HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 18, 16011-16021, May 3, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/18/16011    most recent M105594200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Yoo, S. H. Articles by Shim, C. S. Search for Related Content PubMed PubMed Citation Articles by Yoo, S. H. Articles by Shim, C. S. Social Bookmarking                      What's this?

Localization of the Secretory Granule Marker Protein Chromogranin B in the Nucleus

POTENTIAL ROLE IN TRANSCRIPTION CONTROL^*

Seung Hyun Yoo, Soon Hee You, Moon Kyung Kang, Yang Hoon Huh, Choong Sik Lee, and Chan Seob Shim

From the National Creative Research Initiative Center for Secretory Granule Research, Korea Advanced Institute of Science and Technology, Yu Sung Gu, Dae Jeon, Korea 305-701

Received for publication, June 18, 2001, and in revised form, January 15, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Chromogranins A (CGA) and B (CGB) are two major Ca²⁺ storage proteins of the secretory granules of neuroendocrine cells. Nevertheless, we found in the present study that CGB was also localized in the nucleus. In immunogold electron microscopy using bovine adrenal medullary chromaffin cells, it was found that the number of CGB-labeled gold particles localized per µm² of the nucleus was equivalent to 20% that of CGB-labeled gold particles localized per µm² of the secretory granules. Considering that CGB is estimated to exist in the 0.1-0.2-mM range in the secretory granules of bovine chromaffin cells, 20% of these amounts to 20-40 µM. In addition, transfection of CGA and CGB into nonneuroendocrine COS-7 and NIH3T3 cells repeatedly indicated the nuclear localization of CGB in addition to its usual localization in the cytoplasm. Moreover, immunoblot and immunogold electron microscopy analyses of neuroendocrine PC12 cells also showed the existence of endogenous CGB in both the cytosol and the nucleus. Nuclear routing of CGB did not appear to depend entirely upon the nuclear localization signal as some of the nuclear localization signal mutant CGB were still targeted to the nucleus. In gene array assay, CGB was shown to either induce or suppress transcription of many genes including those of transcription factors. Of these we have analyzed eight genes, four induced (zinc finger protein, MEF2C, hCRP2, abLIM) and four suppressed (hcKrox, T3-receptor, troponin C, integrin) using the quantitative reverse transcription-PCR method and spectrophotometry to determine the transcription levels of each mRNA. CGB was shown to increase the transcription of zinc finger protein, MEF2C, hCRP2, and abLIM by 2.5-5-fold while suppressing that of hcKrox, T3-receptor, troponin C, and integrin by 60-75%. Given that MEF2C and hcKrox genes are transcription factors, these results pointed to the transcription control role of CGB in the nucleus.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The secretory granules of neuroendocrine cells are loaded with hormones, neurotransmitters, and ions such as Ca²⁺, Mg²⁺, and Zn²⁺ along with peptides and proteins of which chromogranins A and B are the most abundant (1-5). Chromogranins A and B are acidic proteins (1-5) with acidic residues constituting 25-30% of the amino acid residues (6-12), and this high content of negatively charged amino acid residues is thought to be responsible for the high capacity, low affinity Ca²⁺ binding property of chromogranins (13, 14), binding 32-93 mol of Ca²⁺/mol (14, 15).

The comparison of the amino acid sequences of CGA¹ (6-8) and CGB (9-12) shows little sequence homology except the two conserved regions, one near the N-terminal region bordered by two cysteine residues (residues 17-38 in bovine CGA and 16-37 in bovine CGB) and the other the C-terminal region (residues 409-431 in bovine CGA and 604-626 in bovine CGB). Despite the differences in amino acid sequences, chromogranins A and B and secretogranin II (also called chromogranin C) were shown to aggregate in an acidic pH and high calcium environment (16-20), the condition found in the trans-Golgi network. Nevertheless, there was a big difference in the pH- and Ca²⁺-dependent aggregation properties of these two proteins; the aggregation of CGB being at least two orders of magnitude more sensitive to Ca²⁺ than CGA (20). Moreover, unlike CGA, which dimerized at pH 7.5 and tetramerized at pH 5.5 (21, 22), purified CGB appeared to exist in a monomeric state (20).

We have shown previously that CGA and CGB, as well as most of the secretory vesicle matrix proteins, not only aggregated in the presence of Ca²⁺ at the intravesicular pH 5.5 but also bound to several integral membrane proteins of the secretory granule, including the IP₃R (23, 24). Some of the vesicle matrix proteins that failed to bind to the vesicle membrane were shown to bind instead to CGA, thus ensuring their interaction with the vesicle membrane (25). Hence, in view of the chromogranins' ability to interact with both the vesicle matrix proteins and the vesicle membrane, the roles of CGA and CGB in the selective aggregation and the sorting of potential vesicle matrix proteins to the secretory granules appear to be essential in secretory granule biogenesis (25, 26). Thus, chromogranins A and B have been suggested to play key roles in secretory granule biogenesis (5, 25, 26). It was indeed reported recently that CGA functions as an on/off switch for secretory granule biogenesis in PC12 cells (27). Using the antisense RNA technique and PC12 cells, Kim et al. (27) showed that the number of secretory granules formed is directly related to the amount of CGA expressed in PC12 cells. It was further shown that the secretory granule formation could be induced in nonneuroendocrine cells, which normally don't contain any secretory granules, by expressing CGA in these cells.

