NT Agonist Regulates Expression of Nuclear High-affinity Neurotensin Receptors

Electron Microscopy and Tissue Preparation

Sprague-Dawley rats (n = 3) were maintained on a 12-hr light/dark cycle and fed ad libitum. Rats were deeply anesthetized with pentobarbital (70 mg/kg) and perfused transaortically with 30 ml heparin (100 U/ml), followed sequentially by a mixture of 50 ml of 3.75% acrolein and 2% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4 (PB), and by 200 ml of 2% paraformaldehyde in PB. The brains were fixed for 30 min in 2% paraformaldehyde in PB. Coronal sections (40 μm) were cut on a vibratome and collected in PB. Immunolabeling was performed by a procedure previously described (Chan et al. 1990; Boudin et al. 1998). Ultrathin sections for gold staining were collected from the surface of blocks of the substantia nigra, counterstained with lead citrate and uranyl acetate, and examined with a JEOL 100CX electron microscope.

Electron Microscopy and Quantitative Analysis

We analyzed three rat brains. For each rat, two to four sections from different blocks were systematically scanned. Microscopic fields with cell bodies were photographed at an original magnification of ×8300. All silver-enhanced gold granules were classified according to the type of element and the ultrastructure with which they were associated. For perikarya, they were classified as plasma membrane, cytosol (labeling apparently not associated with any organelle), Golgi apparatus/endoplasmic reticulum, vesicle, mitochondria, nuclear envelope, or nuclear soma. We measured the surface of the nucleus and cytoplasm of each cell body on the film negatives using a computer-based image analysis system associated with a CCD camera and the software Historag (Biocom; Les Ulis, France). We analyzed a total of 264 micrographs representing 75 cell bodies and 1119 gold particles.

Cell Culture

Human lung cancer cells (LNM35) were grown in RPMI 1640 (Invitrogen; Cergy Pontoise, France). CHO cells were grown in MEM α medium without ribonucleosides and deoxyribonucleosides (Invitrogen). All media were supplemented with 10% fetal calf serum and 2 mM l-glutamine (Invitrogen). For stably transfected CHO cells, 250 μg/ml of geneticin (G418) was added to maintain the transgene. Cells were incubated at 37C in a humidified 5% CO₂ atmosphere. At confluence, cells were routinely dispersed in trypsin-EDTA 0.25% and subcultured at 1:10 dilution.

Stable Cell Line Establishment

The expression plasmid pNT1-EGFP containing the rat NT-1 receptor coding region and the modified form of EGFP from Aequorea victoria, directed by the CMV promoter, was provided by Dr. Llorens-Cortes (Lenkei et al. 2000). pNT1-EGFP was stably transfected in CHO cells as described by Boudin et al. (1995). Stable transfectants were selected with G418 (1000 μg/ml) and colonies were screened for EGFP expression, using a Nikon Diaphot inverted fluorescence microscope at a ×40 magnification, and by [¹²⁵I]-NT-binding assays (Boudin et al. 1995).

Isolation of Membrane-depleted Nuclei

Purified nuclei were prepared using a modification of the procedure previously described (Facy et al. 1990). Cells were washed twice in cold PBS and collected by low-speed centrifugation (300 × g for 7 min). The pellet was washed in 20 ml TKCM buffer containing 10 mM Tris-HCl (pH 7.3), 20 mM KCl, 2 mM CaCl₂, 2 mM MgCl₂, 0.2 mM spermidine, and collected by centrifugation. The pellet was resuspended in 20 ml TKCM buffer and incubated on ice for 1 hr. Cells were homogenized twice with a Potter-Elvehjem tissue grinder in the presence of a mixture of protease inhibitors for mammalian cell and tissue extracts (Sigma-Aldrich; St-Quentin, France), and then centrifuged at 300 × g for 7 min. The cell suspension was layered onto 2 M sucrose, 50 mM Tris-HCl, pH 7.4, and centrifuged at 25,000 × g for 30 min. Nuclei were collected from the bottom of the tubes, and cell membrane homogenates were collected at the interface.

