S. O. and Inturrisi C. E. (1992) Time course of enkephalin mRNA and peptides in cultured rat adrenal medulla. Molec. Brain Res. 12, 173–180. (Accepted 15 January 1997)