We have extended here the chromogranin study and found that CGB was also localized in the nucleus in addition to its usual presence in the secretory granules. Although other secretory granule resident proteins proenkephalin and corticotrophin-releasing hormone have also been found in the nucleus before (28, 29), this is the first time the secretory granule marker protein chromogranin is found in the nucleus, opening new possibilities for the role of chromogranin in the nucleus. One of the nuclear roles of CGB appears to be control of the transcription of many genes, including those for transcription factors.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Antibodies-- The polyclonal anti-rabbit CGA and CGB antibodies were raised against intact bovine CGA and recombinant CGB. The monoclonal antibodies for CGA and CGB were produced using the bovine adrenal medullary chromaffin granule lysates as the antigen. The monoclonal antibody for green fluorescent protein (GFP) was purchased from Santa Cruz Biotechnology. The antibodies for the ER marker protein calnexin and the nucleus marker protein histone-4 were obtained from Calbiochem and Upstate Biotechnology, respectively. The monoclonal hemagglutinin (HA) and His₆ antibodies were from Roche Molecular Biochemicals. The horseradish peroxidase-linked anti-rabbit antibody was from Amersham Biosciences.

Immunocytochemical Localization of CGA and CGB-- For the immunogold electron microscopic study of chromaffin cells, the tissue samples from bovine adrenal medulla were fixed for 2 h at 4 °C in PBS containing 0.1% glutaraldehyde, 4% paraformaldehyde, and 3.5% sucrose. After three washes in PBS, the tissues were postfixed with 1% osmium tetroxide on ice for 2 h, washed three times, and stained en block with 0.5% uranyl acetate, all in PBS. The tissues were then embedded in Epon 812 after dehydration in an ethanol series. Ultrathin sections were collected on Formvar/carbon-coated nickel grids, which were then floated on drops of freshly prepared 3% sodium metaperiodate (30) for 30 min. The immunogold labeling procedure was modified from Spector et al. (31) and the manufacturer's recommended protocol (British Biocell International). After etching and washing, the grids were placed on 50-µl droplets of solution A (phosphate saline solution, pH 8.2, containing 4% normal goat serum, 1% bovine serum albumin, 0.1% Tween 20, 0.1% sodium azide) for 30 min. Grids were then incubated for 2 h at room temperature in a humidified chamber on 50-µl droplets of the anti-rabbit CGA or CGB antibody appropriately diluted in solution B (solution A but with 1% normal goat serum) followed by rinses in solution B. The grids were reacted with the 10-nm gold-conjugated goat anti-rabbit IgG diluted in solution A. Controls for the specificity of the CGA and CGB immunogold labeling included 1) omitting the primary antibody and 2) replacing the primary antibody with the preimmune serum. After washes in PBS and deionized water, the grids were stained with uranyl acetate (7 min) and lead citrate (2 min) and were viewed with a Zeiss EM912 electron microscope.

For the immunogold EM study of PC12 cells, PC12 cells that had been grown on a culture dish were rinsed with PBS followed by fixation in PBS containing 0.1% glutaraldehyde, 4% paraformaldehyde and 3.5% sucrose for 1 h at 4 °C. The cells were then scraped from the culture dish and centrifuged to obtain the cell pellet that was later embedded in 1% agar in PBS. The cell blocks were then washed three times in PBS followed by postfixation with 1% osmium tetroxide on ice for 2 h. The remaining steps followed the procedure described above for the adrenal chromaffin cells.

Construction of Expression Vectors-- For the expression vector construction, the cDNAs for CGA and CGB were prepared by PCR using bovine cDNA as a template, and the PCR products containing the full coding sequences were subcloned into the EcoRI/XbaI site of pCI-neomycin mammalian expression vector (Promega), in which transcription of the cloned gene was under the direction of the constitutively active cytomegalovirus promoter. The PCR primers were designed to include the HA- and His₆-tags at the C-terminal ends of CGA and CGB, respectively. For chromogranin-GFP fusion proteins, the reading frames of CGA and CGB without the stop codons were subcloned into the BglII/SalI site of pd2EGFPN1 vector (CLONTECH). The PCR products were produced using the following oligodeoxynucleotides: 5'-primer, 5'-GCAGATCTGCCTGGAGCGAGCAGTCCA-3', and 3'-primer, 5'-GAGTACTCTCAGCCCCGCCGAAGCTCCTCCA-3', for the pd2CGA-EGFP construct, and 5'-primer, 5'-GCAGATCTGGACGAGCGAGGCCAT-3', and 3'-primer, 5'-GAGTACTCTTAGCCCCTTCGGGTACCACTGA-3', for the pd2CGB-EGFP construct. The circular plasmid cDNAs for transfection were prepared using a Qiagen maxi-preparation kit. Deletion mutants for CGB were constructed using GeneEditor in vitro site-directed mutagenesis system (Promega).

Cell Culture and Transient Transfection-- All culture reagents and powdered media were purchased from Invitrogen. COS-7, NIH3T3, and PC12 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum. Transient transfection was performed with 70-80% confluent cultures. The cells were transfected with the circular plasmid DNAs using LipofectAMINE-plus transfection reagent (Invitrogen). Briefly, the cells were plated at a density of 5 × 10⁵ cells per well (100-mm diameter) and were cultured for an additional 24 h. Four µg of plasmid DNA in 20 µl of LipofectAMINE plus reagent was mixed with 750 µl of OPTI-MEM I medium and incubated for 15 min at room temperature. In addition, 30 µl of LipofectAMINE reagent was mixed with 750 µl of OPTI-MEM I and incubated for 15 min. The mixture was then added into a culture plate containing 5 ml of OPTI-MEM I medium. The transfection was performed for 3 h at 37 °C. After transfection, the medium was replaced with fresh prewarmed culture medium and was further incubated for 72 h. In our culture condition, about 40-50% of COS-7 and 70-80% of NIH3T3 cells were transfected. The pCI-neomycin vector was used as an empty vector.