Chromatin Preparation

Nuclei were resuspended in 1 ml of 1 mM Tris-HCl, pH 7.9. The nuclear suspension was incubated on ice for 1 hr before homogenization. The chromatin was separated from the nuclear membranes by centrifugation at 12,000 × g for 30 min.

Control of the Purity of Membrane-depleted Nuclei

CHO-NT1 cells were labeled with a mix containing 7.5 μCi/ml of l-[³⁵S]-methionine and l-[³⁵S]-cysteine (Amersham Biosciences; Orsay, France) for 16 hr. Membrane homogenates and nuclei were prepared as described above and the radioactivity of each fraction was counted. In parallel, we prepared an unlabeled cell suspension. Before the ultracen-trifugation step, we added either 3 × 10⁶ cpm of labeled nuclei or 7 × 10⁷ cpm of labeled membrane. The mixture was layered onto a sucrose layer and centrifuged as described above. The radioactivity was counted in the membrane homogenates and in the nuclear fraction from both samples.

Enzymatic Markers

The purity of the nuclear fraction was assessed by assaying selected enzymatic marker activity for each cell fraction. The activity of 5' nucleotidase was performed using a Sigma kit based on the method developed by Arkesteijn (1976). The ouabain-sensitive Na⁺-K⁺-ATPase was assessed by the method of Gache et al. (1976). Ouabain-sensitive Na⁺-K⁺-ATPase activity was defined as the activity inhibited by 1 mM ouabain. The K⁺-EGTA ATPase activity was measured as an enzymatic marker of the endoplasmic reticular membranes (Gache et al. 1976).

Binding Studies

Nuclei, cell membrane homogenates, and chromatin were washed with 50 mM Tris-HCl, pH 7.4, and were immediately used for [¹²⁵I]-NT-binding and enzymatic assays. Protein contents were estimated by the method of Bradford using bovine serum albumin as standard (Bradford 1976). Radioligand binding studies were carried out either on cell membrane homogenates or on nuclei. Binding studies were performed as follows: 40 μg of protein from cell membrane homogenates or 80 μg of protein from nuclei was incubated with 0.5 nM [¹²⁵I]-NT in a final volume of 250 μl of buffer A (50 mM Tris, pH 7.4, 0.2% BSA, and 0.8 mM 1.10-phenantroline, 1.6 mM MgCl₂) for 45 min at room temperature (RT). Nonspecific binding was measured in the presence of 1 μM unlabeled NT. The competition binding was carried out under the same conditions, using a range of unlabeled NT concentrations (10^–6 to 10^–11 M). Binding assays were terminated by addition of ice-cold 50 mM Tris-HCl (pH 7.4) supplemented with 0.2% BSA, followed by filtration through glass microfiber filters (GF/B; Whatman, Maidstone, UK) preincubated in 0.2% polyethylenimine (Sigma-Aldrich). After washing three times with 5 ml ice-cold buffer, the radioactivity retained on the filters was counted in a γ-counter (Wallac model 1470 Wizard). Binding affinities and densities (K_d and B_max) were estimated by the Ligand EBDA program.

Immunofluorescent Staining of Nuclei for NT-1 Receptor

Fifty μl of purified nuclei was sedimented onto slides with a Cytospin centrifuge. Nuclei were fixed in 2% paraformaldehyde for 30 min. Fixed nuclei were washed three times with PBS at RT before proceeding to immunocytochemistry. Nonspecific binding was blocked with 0.1% Triton X-100/PBS containing 10% normal rabbit serum (NRS) for 1 hr at RT. Nuclei were then washed three times with PBS and incubated for 1 hr at RT with a goat polyclonal anti-NT-1 receptor antibody (Santa Cruz Biotechnology; Santa Cruz, CA) diluted 1:25 in PBS containing 0.1% Triton X-100 and 3% NRS (buffer A). Nuclei were then rinsed three times with PBS and incubated for 45 min at RT with rabbit anti-goat-Texas Red secondary antibodies (Jackson ImmunoResearch Laboratories; West Grove, PA) diluted 1:300 in buffer A. After three final rinses in PBS, slides were mounted in Glycergel (DAKO-Cytomation; Trappes, France). Appropriate controls devoid of NT-1 receptor antibody were performed in parallel to determine nonspecific staining.