Immunofluorescent Labeling of the Cells-- For immunofluorescent labeling, COS-7 cells were plated onto a 4-well slide chamber (Lab-Tek, Nalge Nunc Inc.) and were cultured to 60-70% confluency. The cells were then transiently transfected with the CGA- or CGB-expression vector. Seventy-two h after transfection, the cells were washed three times with ice-cold PBS and were fixed with 3.7% paraformaldehyde in PBS, pH 7.4, for 10 min. The slides were then treated with permeabilization solution (0.1% Triton X-100 in PBS) for 5 min. After several washes with PBS, the cells were blocked with 3% bovine serum albumin in PBS for 1 h. The antibodies against His₆ (1:100) and HA (1:100) were applied, and the slides were incubated for an additional 1 h at room temperature. The cells were then differentially labeled with the fluorescein-conjugated anti-mouse IgG and the tetramethylrhodamine isothiocyanate-conjugated anti-rabbit IgG. Following several washes with PBS, the slides were mounted with a mounting medium. Immunofluorescence was examined using a Zeiss fluorescence microscope (Axiovert S100), and images were captured and processed with a MetaMorph image analyzer (Universal Imaging Co.).

Extraction of Cellular Proteins and Western Analysis-- To obtain the total cell lysates from the transfected cells, ~1-2 × 10⁹ cells were washed twice with ice-cold PBS and lysed in RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 5 mM EDTA, 1% Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, and 20 µg/ml aprotinin/leupeptin mix). Then the extracts were incubated for 20 min on ice, and the cell debris was removed by centrifugation at 22,000 × g for 10 min at 4 °C. To obtain the cytosolic and nuclear extracts, the harvested cells were lysed by buffer A (10 mM HEPES, pH 7.9, 10 mM KCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, and 0.5% Nonidet P-40). The lysates were then pelleted by centrifugation at 2,000 × g to separate the cytosolic supernatant and the nuclear pellet. The supernatant was used as the cytosolic proteins, but the nuclear pellet was washed twice with buffer A and was lysed in 50 µl of RIPA buffer. After incubation for 20 min on ice, the nuclear debris was removed by centrifugation at 22,000 × g for 10 min at 4 °C, and the supernatant was used as the nuclear proteins. With this method up to 0.5 mg of the cytosolic or nuclear proteins was obtained. The proteins (10-50 µg of each) were then resolved by SDS-PAGE, and the immunoblot was performed using an ECL detection system (Amersham Biosciences).

Analysis of mRNA Expression by the cDNA Array and Quantitative RT-PCR Methods-- Human neuroblastoma SK-N-AS cells (ATCC, CRL-2137) were grown to a density of 1 × 10⁶ cells per well (100-mm diameter) and were transfected with pdEGFP (control vector), pd2CGA-EGFP (CGA vector), or pd2CGB-EGFP (CGB vector). After transfection, the cells were cultured for an additional 36 h. The mRNAs were then extracted from the cells and were converted into fluorescent nucleotide analog (Cyanine 3- or Cyanine 5-dUTP)-labeled cDNAs. These cDNAs were pooled together and simultaneously hybridized to a glass cDNA microarray slide (MICROMAX Human cDNA Microarray System I (2400 genes)) from PerkinElmer Life Sciences. Analysis of the mRNA expression was performed by scanning the microarray slide with a Gene Pix 4000A scanner (Axon Instruments). To determine the relative amount of each mRNA expressed in control, CGA-, and CGB-transfected cells, each mRNA was first quantitatively converted to cDNA by quantitative RT-PCR method using the total RNA extracted and the Platinum Quantitative RT-PCR thermoscript one-step system from Invitrogen. Then, the RT-PCR products were separated on 2% agarose gels followed by excision of the RT-PCR product bands. The cDNA products in the gel slices were extracted using the GENECLEAN turbo kit from BIO 101 (Vista), and the amount of each product was determined by measuring the A₂₆₀ values using a Beckman spectrophotometer. A series of known amounts of DNA was also separated on agarose gels followed by elution of these DNAs from the gel slices. The A₂₆₀ readings of these DNAs were used as standards in estimating the amount of each RT-PCR product. The name of the genes, primer pairs, annealing temperatures, number of PCR cycles, and the RNA amounts used are shown in Table II.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In view of the abundant presence of CGA and CGB in the adrenal chromaffin cells, we investigated the possibility of the presence of endogenous CGB in the nucleus of bovine adrenal medullary chromaffin cells using immunogold electron microscopy (Fig. 1). As shown in Fig. 1A, the CGA-labeled gold particles were primarily localized in the secretory granules with some in the endoplasmic reticulum. But virtually no CGA-labeling gold particles were found in the mitochondria. In contrast, the CGB-labeled gold particles localized not only in the secretory granules but also in the nucleus (Fig. 1B). Like the result in Fig. 1A, the chromogranin B-labeled gold particles were not found in the mitochondria. In control experiments, omission of the primary antibody or preimmune treatment in place of the primary antibody almost completely eliminated the chromogranin-labeled gold particles (Fig. 1C)

View larger version (76K): [in this window] [in a new window]   Fig. 1.   Immunogold electron microscopy showing the localization of CGA and CGB in bovine adrenal medullary chromaffin cells. Bovine adrenal medullary chromaffin cells were immunolabeled for CGA (A) and CGB (B) (10 nm gold) with the affinity purified CGA and CGB antibodies, respectively. Identical experiments were carried out either in the absence of the primary antibody or with the preimmune serum in place of the primary antibody (C). The CGA-labeling gold particles are primarily localized in the secretory granules (SG) with some in the endoplasmic reticulum (rer) but not in the mitochondria (M) or nucleus (Nu) (A). However, the CGB-labeling gold particles are localized both in the secretory granules and the nucleus. Some of the CGB-labeling gold particles are shown in the endoplasmic reticulum but not in the mitochondria (B). In the control experiments without the primary antibody no gold particles were seen in the secretory granules, nucleus, or the mitochondria (C). Bar = 200 nm.

To further evaluate the relative abundance of CGB in the nucleus, we examined fifteen different EM images, which had been prepared from seven different tissue samples and counted the total number of CGB-labeled gold particles in the secretory granules, nucleus, and mitochondria (Table I). As shown in Table I, 1027 CGB-labeled gold particles were found in 6.42 µm² of the secretory granule area thus averaging 160 CGB-labeled gold particles per µm² of the secretory granule area. In the same EM images 636 CGB-labeled gold particles were found in 19.47 µm² of the nuclear area averaging 33 CGB-labeled gold particles per µm², whereas 12 gold particles were localized in 3.51 µm² of mitochondria averaging three gold particles per µm². In light of the fact that two-three gold particles were consistently found per µm² of adrenal chromaffin cells in the control EM images, the three CGB-labeled gold particles found per µm² of mitochondria are considered to result from nonspecific interactions.

Similar to the CGB-immunogold study, fifteen different images from five different tissue samples were also examined for the presence of CGA-labeled gold particles in the secretory granules, nucleus, and mitochondria (Table I). The CGA-labeled gold particles were found virtually in all the secretory granules, averaging 437 CGA-labeled gold particles per µm² of the secretory granule, whereas 2-3 gold particles each were found per µm² of the nucleus and of the mitochondria. Again, the two-three gold particles that were found per µm² of the nucleus or mitochondria are identical to the number of gold particles found in the absence of the primary antibody. This result is in contrast to that obtained with the CGB-labeled gold particles, which clearly demonstrated the presence of CGB in the nucleus.

To determine whether transfected CGB can be routed to the nucleus in nonneuroendocrine cells, we have constructed CGA and CGB expression vectors and transfected them into COS-7 cells. The bovine chromogranins A and B that were used in the present experiments are shown in Fig. 2A. Two conserved regions (near the N-terminal and the C-terminal regions) are indicated as dashed boxes in Fig. 2A. For the immunolabeling and immunoblotting procedures, we tagged CGA and CGB at the C-terminal ends either with HA and His₆, respectively, or with GFP. When chromogranins A and B were introduced into COS-7 cells, the expression of CGA and CGB in the cells was confirmed by immunoblot analysis using either the tagging peptide-specific or chromogranin-specific antibody, which indicated a normal expression of the transfected chromogranins in COS-7 cells (Fig. 2B). In the cells that had been transfected with the control vector (pCI-neo), no band was present.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Construction and expression of bovine CGA and CGB in transiently transfected COS-7 cells. A, schematic diagrams for bovine CGA and CGB used in the transfection analysis. For differential expression, CGA and CGB were tagged with HA (CGA-HA) and His6 (CGB-His) at the C-terminal ends, respectively. The location of highly conserved regions is indicated as gray boxes. Solid boxes at the N-terminal ends of CGA and CGB indicate the leader sequence (L). B, Western analysis of CGA and CGB in the COS-7 cell extracts. The total protein extracts resolved on an 8% SDS-gel were probed serially with the anti-HA and anti-CGA antibodies for CGA (left two panels) and with the anti-His and anti-CGB antibodies for CGB (right two panels). The extract from the pCI-neomycin vector-transfected cells was used as a control.

The nuclear localization of CGB was also evident when the CGB-GFP fusion protein was expressed in COS-7 cells (Fig. 3C). Transfection of the cells with GFP only indicated the expression of GFP throughout the cell with a bit brighter fluorescence in the nuclear area (Fig. 3A). The diffuse fluorescence indicates that GFP is localized both in the cytoplasm and the nucleus, and the brighter fluorescence in the nuclear area probably reflects a greater depth of view of the nuclear area when viewed with a fluorescence microscope. But the CGA-GFP expression was limited to punctate localization in the cytoplasm with no localization in the nucleus (Fig. 3B). The punctate localization of CGA-GFP suggests granular localization of CGA-GFP. Though nonneuroendocrine COS-7 cells do not contain secretory granules, it appears apparent that granular structures were found in the CGA-transfected cells, which is consistent with the published results that demonstrated the secretory granule formation in the CGA-transfected nonneuroendocrine cells (27). In contrast, the CGB-GFP expression was evident in both the cytoplasm and the nucleus (Fig. 3C). The cytoplasmic CGB-GFP fluorescence was shown in punctate structures, suggesting the localization of CGB-GFP in granular structures, whereas the nuclear fluorescence did not appear in punctate structures.

View larger version (36K): [in this window] [in a new window]   Fig. 3.   Localization of CGA-GFP and CGB-GFP in COS-7 cells. The C-terminally tagged chromogranin-GFP fusion proteins were expressed in COS-7 cells. The CTL-GFP (A) indicates transfection of GFP alone (control) while CGA-GFP (B) and CGB-GFP (C) indicate COS-7 cells transiently transfected with CGA- and CGB-GFP, respectively.