RNA Extraction and Reverse Transcription-polymerase Chain Reaction (RT-PCR)

Total RNA was extracted from LNM35 cells using the acidic phenol/chloroform guanidine thiocyanate method (Chirgwin et al. 1979). One hundred ng of total LNM35 RNA was reverse-transcribed in a 30-μl reaction mixture containing 20 mM Tris-HCl, pH 8.3, 50 mM KCl, 5 mM MgCl₂, 10 mM dithiothreitol, 1 mM each of dNTP, 24 U RNasin (Promega; Charbonnière, France), 1 μg/μl of each oligo (dN) and oligo (dT) (Amersham Biosciences; Orsay, France) for NT transcript, or primer RT-NT-1 (5'-GCTGACGTAGAAGAG-3') for NT-1 receptor transcript, and 200 U Moloney murine leukemia virus reverse transcriptase (Invitrogen; Cergy Pontoise, France) at 37C for 1 hr. The PCR amplification was performed on 1:5 (v/v) dilution of the RT reaction in a mixture containing 16 mM Tris-HCl, pH 8.3, 40 mM KCl, 1 mM MgCl₂, 0.2 mM of each dNTP, 25 pmol of the sense primer, 25 pmol of the antisense primer and 1 U Taq polymerase (Applied Biosystems; Les Ulis, France) in a final volume of 50 μl. The PCR primers had the following sequences: S-NT-1 5'-CGTGGAGCTGTACAACTTCA-3' and AS-NT-1 5'-CAGCCAGCAGACCACAAAGG-3' for NT-1 receptor; S-NT 5'-AAGCACATGTTCCCTCTT-3' and AS-NT 5'-CATACAGCTGCCGTTTCAGA-3' for NT (Invitrogen). The amplicon sizes of NT-1 receptor, and NT were 590 and 446 nucleotides, respectively. The amplification profile was divided into 35 cycles of denaturation at 94C for 30 sec, annealing at 57C for 45 sec, and extension at 72C for 45 sec. PCR products were electrophoresed on 1% agarose gels in 90 mM Tris borate and 2 mM EDTA buffer. We routinely introduced a 100 bp DNA ladder (Invitrogen) as a size marker. Gels were stained with ethidium bromide and photographed under a UV lamp.

Confocal Microscopy

Confocal microscope analysis was carried out using the TCS SP Leica microscope (Lasertechnik) equipped with a ×63 objective (plan apo; NA 1.4). A focal series was collected for each specimen every 0.5 μm for nuclei from the CHO-NT1-EGFP cell line. Each confocal image shown here corresponded to the middle of seven optical serial sections. For each optical section, double fluorescence was simultaneously acquired using a krypton-argon mixed-gas laser adjusted to 488 nm for GFP and to 568 nm for TRITC. The variable center of spectrophotometers was adjusted to recover green (500–550 nm) and red (580–630 nm) fluorescence. The signal was treated with line averaging to integrate the signal collected over four lines in order to reduce signal noise. Selected paired sections were then processed to produce a single overlay image (color merged) using a PC computer equipped with Photoshop software (version 6.0) (Adobe; Tucson, AZ). To quantify the intensity of nuclear labeling, Leica TCS-NT software was used. For each experiment, the middle optical sections of 10–15 nuclei from control cells and 10–15 nuclei from treated cells were studied with the same confocal acquisition parameters. An ellipse was drawn around each nucleus to select the region to analyze and the activated pixel intensity was measured. The results provided include the area and the mean intensity of the selected region.