To determine the subcellular localization of transfected CGA and CGB in COS-7 and NIH3T3 cells, immunoblot analysis of the protein extracts of the chromogranin-transfected COS-7 and NIH3T3 cells was carried out (Fig. 4). As shown in Fig. 4A, CGB was detected in the nucleus of the CGB-transfected COS-7 cells, and its level was similar to that of the cytosol, but CGA was not detected in the nuclear extract of CGA-transfected COS-7 cells. Similarly, CGB was detected in the nuclear extract of the CGB-transfected NIH3T3 cells (Fig. 4B), but CGA was not detected in the nucleus of CGA-transfected NIH3T3 cells. The purity of the cytosolic and nuclear protein extracts was ensured by examining the existence of the ER marker protein calnexin in the cytosolic proteins and of the nucleus marker protein histone-4 in the nuclear proteins (Fig. 4, A and B).

View larger version (20K): [in this window] [in a new window]   Fig. 4.   Localization of transfected CGA and CGB in COS-7 and NIH3T3 cells. The presence of CGA and CGB in the protein extracts of total (T), cytosolic (C), and nuclear (N) fractions of COS-7 cells transfected with CGA-GFP and CGB-GFP (A), respectively, and of NIH3T3 cells transfected with CGA-His and CGB-His (B), respectively, was analyzed by immunoblotting using the monoclonal CGA and CGB antibodies. Separation of the cytosolic and nuclear proteins was also ensured by immunoblotting with the polyclonal nuclear marker protein histone-4 and ER marker protein calnexin antibodies. 50 µg of the protein extract per lane was loaded for CGA and CGB immunoblots, but 10 µg of the proteins was loaded for the calnexin and histone-4 immunoblots.

To determine whether the nuclear routing of CGB is due to the overexpression of CGB in the CGB-transfected cells, the expression levels of transfected CGA and CGB were determined by measuring the expression levels of the GFP, which had been tagged to both CGA and CGB, in the CGA-GFP- and CGB-GFP-transfected cells (Fig. 5A). As shown in Fig. 5A, the CGB-GFP expression levels were only one-third or less those of CGA-GFP in NIH3T3 and COS-7 cells. This indicated that the amount of transfected CGB in the chromogranin-transfected NIH3T3 and COS-7 cells is one-third or less that of transfected CGA. Therefore, the nuclear localization of CGB cannot be due to the overexpression of CGB in these cells. Even after taking the larger molecular size of CGB (71 kDa) compared with that of CGA (48 kDa) into consideration, it is obvious that the nuclear routing of CGB is not the result of overexpression of CGB.

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Expression levels of CGA-GFP and CGB-GFP and the separate localization of CGA and CGB in the cells cotransfected with CGA-HA and CGB-His. A, the presence of GFP in the total protein extracts (50 µg/lane) of NIH3T3 and COS-7 cells each transfected with CGA-GFP and CGB-GFP, respectively, was analyzed by immunoblotting using the monoclonal GFP antibody. B, the presence of CGA and CGB in the protein extracts (50 µg/lane) of total (T), cytosolic (C), and nuclear (N) fractions of COS-7 cells cotransfected with CGA-HA and CGB-His (CGA+CGB) was analyzed using the HA- (CGA-HA) and His- (CGB-His) specific antibodies.

To further examine whether the nuclear routing of CGB can still occur in the nonneuroendocrine cells transfected with both CGA and CGB, CGA-HA and CGB-His were cotransfected into COS-7 cells. The coexpression of CGA and CGB in the same cells was confirmed by labeling the expressed CGA with fluorescein isothiocyanate and CGB with tetramethylrhodamine isothiocyanate, respectively. The immunoblot analysis of the presence of CGA and CGB in the protein extracts of these cells showed the targeting of CGB both to the nucleus and to the cytoplasm and of CGA to the cytoplasm only (Fig. 5B). Identical results were also obtained with the CGA construct tagged with His and the CGB construct tagged with HA. This result indicated that a nuclear routing mechanism that carries CGB, not CGA, to the nucleus is in operation.

In view of the fact that PC12 cells contain endogenous CGB (32), we explored the possibility of detecting endogenous CGB in the nucleus of neuroendocrine PC12 cells by immunoblot analysis (Fig. 6). As shown in Fig. 6, endogenous CGB was detected in both the cytosol and the nucleus of these cells although the CGB level in the nucleus was approximately one-third to one-fourth that of cytoplasm. However, unlike CGB, which was detected both in the cytoplasm and in the nucleus of nontransfected PC12 cells, CGA was detected in the cytoplasm but not in the nucleus. The immunoblot analysis of the protein extracts from these cells with the antibodies for the nucleus marker protein histone-4 and the endoplasmic reticulum marker calnexin ensured the lack of cross-contamination of these protein extracts.

View larger version (66K): [in this window] [in a new window]   Fig. 6.   Presence of endogenous CGB in PC12 cells. The presence of endogenous CGA or CGB in the protein extracts of total (T), cytosolic (C), and nuclear (N) fractions of PC12 cells was analyzed by immunoblotting using the monoclonal CGA and CGB antibodies and the polyclonal nuclear marker protein histone-4 and ER marker protein calnexin antibodies. 50 µg of the protein per lane was loaded for the CGA and CGB immunoblots, but 10 µg of the proteins was loaded for the calnexin and histone-4 immunoblots.

The immunogold electron microscopy of PC12 cells also indicated the presence of CGB in the nucleus (Fig. 7). Fig. 7A shows the CGB-labeling gold particles in the secretory granules and the nucleus but not in the mitochondria. In the same experiments, omission of the primary antibody or treatment with the preimmune serum in place of the primary antibody eliminated the gold particles almost completely (Fig. 7B), indicating the specific nature of the CGB-labeling immunogold EM results.