Nuclear Distribution of NT-1 Receptor Immunoreactivity in Neuron Cell Bodies of Rat Substantia Nigra

Substantia nigra (SN) contains a dense network of NT-containing axon terminals interspersed among dopaminergic neurons, which highly express NT-1 receptor (Boudin et al. 1998). In SN neurons, NT-1 receptor immunoreactivity is mostly intracellular and is primarily located on dendrites, axons, and axon terminals. NT-1 receptor immunoreactivity found on nerve cell bodies represents 10.6% of total labeling (Boudin et al. 1998). Using electron microscopy, we examined the detailed distribution of NT-1 receptor immunoreactivity in SN cell bodies (Figure 1) and observed that 30% of the gold particle grains were associated with the nuclear compartment. In the nucleus, 87 ± 4.4% of the gold grains were found inside of the nucleus, whereas 13 ± 4.6% were distributed on the nuclear envelope. The significant proportion of NT-1 receptors inside the nucleus led us to study additional cell models to determine if the intracellular localization was related to the presence of the agonist in the extracellular medium.

Purity and Characterization of the Subcellular Fractions Isolated from CHO-NT1 Cell Line

Several detection procedures were employed to ensure purity of the subcellular fractions isolated from the CHO-NT1 cell line. First, unlabeled cell suspensions were purposely contaminated with labeled nuclei or labeled membrane homogenates purified from cell cultures treated with l-[³⁵S]-methionine and l-[³⁵S]-cysteine, as described in the experimental procedures. This technique revealed a slight contamination (4%) of the nuclear fraction by cytosolic and plasma membranes (Table 1). Our approach to prepare nuclei has been to adopt a gentle homogenization method and to avoid detergents. This technique resulted in a large loss of nuclei but permitted the subsequent use of binding assays. Equally, the sucrose concentration employed in the cell suspension fractionation determined the quality and the quantity of purified nuclei. To quantitatively diminish the membrane contamination within the nuclear fraction, we increased the sucrose molarity, resulting in a low quantity of sedimented nuclei.

OPEN IN VIEWER

Figure 1

Electron micrograph showing the immunoreactivity of immunogold-labeled NT-1 receptor in neuron cell bodies of the substantia nigra. Gold particles were seen in association with the nuclear envelope (NE) (arrows) and were also present inside the nucleus (N2) (arrowheads). Some immunogold particles were located on the plasma membrane (PM). Others were located in the cytoplasm in association with mitochondria (M), and others in vesicles (V). N1, nucleolus. Bar = 2.3 μm.

The purity of the nuclear fraction was also assessed by assaying the enzymatic activity for selected markers in the cellular fractions. Table 2 shows the activity of marker enzymes in nuclei prepared from CHO-NT1 cells. Activity of the ouabain-sensitive Na⁺-K⁺-ATP-ase and K⁺-EGTA ATPase was undetectable in the nuclear fraction. A second marker for the plasma membrane, 5'-nucleotidase, revealed an average 6% contamination of the nuclei by plasma membrane components. Taken together, we concluded that the nuclear preparations were only marginally contaminated with 5% of plasma membrane, based on the average of the two procedures.

OPEN IN VIEWER

Table 1

Estimation of contamination between subcellular fractions

	Presence of radioactive nuclei (%)	Presence of radioactive membranes (%)
Membrane homogenates	17	42
Nuclear fraction	14	4

OPEN IN VIEWER

Table 2

Specific enzymatic activities in membrane homogenates and nuclear fraction a

	Membrane homogenates	Nuclear fraction
Enzyme activities	(μmoles product formed/mg protein/hr)
K+-EGTA ATPase	2.89 ± 1.63	ND
Na+-K+-ATPase (ouabain-sensitive)	2.40 ± 0.90	ND
5′-Nucleotidase (UI)	9.27 ± 3.02	0.67 ± 0.33

^aND, not detected. Results are means ± SEM of three separate experiments.