View larger version (114K): [in this window] [in a new window]   Fig. 7.   Immunogold electron microscopy showing the localization of CGB in PC12 cells. PC12 cells were immunolabeled for CGB (10 nm gold) with the affinity-purified CGB antibody (A). The CGB-labeling gold particles are localized both in the secretory granules (SG) and in the nucleus (Nu). Some of the CGB-labeling gold particles are shown in the endoplasmic reticulum (rer) but not in the mitochondria (M). Either omission or replacement of the primary antibody with the preimmune serum eliminated virtually all the gold particles in the same PC12 cells (B). Bar = 200 nm.

To determine whether there exists the nuclear localization signal (NLS) sequence in CGB we subjected the bovine CGB sequence to the PSORT II program (33) and found that bovine CGB contains a putative NLS sequence (34), Pro-Glu-Val-Asp-Lys-Arg-Arg (PEVDKRR) starting at residue 235 (Fig. 8A). This type of NLS starts with Pro and followed by, within three residues, a basic segment containing three of four Lys/Arg residues (reviewed in Ref. 35). Thus we tested whether this conserved sequence was responsible for the nuclear localization of CGB by introducing substitution mutations into the putative NLS sequence (Fig. 8A). In this mutant, proline and the critical three basic residues were substituted to serine and hydrophobic residues, resulting in Ser-Glu-Val-Asp-Leu-Gln-Leu (SEVDLQL). The expression of the transfected NLS mutant in COS-7 cells was confirmed by immunoblot analysis (Fig. 8B), and the proteins from the whole cells, cytosol, and the nucleus were examined for the presence of CGB (Fig. 8C). Similar to that of wild type CGB, the NLS mutant also expressed CGB both in the cytoplasm and in the nucleus. However, the relative amount of CGB targeted to the nucleus appeared to be significantly smaller than that shown in the wild type. This result indicated that the putative NLS sequence is not exclusively responsible for the nuclear routing of CGB, thereby suggesting the operation of additional regions of CGB or of factors in targeting CGB to the nucleus.

View larger version (32K): [in this window] [in a new window]   Fig. 8.   Role of the putative NLS in the nuclear localization of CGB. A, a putative NLS sequence, PEVDKRR (residues 235-242), of wild type CGB (CGB-wt), and an NLS mutant (CGB-NLS). B, immunoblot analysis of wild type CGB and the NLS mutant expressed in COS-7 cells. C, the presence of CGB in the protein extracts of total (T), cytosolic (C), and nuclear (N) fractions of COS-7 cells transfected with the wild type CGB (CGB-wt) and the NLS mutant (CGB-NLS), respectively, was analyzed by immunoblot using the affinity-purified CGB antibody. Separation of the cytosolic and nuclear proteins was also ensured by immunoblotting with the polyclonal nuclear marker protein histone-4 and ER marker protein calnexin antibodies. 50 µg of the protein extract per lane was loaded for the CGB immunoblots, but 10 µg of proteins was loaded for the calnexin and histone-4 immunoblots.

To investigate the potential roles of CGB in transcription control, human neuroblastoma cell line SK-N-AS cells (ATCC, CRL-2137) were transfected with CGB (pd2CGB-EGFP), and total RNA from the control (vector pd2EGFP only) and CGB-transfected cells were extracted 36 h after transfection. Analyses of the expressed mRNA levels using MICROMAX Human cDNA Microarray System I (2400 genes) from NEN Life Science, which contains the cDNAs of 2400 human genes, indicated that the CGB transfection affected transcription of more than 40 genes either by induction or by suppression (not shown). Of these we have chosen eight genes, four from the induced and four from the suppressed genes, for quantitative evaluation of the expression levels of each mRNA using the quantitative RT-PCR method and spectrophotometry after the extraction of the PCR products (Fig. 9). The cell densities of SK-N-AS cells that had been transfected with vector only, CGA, or CGB did not differ from each other when the RNAs were extracted for analysis. The quantitative PCR was performed using serial dilutions to assure a linear amplification of the target and control genes (36, 37). As shown in the example of MEF2C, hcKrox, and actin (Fig. 10), the amount of RT-PCR product was proportional to the amount of total RNA present in the reaction samples. Nevertheless, the relative ratio of the RT-PCR products of each target gene in the three groups remained constant, indicating that the amount of RT-PCR product is an accurate reflection of the amount of each mRNA present in the total RNA sample. Further, the amount of each PCR product also increased in accordance with the increase in cycle numbers. Again, the relative ratio of the amount of each mRNA in the three groups also remained unchanged regardless of the PCR cycle numbers, further confirming the validity of this method in quantification of the relative mRNA amounts of target genes. Moreover, the amount of mRNA for an internal control actin remained the same in all three groups during the quantitative RT-PCR reactions (Fig. 10C). Hence, these results clearly indicate that the quantitative RT-PCR method used in the present experiment accurately shows the relative abundance of an RNA species in different RNA samples. By these methods we found that CGB increased the mRNA level of zinc finger protein ~5-fold and those of MADS/MEF2-family transcription factor (MEF2C), cysteine-rich protein2 (hCRP2), and actin-binding double-zinc-finger protein (abLIM) by 2.5-3.5-fold while decreasing the mRNA levels of Kruppel-related zinc finger-containing transcription factor (hcKrox), slow skeletal troponin C (troponin C), and integrin 4 subunit (integrin) by 70-75% and that of T3 receptor-associating cofactor-1 (T3-receptor) by 60-65%. Nevertheless, transfection of the cells with either vector alone (pd2EGFP) or CGA (pd2CGA-EGFP) did not change the mRNA levels of the genes studied, further indicating the specific effect of CGB on transcription of many genes. The mRNA expression level of internal control human -actin always remained the same between the control and experimental groups (Fig. 9). These experiments were carried out four times, and similar results were obtained in all four experiments.

View larger version (35K): [in this window] [in a new window]   Fig. 9.   Quantitative RT-PCR analysis of the mRNA expression levels of eight genes in control, CGA-, and CGB-transfected SK-N-AS cells. Human neuroblastoma SK-N-AS cells were transfected with vector only (V), CGA (A), and CGB (B), and the total RNA was extracted and analyzed by RT-PCR for the mRNA expression of the eight indicated genes (four from the induced and four from the suppressed genes) using the primers and the RT-PCR conditions shown in Table II. A, four induced genes: zinc finger protein (ZFP), MADS/MEF2-family transcription factor (MEF2C), cysteine-rich protein2 (hCRP2), and actin-binding double zinc-finger protein (abLIM). B, four suppressed genes: Kruppel-related zinc finger-containing transcription factor (hcKrox), T3 receptor-associating cofactor-1 (T3-receptor), slow skeletal troponin C (troponin C), and integrin 4 subunit (integrin). Human -actin mRNA expression was used as an internal control for both A and B.

View larger version (26K): [in this window] [in a new window]   Fig. 10.   Linear amplification of mRNA by quantitative RT-PCR. The relative amounts of mRNAs for MEF2C (A), hcKrox (B), and actin (C) in total RNA from human neuroblastoma SK-N-AS cells that had been transfected with vector (control), CGA, and CGB were determined by quantitative RT-PCR and spectrophotometry as described under the "Experimental Procedures" and in the conditions described in Table II. The amount of RT-PCR product was shown to increase as a function of the increasing amount of RNA present in the reaction mixture (left panels) and of the increasing number of the reaction cycles (right panels).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

The present results indicate that the secretory granule marker protein CGB is not only present in the nucleus of neuroendocrine adrenal medullary chromaffin cells and PC12 cells but also routed to the nucleus of CGB-transfected nonneuroendocrine COS-7 and NIH3T3 cells. As shown in Fig. 1 and summarized in Table I, the number of CGB-labeled gold particles localized per µm² of the secretory granules of bovine adrenal chromaffin cells was 160 compared with 437 CGA-labeled gold particles for the same unit area. In line with the fact that bovine adrenal chromaffin cells contain ~9-fold more CGA than CGB in the secretory granules (1-5), the CGA-labeled gold particles always outnumbered the CGB-labeled gold particles in the secretory granules although we have used more diluted CGA antibody in the immunogold EM experiments. However, 33 CGB-labeled gold particles were localized per µm² of the nucleus, equivalent to 20% of the number of CGB-labeled gold particles localized per µm² of the secretory granules of adrenal chromaffin cells. Considering that CGB is the second most abundant protein in the secretory granules of bovine adrenal chromaffin cells, which is estimated to exist in the 0.1-0.2-mM range (1-5), 20% of these concentrations amounts to 20-40 µM. Further, given the large volume occupied by the nucleus in the cell, 20-40 µM of CGB in the nucleoplasm would represent a large number of CGB molecules, thus suggesting an active transport of CGB into the nucleus and implying potentially important roles of CGB in the nucleus.

In line with the granular location of chromogranins, the immunolabeling experiments repeatedly indicated punctate localization of CGA and CGB in the cytoplasm of nonneuroendocrine cells such as COS-7 and NIH3T3 cells (Fig. 3). These punctate structures appear to suggest the formation of secretory granules in these cells as a result of CGA or CGB transfection. The example of secretory granule formation in nonneuroendocrine cells as a result of CGA transfection has indeed been shown recently (27); using nonneuroendocrine fibroblast CV-1 cells, Kim et al. (27) demonstrated the formation of secretory granules in these cells as a result of CGA transfection. Moreover, they also showed that down-regulation of CGA expression in PC12 cells leads to a profound loss of secretory granules in this neuroendocrine cell. These results demonstrated that CGA could function as an on/off switch controlling the secretory granule formation in the cells. Despite the secretory granule-forming effect of CGA, they failed to see the effect of CGB on secretory granule biogenesis in PC12 or CV-1 cells (27). Nevertheless, our present results in Fig. 3 show formation of punctate structures in the cytoplasm of CGA- or CGB-transfected nonneuroendocrine COS-7 cells, suggesting the secretory granule formation in these cells. The punctate staining of the cytoplasm by transfected CGB-GFP has also previously been shown in nonendocrine HeLa cells, implying the localization of CGB-GFP in vesicular structures (38). Taken together, it appears that CGA and CGB possess an intrinsic ability to encircle themselves with the secretory granule components regardless of the neuroendocrine nature of the cells in which they are expressed.

In addition to the fluorescence results that showed the presence of CGB not only in the cytoplasm but also in the nucleus (Fig. 3), the immunoblot results also clearly showed the nuclear routing of transfected CGB in COS-7 and NIH3T3 cells (Figs. 4 and 5). Further, the comparison of the expression levels of GFP, which had been tagged to both CGA and CGB, demonstrated that the level of CGB-GFP expression is approximately one-third or less than that of CGA-GFP expression in these cells (Fig. 5A), clearly indicating that the nuclear localization of CGB is not due to the overexpression of CGB. Furthermore, the nuclear localization of CGB in the cells that had been cotransfected with CGA and CGB (Fig. 5B) also precluded the possibility of the overcrowding of the transport route being the reason for the exclusive routing of CGB to the nucleus.