NT-1 Receptors and Neurotensin-binding Sites Are Found in Purified Nuclei from Cell Lines Overexpressing NT-1 Receptor

Cytoplasmic membranes obtained from the CHO-NT1 cell line were previously shown to exhibit a Kd of 0.44 ± 0.01 nM and a B_max of 1960 fmol/mg protein (Boudin et al. 1995). [¹²⁵I]-NT binding analyses carried out on cell membrane homogenates, purified nuclei, and chromatin from CHO-NT1 cells exhibited a high specific binding of 98 ± 0.6%, 94.6 ± 1.5%, and 74.6 ± 4.2%, respectively. Scatchard analyses were performed on cell membrane homogenates and purified nuclei (Figure 2A). Calculations from Scatchard plots showed a lower affinity in the purified nuclei fraction than in the cell membrane homogenates, with a Kd of 1.2 ± 0.3 nM vs 0.3 ± 0.2 nM, respectively. The Hill coefficients from nuclear and cell membrane fractions were 1 ± 0.1 and 1.1 ± 0.3, respectively, indicating a single population of binding sites. The binding capacity of the cell membrane homogenates fraction was 3808 ±788 fmol/mg protein. Surprisingly, the B_max of purified nuclei appeared to be very high, 1836 ± 406 fmol/mg protein. However, binding experiments performed in the presence or absence of Gpp(NH)p showed that the nuclear NT-1 receptor population was not sensitive to Gpp(NH)p and therefore was not coupled to G-protein (Figure 2B). In contrast, the cell membrane fraction is coupled to G-protein because it is sensitive to Gpp(NH)p (Figure 2B). Significant specific binding was also present in the chromatin fraction, but the amount of recovered material did not permit further analysis (data not shown).

To visualize localization of the NT-1 receptor in the nuclei, we used a CHO-NT1-EGFP cell line expressing the rat NT-1 receptor fused with the EGFP (Lenkei et al. 2000). Purified nuclei were exposed to confocal microscopy and an image gallery was sequentially acquired. Similarly to what was detected by immunocytochemistry in rat SN, Figure 3 illustrates that the NT-1 receptor-EGFP was found at the nuclear membrane and in the nuclear soma. NT-1 receptor-EGFP was detected at the center of the nucleus, slices 2, 3, and 4.

OPEN IN VIEWER

Figure 2

[¹²⁵I]-NT-binding parameters of CHO-NT1 subcellular fractions. (A) Scatchard analysis of specific [¹²⁵I]-NT binding to nuclear fraction (∘) and cell membrane homogenates (•) prepared from CHO-NT1 cells. (B) [¹²⁵I]-NT binding assays to cell membrane homogenates and nuclear fraction from CHO-NT1 cells performed in the presence and absence of 1 μM Gpp(NH)p. Results are the means ± SEM of three separate experiments.

Effects of NT Agonist Treatment on NT-1 Receptor Nuclear Localization

We hypothesized that the presence of the NT-1 receptor in SN cell nuclear bodies was related to the dense NT innervations of this brain area. We therefore investigated the role of an NT agonist, JMV 449, on the NT-1 receptor nuclear localization in CHO-NT1-EGFP cells treated with 1 μM JMV 449 for 6 hr. Nuclei from treated and untreated CHO-NT1-EGFP cells were labeled for NT-1 receptor immunoreactivity and visualized by confocal microscopy. First, we confirmed that the NT-1 receptor is still fused to EGFP by immunocytochemistry by using an antibody directed against the N-terminal portion of the NT-1 receptor. As shown in Figure 4A, (above), an overlap between EGFP labeling (green) and NT-1 receptor immunoreactivity (red) was observed. These results confirm the presence of the NT-1 receptor inside the nucleus and in the nuclear envelope. In nuclei from cells treated with the agonist, we observed that both EGFP (green) and NT-1 receptor (red) labelings were always more intense compared to control nuclei (Figure 4A). Quantification of EGFP intensity was performed on the central slice after each image acquired by confocal microscopy. Figure 4B shows that, after JMV 449 treatment, the EGFP (green) intensity is increased by twofold, resulting in an accumulation of NT-1 receptors both in nuclear soma and in the nuclear envelope.

OPEN IN VIEWER

Figure 3

Image gallery from confocal microscopy of purified nuclei isolated from CHO-NT1-EGFP cells. A focal series of seven optical sections was collected for each nucleus every 0.5 μm. Bar = 1.7 μm.

NT-binding Sites from Purified Nuclei of Cells Endogenously Expressing NT and NT-1 Receptor

The result observed in CHO-NT1-EGFP cells confirmed the hypothesis that NT-1 receptor nuclear localization was the consequence of NT stimulation. We then investigated localization of the NT-1 receptor in a cell line having a native NT autocrine regulation. The human lung cancer cell line LNM35 exhibits high expression of both NT and the NT-1 receptor, as suggested by the PCR amplicon intensity for NT and NT-1 receptor transcripts (Figure 5A). Binding studies performed on LNM35 crude membrane extracts exhibited an 80 ± 4.8% specific binding for [¹²⁵I]-NT. Scatchard analysis showed a single population of sites (Figure 5B). The Kd value and the Hill coefficient (0.22 ± 0.07 nM and 1.04 ± 0.06 nM, respectively) were similar to those obtained for cloned and endogenous human high affinity NT-1 receptors (Vita et al. 1993; Najimi et al. 1998). The NT binding capacity found in LNM35 crude membrane extracts was 98.3 ± 23.4 fmol/mg protein. To identify the presence of the nuclear NT-1 receptor in LNM35 cells, subcellular fractionation was carried out as described in the experimental procedures. [¹²⁵I]-NT binding analyses were performed on cell membrane homogenates, purified nuclei, and chromatin. Specific [¹²⁵I]-NT-binding was detected in cell membrane homogenates (80.6 ± 4%), in the nuclear fraction (32.4 ± 2.4%), and in chromatin (72.7 ± 3.5%), indicating the presence of NT-1 receptor-specific sites in these cell fractions. Unfortunately, the recovery of purified nuclei from LNM35 was very poor, and the level of NT nuclear sites was low (16,210 ± 3690 cpm/mg protein). Therefore, proper Scatchard analyses could not be performed on purified nuclei. Nevertheless, to further characterize the NT-1 receptor localized in LNM35 purified nuclei, binding experiments were done in the presence or absence of Gpp(NH)p. As shown in Figure 5C, the specific binding was affected by the non-hydrolysis analogues of GTP in the nuclear fraction, indicating that the nuclear sites were coupled to G-protein. The cell membrane homogenate fraction was also sensitive to Gpp(NH)p. The specific [¹²⁵I]-NT binding decreased by threefold in comparison to control (Figure 5C). These results argue for the presence of G-protein-coupled NT-1 receptor in purified LNM35 nuclei. To evaluate the relationship between NT transmission and the NT-1 receptor localization in the nuclear compartment, LMN35 cells were treated for 48 hr with 10^–5 M SR 48692, an NT-1 receptor antagonist (Gully et al. 1993). By blocking the neurotensinergic transmission, nuclear expression of the NT-1 receptor strongly declined in the LNM35 cells (Figure 5D). Indeed, when cells were treated with SR 48692, the NT-binding capacity in the purified nuclei and in the cell membrane homogenates was considerably decreased, by 72% and 90%, compared to their respective controls (Figure 5D).

OPEN IN VIEWER

Figure 4

Effect of an NT agonist on the nuclear NT-1 receptor. CHO-NT1-EGFP cells were treated or not with 1 μM JMV 449 at 37C for 6 hr and subcellular fractions were prepared. (A) Nuclei were fixed with paraformaldehyde, permeabilized with 0.1% Triton X-100, and incubated with goat polyclonal anti-NT-1 receptor, followed by Texas Red-conjugated rabbit anti-goat to detect NT-1 receptors as described in Materials and Methods. Images were acquired by confocal microscopy and each picture corresponds to the central section of a serial xy optical section of 0.5 μm. The NT-1 receptor column shows the NT-1 receptor immunocytochemistry and its recovered red fluorescence. The third column displays the overlay from the confocal images of nuclei from control or treated cells recovered in green and red; co-localization is identified as yellow. The fourth panel represents an xz optical section perpendicular to the xy optical section. Bar = 1.7 μm. (B) Quantitative analysis of the nuclear labeling in nuclei from CHO-NT1-EGFP cells treated or not with JMV 449 for 6 hr. 4p<0.01 vs control.