In view of the fact that PC12 cells already contain large amounts of intrinsic CGB (32, 39), the immunoblot analysis of endogenous chromogranins in PC12 cells has indeed shown the existence of CGB in the cytoplasm and nucleus, whereas CGA was detected only in the cytoplasm (Fig. 6). The relative expression level of endogenous CGB, i.e. the amount of CGB over the total proteins, in the cytoplasm was ~3-fold higher than that in the nucleus of PC12 cells (Fig. 6). Given the widespread presence of CGB in the cytoplasm of PC12 cells, it appeared that CGB is also widely present in the nucleus. The immunogold EM results of PC12 cells (Fig. 7) also showed the widespread presence of CGB in the nucleus, further confirming the results obtained with the adrenal chromaffin cells (Fig. 1).

Moreover, given the clear targeting of CGB to the nucleus, we looked for the presence of the NLS in bovine CGB and found that bovine CGB contains a putative NLS sequence, Pro-Glu-Val-Asp-Lys-Arg-Arg (PEVDKRR) (residues 235-241), which is lacking in CGA. This type of NLS starts with Pro and followed by, within 3 residues, a basic segment containing three of four Lys/Arg residues (reviewed in Ref. 35). A similar sequence was also observed in various nuclear proteins. The NLS mutation results indicated that substitution of the NLS sequence decreased the amount of CGB targeted to the nucleus but failed to completely prevent CGB from moving into the nucleus (Fig. 8C), suggesting nuclear targeting roles by either other regions of CGB in addition to the NLS sequence or other hitherto unknown factors.

In view of our recent finding that CGB tightly interacts with one of the integral secretory granule membrane proteins, the IP₃R (23, 24), and in view of the fact that the IP₃Rs are also localized in the nucleus (40-42), it may be possible for CGB to move into the nucleus through its interaction, involving at least in part the conserved near N-terminal region with the IP₃Rs headed for the nucleus. Nevertheless, the potential cotranslocation property of CGB is not expected to be shared with CGA due to the lack of CGA interaction with the IP₃R at a near physiological pH 7.5 (24).

Other examples of nuclear localization of secretory proteins include proenkephalin (28) and corticotrophin-releasing hormone (29). In the case of proenkephalin, the absence of the signal peptide led proenkephalin to both the nucleus and the secretory granules, whereas proenkephalin with the signal peptide was exclusively routed to the secretory granules (28). Furthermore, the presence of the signal sequence has also been shown to be sufficient for GFP to be routed to the secretory granules (43), underscoring the importance of the signal sequence in routing secretory proteins to the secretory granules. In the case of proenkephalin, alternate transcriptions at different sites are known to occur (44), potentially resulting in many different proenkephalin translation products. Some proenkephalin products hence will be without the signal sequence, enabling them to enter into both the nucleus and the secretory granules. However, unlike the proenkephalins, no alternate transcriptions or different translation products are known to exist for CGB thus far. Further, since the transfected CGB that has been used in the present study contained the signal sequence (Fig. 2), it is not known at present whether the nuclear CGB had originally contained the signal sequence, as is the case with CGB that is routed to the secretory granules.

Moreover, the absence of CGA, another member of the chromogranin family with a high capacity, low affinity Ca²⁺ binding property in the nucleus further underscores the specific nature of nuclear translocation of CGB and appears to foretell exciting new roles of CGB in the nucleus. Indeed analyses of the expressed mRNA levels using a human cDNA microarray system, which contained the cDNAs of 2400 human genes, showed that the CGB transfection affected transcription of more than 40 genes either by induction or by suppression. The quantitative evaluation of the mRNA expression levels of eight of these genes (Fig. 9) indicated that CGB increased the mRNA levels of zinc finger protein, MADS/MEF2-family transcription factor (MEF2C), cysteine-rich protein2 (hCRP2), and actin-binding double-zinc-finger protein (abLIM) by 2.5-5-fold while decreasing the mRNA levels of Kruppel-related zinc finger-containing transcription factor (hcKrox), T3 receptor-associating cofactor-1 (T3-receptor), slow skeletal troponin C (troponin C), and integrin 4 subunit (integrin) by 60-75%. In light of the fact that MEF2C and hcKrox are transcription factors (45-48), these results demonstrate the transcription control role of CGB in the nucleus. Analogous to nuclear proenkephalin, which is known to participate in the growth arrest and differentiation of cells in which they are expressed (28), one of the nuclear roles of CGB is the control of the transcription of many genes, including those of transcription factors by induction and suppression.

	ACKNOWLEDGEMENT

We thank Dr. Joseph P. Albanesi (University of Texas Southwestern Medical Center) for comments on the manuscript and Y. J. Jung and S. U. Kang for help with the experiments.

	FOOTNOTES

^* This work was supported by the Creative Research Initiatives Program of the Ministry of Science and Technology of Korea.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 82-42-869-8279; Fax: 82-42-869-8280; E-mail: shyoo@ kaist.ac.kr.

Published, JBC Papers in Press, February 19, 2002, DOI 10.1074/jbc.M105594200

	ABBREVIATIONS

The abbreviations used are: CGA, chromogranin A; CGB, chromogranin B; IP₃R, inositol 1,4,5-trisphosphate receptor; GFP, green fluorescent protein; ER, endoplasmic reticulum; HA, hemagglutinin; PBS, phosphate-buffered saline; EM, electron microscope; NLS, nuclear localization signal.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES